The Geological Record of

Neoproterozoic Glaciations
 Geological Society Memoirs
The Geological Society of London
Books Editorial Committee

Chief Editor
Bob Pankhurst (UK)

Society Books Editors

John Gregory (UK)
Jim Griffiths (UK)
John Howe (UK)
Howard Johnson (UK)
Rick Law (USA)
Phil Leat (UK)
Nick Robins (UK)
Randell Stephenson (UK)

Society Books Advisors

Eric Buffetaut (France)
Jonathan Craig (Italy)
Tom McCann (Germany)
Mario Parise (Italy)
Satish-Kumar (Japan)
Gonzalo Veiga (Argentina)
Maarten de Wit (South Africa)

IUGS/GSL publishing agreement

This volume is published under an agreement between the International Union of Geological Sciences and the Geological Society of London and arises
from IGCP project number 512.
GSL is the publisher of choice for books related to IUGS activities, and the IUGS receives a royalty for all books published under this agreement.
Books published under this agreement are subject to the Society’s standard rigorous proposal and manuscript review procedures.

It is recommended that reference to all or part of this book should be made in one of the following ways:

Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) 2011. The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36.

Hoffman, P. F., Macdonald, F. A. & Halverson, G. P. 2011. Chemical sediments associated with Neoproterozoic glaciation: iron formation, cap
carbonate, barite and phosphorite. In: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic
Glaciations. Geological Society, London, Memoirs, 36, 67–80.
 GEOLOGICAL SOCIETY MEMOIR NO. 36

The Geological Record of Neoproterozoic Glaciations

EDITED BY

EMMANUELLE ARNAUD
University of Guelph, Canada

GALEN P. HALVERSON
McGill University, Canada

and

GRAHAM SHIELDS-ZHOU
University College London, UK

2011
Published by
The Geological Society
London
 THE GEOLOGICAL SOCIETY

The Geological Society of London (GSL) was founded in 1807. It is the oldest national geological society in the world and the largest in Europe. It was incorporated
under Royal Charter in 1825 and is Registered Charity 210161.
The Society is the UK national learned and professional society for geology with a worldwide Fellowship (FGS) of over 10 000. The Society has the power to
confer Chartered status on suitably qualified Fellows, and about 2000 of the Fellowship carry the title (CGeol). Chartered Geologists may also obtain the equivalent
European title, European Geologist (EurGeol). One fifth of the Society’s fellowship resides outside the UK. To find out more about the Society, log on to
www.geolsoc.org.uk.
The Geological Society Publishing House (Bath, UK) produces the Society’s international journals and books, and acts as European distributor for selected
publications of the American Association of Petroleum Geologists (AAPG), the Indonesian Petroleum Association (IPA), the Geological Society of America (GSA), the
Society for Sedimentary Geology (SEPM) and the Geologists’ Association (GA). Joint marketing agreements ensure that GSL Fellows may purchase these societies’
publications at a discount. The Society’s online bookshop (accessible from www.geolsoc.org.uk) offers secure book purchasing with your credit or debit card.
To find out about joining the Society and benefiting from substantial discounts on publications of GSL and other societies worldwide, consult www.geolsoc.org.uk,
or contact the Fellowship Department at: The Geological Society, Burlington House, Piccadilly, London W1J 0BG: Tel. þ 44 (0)20 7434 9944; Fax þ 44 (0)20 7439
8975; E-mail: enquiries@geolsoc.org.uk.
For information about the Society’s meetings, consult Events on www.geolsoc.org.uk. To find out more about the Society’s Corporate Affiliates Scheme, write to
enquiries@geolsoc.org.uk.

Published by The Geological Society from:

The Geological Society Publishing House, Unit 7, Brassmill Enterprise Centre, Brassmill Lane, Bath BA1 3JN, UK
(Orders: Tel. þ44 (0)1225 445046, Fax þ44 (0)1225 442836)
Online bookshop: www.geolsoc.org.uk/bookshop
The publishers make no representation, express or implied, with regard to the accuracy of the information contained in this book and cannot accept any legal
responsibility for any errors or omissions that may be made.
# The Geological Society of London 2011. All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission. No
paragraph of this publication may be reproduced, copied or transmitted save with the provisions of The Copyright Licensing Agency Ltd, Saffron House, 6 –10 Kirby
Street, London EC1N 8TS, UK. Users registered with the Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923, USA: the item-fee code for this
publication is 0435-4052/11/$15.00.

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library.
ISBN 978-1-86239-334-9

Distributors
For details of international agents and distributors see:
www.geolsoc.org.uk/agentsdistributors
Typeset by Techset Composition Ltd, Salisbury, UK
Printed by CPI Antony Rowe, Chippenham, UK
Mexico Hut, 1978. Clockwise from the back left: Nick Cox (Motorboat Skipper and Engineer), Tim Druitt (Geologist),
Robin (“Bruce”) Davies (Motorboat Skipper and Engineer), Brian Harland (Expedition Leader), James Carter
(undergraduate field assistant), Mike Hambrey (Geologist) and Paul Waddams (PhD Student), Ny-Ålesund, Svalbard

The editors dedicate this Memoir to Mike Hambrey and Brian Harland for their
pioneering work in Neoproterozoic glacial geology and for their comprehensive 1981
volume Earth’s Pre-Pleistocene Glacial Record, which inspired the present work.
 Contents
List of Reviewers ix

Introductory chapters

ARNAUD, E., HALVERSON, G. P. & SHIELDS-ZHOU, G. The geological record of Neoproterozoic ice ages 1

HOFFMAN, P. F. A history of Neoproterozoic glacial geology, 1871– 1997 17

ARNAUD, E. & ETIENNE, J. L. Recognition of glacial influence in Neoproterozoic sedimentary successions 39

HALVERSON, G. P. & SHIELDS-ZHOU, G. Chemostratigraphy and the Neoproterozoic glaciations 51

HOFFMAN, P. F., MACDONALD, F. A. & HALVERSON, G. P. Chemical sediments associated with Neoproterozoic glaciation: iron formation, cap 67
carbonate, barite and phosphorite

BAHLBURG, H. & DOBRZINSKI, N. A review of the Chemical Index of Alteration (CIA) and its application to the study of Neoproterozoic 81
glacial deposits and climate transitions

EVANS, D. A. D. & RAUB, T. D. Neoproterozoic glacial palaeolatitudes: a global update 93

GREY, K., HILL, A. C. & CALVER, C. Biostratigraphy and stratigraphic subdivision of Cryogenian successions of Australia in a global context 113

CONDON, D. J. & BOWRING, S. A. A user’s guide to Neoproterozoic geochronology 135

GODDÉRIS, Y., LE HIR, G. & DONNADIEU, Y. Modelling the Snowball Earth 151

Africa

SHIELDS-ZHOU, G. A., DEYNOUX, M. & OCH, L. The record of Neoproterozoic glaciation in the Taoudéni Basin, NW Africa 163

MASTER, S. & WENDORFF, M. Neoproterozoic glaciogenic diamictites of the Katanga Supergroup, Central Africa 173

TAIT, J., DELPOMDOR, F., PRÉAT, A., TACK, L., STRAATHOF, G. & NKULA, V. K. Neoproterozoic sequences of the West Congo and 185
Lindi/Ubangi Supergroups in the Congo Craton, Central Africa

HOFFMAN, P. F. Glaciogenic and associated strata of the Otavi carbonate platform and foreslope, northern Namibia: evidence for large 195
base-level and glacioeustatic changes

PRAVE, A. R., HOFFMANN, K.-H., HEGENBERGER, W. & FALLICK, A. E. The Witvlei Group of East-Central Namibia 211

FRIMMEL, H. E. The Chameis Gate Member, Chameis Group, Marmora Terrane, Namibia 217

FRIMMEL, H. E. The Kaigas and Numees formations, Port Nolloth Group, in South Africa and Namibia 223

FRIMMEL, H. E. The Karoetjes Kop and Bloupoort formations, Gifberg Group, South Africa 233

Eurasia –Nubian Shield

ALLEN, P. A., RIEU, R., ETIENNE, J. L., MATTER, A. & COZZI, A. The Ayn Formation of the Mirbat Group, Dhofar, Oman 239

ALLEN, P. A., LEATHER, J., BRASIER, M. D., RIEU, R., MCCARRON, M., LE GUERROUÉ, E., ETIENNE, J. L. & COZZI, A. The Abu Mahara 251
Group (Ghubrah and Fiq formations), Jabal Akhdar, Oman

MILLER, N. R., AVIGAD, D., STERN, R. J. & BEYTH, M. The Tambien Group, Northern Ethiopia (Tigre) 263

STERN, R. J., JOHNSON, P. R., ALI, K. A. & MUKHERJEE, S. K. Evidence for Early and Mid-Cryogenian glaciation in the Northern 277
Arabian –Nubian Shield (Egypt, Sudan, and western Arabia)

CHUMAKOV, N. M. Glacial deposits of the Bokson Group, East Sayan Mountains, Buryatian Republic, Russian Federation 285

CHUMAKOV, N. M. The Neoproterozoic glacial formations of the North and Middle Urals 289
 CONTENTS vii

CHUMAKOV, N. M. Glacial deposits of the Nichatka Formation, Chara River basin and review of Upper Precambrian diamictites 297
of Central Siberia

CHUMAKOV, N. M. Glacial deposits of the Baykonur Formation, Kazakhstan and Kyrgyzstan 303

CHUMAKOV, N. M., POKROVSKY, B. G. & MELEZHIK, V. A. The glaciogenic Bol’shoy Patom Formation, Lena River, central Siberia 309

SOVETOV, J. K. Late Cryogenian (Vendian) glaciogenic deposits in the Marnya Formation, Oselok Group, in the foothills of the 317
East Sayan Range, southwestern Siberian Craton

MACDONALD, F. A. The Tsagaan Oloom Formation, southwestern Mongolia 331

MACDONALD, F. A. & JONES, D. S. The Khubsugul Group, Northern Mongolia 339

ETIENNE, J. L., ALLEN, P. A., LE GUERROUÉ, E., HEAMAN, L., GHOSH, S. K. & ISLAM, R. The Blaini Formation of the Lesser Himalaya, NW India 347

ZHANG, Q.-R., CHU, X.-L. & FENG, L.-J. Neoproterozoic glacial records in the Yangtze Region, China 357

ZHU, M. & WANG, H. Neoproterozoic glaciogenic diamictites of the Tarim Block, NW China 367

North America

MACDONALD, F. A. The Hula Hula Diamictite and Katakturuk Dolomite, Arctic Alaska 379

MACDONALD, F. A. & COHEN, P. A. The Tatonduk inlier, Alaska –Yukon border 389

HOFFMAN, P. F. & HALVERSON, G. P. Neoproterozoic glacial record in the Mackenzie Mountains, northern Canadian Cordillera 397

SMITH, M. D., ARNAUD, E., ARNOTT, R. W. C. & ROSS, G. M. The record of Neoproterozoic glaciations in the Windermere 413
Supergroup, southern Canadian Cordillera

LINK, P. K. & CHRISTIE-BLICK, N. Neoproterozoic strata of southeastern Idaho and Utah: record of Cryogenian rifting and glaciation 425

LUND, K., ALEINIKOFF, J. N. & EVANS, K. V. The Edwardsburg Formation and related rocks, Windermere Supergroup, central Idaho, USA 437

MROFKA, D. & KENNEDY, M. The Kingston Peak Formation in the eastern Death Valley region 449

PETTERSON, R., PRAVE, A. R. & WERNICKE, B. P. Glaciogenic and related strata of the Neoproterozoic Kingston Peak Formation in the 459
Panamint Range, Death Valley region, California

CARTO, S. L. & EYLES, N. The deep-marine glaciogenic Gaskiers Formation, Newfoundland, Canada 467

CARTO, S. L. & EYLES, N. The Squantum Member of the Boston Basin, Massachusetts, USA 475

South America

CHEW, D. & KIRKLAND, C. The Chiquerı́o Formation, southern Peru 481

ALVARENGA, C. J. S., BOGGIANI, P. C., BABINSKI, M., DARDENNE, M. A., FIGUEIREDO, M. F., DANTAS, E. L., UHLEIN, A., SANTOS, R. V., 487
SIAL, A. N. & TROMPETTE, R. Glacially influenced sedimentation of the Puga Formation, Cuiabá Group and Jacadigo Group, and
associated carbonates of the Araras and Corumbá groups, Paraguay Belt, Brazil

FIGUEIREDO, M. F., BABINSKI, M. & ALVARENGA, C. J. S. The Serra Azul Formation, Paraguay Belt, Brazil 499

GUIMARÃES, J. T., MISI, A., PEDREIRA, A. J. & DOMINGUEZ, J. M. L. The Bebedouro Formation, Una Group, Bahia (Brazil) 503

MISI, A., KAUFMAN, A. J., AZMY, K., DARDENNE, M. A., SIAL, A. N. & DE OLIVEIRA, T. F. Neoproterozoic successions of the São Francisco 509
Craton, Brazil: the Bambuı́, Una, Vazante and Vaza Barris/Miaba groups and their glaciogenic deposits

PEDROSA-SOARES, A. C., BABINSKI, M., NOCE, C., MARTINS, M., QUEIROGA, G. & VILELA, F. The Neoproterozoic Macaúbas Group, 523
Araçuaı́ orogen, SE Brazil

ROCHA-CAMPOS, A. C., DE BRITO NEVES, B. B., BABINSKI, M., DOS SANTOS, P. R., DE OLIVEIRA, S. M. B. & ROMANO, A. Moema laminites: 535
a newly recognized Neoproterozoic (?) glaciogenic unit, São Francisco Basin, Brazil
viii CONTENTS

UHLEIN, A., ALVARENGA, C. J. S., DARDENNE, M. A. & TROMPETTE, R. R. The glaciogenic Jequitaı́ Formation, southeastern Brazil 541

PAZOS, P. J., RAPALINI, A. E., BETTUCCI, L. S. & TÓFALO, O. R. The Playa Hermosa Formation, Playa Verde Basin, Uruguay 547

PECOITS, E., GINGRAS, M. K. & KONHAUSER, K. O. Las Ventanas and San Carlos formations, Maldonado Group, Uruguay 555

PAZOS, P. J. & RAPALINI, A. The controversial stratigraphy of the glacial deposits in the Tandilia System, Argentina 565

Europe

HALVERSON, G. P. Glacial sediments and associated strata of the Polarisbreen Group, northeastern Svalbard 571

STOUGE, S., CHRISTIANSEN, J. L., HARPER, D. A. T., HOUMARK-NIELSEN, M., KRISTIANSEN, K., MACNIOCAILL, C. & BUCHARDT-WESTERGÅRD, B. 581
Neoproterozoic (Cryogenian–Ediacaran) deposits in East and North-East Greenland

RICE, A. H. N., EDWARDS, M. B., HANSEN, T. A., ARNAUD, E. & HALVERSON, G. P. Glaciogenic rocks of the Neoproterozoic Smalfjord and 593
Mortensnes formations, Vestertana Group, E. Finnmark, Norway

STODT, F., RICE, A. H. N., BJÖRKLUND, L., BAX, G., HALVERSON, G. P. & PHARAOH, T. C. Evidence of late Neoproterozoic glaciation in the 603
Caledonides of NW Scandinavia

NYSTUEN, J. P. & LAMMINEN, J. T. Neoproterozoic glaciation of South Norway: from continental interior to rift and pericratonic basins in 613
western Baltica

KUMPULAINEN, R. A. & GREILING, R. O. Evidence for late Neoproterozoic glaciation in the central Scandinavian Caledonides 623

KUMPULAINEN, R. A. The Neoproterozoic glaciogenic Lillfjället Formation, southern Swedish Caledonides 629

ARNAUD, E. & FAIRCHILD, I. J. The Port Askaig Formation, Dalradian Supergroup, Scotland 635

PRAVE, A. R. & FALLICK, A. E. The Neoproterozoic glaciogenic deposits of Scotland and Ireland 643

Australia

CALVER, C. R. Neoproterozoic glacial deposits of Tasmania 649

CORKERON, M. Neoproterozoic glacial deposits of the Kimberly Region and northwestern Northern Territory, Australia 659

GOSTIN, V. A., MCKIRDY, D. M., WEBSTER, L. J. & WILLIAMS, G. E. Mid-Ediacaran ice-rafting in the Adelaide Geosyncline and 673
Officer Basin, South Australia

HILL, A. C., HAINES, P. W. & GREY, K. Neoproterozoic glacial deposits of central Australia 677

JENKINS, R. J. F. Billy Springs glaciation, South Australia 693

PREISS, W. V., GOSTIN, V. A., MCKIRDY, D. M., ASHLEY, P. M., WILLIAMS, G. E. & SCHMIDT, P. W. The glacial succession of 701
Sturtian age in South Australia: the Yudnamutana Subgroup

WILLIAMS, G. E., GOSTIN, V. A., MCKIRDY, D. M., PREISS, W. V. & SCHMIDT, P. W. The Elatina glaciation (late Cryogenian), 713
South Australia

Index 723
 List of Reviewers
Philip Allen Nick Eyles
Imperial College London, UK University of Toronto at Scarborough, Canada
Carlos Alvarenga Ian Fairchild
University of Brasilia, Brazil University of Birmingham, UK
Jose-Javier Alvaro Mikhail Fedonkin
Centro de Astrobiologia, Madrid, Spain Paleontological Institute, Moscow, Russia
John Arthurs Eric Font
Holywood, County Down, Ireland University of Lisbon, Portugal
Dov Avigad Hartwig Frimmel
The Hebrew University of Jerusalem, Israel University of Würzburg, Germany
Marly Babinsky Claudio Gaucher
University of Sao Paulo, Brazil Universidad de la República, Uruguay
Heinrich Bahlburg Vic Gostin
Westfälische Wilhelms-Universität, Muenster, Germany University of Adelaide, Australia
Richard Bailey Dimitriy Grazhdankin
Northeastern University, USA Institute of Petroleum Geology and Geophysics, Novosibirsk,
Dhiraj Banerjee Russia
University of Delhi, India Mike Hambrey
Angelo Miguel Stipp Basei Aberystwyth University, Wales, UK
University of Sao Paulo, Brazil Richard Hanson
Julie Bartley Texas Christian University, USA
State University of West Georgia Christoph Heubeck
Nicholas Beukes Freie Universität Berlin, Germany
University of Johannesburgh, South Africa Andrew Hill
Knut Bjorlykke Centro de Astrobiologia, Madrid, Spain
University of Oslo, Norway Rick Hiscott
Robert Blodgett Memorial University, Canada
United States Geological Survey, USA Paul Hoffman
Nick Butterfield Harvard University, USA/University of Victoria, Canada
Cambridge, UK John Howe
Clive Calver Scottish Marine Institute, UK
Mineral Resources, Tasmania Ganqing Jiang
Nikolay Chumakov University of Las Vegas, USA
Geological Institute, Russian Academy of Sciences Peter Johnson
Allan Collins Washington, DC, USA
University of Adelaide, Australia Sandra Kamo
Maurice Colpron University of Toronto, Canada
Yukon Geological Survey, Canada A. F. Kamona
Frank Corsetti University of Namibia, Namibia
University of Southern California, USA Alan Jay Kaufman
Lucieth Cruz Viera University of Maryland, USA
University of Sao Paulo, Brazil Brian Kendall
Bley Benjamin de Brito Neves Arizona State University, USA
University of Sao Paulo, Brazil Andy Knoll
Max Deynoux Harvard University, USA
CNRS, Strasbourg, France Risto Kumpulainen
Nicole Dobrzinsky Stockholm University, Sweden
Panterra Geoconsultants B. V., The Netherlands Alexandre Kuzmichev
Eugene Domack Geological Institute, Russian Academy of Sciences, Russia
Hamilton College, USA
Anton Kuznetsov
James Etienne Institute of Precambrian Geology and Geochronology,
Neftex, UK St. Petersburg, Russia
David J. A. Evans Erwan Le Guerroué
Durham University, UK Université de Rennes, France
x LIST OF REVIEWERS

Zheng-Xiang Li Victor Ramos

Curtin University of Technology, Australia University of Buenos Aires, Argentina
John Lindsay Pete Reid
Universities Space Research Association, USA Petratherm, Australia
Paul Link Claudio Riccomini
Idaho State University, USA University of Sao Paulo, Brazil
Karen Lund
United States Geological Survey, USA Hugh Rice
University of Vienna, Austria
Francis Macdonald
Harvard University, USA Ruben Rieu
Repsol YPF, Exploration and Production, Spain
Adam Maloof
Princeton University, USA Anthony Rocha-Campos
University of Sao Paulo, Brazil
Marcelo A. Martins Neto
University of Ouro Preto, Brazil Julius Sovetov
Andrey Maslov Institute of Oil and Gas Geology and Geophysics, Novosibirsk,
Institute of Geology and Geochemistry, Ekaterinburg, Russia Russia

Bill McClelland Anthony Spencer

University of Idaho, USA Statoil, Norway
Roy Miller Doug Sprinkel
Windhoek, Namibia Utah Geological Survey, USA
Aroldo Misi Robert Strachan
Federal University of Bahia, Brazil University of Portsmouth, UK
Malgorzata Moczydlowska-Vidal Nick Swanson-Hysell
Uppsala University, Sweden Princeton University, USA
Paul Myrow
Colorado College, USA Eric Thover
University of Western Australia
Afonso Noguiera
Universidade Federal do Pará, Brazil Elizabeth Turner
Laurentian University, Canada
Robert Pankhurst
British Geological Survey, UK Rob Van der Voo
University of Michigan, USA
Sandra Passchier
Montclair University, USA Pat Vickers-Rich
Pablo Pazos Monash University, Australia
University of Buenos Aires, Argentina Hilmar von Eynatten
Vic Pease University of Göttingen, Germany
University of Stockholm, Sweden
Stephen Warren
Sergei Pisarevsky University of Washington, USA
The University of Western Australia, Australia
George Williams
Boris Pokrovsky University of Adelaide, Australia
Geological Institute, Moscow, Russia
Gary Yeo
Michael Pope Denison Mines Corporation, Canada
Washington State University, USA
Susannah Porter Grant Young
University of California-Santa Barbara, USA University of Western Ontario, Canada

Anthony Prave Chuanming Zhou

University of St. Andrews, UK Nanjing Institute of Geology and Paleontology, Chinese Academy
of Science, China
Wolfgang Preiss
Department of Primary Industries and Resources of South Australia, Udo Zimmermann
Australia University of Stavanger, Norway
 Chapter 1

The geological record of Neoproterozoic ice ages

EMMANUELLE ARNAUD1*, GALEN P. HALVERSON2,3 & GRAHAM SHIELDS-ZHOU4

1
School of Environmental Sciences, University of Guelph, Guelph, Ontario N1G 2W1, Canada
2
School of Earth and Environmental Sciences, The University of Adelaide, North Terrace, Adelaide, SA 5005, Australia
3
Present address: Department of Earth and Planetary Sciences, McGill University, 3450 University Street, Montreal, Quebec,
H3A 2A7, Canada
4
Department of Earth Sciences, University College London, Gower Street, London WC1E 6BT, UK
*Corresponding author (e-mail: earnaud@uoguelph.ca)

The IUGS- and UNESCO-funded International Geoscience Pro-
gramme Project #512 (Neoproterozoic Ice Ages) was conceived
to contribute towards a global synthesis of current geological
data on the number, duration, extent, causes and consequences
of glacial episodes during the Neoproterozoic Era. IGCP 512
attracted more than 200 scientists from over 30 countries, many
of whom provided their regional and specialist expertise on Neo-
proterozoic successions around the world to the realization of
this volume. IGCP 512 focused on integrating various aspects of
Neoproterozoic geology: geochronology, geochemistry, sedimen-
tary geology, biostratigraphy, palaeomagnetism and economic
geology. At its inaugural meeting on 27 August 2005 during the
International Association of Sedimentology conference on
glacial processes and products in Aberystwyth, Wales, IGCP 512
members decided to produce a volume that summarized existing
data sets in a form similar to Earth’s Pre-Pleistocene Glacial
Record by Hambrey & Harland (1981).
An enormous amount of work has been carried out in the 12
years since the publication of Hoffman et al.’s (1998) paper on
the Snowball Earth hypothesis for Neoproterozoic glaciation
(Fairchild & Kennedy 2007). The Snowball Earth hypothesis
and, more generally, Neoproterozoic climate, have been the
topic of numerous special volumes, special sessions, a dedicated
conference in Ascona (Switzerland) in 2006 (Shields 2006), and
numerous documentaries. Motivated by this intense worldwide
interest in the Neoproterozoic glaciations and an exploding
body of research into the topic, this volume synthesizes the
state-of-the-art in this now highly multidisciplinary research
field. It is intended to facilitate the integration of data sets,
inspire new research projects, and inform ongoing work into the
definition and subdivision of the Neoproterozoic timescale,
including selection of the Global Stratotype Section and Point
(GSSP) for the base of the Cryogenian Period. Despite such
lofty aims, any book such as this cannot claim to be complete,
and there are indeed many gaps in our knowledge and also in
this book’s coverage, some of which are outlined in this Introduc-
tion and throughout the volume.
Book organization, format and terminology
This book contains ten introductory overview chapters followed
by 60 site- or succession-specific chapters. The multidisciplinary
overview chapters provide reviews of the study and interpretation
of Neoproterozoic glaciations. The first chapter by Hoffman
reviews the history of research (1871 – 1997) into late Precambrian
glaciations from the first recorded discovery of Neoproterozoic
glaciogenic rocks at Port Askaig on Islay, SW Scotland in 1871
to just before the renewal of interest in their significance after
1998. Arnaud & Etienne provide a ‘user’s guide’ to the identifi-
cation of glacial influence in the rock record, with emphasis on the
processes and sedimentary products found in various glaciated
basins as well as some common issues encountered with determin-
ing the palaeoclimatic significance of commonly used indicators of
glacial palaeoenvironmental conditions. Geochemistry, in particu-
lar, isotope chemostratigraphy, has been key to the recent revival
of interest in Neoproterozoic climate change and underpins
models of both glaciation and global correlations. Halverson &
Shields-Zhou review Neoproterozoic chemostratigraphic records
with a focus on how they have been applied to palaeoenvironmen-
tal studies and to constrain both the number and relative timing of
glacial events.
Hoffman et al. then review the occurrence of chemical sedi-
ments and their depositional environments associated with the
Neoproterozoic glaciations. In particular, the role of the enigmatic,
post-glacial ‘cap carbonates’ and the reoccurrence of iron for-
mations after a billion-year hiatus are explored here. Climate
affects how rocks are weathered, and this in turn influences the
global carbon cycle, a key factor in modulating global climate
change. Bahlburg & Dobrzinski offer in this regard a more
specific review of the application of the Chemical Index of Altera-
tion (CIA) to Neoproterozoic glacial deposits. If the Snowball
Earth hypothesis remains contentious, the following chapter
should at least dispel any doubt that Neoproterozoic glaciation
reached the tropics. In their review of palaeolatitudes of Neopro-
terozoic glacial deposits, Evans & Raub contribute a comprehen-
sive compilation of locations and available palaeomagnetic
data for Neoproterozoic glacial deposits worldwide. Condon &
Bowring also provide a much needed user’s guide to Neoprotero-
zoic geochronology, with emphasis on the most commonly used
techniques, their strengths and weaknesses and the source of uncer-
tainties that should be considered when using these data to con-
strain the timing of climatic changes. In addition, Condon &
Bowring provide a summary of the key geochronological con-
straints on the timing and duration of glaciations and other signifi-
cant Neoproterozoic events related to isotopic excursions and
biological evolution. Grey et al. tackle the Neoproterozoic fossil
record, and in particular the question of how climate change
affected life on Earth. In their detailed review, they formulate a
Cryogenian biostratigraphy based on Australian data, which can
potentially be used as a template for such studies, generally still
in their infancy, in other regions and for other Proterozoic time
periods. In the final overview chapter, Goddéris et al. provide a
critical review of the widely differing climate models that have
been used to investigate the onset and melting of a Snowball
Earth and highlight that whereas initiation of a Snowball Earth is

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 1 –16. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.1
2 E. ARNAUD ET AL.
not difficult to accomplish in many models, how the Earth escapes
from the icy grip of a snowball requires more study from a model-
ling perspective.
The remaining chapters, organized by current geography,
address specific sites from around the world where Neoproterozoic
deposits have been studied. The purpose of these chapters is to
provide a summary of available data and key references; they
are not comprehensive reviews. Each chapter was intended to
follow a consistent format in order to cover concisely the following
topics and to facilitate cross chapter comparisons: structural frame-
work, regional stratigraphy, sedimentary characteristics of glacial
strata and associated deposits, boundary relations with non-glacial
strata, chemostratigraphy, palaeomagnetism and palaeogeo-
graphy, geochronology and a discussion outlining interpretations
that can be inferred from these data. The structural framework
section was designed to include a description of the overall struc-
tural and tectonic setting (such as cratons, types of sedimentary
basin, regional-scale folds and faults) as well as the degree to
which the sections have been modified by post-depositional tecton-
ism and metamorphism. Authors were also asked to discuss the
history of basin development in the region. The stratigraphy
section was meant to provide an overview of the relevant strati-
graphic units with comments about any lateral variations. The gla-
ciogenic and associated strata section was designed to include
descriptions of typical sedimentary characteristics of the glacio-
genic and associated strata such as ironstones and carbonates,
with the nature of contact with the overlying and underlying
non-glacial units described in the following boundary relations
section. The chemostratigraphy section was meant to include geo-
chemical data within a stratigraphic context, including CIA values,
d13C, d18O, d34S, 87Sr/86Sr, and other available isotopic data.
Some authors also included analyses of Rare Earth Elements,
whole rock geochemistry, major and trace elements, Nd data,
and Ce, Eu and Ir anomalies. The other characteristics section
was designed to cover economic deposits, biomarkers or any
other notable feature not covered in the other sections. The palaeo-
latitude and palaeogeography section was primarily designed to
discuss the palaeomagnetic data available for that succession in
order to consider palaeogeographic location of the deposits at
the time of deposition. Some authors also included provenance
data in the context of palaeogeography based on lithology or geo-
chemistry. The geochronological constraints section was designed
to report any available radiometric or biostratigraphic data that
constrains the minimum or maximum age of the succession.
These dates could come from the glacial deposits themselves or
from associated strata. Authors were also asked to discuss regional
stratigraphic correlations based on radiometric data, chemostrati-
graphy, lithostratigraphy and/or biostratigraphy. Global strati-
graphic correlations were discouraged as these are discussed in
the introductory chapters by Condon & Bowring (geochronology)
and Halverson & Shields-Zhou (chemostratigraphy). Lastly, the
discussion section was designed to include interpretations of
available datasets with respect to palaeoenvironmental conditions,
timing of climate change and palaeogeography, showing how
Fig. 1.1. Map showing distribution of
Neoproterozoic glaciogenic successions
within Africa, as covered in this volume. (A)
Taoudeni Basin, NW Africa; (B) Katanga
Supergroup, central Africa; (C) West Congo
and Lindi/Ubangi Supergroups, central
Africa; (D) Otavi Group, northern Namibia;
(E) Witvlei Group, East-Central Namibia;
(F) The Chameis Gate Member, Namibia;
(G) The Kaigas and Numees formations,
South Africa and Namibia, (H) Karoetjes
Kop and Bloupoort formations, South
Africa.
Table 1.1. Summary of Neoproterozoic data sets from Africa

Lead author Glaciogenic strata Structural Chemostratigraphy Geobiology Economic Palaeomagnetism Geochronology
framework
d13C d18O d34S 87
Sr/86Sr Other U–Pb U– Pb Other
detrital direct

Shields-Zhou Basal diamictite of the Foreland basin Y Y Y Y Scarce Barite, BIF, Y K– Ar,
triad sequence, petroleum Rb– Sr
Taoudeni Basin, source rocks
diamictite of the triad
sequence, Volta Basin
Master Grand Conglomerat Fm., Synrift, foreland Y Ir Y Fe, Cu– Co, Zn– Y Y Y
Petit Conglomerat Fm. basin Pb–Cu –Ge–
Ga, Cu –Au,

INTRODUCTION
Ba–Fe, Pb –Zn
Tait Lower and Upper Passive margin Y Y Y Y Ar/Ar
Diamictite Formations,
West Congo
Supergroup;
diamictites of the
Lindi/Ubangi
Supergroup
Hoffman Chuos Fm., Ghaub Fm. Passive margin Y Y Y Y [Fe, Mn, Ba], Y
d11B, d44Ca
Prave Blaubeker Fm. Rift to passive Y Y Y Y reset Y
Frimmel Kaigas Fm., Numees Fm. Failed rift; passive; Y Y Y Y Pb, Zn Y Y Pb/Pb,
back-arc? Ar/Ar
Frimmel Chameis Gate Mb., Back-arc? Y Y Y Limited trace Pb/Pb,
Chameis Group elements Ar/Ar
Frimmel Karoetjes Kop Fm.; Rift; pasive margin? Y Y Y Limited trace Calcitic marble;
Bloupoort Fm. elements Cu, Fe, Mn;
Fe oxides

Note: ‘U –Pb direct’ means dates from volcanic intrusions or tuffs anywhere in the succession, not necessarily from the glaciogenic strata itself. ‘U– Pb detrital’ means dates from detrital zircons from within the succession. Data in
presumed correlative successions were not included.

3
4 E. ARNAUD ET AL.

the tectonic, stratigraphic, sedimentological, isotopic, palaeomag-
netic and geochronological data outlined in previous sections
support such interpretations. Whereas authors clearly favoured
certain interpretations, they were encouraged to discuss competing
interpretations and continued controversies. In an attempt to avoid
redundancy and maintain consistency, photographs were omitted
from site-specific chapters, and are made available in a companion
online photo atlas (http://neoproterozoic-glaciations.weebly.
com). Although this format was generally followed, some flexi-
bility was required to accommodate the complexity of certain
areas and the widely differing perspectives that reflect the diversity
of authors contributing to this volume.
Authors were asked to avoid interpretive terms, such as tillite,
varvite, and cap carbonate, in preference to descriptive terms,
except in the discussion section where interpretations are pre-
sented. More specifically, the non-genetic term diamictite
was meant to be used to describe poorly sorted materials that
contain a mixture of gravel-, sand- and mud-sized particles. For
example, the term tillite was meant to be used in the discussion
section in referring to poorly-sorted materials that were demonstra-
bly deposited by ice without subsequent disaggregation and flow.
The use of the term varvites or varves for rhythmically laminated
mudstone that record seasonal fluctuations in ice cover was
discouraged considering the lack of chronological control on
Neoproterozoic successions and the inability to demonstrate seaso-
nal cyclicity. In the discussion section, authors were asked, where
possible, to distinguish between the cap carbonate sequence that
overlies the glacial deposits and the cap dolostone that occupies
the basal transgressive tract of the post-glacial sequence.
Although the focus of the book is on Cryogenian glaciation,
these glacial deposits and associated strata are best understood
within the context of their overall stratigraphic successions.
Therefore, several site chapters also review associated later Neo-
proterozoic, Ediacaran-age deposits to highlight the carbonate
strata associated with the late Cryogenian glacial deposits and
the evolution of early animals that followed the late Cryogenian
and mid-Ediacaran glaciations.
Every effort was made to include all the sites where Neoproter-
ozoic glaciogenic successions have been studied. In some cases,
multiple distinct glaciogenic successions within a single region
were grouped into a single chapter, whereas in others, they were
treated in separate chapters. For the Adelaide rift-basin in South
Australia, where extensive work has been done on each of the
older (Sturtian and equivalent) and younger (Elatina and equival-
ent) glacial successions, the reviews are split into two chapters –
one for each glaciation.
Unfortunately, not every site known to preserve evidence of
Neoproterozoic glaciation is included in this volume. These

50

F I
G 40
J E
L
H
D K 30

O
20
N
C B M
A 10

N
0
0 1000 2000
km
40 50 60 70 80 90 100 110 120 130
Fig. 1.2. Map showing distribution of Neoproterozoic glaciogenic successions within Eurasia and the Nubian Shield, as covered in this volume. (A) Ayn Fm., Oman;
(B) Abu Mahara Group, Oman; (C) Tambien Group, northern Ethiopia; (D) northern Arabian– Nubian Shield (Egypt, Sudan and western Arabia); (E) Bokson Group,
East Sayan Mountains, Russian Federation; (F) North and Middle Urals, Russian Federation; (G) Nichatka Fm., central Siberia; (H) Baykonur Fm., Kazakhstan and
Kyrgystan; (I) Bol’shoy Patom Fm., central Siberia; (J) Marnya Fm., foothills of East Sayan mountains; (K) Tsagaan Oloom Fm., SW Mongolia; (L) Khubsugul Group,
northern Mongolia; (M) Blaini Fm., NW India; (N) Yangtze region, China; (O) Tarim Block, NW China.
Table 1.2. Summary of Neoproterozoic data sets in Eurasia and the Nubian Shield

Lead author Glaciogenic strata Structural Chemostratigraphy Geobiology Economic Palaeomagnetism Geochronology
framework
d13C d18O d34S 87
Sr/86Sr Other U– Pb U– Pb Other
detrital direct

Allen Ayn Fm., Shareef Fm. Rift Y CIA, MIA Y, unreliable Y Rb –Sr, K– Ar
Allen Ghubrah Fm., Fiq Fm. Rift Y CIA, MIA Y, unreliable Y Y
Stern Meriti Gp.?, Mahd Arc–back-arc, REE, Ce, Eu BIF, Au, Cu, Y
Gp.?, Atud & oceanic basin Zn, Ar,
Nuwaybah
diamictites
Miller Didikama Fm., Matheos Arc, intra-oceanic Y Y Y CIA, PIA, Y Y Sm–Nd, Pb/
Fm. (Tambien Gp.) platform d13C_TOC, Pb, Rb– Sr
Sr
Chumakov Zabit Fm., Kushatay Foreland basin Y Y Y Bauxite, Rb –Sr
Fm. (Bokson Gp.) Phosphorite
Chumakov Churochnaya Fm., Passive margin Y Y Y Y K –Ar, Sm–
Tany/ Koyva/Wil’va Nd, Rb –Sr,
formations, Lower Pb/Pb

INTRODUCTION
Starye Pechi
Subformation
Chumakov Nichatka Fm. Passive margin Y Y Y
Chumakov Baykonur Unknown Y Y Vanadium, Y
phosphorite
Chumakov Bol’shoy Patom Fm. Passive margin Y Y Y Y, reset K –Ar,Rb– Sr
Sovetov Marnya Fm. (Oselok Foreland basin Y Y Y Y Y, overprinted Ar/Ar
Gp.)
Macdonald Tsagaan Oloom Fm. Passive margin? Y Y Ce Y Y Y
Macdonald Ongoluk Fm., Khesen Rift? Y Y Phosphorite
Fm. (Khubsugul Gp.)
Etienne Blaini Fm. Passive margin Y Y Y Unreliable Y Y
Zhang Chang’an/Fulu Fm Rift Y Y TOC Y (Biomarker) Manganese, Y Y Nd (tDM)
(Jiangkou Gp); BIF
Nantuo Fm.
Zhu Polong Fm. Yutang/ Foreland basin Y Y Y Y
Yulmeinak diamictite
(Tarim Basin);
Beiyixi, Altungol,
Tereeken and
Hankalchough
diamictites (Tarim
Block); and others

Note: ‘U –Pb direct’ means dates from volcanic intrusions or tuffs anywhere in the succession, not necessarily from the glaciogenic strata itself. ‘U– Pb detrital’ means dates from detrital zircons from within the succession. Data in
presumed correlative successions were not included.

5
6 E. ARNAUD ET AL.
lacunae are due either to the inability to find an appropriate author
or because little or no work has been published in that region since
the publication of Hambrey & Harland (1981). In such cases,
readers are referred to that earlier volume for more information
or, where appropriate, to more recent literature referenced in
chapters covering neighbouring basins. Where possible, these
overlooked successions were incorporated into chapters covering
nearby regions or related deposits. Most are also listed in Evans
& Raub’s review of the palaeomagnetic database.
The current knowledge base
General trends
Many of the successions and regions covered in this book were
reviewed in Hambrey & Harland (1981). Since that time, the great-
est advances in our knowledge have come from the fields of
geochemistry, palaeomagnetism and geochronology, although
palaeoenvironmental interpretations of many glaciogenic and
overlying units have also been variably reinforced and challenged
in recent years. One of the novel aspects of the Snowball Earth
hypothesis was its integration of various geological data sets,
specifically bringing attention to the geochemical signatures
associated with the Neoproterozoic glacial deposits. As a result,
geochemical analyses of associated carbonate rocks proliferated
from the late 1990s onwards. Although the main focus has been
in acquiring traditional stable isotope (namely d13C and d34S)
and 87Sr/86Sr data sets to reconstruct ocean compositions and
terrestrial weathering processes (Halverson & Shields-Zhou
2011), these rocks have also proven fertile grounds for novel iso-
topic approaches such as Ca and B isotopes, and multiple O
isotopes (e.g. Kasemann et al. 2004; Bao et al. 2008). In addition,
the hypothesis generated renewed interest in the idea of global syn-
chronous glaciation, its onset and initiation, which in turn
prompted focused efforts to provide additional geochronological
constraints. The development of new techniques such as the use
of Re –Os in dating organic shales, the proliferation of SHRIMP
and LA-ICP-MS analyses, and the refinement of the ID-TIMS
U –Pb dating technique have simultaneously refined some age
constraints and seriously challenged conventional age models for
key Neoproterozoic successions, particularly in Australia. The
glacial deposits themselves, as the most direct record of the
global glaciations, have come under much greater scrutiny, and
many units have been re-evaluated using facies and sequence stra-
tigraphic analysis, incorporating advances in our understanding of
glaciated basins made in the last 30 years. The role of tectonics in
modulating climate change during this time period and the role
of basin setting in controlling the preservation and nature of the
sedimentary record has also been explored further (e.g. Eyles &
Januszczak 2004; Allen 2007; Stern 2008).
Some workers have focused on sites that previously had rela-
tively little to no data, such as Mongolia, Alaska, Ethiopia, Egypt
and western Arabia and some parts of Australia, Russian Federa-
tion and South America, while some of the classic Neoproterozoic
sections have been the subject of additional study by several
Fig. 1.3. Map showing distribution of
Neoproterozoic glaciogenic successions
within North America, as covered in this
volume. (A) Hula Hula diamictite, Arctic
Alaska; (B) Tatonduk Inlier, Alaska–Yukon
border; (C) Windermere Supergroup,
Mackenzie Mountains, Canada; (D)
Windermere Supergroup, southern Canadian
Cordillera; (E) SE Idaho and Utah, USA; (F)
Edwardsburg Fm., central Idaho, USA; (G)
Kingston Peak Fm., eastern Death Valley
region, USA; (H) Kingston Peak Fm.,
Panamint Range, USA; (I) Gaskiers Fm.,
Newfoundland, Canada; (J) Squantum
Member, Boston Basin, USA.
Table 1.3. Summary of Neoproterozoic data sets in North America

Lead author Glaciogenic strata Structural Geochemistry Geobiology Economic Palaeomagnetism Geochronology
framework
d13C d18O d34S 87
Sr/86Sr Other U –Pb U– Pb Other
detrital direct

Macdonald Hula Hula Diamictite Unknown Y Y Pb/Pb

(detrital),
Rb– Sr
Macdonald Rapitan Gp.; Hay Rift to passive Y Y Y Y Ar/Ar
Creek Gp.
Hoffman Rapitan Gp.; Ice Rift to passive Y Y Y Y d57Fe, Ce/Ce*, Y Fe Y (Rap. Gp.) Pb/Pb
Brook Fm. d44Ca, Elemental

INTRODUCTION
(Stelfox Mb.) analysis, REE
Smith Toby Fm., Vreeland Rift; controversial Y Y Y TOC, Mo, V/Cr Y Cu Y Re–Os,
Fm (Windermere upper WSG Sm–Nd
Supergroup)
Link Pocatello Fm. Rift to passive Y CIA, TOC, Nd, Sr Y Y Y
(Idaho), Mineral
Fork Fm. (Utah)
Lund Edwardsburg Fm., Extensional Whole-rock Y
Moores Lake Fm. (inferred) geochem.
Mrofka Kingston Peak Fm. Extensional Y Y Y Y
(Eastern Death
Valley)
Petterson Kingston Peak Fm. Rift Y U, U– Th,
(Panamint Range) Iron ore
Carto Gaskiers Fm. Arc Y FeHR/FeT Y Y Y Y
Carto Squantum Mb. Extensional-rift, CIA Y Y Y Pb/Pb
intra-arc

Note: ‘U –Pb direct’ means dates from volcanic intrusions or tuffs anywhere in the succession, not necessarily from the glaciogenic strata itself. ‘U– Pb detrital’ means dates from detrital zircons from within the succession. Data in
presumed correlative successions were not included.

7
8 E. ARNAUD ET AL.
research groups, notably Svalbard, northern Norway, Scotland/
Ireland, the North American Cordillera (Death Valley, Idaho and
western Canada), Namibia, various regions in Australia, and
China. As a result, some of these classic localities have become
‘type’ localities for investigating Neoproterozoic glaciations and
their aftermath.
With the increasingly global focus in research into the Neopro-
terozoic and ongoing efforts to define and normalize chronostrati-
graphic units in the Neoproterozoic, the terminology used to refer
to Neoproterozoic glaciations and, implicitly, their age and global
correlations has significantly evolved. For example, now largely
absent from the international lexicon are such previously main-
stream terms as Sinian, Vendian and Varangerian, which refer
to overlapping subsets of Neoproterozoic time spanning glacial
epochs. Because these intervals were defined based on regional
stratigraphic sequences with little basis other than the occurrence
of glacial deposits, their use was inconsistent and often confusing.
New informal terms have emerged in their place and partly as
a result of the intense efforts to determine the number of Neo-
proterozoic glacial epochs. Specifically, the Neoproterozoic
glaciations are now commonly identified by their relative timing
and assumed correlation with well-studied localities. Mid-
Ediacaran glacial deposits are widely referred to as Gaskiers,
based on the eponymous glaciogenic formation on the Avalon
Peninsula of southeastern Newfoundland, which occurs just
below the first appearance of Ediacaran fossils c. 575 Ma (Nar-
bonne & Gehling 2003). Older, Cryogenian-aged glacial deposits
are commonly referred to as Sturtian (older) and Marinoan
(younger) (e.g. Kennedy et al. 1998; Halverson et al. 2005) in an
adaptation of terms originally intended for specific sequences in
the Adelaidean rift-basin of South Australia (Preiss 2000).
Despite their proliferation and appeal, the application of these
terms usually hinges on inferred inter-regional correlations that
are commonly controversial and often poorly backed up by radio-
metric data. Although biostratigraphy (Grey et al. 2011), chemos-
tratigraphy (Halverson & Shields-Zhou 2011) and magnetic
stratigraphy (Evans & Raub 2011) are increasingly able to
resolve relative age assignments and correlations, we have
favoured reference to the emerging Neoproterozoic timescale
(i.e. Cryogenian and Ediacaran) over the terms Sturtian, Marinoan
Fig. 1.4. Map showing distribution of
Neoproterozoic glaciogenic successions
within South America, as covered in this
volume. (A) Chiquerio Fm., Peru; (B) Puga
Fm., Paraguay Belt, Brazil; (C) Serra Azul
Fm., Paraguay Belt, Brazil; (D) Bebedouro
Fm., Brazil; (E) São Francisco Craton,
Brazil; (F) The Macaúbas Group, SE Brazil;
(G) Moema laminites, São Francisco basin,
Brazil; (H) Jequitaı́ Fm., SE Brazil; (I) Playa
Hermosa Fm., Uruguay; (J) Last Ventanas
and San Carlos formations, Uruguay;
(K) Tandilia system, Argentina.
Table 1.4. Summary of Neoproterozoic data sets in South America

Lead author Glaciogenic strata Structural Chemostratigraphy Geobiology Economic Palaeomagnetism Geochronology
framework
d13C d18O d34S 87
Sr/86Sr Other U –Pb U–Pb Other
detrital direct

Chew Chiquerı́o Fm. Extensional, Y Y Fe, Cu, Au Y Y

not well
constrained
Alvarenga Puga Fm., Cuiabá Gp, Extensional Y Y Y Y Fe (BIF), Mn Y Y Pb/Pb, Nd
Jacadigo Gp.
Figueiredo Serra Azul Fm. Passive margin Ar/Ar, K– Ar,
Sm–Nd
Guimaraes Bebedouro Fm. Extensional Y Y Barite nodules; Y Pb/Pb
phosphate

INTRODUCTION
Misi Bambuı́ Gp., Una Not well Y Y Y (ltd) Fe– Mn, Y Re– Os, Pb/
Gp., Vazante Gp., constrained Phosphorite, Pb, Sm–Nd
Vaza Barris/Miaba fluorite, barite,
Gp. Pb–Zn, Zn
Pedrosa-Soares Serra do Catuni Fm., Rift to passive Mn, Fe, diamonds Y Y
Nova Aurora Fm., margin
Lower Chapada
Acaua Fm.
Rocha-Campos Moema Laminites Intracratonic basin Major and trace Kaolinite
elements, (ceramic)
REE, Eu, Ce,
Uhlein Jequitaı́ Fm. Cratonic/ Y Y Y Y Rb– Sr, Pb/Pb
intracratonic?
Pazos Playa Hermosa Fm. Not well Preliminary Ar/Ar, K– Ar,
constrained Rb– Sr
Pecoits Las Ventanas Fm., Strike–slip basin Y Y Y K– Ar, Rb –Sr,
San Carlos Fm. Ar/Ar
Pazos Sierra del Volcán Fm. Not well Y Y Y? Y Rb– Sr
constrained

Note: ‘U –Pb direct’ means dates from volcanic intrusions or tuffs anywhere in the succession, not necessarily from the glaciogenic strata itself. ‘U– Pb detrital’ means dates from detrital zircons from within the succession. Data in
presumed correlative successions were not included.

9
10 E. ARNAUD ET AL.
and Gaskiers in this volume in recognition that ages and corre-
lations of the Neoproterozoic glacial deposits remain controver-
sial. This said, the fact that the lower boundary of the newly
defined Ediacaran Period is defined as the contact between the gla-
ciogenic Elatina Fm. (¼Marinoan) and the overlying Nuccaleena
cap carbonate (Knoll et al. 2006) is suggestive of the chronostrati-
graphic importance of at least one of these Neoproterozoic
glaciations.
The definition of this boundary was bolstered by several new,
well placed and precise radiometric ages on basal Ediacaran cap
dolostones (i.e. the lower, transgressive, dolomite facies of the
cap carbonates; Hoffman et al. 2011) that strengthen the case
for their global age equivalence. The trickle of new radiometric
ages is generally tightening the age constraints on this late
Cryogenian glaciation and the other glacial epochs and at the
same time calibrating and helping to refine the secular evolution
of seawater geochemical proxies (Halverson & Shields-Zhou
2011). New ages have demonstrated that some putative ‘Cryogen-
ian’ glacial deposits are in fact post-Cryogenian in age, such as the
Gaskiers unit in Newfoundland, while the purported ‘Cambrian’
glaciation in the Taoudeni Basin is now regarded unequivocally
as Cryogenian in age. Recent U –Pb and increasingly Re –Os
data (Condon & Bowring 2011) confirm the division of Cryogen-
ian glaciations into two main episodes (c. 720–c. 660 Ma and
c. 650 –635 Ma), although older diamictites of unconfirmed
origin may hint at localized earlier Cryogenian glaciation. In con-
trast to the now reasonably well constrained late Cryogenian gla-
ciation, the wide range of dates obtained from older Cryogenian
glacial deposits long thought to be correlative (mainly in Australia
and the North American cordillera), have called into question their
equivalence and have been used to argue against global glaciation.
Whereas the available data allow for a single, long-lived earlier
Cryogenian glaciation, it is clear that the implications and inter-
pretations of these many new ages are contentious and rapidly
evolving. Therefore, references throughout the book to early,
earlier or middle Cryogenian glaciation should not be assumed
to indicate synchronous glaciation.
In recent years, with resource exploitation reaching unprece-
dented peaks, there has also been increasing interest in the econ-
omic potential of Neoproterozoic strata. Ediacaran rocks comprise
petroleum source rocks in China and Oman, and this has triggered
interest in correlative post-glacial successions in Africa, India,
Australia, Brazil and Russia. Phosphorus resources have also
become a vulnerable commodity worldwide as Moroccan deposits
Fig. 1.5. Map showing distribution of
Neoproterozoic glaciogenic successions
within Europe, as covered in this volume.
(A) Polarisbreen Group, NE Svalbard; (B)
NE Greenland; (C) Smalfjord and
Mortensnes formations, Finnmark, Norway;
(D) Caledonides, NW Scandinavia; (E)
Moelv and Koppang formations, southern
Norway; (F) Caledonides, central
Scandinavia; (G) Lillfjället Fm., southern
Swedish Caledonides; (H) Port Askaig Fm.,
Scotland; (I) Scotland and Ireland.
Table 1.5. Summary of Neoproterozoic data sets in Europe

Lead author Glaciogenic strata Structural Chemostratigraphy Geobiology Economic Palaeomagnetism Geochronology
framework
d13C d18O d34S 87
Sr/86Sr Other U–Pb U –Pb Other
detrital direct

Halverson Petrovbreen Mb., Thermally Y Y Y D17O Y

Wilsonbreen Fm. subsiding
(Polarisbreen Group) margin
Stouge Ulvesø Fm., Storeelv Fm. Rift to passive Y TC, TOC, TS Y Y/unreliable?
and Støvfanget Fm.
Rice Smalfjord Fm., Rift to passive Y Y Y Rb –Sr
Mortensnes Fm.

INTRODUCTION
Stodt Possible correlatives of Passive margin Y (ltd) Y
the Mortensnes Fm.
Nystuen Moelv Fm., Koppang Rift; pericratonic Y (ltd) Y (ltd) Lu–Hf Y Y Rb –Sr, Re– Os
Fm. shelf
Kumpulainen Längmarkberg Fm., Rift to passive
Risback Gp. margin
Diamictite?
Kumpulainen Lillfjället Fm. Rift to passive Ar/Ar
margin?
Arnaud Port Askaig Fm. Extensional Y Y CIA Y Pb– Zn, magnetite Y Pb/Pb, Sm– Nd
Prave Dalradian Supergroup Rift Y Y Y Y
overview-Port
Askaig Fm.;
Stralinchy-Reelan;
MacDuff/Loch na
Cille/Inishowen
boulder beds

Note: ‘U –Pb direct’ means dates from volcanic intrusions or tuffs anywhere in the succession, not necessarily from the glaciogenic strata itself. ‘U– Pb detrital’ means dates from detrital zircons from within the succession. Data in
presumed correlative successions were not included.

11
12 E. ARNAUD ET AL.
become the only widely traded source of phosphorus for fertilizer.
Ediacaran-age phosphorites of China are the only other world-
class phosphorite resource, sparking interest in possibly correla-
tive, glacially associated phosphorites in West Africa, South
America and Australia. In addition to being an oceanographic
curiosity of the Cryogenian Period, the widespread deposition
of iron and manganese deposits may also have economic impli-
cations for those regions where they are found (South America,
northwestern Canada, India, China and Australia; Hoffman
et al. 2011).
Significant advances have been made in all regions since the
publication of Hambrey & Harland (1981). However, with new
data always come new questions and controversies. Some of the
notable advances and contentious issues are highlighted below.
Readers interested in specific glaciogenic successions are encour-
aged to read chapters from neighbouring sites as controversies are
sometimes most apparent when viewing the geological record
from a regional vantage point.
Africa
Neoproterozoic glaciogenic successions reported from Africa
are located primarily in the NW, central and southern regions,
with a cluster of sites in Namibia (Fig. 1.1, Table 1.1). These
occur in a wide range of tectonic settings, with few sites contain-
ing biostratigraphic or palaeomagnetic data. Geochemical and
geochronological studies have been carried out at most sites,
with the most complete data sets found in the Taoudeni Basin in
NW Africa and the Otavi Group of Namibia. The glacial deposits
in the former (Shields-Zhou et al. 2011) remain significant for
being among the few to preserve unequivocal terrestrial glacial
deposits, which are now known to be age-equivalent to other end-
Cryogenian glaciogenic units around the world. Robust geochro-
nological constraints spanning the two glaciogenic intervals of
the Otavi Group are particularly notable. The Otavi Group has
been the subject of intense study over the past two decades, and
despite lingering controversy over interpretation of glacial depos-
its in the succession (Hoffman 2011), plays a central role in debates
over the Snowball Earth hypothesis. Prave et al. (2011) have pro-
vided a useful review of the hitherto poorly documented Witvlei
Group in east-central Namibia (northern Kalahari craton), which
provides a firmer basis for correlation with the numerous Neopro-
terozoic glacial deposits elsewhere on the Kalahari craton (see the
chapters by Frimmel). Whereas few geochronological or geochem-
ical data had been available from the West Congo belt until
recently, new research has provided key data that strengthen
glacial interpretations and correlations across the belt and else-
where on the Congo craton (Tait et al. 2011), such as the Katangan
Supergroup, best known for its world class Cu deposits (Master &
Wendorff 2011).
Eurasia –Nubian Shield
This region encompasses Neoproterozoic glaciogenic successions
found throughout the Russian Federation, Mongolia, China and
India, as well as those from Oman, the shores of the Red Sea
and Ethiopia (Fig. 1.2, Table 1.2). Excellent preservation of Cryo-
genian glacial deposits in northern Oman have allowed detailed
sedimentological and sequence stratigraphic analysis to be
carried out over the last several decades. The results have shed
light on the environmental conditions during Neoproterozoic gla-
ciations, with some more recent geochemical and geochronologi-
cal analyses contributing a relatively comprehensive data set for
this region. Unfortunately, palaeomagnetic studies in Oman have
thus far yielded no reliable palaeolatitudes. In contrast, studies in
Ethiopia and the shores of the Red Sea have focused on geochem-
ical and geochronological aspects of Neoproterozoic geology,
making this region an ideal candidate for future work in sedimen-
tology and stratigraphy to confirm glaciogenic conditions existed
in these regions and maximize the impact of these new radiometric
ages. The basins of the Arabian –Nubian Shield are unique in that
they occur in arc or back-arc basins, a relatively rare tectonic
setting for glaciated Neoproterozoic basins.
Glaciogenic successions in the Russian Federation and neigh-
bouring Mongolia, Kazakhstan and Kyrgyzstan are found
within primarily passive margin and foreland basin settings, with
Fig. 1.6. Map showing distribution of
Neoproterozoic glaciogenic successions
within Australia, as covered in this volume.
(A) Tasmania; (B) the Kimberly region and
NW Northern Territory; (C) Mid-Ediacaran
ice rafted deposits, Adelaide Geosyncline
and Officer Basin; (D) Central Australia; (E)
Billy Springs Fm.; (F) Yudnamutana
Subgroup (Sturtian); (G) Elatina Fm.
Table 1.6. Summary of Neoproterozoic data sets in Australia

Lead author Glaciogenic strata Structural Chemostratigraphy Geobiology Economic Palaeomagnetism Geochronology
framework
d13C d18O d34S 87
Sr/86Sr Other U– Pb detrital U–Pb Other
direct

Calver Cottons Breccia, Julius Rift, epicratonic Y Y Y Y Calcic scheelite Unpublished Y Y Re–Os,
R. Mbr., Croles Hills basin skarns Nd–Sm
Diamictite; Wedge
R. beds, Cotcase Ck. Fm.
Corkeron Walsh Fm., Landrigan/ Unknown Y Y Unreliable Y Y Rb– Sr
Egan Fms., Fargoo/

INTRODUCTION
Moonlight Valley Fm.
Gostin Bunyeroo Fm., Dey Dey Thermally Y Atomic H/C Y Y Y Rb– Sr
Mudstone subsiding basin; ratios
platformal
Hill Glaciogenic deposits of Intracratonic sag Y Y Y Y Y Re–Os
Central Australia basin
(Officer, Amadeus,
Ngalia, Georgina Basins
Jenkins Billy Springs Fm. Rift complex Y Unpub. Y Y (ltd)
Preiss Yudnamutana Subgroup Rift Y Y Y Y Elemental Y Rb– Sr,
(Sturt and correlatives) analysis, REE, Re–Os
Ironstone
geochemistry
Williams Yerelina Subgroup, Rift Y Y Y Y Y Re–Os
Elatina Fm.

Note: ‘U –Pb direct’ means dates from volcanic intrusions or tuffs anywhere in the succession, not necessarily from the glaciogenic strata itself. ‘U– Pb detrital’ means dates from detrital zircons from within the succession. Data in
presumed correlative successions were not included.

13
14 E. ARNAUD ET AL.
geochemical data (primarily carbon, but also oxygen and strontium
isotopes in some cases) as well as important biostratigraphic data.
Interestingly, these studies have led to some contradictory strati-
graphic correlations across neighbouring basins (see chapters on
the Bokson Group and Khubsugul Basin). Chumakov (2011)
suggests that two diamictite-bearing horizons within the Bokson
Group in the East Sayan Mountains are likely Ediacaran in age
based on biostratigraphic data, whereas Macdonald & Jones
(2011) suggest that the two horizons in the neighbouring Mongo-
lian basin correspond to an early and late Cryogenian glacial
episode. Dispute also remains over the glacial origin of the
younger Mongolian diamictite-bearing unit. Nonetheless, the tec-
tonic setting of the region makes these basins suitable targets for
future geochronological studies, which may help to resolve these
stratigraphic inconsistencies. The chapter on the Marnya Fm. in
the East Sayan Mountains (Sovetov 2011) is notable for its review
of regional stratigraphic correlation and useful comparison of the
Russian stratigraphic scheme with that of the more widely used
International Subcommission on Stratigraphy (see also Table
27.2). Palaeogeographic reconstructions in northern Eurasia are
hampered by the lack of reliable palaeomagnetic data, although
ongoing studies in Mongolia show some promise.
The Blaini Fm. of India did not receive much attention until
the more recent studies reported in this volume, which include
new sedimentological and geochemical data. In contrast, the gla-
ciogenic successions of the Yangtze region (rift setting) and
Tarim Block (foreland basin) have a long history of Neo-
proterozoic research that has resulted in extensive data sets for
geochemistry, geobiology and biostratigraphy, palaeomagnetism
and geochronology. Notably, a U– Pb zircon age from an ash
bed in the Yangtze region has provided a definitive date for the
end of the late Cryogenian glaciation (c. 635 Ma), whereas max-
imum and syn-glacial ages in China and elsewhere have shown
that this late Cryogenian glaciation lasted ,10 Ma (Condon &
Bowring 2011). In addition, palaeomagnetic data indicating low
palaeolatitudes for the Nantuo diamictite have one of the highest
reliability ratings of available palaeomagnetic data, as discussed
by Evans & Raub (2011). Lastly, novel geobiological and biostra-
tigraphic studies have demonstrated the preservation of animal
embryos in the Ediacaran-age Doushantuo Fm. of the Yangtze
region (Hagadorn et al. 2006).
North America
Many of the Neoproterozoic glaciogenic successions in North
America are found along the length of the North American
Cordillera in rift basins and passive margin settings, whereas
others are found on exotic terranes in Arctic Alaska and along
the eastern margin of North America (Fig. 1.3, Table 1.3). The
Cordilleran sections generally have a long history of research,
but have enjoyed renewed interest and have been the source of
much new data. While many researchers have attempted to
correlate these widely separated sections, recent studies have
highlighted the possible diachroneity of Neoproterozoic glacia-
tions and rifting related to Rodinian break-up. New additions to
the North American database include the preliminary results of
ongoing multidisciplinary studies in Alaska and the Yukon
including geochronological and palaeomagnetic data (Macdonald
2011; Macdonald & Cohen 2011); new geochronological data
from Idaho (Link & Christie-Blick 2011; Lund et al. 2011) and
the Canadian Cordillera (Smith et al. 2011; Macdonald et al.
2010); detailed mapping and geochemical data from the Death
Valley region; a better understanding of the tectonic development
of the southern Cordillera margin based on work in Idaho,
Utah and Death Valley; the analysis of a deepwater ‘cap’ car-
bonate in the southern Canadian Cordillera; and influential
studies on the impressive Ediacaran biota and ocean redox in
Newfoundland.
South America
Neoproterozoic glaciogenic successions in South America are
primarily clustered in the Paraguay Belt and on the São
Franscisco craton of Brazil, with other sections reported from
Uruguay, Argentina and Peru (Fig. 1.4, Table 1.4). The Araras
Group in the northern Paraguay Belt, in particular, has been the
subject of interdisciplinary studies and preserves some of the
best evidence for Ediacaran glaciation in South America (Figueir-
edo et al. 2011). New entries to the South American database
include the sections in Peru, Uruguay and Argentina as well as
the Moema laminites in Brazil. Despite some new geochronologi-
cal data, there remain significant challenges in regional strati-
graphic correlation schemes, limited information regarding
palaeolatitude, and uncertainty regarding tectonic setting and
glacial origin of some of the units described herein. These chal-
lenges are manifested in significant controversies over the corre-
lation and ages of Proterozoic glacial deposits across Brazil
(Misi et al. 2011). Despite uncertainties in age, the correlations
between the Bambuı́ and Una groups have been strengthened
(Guimarães et al. 2011).
Europe
The Neoproterozoic glaciogenic successions of Europe described
here are primarily those found in extensional and passive margin
basins of Scandinavia, Svalbard, Greenland and the British Isles
(Fig. 1.5, Table 1.5). Although most of these sites were described
in Hambrey & Harland (1981), new data have become available in
recent decades, including additional geochemical data from most
of these successions, additional information on associated carbon-
ate strata and some additional geochronological constraints. Most
successions in the region have been significantly impacted by
Caledonian tectonics and hence lack reliable palaeomagnetic
data directly related to the glacial deposits. Robust geochronologi-
cal constraints are also largely lacking, making regional strati-
graphic correlations relatively tenuous, despite some advances
using chemostratigraphic data. Interestingly, ongoing studies in
Greenland have led some researchers to propose that the Greenland
and Svalbard sections are not as intimately connected palaeogeo-
graphically as was previously proposed.
Australia
The Neoproterozoic glaciogenic successions in Australia covered
in this volume are very similar to the entries found in Hambrey
& Harland (1981) over 30 years ago (Fig. 1.6, Table 1.6).
Recent advances include the collection of additional geochemical
and biostratigraphic data, continued state-of-the-art palaeomagnet-
ism research with confirmation of low-latitude glaciation, and
additional geochronological constraints that have prompted abun-
dant discussion about the duration and timing of the older Cryo-
genian glaciation. Entries for South Australia are focused on
individual glacial events (Preiss et al. 2011; Williams et al.
2011), with new evidence for Ediacaran-aged glaciation within
the Adelaidean rift-basin (Gostin et al. 2011; Jenkins 2011). Cor-
keron, Hill et al. and Calver provide timely and useful syntheses of
the separate Neoproterozoic successions in the Kimberly region,
‘Centralian Superbasin’ and Tasmania, respectively.
Calibrating Neoproterozoic change
One of the prominent developments in stratigraphy over the last
decade has been the attempt to establish consistent, rock-based
periods of geological time beneath the Cambrian Period. Neopro-
terozoic glaciation played a pivotal role in this when the Ediacaran
 INTRODUCTION
Period, which was the first period of the geological timescale to be
established for a hundred years, was defined as beginning directly
after the widespread, end-Cryogenian (¼ Elatina or Marinoan) gla-
ciation. Now, international efforts are under way to further subdi-
vide the Ediacaran Period and define the underlying Cryogenian
Period. In this endeavour, the integrated approach taken by many
of the authors in the present volume will be key as a global strati-
graphic framework can only be established in the Precambrian by
meshing together sparse fossil information with high-resolution
chemostratigraphy, strategically placed age constraints and litho-
stratigraphic markers.
At present, the base of the Cryogenian Period is defined
chronometrically at 850 Ma, but few age constraints exist to ident-
ify this level in sedimentary successions from around the world.
Although, the base of a new chronostratigraphically defined Cryo-
genian Period is yet to be established, any GSSP is likely to be
placed ‘beneath the oldest clearly glaciogenic deposits in a Neo-
proterozoic succession’ at an outcrop horizon with ‘proven poten-
tial for global C- and Sr-isotope stratigraphic correlation and
preferably be amenable to microfossil biostratigraphy, isotope
geochronology and other forms of global correlation such as mag-
netostratigraphy’ (Neoproterozoic Subcommission 2009 Annual
Report). This emphasis on glacial strata will likely lead to a
shorter Cryogenian Period more in line with its Phanerozoic
counterparts, and with a base close in age to the Dunn et al.
(1971) estimate of c. 750 Ma. The unique Neoproterozoic palaeo-
climate, which underpins this volume, has already cemented the
place of Neoproterozoic ice ages in the international geological
timescale (Ogg et al. 2008) and will continue to be central in the
subdivision of Proterozoic time.
Future work
This memoir was inspired by Earth’s Pre-Pleistocene Glacial
Record (Hambrey & Harland 1981), and it is hoped that it can
serve similarly as a long-lasting reference work that provides
useful summaries of most Neoproterozoic glacial successions,
while also highlighting the state-of-the art in the study of these
rocks. The large number of chapters and impressive list of contri-
butors highlight the extraordinary interest in the Neoproterozoic
and the vast amount of research carried out over the past 30
years. At the same time, the results of this research have revealed
gaping holes in our understanding of the Neoproterozoic and
aroused compelling new questions. Many glaciogenic units
described in Hambrey & Harland (1981) have received little or
no attention since that time, such as the widespread diamictites
of the Sino-Korean craton, which are still of uncertain age and
origin. Radiometric age constraints on most glacial units remain
sparse or non-existent, which poses a persistent challenge to inter-
regional correlations. Many important advances have been in the
biostratigraphy of Cryogenian and Ediacaran successions, but
most sections remain poorly studied. In addition, a welter of excit-
ing new geochemical methods, such as the D47 palaeothermometer
and transition-metal stable isotopes, await rigorous application to
the Neoproterozoic. Clearly, much remains to be done, but in
almost every region of the world there are now interdisciplinary
teams of scientists at work in what has become a global endeavour
not only to explore the extremes of past climate change, but also
to evaluate the link between eukaryotic evolution and global
glaciations.
We wish to thank all the authors who made this collection possible, for agreeing to
write these contributions amidst many other writing commitments, and patiently
awaiting the completion of the volume. We particularly thank N. Chumakov and
Q. R. Zhang for promptly contributing to the original template papers that were
essential in establishing the format for all subsequent regional chapters. We are
also especially grateful to P. Hoffman, F. Macdonald, M. Zhu, R. Stern and
J. Tait for acquiescing to last-minute requests for contributions. We would also
15
like to thank all the reviewers (see list of reviewers) who provided meaningful
and constructive comments and improved the individual chapters. I. Fairchild
kindly provided the cover photo and the historical photograph for the dedication.
E. Arnaud would also like to thank those who provided guidance in developing
the instructions for authors with suggested content required of specific sections:
namely D. Condon, N. Eyles, G. Halverson, B. MacGabhann, P. Martini,
T. Raub, G. Shields-Zhou & R. van der Voo. The following regional coordinators
also assisted in tracking down authors and names of Neoproterozoic glaciogenic
successions in their regions: P. Allen, C. Calver, N. Chumakov, M. Corkeron,
N. Eyles, V. Gostin, P. Hoffman, C. Hoffmann, G. Jiang, K. Kristiansen,
R. Kumpulainen, P. Link, U. Linnemann, J. Meert, N. Miller, P. Pazos,
T. Prave, Q. R. Zhang, H. Rice, T. Rocha Campos, S. Stouge, M. Wendorff
and S. Xiao. We thank J. Howe for his contribution as the Society Books
Editor and for his timely review of this chapter. Angharad Hills at the Geological
Society of London Publishing House kept us on track and provided editorial
assistance throughout the project. This memoir represents a contribution of the
IUGS-UNESCO sponsored International Geoscience Programme (IGCP)
Project # 512.
References
Allen, P. A. 2007. The Huqf Supergroup of Oman: basin development
and context for Neoproterozoic glaciation. Earth-Science Reviews,
84, 139– 185.
Bao, H., Lyons, J. R. & Zhou, C. 2008. Triple oxygen isotope evidence
for elevated CO2 levels after a Neoproterozoic glaciation. Nature,
453, 504–506.
Chumakov, N. M. 2011. Glacial deposits of the Bokson Group, East
Sayan Mountains, Buryatian Republic Russion Federation. In:
Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geo-
logical Record of Neoproterozoic Glaciations. Geological Society,
London, Memoirs, 36, 285–288.
Condon, D. J. & Bowring, S. A. 2011. A user’s guide to Neoproterozoic
geochronology. In: Arnaud, E., Halverson, G. P. & Shields-
Zhou, G. (eds) The Geological Record of Neoproterozoic Glacia-
tions. Geological Society, London, Memoirs, 36, 135– 149.
Dunn, P. R., Thomson, B. P. & Rankama, K. 1971. Late Pre-Cambrian
glaciation in Australia as a stratigraphic boundary. Nature, 231,
498– 502.
Evans, D. A. D. & Raub, T. D. 2011. Neoproterozoic glacial palaeo-
latitudes: a global update. In: Arnaud, E., Halverson, G. P. &
Shields-Zhou, G. (eds) The Geological Record of Neo-
proterozoic Glaciations. Geological Society, London, Memoirs, 36,
93– 112.
Eyles, N. & Januszczak, N. 2004. Interpreting the Neoproterozoic
glacial record; the importance of tectonics. In: Jenkins, A.,
McMenamin, M. A. S., McKey, C. P. & Sohl, L. (eds) The
Extreme Proterozoic: Geology, Geochemistry and Climate. Ameri-
can Geophysical Union, Geophysical Monograph 146, 125–144.
Fairchild, I. J. & Kennedy, M. J. 2007. Neoproterozoic glaciation in the
Earth System. Journal of the Geological Society, London, 164,
895– 921.
Figueiredo, M. F., Babinski, M. & Alvarenga, C. J. S. 2011. The Serra
Azul Formation, Paraguay Belt, Brazil. In: Arnaud, E., Halverson,
G. P. & Shields-Zhou, G. (eds) The Geological Record of Neopro-
terozoic Glaciations. Geological Society, London, Memoirs, 36,
499– 502.
Gostin, V. A., McKirdy, D. M., Webster, L. J. & Williams, G. E.
2011. Mid-Ediacaran ice-rafting in the Adelaide Geosyncline and
Officer Basin, South Australia. In: Arnaud, E., Halverson, G. P.
& Shields-Zhou, G. (eds) The Geological Record of Neopro-
terozoic Glaciations. Geological Society, London, Memoirs, 36,
673– 676.
Grey, K., Hill, A. C. & Calver, C. 2011. Biostratigraphy and
stratigraphic subdivision of Cryogenian successions of Australia
in a global context. In: Arnaud, E., Halverson, G. P. &
Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic
Glaciations. Geological Society, London, Memoirs, 36, 113– 134.
Guimarães, J. T., Misi, A., Pedreira, A. J. & Dominguez, J. M. L. 2011.
The Bebedouro Formation, Una Group, Bahia (Brazil). In: Arnaud,
E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological
16 E. ARNAUD ET AL.
Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 503– 508.
Hagadorn, J. W., Xiao, S. et al. 2006. Integrated X-ray insights into
cellular and subcellular structures of Neoproterozoic animal
embryos. Science, 314, 291–294.
Halverson, G. P. & Shields-Zhou, G. 2011. Chemostratigraphy
and the Neoproterozoic glaciations. In: Arnaud, E., Halverson,
G. P. & Shields-Zhou, G. (eds) The Geological Record of Neopro-
terozoic Glaciations. Geological Society, London, Memoirs, 36,
51 –56.
Halverson, G. P., Hoffman, P. F., Schrag, D., Maloof, A. C. & Rice,
A. H. 2005. Towards a Neoproterozoic composite carbon isotope
record. Geological Society of America Bulletin, 117, 1181–1207.
Hambrey, M. J. & Harland, W. B. (eds) 1981. Earth’s Pre-Pleistocene
Glacial Record. Cambridge Earth Science series. Cambridge Univer-
sity Press, Cambridge.
Hoffman, P. F. 2011. A history of Neoproterozoic glacial geology, 1871–
1997. In: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds)
The Geological Record of Neoproterozoic Glaciations. Geological
Society, London, Memoirs, 36, 17 –37.
Hoffman, P. F., Kaufman, A. J., Halverson, G. P. & Schrag, D. P.
1998. A Neoproterozoic snowball Earth. Science, 281, 1342– 1346.
Hoffman, P. F., Macdonald, F. A. & Halverson, G. P. 2011. Chemical
sediments associated with Neoproterozoic glaciation: iron formation,
cap carbonate, barite and phosphorite. In: Arnaud, E., Halverson,
G. P. & Shields-Zhou, G. (eds) The Geological Record of Neo-
proterozoic Glaciations. Geological Society, London, Memoirs, 36,
67 –80.
Jenkins, R. J. F. 2011. Billy Springs Glaciation, South Australia. In:
Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geo-
logical Record of Neoproterozoic Glaciations. Geological Society,
London, Memoirs, 36, 693–699.
Kasemann, S. A., Hawkesworth, C. J., Prave, A. R., Fallick, A. E. &
Pearson, P. 2005. Boron and calcium isotope composition in Neo-
proterozoic carbonate rocks from Namibia: evidence for extreme
environmental change. Earth and Planetary Sciences Letters, 231,
73 –86.
Kennedy, M. J., Runnegar, B., Prave, A. R., Hoffmann, K. H. &
Arthur, M. 1998. Two or four Neoproterozoic glaciations?
Geology, 26, 1059–1063.
Knoll, A. H., Walter, M. R., Narbonne, G. M. & Christie-Blick, N.
2006. The Ediacaran Period: a new addition to the geologic time
scale. Lethaia, 39, 13 – 30.
Link, P. K. & Christie-Blick, N. 2011. Neoproterozoic strata of South-
eastern Idaho and Utah: record of Cryogenian rifting and glaciation.
In: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The
Geological Record of Neoproterozoic Glaciations. Geological
Society, London, Memoirs, 36, 425–436.
Lund, K., Aleinikoff, J. N. & Evans, K. V. 2011. The Edwardsburg For-
mation and related rocks, Windermere Supergroup, central Idaho,
U.S.A. In: Arnaud, E., Halverson, G. P. & Shields-Zhou, G.
(eds) The Geological Record of Neoproterozoic Glaciations. Geo-
logical Society, London, Memoirs, 36, 437– 447.
Macdonald, F. A. 2011. The Hula Hula diamictite and Katakturuk Dolo-
mite, Arctic Alaska. In: Arnaud, E., Halverson, G. P. & Shields-
Zhou, G. (eds) The Geological Record of Neoproterozoic Glacia-
tions. Geological Society, London, Memoirs, 36, 379–387.
Macdonald, F. A. & Cohen, P. A. 2011. The Tatonduk inlier,
Alaska-Yukon border. In: Arnaud, E., Halverson, G. P. &
Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic
Glaciations. Geological Society, London, Memoirs, 36, 389–396.
Macdonald, F. A. & Jones, D. S. 2011. The Khubsugul Group, northern
Mongolia. In: Arnaud, E., Halverson, G. P. & Shields-Zhou, G.
(eds) The Geological Record of Neoproterozoic Glaciations.
Geological Society, London, Memoirs, 36, 339–345.
Macdonald, F. A., Schmitz, M. D. et al. 2010. Calibrating the Cryo-
genian. Science, 327, 1241–1243.
Master, S. & Wendorff, M. 2011. Neoproterozoic glaciogenic diamic-
tites of the Katanga Supergroup, Central Africa. In: Arnaud, E.,
Halverson, G. P. & Shields-Zhou, G. (eds) The Geological
Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 173–184.
Misi, A., Kaufman, A. J., Azmy, K., Dardenne, M. A., Sial, A. N. &
de Oliveira, T. F. 2011. Neoproterozoic successions of the São
Francisco Craton, Brazil: The Bambuı́, Una, Vazante and Vaza
Barris/Miaba groups and their glaciogenic deposits. In: Arnaud,
E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological
Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 509–522.
Narbonne, G. M. & Gehling, J. G. 2003. Life after snowball: the oldest
complex Ediacaran fossils. Geology, 31, 27 – 30.
Ogg, J. G., Ogg, G. & Gradstein, F. M. 2008. The Concise Geological
Time Scale. Cambridge University Press, Cambridge.
Prave, A. R., Hoffmann, K.-H., Hegenberger, W. & Fallick, A. E.
2011. The Witvlei Group of East-Central Namibia. In: Arnaud, E.,
Halverson, G. P. & Shields-Zhou, G. (eds) The Geological
Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 211–216.
Preiss, W. V. 2000. The Adelaide Geosyncline of South Australia and
its significance in Neoproterozoic continental reconstruction.
Precambrian Research, 100, 21 – 63.
Preiss, W. V., Gostin, V. A., McKirdy, D. M., Ashley, P. M., Wil-
liams, G. E. & Schmidt, P. W. 2011. The glacial succession of
Sturtian age in South Australia: the Yudnamutana Subgroup. In:
Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geo-
logical Record of Neoproterozoic Glaciations. Geological Society,
London, Memoirs, 36, 701–712.
Shields, G. A. 2006. Snowball Earth is dead! Long live Snowball Earth!
Episodes, 29, 287–288.
Shields-Zhou, G. A., Deynoux, M. & Och, L. 2011. The Record of Neo-
proterozoic Glaciation in the Taoudéni Basin, NW Africa. In:
Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geo-
logical Record of Neoproterozoic Glaciations. Geological Society,
London, Memoirs, 36, 163–171.
Smith, M. D., Arnaud, E., Arnott, R. W. C. & Ross, G. M. 2011.
The record of Neoproterozoic glaciation in the Windermere Super-
group, southern Canadian Cordillera. In: Arnaud, E., Halverson,
G. P. & Shields-Zhou, G. (eds) The Geological Record of Neopro-
terozoic Glaciations. Geological Society, London, Memoirs, 36,
413– 423.
Sovetov, J. K. 2011. Late Neoproterozoic (Vendian) glaciogenic deposits
in the Marnya Formation, Oselok Group, in the foothills of the East
Sayan Range, Southwestern Siberian craton. In: Arnaud, E., Hal-
verson, G. P. & Shields-Zhou, G. (eds) The Geological Record
of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 317–329.
Stern, R. J. 2008. Neoproterozoic crustal growth: the solid Earth system
during a critical episode of Earth history. Gondwana Research, 14,
33 – 50.
Tait, J., Delpomdor, F., Préat, A., Tack, L., Straathof, G. & Kanda
Nkula, V. 2011. Neoproterozoic Sequences of the West Congo and
Lindi/Ubangi Supergroups in the Congo Craton, central Africa. In:
Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geo-
logical Record of Neoproterozoic Glaciations. Geological Society,
London, Memoirs, 36, 185–194.
Williams, G. E., Gostin, V. A., McKirdy, D. M., Preiss, W. V. &
Schmidt, P. W. 2011. The Elatina glaciation (late Cryogenian),
South Australia. In: Arnaud, E., Halverson, G. P. & Shields-
Zhou, G. (eds) The Geological Record of Neoproterozoic Glacia-
tions. Geological Society, London, Memoirs, 36, 713–721.
 Chapter 2

A history of Neoproterozoic glacial geology, 1871– 1997

PAUL F. HOFFMAN1,2
1
Department of Earth and Planetary Sciences, Harvard University, Cambridge, MA, 02138, USA
2
School of Earth and Ocean Sciences, University of Victoria, Victoria, BC, V8W 2Y2, Canada (e-mail: paulfhoffman@yahoo.com)

Abstract: Neoproterozoic glacial records have been discovered on 23 palaeocontinents, their rate of discovery changing little since
1871. Yet, half of all the resulting publications appeared since 2000. The history of research before 1998 is described in five stages
defined by publication spikes; subsequent work is not covered because historical perspective is lacking. In stage 1 (1871–1907), ‘Cam-
brian’ (now Neoproterozoic) glaciation was recognized successively in Scotland, Australia, India, Norway, Svalbard and China. Criteria
for recognition included faceted and striated pebbles in matrix-supported conglomerates resting on ice-worn bedrock pavements. In
stage 2 (1908–1940), Neoproterozoic glaciation was shown to have been widespread in Africa, Asia and the Americas. Major textbooks
summarized these findings, but the rejection of continental drift (to account for late Palaeozoic glacial dynamics) put a chill on research.
In stage 3 (1942– 1964), the occurrence of glacial deposits within carbonate successions, as well as nascent palaeomagnetic observations,
suggested that Neoproterozoic glaciers reached sea-level at low palaeolatitudes, but the belated recognition of sediment gravity flowage
caused glacial interpretations to be prematurely abandoned in key areas. In stage 4 (1965– 1981), the extent of Neoproterozoic glaciation
was rethought in light of plate tectonics. Distinctive chemical sediments (iron + manganese formations and cap carbonates) were ident-
ified. In basic climate models, modest lowering of solar luminosity resulted in global glaciation due to ice-albedo feedback, and degla-
ciation due to greenhouse forcing resulted from silicate-weathering feedback in the carbon cycle. Neoproterozoic glacial geologists were
blind to these ideas. In stage 5 (1982– 1997), reliable palaeomagnetic data combined with glacial marine sedimentation models confirmed
that Neoproterozoic ice sheets reached sea level close to the palaeoequator.

The first Neoproterozoic glacial deposits were found in the SW of
Scotland in 1871 and the rate of their discovery on 23 palaeoconti-
nents and microcontinents was essentially linear until 1992
(Fig. 2.1). However, half of all the papers (n ¼ 811) written on
them before the end of 2008 were published in the previous 10
years (Fig. 2.2). This surge in activity is the justification for the
present volume, but this history ends at 1997. We lack historical
perspective on the last decade and the endpoint eliminates the
need to weigh any of the author’s own contributions on this topic.
This chapter begins by placing the recognition of the Port
Askaig boulder beds of Scotland as glaciogenic in its historical
context, coming in 1871 on the heels of the discoveries of late
Palaeozoic glacial deposits in India, Australia and South Africa.
Their inspiration was the Pleistocene glacial controversy of
1837– 1865, which spawned both the orbital and ‘greenhouse’ the-
ories of bidirectional climate change, and established the principal
criteria for the recognition of past glaciation.
The history is divided into five periods punctuated by spikes
in publications (Fig. 2.2) associated with the 10th and 17th
International Geological Congresses in Mexico City (1906) and
Moscow (1937), respectively, the twin palaeoclimate con-
ferences in Newcastle upon Tyne (1963) and Köln (1964), the
IGCP Project 38 volume on Earth’s Pre-Pleistocene Glacial
Record (Hambrey & Harland 1981), and the onset of the present
surge triggered in 1998.
The first period (1871 –1908) saw the recognition of ‘Lower
Cambrian’ glaciogenic formations in SW Scotland (1871), South
Australia (1884), NE Norway (1891), western Svalbard (1898),
South China (1904) and NW India (1908). It ended with overviews
of Late Palaeozoic and Lower Cambrian (‘possibly Pre-
Cambrian’) glaciations by Edgeworth David (1907a, b) at the
10th IGC, and the winning-over of a skeptical Geological
Society in London by a fellow Australian (Howchin 1908).
Faceted and striated clasts in diamictites were the decisive criteria
for glaciogenesis, with a striated subglacial pavement beautifully
exposed beneath ‘Reusch’s Moraine’ (Fig. 2.3) in Finnmark,
Norway, for good measure.
The second period (1909– 1941) saw Neoproterozoic glacial
deposits recognized in southern, central and western Africa,
central and northern Asia, eastern and western North America
and eastern South America (Fig. 2.1, Table 2.1). It was evident
that glaciation ‘at or just before the beginning of the Cambrian’
had been more extreme than in the Pleistocene, possibly equal to
the composite Late Palaeozoic glaciations of Gondwanaland.
These findings were highlighted in a Symposium on Palaeozoic
and Pre-Cambrian Climates at the 17th IGC (1937) in Moscow.
Major textbooks made note of these developments (Köppen &
Wegener 1924; Brooks 1922, 1926; Coleman 1926).
In the third period (1942 – 1964), nascent evidence from
palaeomagnetism and from the occurrence of glacial deposits
atop thick carbonate-dominated successions in the North Atlantic
region and within such successions in central and southern
Africa implied that Neoproterozoic ice sheets had reached sea
level in low palaeolatitudes. This inference met resistance and,
after the discovery of turbidity currents as a cause of graded
bedding (1950), some diamictites were reinterpreted as mass-flow
deposits of non-glacial origin. These issues were discussed at
major conferences on Palaeoclimates in Newcastle-upon-Tyne,
England (1963), and Köln, Germany (1964). The radiation of
macrofauna following soon after the ‘infra-Cambrian’ ice age
was often remarked upon.
The fourth period (1965– 1981) was notable for a detailed
restudy of the classic Port Askaig Tillite, debate over the roles
of true polar wander and plate tectonics in the distribution of
glaciogenic deposits, and the recognition of widespread post-
glacial ‘cap dolomites’ and syn-glacial iron and manganese
formations, features not known from Phanerozoic glaciations.
Climate modelling received a kick-start when simple energy-
balance calculations suggested that the Earth could freeze over
from pole to pole due to ice-albedo feedback in response to lower-
ing of the solar constant by a few percent. Such a ‘white Earth’ dis-
aster was shown to be self-reversing because of negative climate
feedback associated with the geochemical cycle of carbon, but
geologists were unaware of these developments.
In the fifth period (1982 –1997), Neoproterozoic diamictites
were increasingly interpreted in light of glacial marine sedi-
mentation models and, combined with reliable palaeomagnetic
data, proved that ice sheets had reached sea level close to the
palaeoequator. The ‘Snowball Earth’ hypothesis, adapted
from climate and planetary science, was advanced as a

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 17– 37. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.2
18 P. F. HOFFMAN

24 1871 Laurentia (S) overstep the bounds of prudent speculation in suggesting that those erratics
1884 Australia are the reassorted materials of some great Northern Continent that has
22 1887 India yielded to the ceaseless gnawing tooth of time, leaving scattered fragments
1891 Baltica
20 1904 South China
as wreckage of its former greatness, and that the material of which the mass
1910 Avalonia is composed have in time, deeper than we have hitherto suspected, been trans-
18 ported by the agency of ice.’
1913 Laurentia (N)
16 1915 Kalahari
James Thomson, F.G.S. 1871. On the stratified rocks of Islay. Report of the
1929 Congo
14 1930 Siberia 41st Meeting of the British Association for the Advancement of Science,
1930 Sao Francisco Edinburgh, John Murray, London, 110–111.
12 1933 West Africa
1937 Tarim
10 1950 Cadomia
8 1956 Kazakhstan The Port Askaig boulder beds
1959 Amazonia
6 1979 Arequipa
1981 North China The first Neoproterozoic (then ‘Cambrian’) formation to be inter-
4 1982 Arabia
1984 Rio Plata
preted as glaciogenic (see quote above) was the Port Askaig,
2 1990 Mongolia exposed on the east side of Islay (Thomson 1871, 1877), an
0 1992 Iran island in the west of Scotland famous for its single malt whisky.
1880 1900 1920 1940 1960 1980 2000 2003 Nile
2009 Chukotka- The formation is a 750-m-thick succession of 47 mappable
Year of first report Arctic Alaska heterolithic-boulder diamictites with interbeds of crossbedded
sandstone (Kilburn et al. 1965; Spencer 1971; Arnaud 2004). In
Fig. 2.1. Cumulative discovery of Neoproterozoic glaciogenic deposits by his richly detailed sedimentological study, Spencer (1971)
palaeocontinent. See Table 2.1 for locations and references. Note that Laurentia describes the Port Askaig as lying ‘at the same horizon, between
is counted twice because of the large geographical separation between two carbonate-rich formations, for 700 km from NE Scotland to
southern (S) and northern (N) palaeohemisphere deposits of present eastern western Ireland’. It is now thought to be the oldest of three distinct
and western Laurentia, respectively. glaciogenic horizons within the Dalradian Supergroup (McCay
et al. 2006) – older and younger Cryogenian and middle Ediacaran
parsimonious explanation for low-latitude glaciation, associated in age. James Thomson’s speculation (see above) about a ‘great
iron (+manganese) deposits and (later) cap carbonates. Initially Northern Continent’ as the source of the glacial debris turned
ignored, the strong negative reaction to this hypothesis on the out to be true (Cawood et al. 2003), although that continent
part of Neoproterozoic glacial sedimentologists will in future be did survive the ‘gnawing tooth of time’ after all. It is North
seen as the most striking feature of the post-1998 period. America, or more precisely, its pre-Pangaean antecedent Lauren-
tia. It was merely displaced from the Dalradian in the early
Eocene, when Greenland separated from NW Europe. The Port
Prologue: discovery of the Port Askaig Tillite and its Askaig was deposited at palaeolatitudes near 258S according
historical scientific context to reliable palaeomagnetic data from mafic dykes, lavas and
sills of the marginally older (723 –718 Ma) Franklin Igneous
[On the stratified rocks of Islay.] ‘If . . . we compare the embedded boulders of Suite of Arctic Laurentia (Evans 2000). It accumulated on a
granite [in the schist] with the granites found in situ throughout the Highlands, marine shelf, subject to repeated subaerial exposure (Spencer
we feel the necessity of tracing them to another source, and hope we do not 1971; Johnston 1993), situated at the then southeastern margin

1.0
60 60
Cumulative fractional growth AGU
0.8 GM146
55 (n = 811) 55

50 0.6 50

45 0.4 45

40 40
0.2
35 35
0.0 IGCP 38
30 1880 1900 1920 1940 1960 1980 2000 Hambrey & Harland 30
1981
25 25
Newcastle
and Koln
“
20 20
conferences
1870 1880 1890
17th IGC 1963-64
15 15
10th IGC Moscow
Mexico 1937
10 10
1906
5 5

0 0
1890 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010

Fig. 2.2. Growth in the annual number of papers concerning Neoproterozoic glaciation, 1871–2008.
 HISTORY OF NEOPROTEROZOIC GLACIAL GEOLOGY 19

Late Palaeozoic glaciation and the theory of

continental drift

Thomson’s (1871) interpretation of the Port Askaig came at a time

Fig. 2.3. Iconic sketch of the glacially striated pavement beneath the
end-Cryogenian Smalfjord diamictite (‘Reusch’s moraine’) at Bigganjargga,
Varangerfjord, East Finnmark, northern Norway (from Reusch 1891).
of the palaeocontinent. The Port Askaig nicely illustrates the
basic information needed to interpret Neoproterozoic glaciogenic
deposits: their distribution, derivation, depositional environment,
palaeogeographic setting and age.
of growing interest in pre-Pleistocene glaciation. A subglacial
pavement overlain by diamictite in South Australia found in
1859 by A.C. Selwyn proved to be of Permo-Carboniferous age
(Howchin 1912). A Permo-Carboniferous diamictite in the
Talchir coal basin of NE India (Blanford et al. 1859) was shown
to have been deposited by grounded, northward(!)-flowing glaciers
(Fedden 1875; Koken 1907). The late Carboniferous Dwyka Tillite
in Natal, South Africa, rests on a pavement scratched and grooved
by onshore(!)-flowing glaciers (Sutherland 1870). Ironically, these
findings had been inspired by the glacial interpretation of a
Permian-age diamictite in England (Ramsay 1855) that proved
to be non-glacial in origin.
By 1907 it was evident that with continents in fixed relative
positions, no amount of polar wander could prevent early
Permian glaciation in the southern hemisphere from extending
to within 108 latitude of the palaeoequator, while in the northern
hemisphere tropical warmth stretched to the pole (Koken 1907;
Irving 1956). ‘The Permian ice age poses an unsolvable
problem to all models that do not dare to assume horizontal dis-
placements of the continents’, wrote a 32-year-old German
meteorologist (Wegener 1912). With the Atlantic and Indian
oceans closed up, however, all the glacial action could be con-
tained poleward of 458S palaeolatitude, comparable to the
extent of northern hemisphere glaciation in the Pleistocene.
Moreover, a radial pattern of flow could then be inferred,
centred on south-central Africa (Martin 1981). ‘This would
take everything mysterious away from the phenomenon’
(Wegener 1912). The distribution of Late Palaeozoic glacial
deposits was prime motivation for Wegener’s theory of continen-
tal displacement, or drift (Martin 1981). Had he lived, vindication
in the form of plate tectonics would have come to Wegener at the
age of 87.

Table 2.1. First reported occurrences of Neoproterozoic glaciogenic deposits by palaeocontinent

No. Year Palaeocontinent Continent Ref.

1a 1871 Laurentia (south) Europe Thomson (1871)

2 1884 Australia Australasia Woodward (1884)
3 1891 Baltica Reusch (1891)
4 1904 South China Asia Willis (1904)
5 1908 India Holland (1908)
6 1910 Avalonia North America* Sayles & Laforge (1910)
1b 1913 Laurentia (north) North America Hintze (1913)
7 1915 Kalahari Africa Rogers (1915)
8 1929 Congo Beetz (1929)
9 1930 Siberia Nicolaev (1930)
10 1930 São Francisco† South America Moraes Rego (1930)
11 1933 West Africa Baud (1933), Furon (1933)
12 1937 Tarim Norin (1937)
13 1950 Cadomia Wegmann et al. (1950)
14 1956 Kazakhstan Nalivkin (1956)
15 1959 Amazonia Maciel (1959)
16 1979 Arequipa Caldas (1979)
17 1981 North China Mu (1981), Lu et al. (1985)
18 1982 Arabia Gorin et al. (1982)
19 1984 Rio Plata Spalletti & Del Valle (1984)
20 1990 Mongolia Gibsher & Khomentovsky (1990)
21 1992 Iran Hamdi (1992)
22 2003 Nile‡ Miller et al. (2003)
23 2009 Chukotka-Arctic Alaska Macdonald et al. (2009b)

*Before 1980, the Squantum Tillite was thought to be Late Palaeozoic in age.
†
Often considered part of the Congo palaeocontinent.
‡
Western margin of East African orogen.
20 P. F. HOFFMAN
The Pleistocene glacial controversy
Interest in pre-Quaternary glaciation arose in the wake of the
most acrimonious and far-reaching controversy in 19th-century
geology, the brouhaha over the glacial theory for Pleistocene
tills, erratic boulders and associated landforms (moraines, drum-
lins, eskers, kame-and-kettle, fluted ground, grooved bedrock,
crag-and-tail, roches moutonnées, whaleback rocks, cirques and
hanging valleys, horns and arêtes, U-shaped valleys and fjords).
Glacial theory – that most of northern Europe and North
America had been sculpted by enormous and dynamic ice sheets
in the geologically recent past – was independently proposed by
Esmark (1824) in Norway, Dobson (1925) in the USA, Venetz
(1830) and Charpentier (1837) in Switzerland and Bernhardi
(1832) in Germany. None of these papers – published in major
journals by reputable professors (Esmark and Bernhardi), engin-
eers (Venetz and Charpentier) and an industrialist (Dobson) –
raised a ripple. No geologist ever overturned conventional
wisdom with a single paper.
Conventional theory held that these features were products
of floods – meltwater floods resulting from dam bursts in the
Alps (cf. de Saussure, von Humboldt, von Buch, de Beaumont)
or iceberg-laden floods of Arctic origin in northern Europe and
North America (cf. Sefström, Murchison, Hitchcock, Lyell). The
controversy erupted when glacial theory was taken up by a
young, ambitious, energetic and multilingual Swiss palaeo-
ichthyologist (Agassiz 1837, 1840). A recent convert to glacial
theory, Agassiz linked it to the extinction of the boreal megafauna
(e.g. mammoth, mastodon, woolly rhinoceros, giant deer), a
linkage easily disproved (Forbes 1846). He tied the end of the
glacial period to the uplift of the Alps, exactly opposite to the
view of Charpentier, his primary glaciological tutor, and laughable
to those who knew that Alpine orogeny began in the Eocene. Of
the geological establishment, only the former arch-Diluvialist
William Buckland (Oxford University) was quickly won over to
glacial theory.
The climax of Agassiz’s campaign in support of glacial theory
came in 1840 (Davies 1968), when he and Buckland toured
the British Isles, finding varied and widespread evidence of
glacial action (Agassiz 1842). The evidence included tills
(boulder clays) with polished, faceted and striated clasts, moraines
marking the limits of former glaciers, streamlined bedrock and
perched terraces resulting from ice-dammed lakes. At first, Agas-
siz’s findings aroused great interest in the English-speaking world,
but within months of his tour the glacial hypothesis was rejected by
most geologists as unworkable (contrast Lyell 1840, 1841; Hitch-
cock 1841). Thereafter, glacial theory would languish until a new
generation of Scottish geologists undertook systematic surveys
that provided (to their own surprise) overwhelming evidence that
glacial theory was correct after all (Ramsay 1860; Jamieson
1862, 1863, 1865; A. Geikie 1863, 1865; J. Geikie 1874).
Agassiz had to wait 25 years for vindication. Had he lived, Jens
Esmark would have been 99 years old.
Lyell, who had ‘an undeniable penchant for the skilful defence
of lost causes’ (Cunningham 1990, p. 247), never completely gave
up on his decidedly non-uniformitarian iceberg-drift hypothesis.
Conversely, it is insufficiently appreciated today that many of
the criteria still used to distinguish ancient glaciogenic deposits
were clearly spelled out over 170 years ago by Esmark, Dobson,
Venetz, Bernhardi, Charpentier and Agassiz.
The problem of bidirectional climate change and the
discovery of the ‘greenhouse’ effect
The idea of a glacial period, or ‘Ice Age’ (Schimper 1837), was not
controversial to geologists per se: characteristic fauna and flora
now live c. 1000 km farther north of their occurrences in the
‘Drift’ (Smith 1836, 1839; Forbes 1846). However, to physicists
interested in climate, it came as a shock. They had assumed the
the Earth’s climate was slowly cooling over time due to the dissi-
pation of primordial heat in the Solar System according to the
nebular hypothesis. This was consistent with undisputed palaeobo-
tanical evidence (e.g. palm fronds in Switzerland) that most
regions were warmer in the Cretaceous and early Cenozoic than
they are today. If the climate could change dramatically in either
direction, previously unknown factors must be at work in the
climate system.
It fell to the Irish physicist, mountaineer and orator John
Tyndall (1861, 1863) to demonstrate experimentally that certain
gaseous molecules in the atmosphere, notably water vapour and
carbon dioxide (CO2), absorb infrared radiation (‘obscure rays’)
emitted by the Earth but are transparent to sunlight. Fourier
(1824) had previously mentioned the possibility of such a ‘green-
house’ effect, but had rejected it as quantitatively unimportant. For
Tyndall (1863, p. 204), water vapour provides ‘a blanket more
necessary to the vegetable life of England than clothing is to
man’. Radiative energy balance is maintained because the atmos-
phere ‘constitutes a local dam, by which the temperature of
the Earth’s surface is deepened: the dam, however, finally over-
flows, and we give to space all that we receive from the Sun’
(Tyndall 1863, p. 205).
Tyndall understood that water vapour cannot be the ultimate
cause of climate change, because its concentration is itself a
strong function of temperature. It amplifies any given climate
change but cannot be its ultimate cause. On the other hand, the
content of CO2 in the atmosphere, or any of the hydrocarbon
gases, can vary independently. ‘It is not, therefore, necessary to
assume alterations in the density and height of the atmosphere to
account for different amounts of heat being preserved to the
Earth at different times’, wrote Tyndall (1861, p. 277), ‘a slight
change in its variable constituents would suffice for this.
Such changes may in fact have produced all the mutations
of climate which the researches of geologists have revealed’
[italics added].
Unknown to Tyndall, a prematurely deceased French ceramicist
had already established the basis for the changes in CO2 that he
required, through the geochemical cycle of carbon (Ebelmen
1845, 1847; see also Hunt 1880; Berner & Maasch 1996).
1871 –1908: pioneering discoveries
Following the first identification of glaciogenic deposits now
known to be Neoproterozoic in age in Scotland (Thomson 1871),
similar discoveries were made in Australia (Woodward 1884),
Norway (Reusch 1891), Svalbard (Garwood & Gregory 1898),
South China (Willis 1904) and India (Holland 1908). The deposits
in the extreme NE of Norway made the biggest initial impact. Hans
Reusch (1852– 1922), then Director of the Norwegian Geological
Survey, found them around the head of Varangerfjorden, not far
from the Russian border. Reusch (1891) describes hills underlain
by up to 50 m of non-stratified, matrix-supported conglomerate
in which cobbles of Archaean gneiss and granite predominate,
but which also carry smaller clasts of dolomite with facets and non-
parallel striations identical to those found in Quaternary tills
throughout Norway. On the coast of the fjord near the lighthouse
at Bigganjargga, about 40 minutes’ walk east from Karlebotn,
Reusch (1891) found an isolated ridge of diamictite c. 70 m
long  8 m wide  3 m high (Bjørlykke 1967; Edwards 1975).
The ridge belongs to the older (Smalfjord Formation) of two gla-
ciogenic intervals in the region, and is onlapped and draped by
shallow-marine sandstone. Just above the waterline of the fjord,
the diamictite has eroded away to reveal a glacial pavement on
crossbedded quartzite carrying glacial striations in two main orien-
tations (Fig. 2.3). The less distinct (east – west) set of striations
preserves lateral ridges and streaks of cataclasite, which suggest
that the overlying diamictite is the melt-out tillite of a stagnant
 HISTORY OF NEOPROTEROZOIC GLACIAL GEOLOGY
ice-cored moraine (Edwards 1975). ‘Reusch’s Moraine’ has been
restudied by generations of geologists (Strahan 1897; Schiøtz
1898; Dal 1900; Holtedahl 1918; Rosendahl 1931, 1945; Føyn
1937; von Gaertner 1943; Crowell 1964; Reading & Walker
1966; Bjørlykke 1967; Edwards 1975, 1997; Jensen & Wulff-
Pedersen 1996; Rice & Hofmann 2000; Bestmann et al. 2006;
Arnaud 2008) and, despite differences in interpretation, it
remains an icon of Neoproterozoic glacial geology.
Although they were not the first to recognize glacial action in
‘Cambrian’ (later Adelaidean) strata of South Australia, Walter
Howchin in Adelaide and Edgeworth David in Sydney combined
to bring its extent and importance to wide attention (Cooper
2009). Howchin (1901, 1903, 1908) traced the older of two
glaciogenic intervals from its type section in the Sturt River
gorge near Adelaide over an area of nearly 800  400 km2
(Sprigg 1986), suggesting that ‘the ice gathered on a plateau of
comparatively low relief’ (David 1907b). The younger (Marinoan)
glaciogenic interval was first recognized by Jack (1913; see also
Mawson 1949b; Preiss 1987). At first, Howchin’s glacial inter-
pretation was greeted skeptically in Australia, but his convincing
photographs of striated clasts (Howchin 1908) were accepted
without dissent by the Geological Society in London. If
Howchin was responsible for much of the legwork in South
Australia, David was intrumental in getting the word out overseas
(Cooper 2009). Reports of the 10th IGC (Mexico City,
1906) include his 46-page synthesis (David 1907a), as well as
two shorter papers (one in French) on pre-Cenozoic glacial
epochs globally, with emphasis on the Australian glacial record.
He also visited India, Great Britain and the USA during his 1906
travels and doubtless raised the awareness of ‘Cambrian’
glaciation.
In the lower gorges of the Yangtze River in western Hubei
Province, South China, Bailey Willis and Eliot Blackwelder
described a body of ‘till’ of probable ‘Lower Cambrian’ age that
carries heterogeneous boulders exhibiting polish and striae of
‘unquestionably glacial’ origin, based on examination of speci-
mens by Quaternary glacial geologist T.C. Chamberlin (Willis
1904). The Nantuo tillite (Lee & Chao 1924) extends for
1200 km from NE to SW across the South China platform (Lee
1936; Lee & Lee 1940) and represents the younger of two discrete
Cryogenian glaciations in the region (Wang et al. 1981). Three
years later, a similar and likely correlative body (Jiang et al.
2003) was described in the Blaini Formation of the Lesser
Himalaya, northern India (Holland 1908).
1909 – 1941: period of globalization
The number of palaeocontinents subject to glaciation in what is
now known to be Neoproterozoic time grew from five in 1908 to
fourteen in 1937 (Fig. 2.1, Table 2.1), with new finds in eastern
and western North America, central and northern Asia, eastern
South America, and southern, central and western Africa. By the
time of the 17th IGC in Moscow in 1937, Neoproterozoic glacia-
tion was known on every continent except Antarctica.
In 1913, a boulder slate at Big Cottonwood Canyon in the
Wasatch Mountains of Utah was interpreted as glacial in origin
(Hintze 1913; see also Blackwelder 1932). Earlier, the Squantum
‘Tillite’ member of the Roxbury Conglomerate in the Boston
Basin of eastern Massachusetts was interpreted as glaciogenic
(Sayles & LaForge 1910; Sayles 1914). This unit has a checquered
history. As a result of miscorrelation, it was long considered
to represent the only Carboniferous glaciation in the northern
hemisphere. Its glacial origin was repeatedly questioned (e.g.
Dott 1961) and later its age. Only after 1980 was its mid-Ediacaran
age recognized through geological mapping, geochronology
and palynology (Lenk et al. 1982; Thompson & Bowring
2000). In Ediacaran time, the Boston Basin was part of the
ribbon continent Avalonia –Cadomia, located off North Africa. It
21
did not become part of Laurentia until the late Silurian (Wilson
1966).
In southern Africa, glaciogenic diamictites of Neoproterozoic
age were first recognized in the Numees Formation of northwes-
ternmost South Africa and adjacent Namibia (Rogers 1915).
‘Anyone familiar with the southern Dwyka [Tillite] who was
suddenly put down on the Numees beds would feel sure he was
on the Dwyka again’ comments Rogers (1915, p. 89). In fact,
distal facies of the Dwyka (Carboniferous) rest unconformably
on the Numees, ‘which resembles the Dwyka tillite of the southern
Karoo more closely than the Dwyka beds in the same neighbour-
hood do’ (Rogers 1915, p. 90).
There were no new discoveries between 1915 and 1929 – who
knows how many careers were cut down in World War I? – but
four book-length syntheses of ancient climates appeared in rapid
succession (Brooks 1922, 1926; Köppen & Wegener 1924;
Coleman 1926). In the more theoretical of his books, the English
meteorologist Brooks (1926) explicitly recognized and attempted
to rationalize the existence of two distinct climate states,
‘glacial’ (eo-Cambrian, late Palaeozoic, late Cenozoic) and ‘non-
glacial’ (most of Phanerozoic time). His names are preferable to
the later terms ‘greenhouse’ and ‘icehouse’ used synonymously.
Köppen & Wegener (1924) is justly famous – despite having
never been translated into English – for its detailed treatment of
Phanerozoic palaeoclimates in terms of continental drift, as well
as for inserting the first of Milankovic’s graphs of northern-
hemisphere summer insolation calculated for different latitudes
over the past 650 000 years. It devotes three pages to the Neopro-
terozoic (‘Kambrium’) glacial record. Coleman (1926), who had
earlier proposed a glacial origin for the ‘slate conglomerates’ of
Huronian age (early Palaeoproterozoic) in Ontario, Canada,
when they were thought to be the oldest known sedimentary
rocks (Coleman 1907), devotes a full chapter to glaciation ‘at
or just before the beginning of the Cambrian’ (Coleman 1926).
He concludes that the eo-Cambrian glaciation was more extreme
than the Pleistocene and possibly equal to the composite late
Palaeozoic glaciations, which were known to be diachronous,
younging from west to east (Du Toit 1922). The same conclusion
was reached by Kulling (1934), who updated the global
picture (Table 2.1) at the end of a detailed account of Neoprotero-
zoic stratigraphy in the NE of the Svalbard Archipelago of
Arctic Europe.
The pace of new discoveries picked up again from 1929 to
1937 with the recognition of glaciogenic formations in the
Katangan Series of southern Zaire (Beetz 1929), beneath the
Bambuı́ Group on the São Francisco craton of central Brazil (de
Moraes Rego 1930; see also Isotta et al. 1969), on the Yenisey
Ridge at the western margin of the Siberian craton (Nikolaev
1930), in the Taoudeni Basin of the West African craton (Baud
1933; Furon 1933) and in the central Tien Shan of NW China
(Norin 1937).
A major symposium on Palaeozoic and Pre-Cambrian climates
was held at the 17th IGC in Moscow in 1937 (published in 1940
as Report 6). Ten substantial papers describe Neoproterozoic
glacial deposits in different parts of Asia, Africa, Australia,
Europe and North America. Highlights include first-hand accounts
of various pre-Dwyka glacial deposits throughout southern Africa
(Gevers & Beetz 1940), the vast extent of the Nantuo Tillite in
South China (Lee & Lee 1940) and the existence of glacial deposits
in central Africa close to the equator (Davies 1940; Robert 1940).
A consensus emerged that ‘eo-Cambrian’ glaciation was matched
only in the late Palaeozoic, and that invertebrate animals arose fol-
lowing retreat of the ice sheets.
However, the rejection of continental drift in the late 1920s
following intense debate (Newman 1995; Oreskes 1999) had a
chilling effect on pre-Pleistocene palaeoclimate research. If late
Palaeozoic glaciation was allowed to remain ‘a hopeless riddle’,
as Wegener put it, what chance was there to understand glacial
periods twice their age?
22 P. F. HOFFMAN
1942– 1964: rebutting challenges
The consensus that emerged from the Moscow Congress of 1937
carried over initially into the postwar period. In a 1948 Royal
Society lecture in Sydney, Australia, the normally reserved Sir
Douglas Mawson (Sprigg 1986) proclaimed that ‘the world must
then have experienced its greatest Ice Age’ and speculated that
prolonged genetic isolation during the period of refrigeration con-
tributed to the subsequent Cambrian radiation of life (Mawson
1949a). However, Mawson was a fixist and therefore unduly
impressed by the evidence for glaciation in equatorial Africa
(Beetz 1929; Davies 1939, 1940).
But what were mobilists to do? Plotting ‘infra-Cambrian’
glacial deposits on Pangaea reconstructions, assuming them to
be closer to reality than the present geography, still left some
deposits close to the palaeoequator (cf. Harland 1964a, b;
Harland & Rudwick 1964). On the other hand, if the continents
were tightly clustered, they might conceivably have crossed the
poles diachronously in infra-Cambrian time due to polar wander-
ing and/or continental drift. In that case, the extent of glacial
deposits could far exceed the extent of ice at any time. What was
needed were means of determining the palaeolatitudes of individ-
ual deposits, so that the ice extent inferred would not depend
on correlation.
Palaeomagnetism and the meridional extent of
Neoproterozoic ice sheets
In the early 1950s, statistical corroboration of the geocentric –
axial –dipole hypothesis by Jan Hospers (Frankel 1987; Irving
2008) had opened the way for palaeomagnetic and palaeoclimatic
testing of polar wandering and continental drift (Irving 1956, 1959;
Runcorn 1961). Wegener and Du Toit were soon vindicated, drift
was unavoidable, and the revolution in the interpretation of ocean
basins followed in 1962–1967. Palaeomagnetism offered a means
by which the palaeolatitudes of ancient glacial deposits could be
determined. Irving (1957a) had shown that the average magnetic
inclination in Carboniferous glacial varves in Australia is subver-
tical, implying high-latitude deposition, consistent with Wegener’s
(1912, 1929) assumption based on their glacial origin. Preservation
of primary (depositional) natural remnant magnetization in red
clastic sediments of Precambrian age having been demonstrated
(Irving 1957b; Irving & Runcorn 1957), it fell to Harland &
Bidgood (1959) to first attempt this method in infra-Cambrian
glacial deposits. Their preliminary results (from a small subset
of samples) implied equatorial palaeolatitudes for glaciogenic for-
mations in southern Norway and East Greenland. Harland’s
palaeomagnetic work lacked the sophistication and reach of Run-
corn’s group three miles to the west (there was little contact
between the two Cambridge University laboratories) and it
would be 27 years before a truly reliable palaeomagnetic result
was obtained for a Neoproterozoic periglacial marine formation
(Embleton & Williams 1986).
Palaeomagnetists have long been interested in the palaeolati-
tudes of climate-sensitive sedimentary facies (e.g. thick carbon-
ates, evaporites, redbeds, aeolianites, tillites) as a test of the
hypothesis that the time-averaged geomagnetic field closely
approximated a geocentric axial dipole over Phanerozoic and
Proterozoic times (Blackett 1961; Opdyke 1962; Briden & Irving
1964; Briden 1970; Evans 2006). Thick shallow-water carbonates
are today found only equatorwards of 358 latitude (Rodgers 1957)
and their zonal distribution did not change between glacial and
non-glacial periods of the Phanerozoic (Briden 1970; Ziegler
et al. 1984; Opdyke & Wilkinson 1990; Witzke 1990; Kiessling
2001). There are two reasons for this. First, the saturation state
of seawater with respect to calcium carbonate or dolomite varies
with the relative, not the absolute, temperature. Second, in
glacial periods, higher alkalinity fluxes due to glacial erosion
in high latitudes are compensated by reduced weathering rates in
the cooler tropics. The occurrence of glacial marine strata directly
above or sandwiched between thick carbonate successions in
different areas (e.g. Harland & Wilson 1956; Ziegler 1960; Katz
1961; Martin 1965a) suggested that, unless the glaciogenic and
carbonate strata are everywhere separated by stratigraphic gaps
of large magnitude, tidewater glaciers existed in the warmest
parts of the surface ocean, consistent with nascent palaeomagnetic
evidence for low palaeolatitudes (Girdler 1964). If glaciers
occurred in relatively warm areas, including marine platforms
where no mountains existed, higher latitudes and elevations must
also have been glaciated. One is compelled to infer glaciation at
all palaeolatitudes, independent of any assumption with respect
to correlation (Harland 1964a, b). This is the crux of the Neopro-
terozoic glacial conundrum.
The Newcastle and Köln palaeoclimate conferences
of 1963 – 1964
The spectre of glaciation at low palaeolatitudes caused some to
question the glacial origin of Neoproterozoic ‘pebbly mudstones’
(Crowell 1957; Schermerhorn & Stanton 1963; Schermerhorn
1974). Since the recognition of turbidity current as a cause of
graded bedding (Kuenen & Migliorini 1950; Natland & Kuenen
1951), mass-flows loomed as alternatives to glaciers in some
areas. This would not, however, account for the disproportionate
occurrence of diamictites in late Neoproterozoic times. Martin
et al. (1985) suggested that mass-flows at low palaeolatitudes
might have been triggered by glacioeustatic changes driven by
high-latitude glaciation, but late Palaeozoic glaciation did not
result in low-latitude diamictites comparable in setting and
extent to those of the late Neoproterozoic.
These were among the issues discussed at a pair of palaeocli-
mate conferences that proved to be a watershed for Neoproterozoic
glacial geology (Fig. 2.2). The NATO Palaeoclimates Conference
held at Newcastle-upon-Tyne in January 1963 was the first to high-
light the topic since the Moscow symposium in 1937. Girdler
(1964) reviewed the palaeolatitudes of continents during pre-
Mesozoic glacial periods according to existing palaeomagnetic
data. The Permo-Carboniferous and Huronian (early Palaeoproter-
ozoic) glaciations occurred at high palaeolatitudes, affirming the
data, but middle and low palaeolatitudes were registered for the
eo-Cambrian glaciations. There were three possible explanations
(Girdler 1964): the (eo-Cambrian) palaeomagnetic data were
faulty, the rocks are not glacial in origin, or the whole globe was
glaciated. An entire session was devoted to the recognition of
glacial sediments and of till-like deposits of non-glacial origin
(Schwarzbach 1964a; Crowell 1964; Heezen & Hollister 1964).
In the Discussion, geologist Wally Pitcher and others pushed
back against non-glacial interpretations, citing detailed strati-
graphic studies on the Port Askaig boulder beds (Kilburn et al.
1965). Harland (1964a) provided a global synthesis of Precam-
brian glacial deposits and their stratigraphic settings, concluding
that infra-Cambrian glaciation had extended into the marine
environment at low palaeolatitudes, unlike any younger ice age.
An early tectonic ‘mobilist’, his views were informed by extensive
fieldwork on the diamictite-bearing Hecla Hoek succession in
Svalbard (e.g. Harland & Wilson 1956; Wilson & Harland
1964). He urged that such a glaciation provided an exceptional
basis for global temporal correlation. Rudwick (1964) discussed
the glacial aftermath as an ecological cradle for the Cambrian radi-
ation. Four months after the Newcastle meeting, another global
synthesis of Palaeozoic and Precambrian glaciation appeared,
graced with fine images of Neoproterozoic glacial debris from
central Africa (Cahen 1963).
In March 1964, Manfred Schwarzbach convened an inter-
national symposium on palaeoclimates in Köln, simultaneous
with the publication in English translation of his book Climates
 HISTORY OF NEOPROTEROZOIC GLACIAL GEOLOGY
of the Past (Schwarzbach 1964b). The book has a chapter on
eo-Cambrian glaciation, the subject of 7 of 37 papers from the
Köln symposium published in the Geologische Rundschau.
Harland (1964b) again summarized the global infra-Cambrian
glacial record, and Chumakov (1964, 1992) reviewed the
growing number of Precambrian tilloids in the (European)
USSR. Spjeldnaes (1964, p. 38) affirmed the glacial origin of
strata in different parts of Norway and Svalbard, noting that the
eo-Cambrian deposits had been studied ‘with the same methods,
in the same regions and by the same scientists as have the more
easily interpreted Quaternary deposits’.
Perhaps the defining summation of the infra-Cambrian glacial
problem from this period is Harland & Rudwick (1964) in the
popular science magazine The Scientific American – which is
still worth reading.
1965 – 1981: in the wake of the plate tectonic revolution
Plate tectonics revolutionized the Earth sciences, and Neoprotero-
zoic glacial geology was no exception. It freed the distribution of
ancient glacial deposits from the assumption of fixed relative
positions of the continents, as foreseen by Harland (1964a) and
(earlier) Wegener (1912), just as the concept of polar wandering
(Gold 1955) had freed them from fixed position with respect to
the poles. Equally important, plate tectonics provided a basis for
making sense of geochemical cycles, and thereby to understand
the controls on and changes to the chemical compositions of the
oceans and atmosphere (Siever 1968). This opened up the possi-
bility of understanding secular changes in atmospheric CO2 con-
centration – the dream of Ebelmen (1845, 1847), Tyndall (1861,
1863) and Chamberlin (1898, 1899) – and consequent changes
in climate through ‘greenhouse’ radiative forcing. This led to
renewed interest during the period 1965 –1981 in certain chemical
sediments characteristically associated with Neoproterozoic gla-
ciogenic deposits (banded iron- and manganese-formations, and
post-glacial ‘cap’ carbonates). Finally, the period saw the publi-
cation of an influential textbook (Sugden & John 1976) that did
much to invigorate the study of processes of glacial erosion, sedi-
mentation and landscape development.
Centenary of Thomson’s (1891) study of the Port
Askaig Tillite
There could have been no finer centennial celebration than the pub-
lication of Anthony (Tony) Spencer’s Geological Society Memoir
on Late Pre-Cambrian Glaciation in Scotland (Spencer 1971). It
remains the gold standard in Neoproterozoic glacial sedimentol-
ogy. The centenary was also marked by papers symbolizing the
disunity over the hot-button issue of diachronous or synchronous
glaciation (Crawford & Daily 1971; Dunn et al. 1971). If the con-
tinents were tightly clustered, it was conceivable that continental
drift or polar wandering could move them all at different times
through the polar region, as was the case for Gondwanaland in
the Palaeozoic (Du Toit 1922; Wegener 1929; Crowell & Frakes
1970). However, drift alone could not account for Palaeozoic gla-
ciations, because there were long stretches of Palaeozoic time
when Gondwanaland was situated over the South Pole yet no
large ice sheets existed.
Associated chemical sediments: banded iron- and
manganese-formations and cap carbonates
Sedimentary iron- and manganese-formations (BIFs) were the
focus of worldwide mineral exploration for post-World War II
reconstruction. Major deposits in SW Brazil (Dorr 1945;
Almeida 1946; Walde et al. 1981), NW Canada (Ziegler 1960;
23
Young 1976; Yeo 1981), Namibia (Martin 1965a) and Australia
(Whitten 1970) turned out to be intimately associated with Cryo-
genian glaciogenic diamictites. None is closely associated with
volcanics. They are the only large-scale BIFs in the stratigraphic
record younger than the Palaeoproterozoic (1.9 Ga) deposits
around the Superior craton of North America. Martin (1965b)
suggested that ‘this peculiar combination of sediments’ might be
attributed to ‘oxygen deficiency in stagnating bottom waters
caused by an ice cover’ (p. 116).
The presence of a generally rather thin but remarkably continu-
ous layer of carbonate, directly above Neoproterozoic diamictites,
was occasionally noted by early workers (David 1907b; Norin
1937; Robert 1940; Mawson 1949b). These descriptions now
become more specific, noting that the ‘capping dolomite’ is a dis-
tinctive pink or less commonly cream colour, invariably laminated
or thin-bedded, and persists over basement highs where the dia-
mictite cuts out (Dow 1965; see also Biju-Duval & Gariel 1969;
Dunn et al. 1971; Rankama 1973; Deynoux & Trompette 1976;
Plummer 1978; Williams 1979). Williams (1979) was the first to
report stable isotope values (measured by Karlis Muehlenbachs)
for ‘cap dolostones’ (and the first to use the term), from the Kim-
berleys of Western Australia. He notes that their d18O is similar to
other Neoproterozoic dolostones, but their d13C is moderately
depleted. Assuming the dolomite to be primary or early diagenetic,
he concludes that ‘abrupt climatic warming at the close of late
Precambrian glacial epochs is implied’, consistent with the d18O
data (Williams 1979).
Climate models and the ‘white Earth’ instability
The geological literature in this period was fixated on the use or
misuse of glaciation for the division of Neoproterozoic time (e.g.
Harland 1964a; Crawford & Daily 1971; Dunn et al. 1971;
Rankama 1973; Schermerhorn 1977). The cause of glaciation
was barely mentioned – Harland (1964a) suggested diminished
solar forcing and Williams (1972, 1974, 1975) appealed to large
variations in the Earth’s orbital obliquity (see below).
Climate physicists, however, made a startling discovery when
they used radiative enery-balance equations to calculate surface
temperatures as a function of latitude and variable solar forcing,
with simple parameterizations of meridional heat transport and
feedbacks due to snow, ice and clouds. When they reduced solar
irradiance by a few percent, surface temperatures fell below
freezing everywhere due to runaway snow and ice-albedo feed-
back (Eriksson 1968; Sellers 1969, 1990; Budyko 1969; see also
North 1990). Although decadal to millenial solar variability is
only c. 0.1%, solar luminosity is thought to have slowly increased
by 25–30% since 4.5 Ga due to the production of helium in the
Sun’s core. The energy-balance calculations were therefore
assumed to be in error because they predicted that the Earth
should have been permanently frozen: irradiance roughly 25%
higher than present is required to overcome the albedo of an ice-
covered planet, which reflects over 60% of the sunlight it receives.
This problem was a stimulus for the nascent science of climate
modelling. However, the ‘white Earth’ instability (Wetherald &
Manabe 1975) turned out to be a robust feature not only of
simple energy-balance models but also of most atmospheric
general-circulation models.
In 1981, three planetary scientists proposed a solution to the
‘white Earth’ problem (Walker et al. 1981). They appealed to
the geochemical cycle of carbon (which is not accommodated in
physical climate models because of its timescale of c. 106 years).
CO2 is supplied to the ocean and atmosphere by metamorphic-
volcanic outgassing and is consumed by silicate rock weathering.
The latter is temperature-dependent because of the direct effect
of temperature on reaction kinetics and also because moisture
in wet regions, where most weathering occurs, increases
with temperature (Clausius – Clapeyron relationship). The
24 P. F. HOFFMAN
temperature-dependence of weathering provides a negative
climate feedback (abiotic Gaia) that acts as a natural thermostat.
If global temperatures rise (or fall) for any reason, CO2 is con-
sumed at a faster (or slower) rate, thereby limiting the temperature
change. In effect, the feedback slowly adjusts the atmospheric CO2
concentration to maintain a balance between CO2 sources and
sinks. As solar luminosity increased over geological time, atmos-
pheric CO2 concentration adjusted downwards, keeping surface
temperatures within the range suitable for life. This simple hypoth-
esis not only went a long way towards solving the ‘faint young
Sun’ paradox (Sagan & Mullen 1972), it rationalized long-term
climate change while explaining why such changes were limited
in magnitude.
In the penultimate paragraph of their classic paper, Walker et al.
(1981) suggested that silicate weathering feedback had been
responsible for averting a ‘white Earth’. They added, however,
that ‘if a global glaciation were to occur, the rate of silicate
weathering should fall nearly to zero, and carbon dioxide should
accumulate in the atmosphere at whatever rate it is released
from volcanoes. Even the present rate of release would yield 1 bar
of carbon dioxide in only 20 Ma. The resultant large greenhouse
effect should melt the ice cover in a geologically short period of
time’ (Walker et al. 1981). If a ‘white Earth’ did occur, it would
be self-reversing. Accordingly, its occurrence in deep time could
not be ruled out a priori. This, in a nutshell, is the climatological
concept behind the ‘Snowball Earth’ hypothesis (Kirschvink 1992).
Walker et al. (1981) made no reference to ancient glaciation,
Neoproterozoic or otherwise. This was not because they were
unaware of geology. On the contrary, the ‘faint young Sun’
problem was brought into focus for Walker by his participation
in the Precambrian Palaeobiology Research Group (PPRG)
organized by micropalaeontologist J. William Schopf. (Third
author Kasting and palaeomagnetist Joe Kirschvink were PPRG
participants later in the decade.) By 1981, Harland’s great infra-
Cambrian glaciation had fallen off the radar screen. Palaeo-
magnetism, on which high hopes were pinned, had encountered
problems. The foremost was the susceptibility of sediments to low-
temperature chemical remagnetization (e.g. McCabe & Elmore,
1989). Overcoming these problems would require time-consuming
stepwise chemical and thermal demagnetization, studies of mag-
netic mineralogy, ‘field tests’ to constrain the age of remnant
magnetic components, and more sensitive magnetometers.
Earth’s Pre-Pleistocene Glacial Record volume
In 1981, Cambridge University Press published Earth’s Pre-
Pleistocene Glacial Record (Hambrey & Harland 1981), a
product of Project 38 (Pre-Pleistocene Tillite Project) of the Inter-
national Geological Correlation Programme. This sought-after
volume contains 58 unified descriptions of Neoproterozoic glacio-
genic formations worldwide, and is a tribute to Harland’s vision
and Hambrey’s dedication. It was synthesized in Harland (1983)
and Hambrey & Harland (1985). An additional global synthesis
of Neoproterozoic glaciogenic deposits was published by Chuma-
kov (1981).
1982– 1997: the gathering storm
There were many developments during this period – burgeoning
information regarding glaciomarine processes and deposits,
widespread use of carbon isotopes as a tool for correlation,
acquisition of reliable palaeomagnetic constraints on palaeolati-
tudes of proximal glaciomarine deposits, increased awareness of
the sedimentological peculiarites of post-glacial ‘cap’ carbonates,
and growing interest in causal mechanisms for low-latitude
glaciation.
Glaciomarine deposits are more likely than their terrestrial
counterparts to be preserved in the geological record because
they are less susceptible to erosion, but ice-proximal processes
in the marine environment are difficult to study because they
occur underwater and often under ice. Nevertheless, much has
been gleaned from geomorphic and seismic surveys of polar con-
tinental shelves backed up by drilling programmes (e.g. Elverhøi
1984; Powell 1984, 1990; Alley et al. 1989; Barrett 1989;
Boulton 1990; King et al. 1991; Powell & Domack 1995). A
growing emphasis on glaciomarine processes and deposits is
reflected in many books written during this period (e.g. Molnia
1983; Drewry 1986; Dowdeswell & Scourse 1990; Anderson &
Ashley 1991; Hambrey 1994; Benn & Evans 1998; Anderson
1999). Neoproterozoic glacial geology benefited enormously
from these developments, although the complexity of facies archi-
tecture poses a challenge given the limited three-dimensional
exposure of ancient deposits and weak or non-existent constraints
on rates of accumulation.
Beginning in the mid-1950s, oxygen isotopes have been used as
a basic tool for stratigraphic correlation of Quaternary marine
carbonates and for estimating seawater temperatures and global
ice-sheet volumes. Unfortunately, Neoproterozoic carbonates do
not preserve their original mineralogy, and their oxygen isotope
compositions are seriously compromised by aqueous diagenesis.
However, carbonate rocks buffer the isotopic composition of
the relatively low concentration of carbon in aqueous solution.
Consequently, the carbon isotopic compositions of carbonate
rocks resist alteration. Beginning in the mid-1980s, carbon iso-
topes were increasingly used for stratigraphic correlation of Neo-
proterozoic carbonates and to a lesser extent for organic-rich
shales (Knoll et al. 1986; Knoll & Walter 1992; Kaufman et al.
1997). Studies show that Neoproterozoic glaciogenic formations
are typically bracketed by negative isotopic excursions. The excur-
sions are reproduceable and approximately five times larger in
magnitude than spatial or depth variations in the modern ocean.
Because atmospheric CO2 concentrations were likely higher in
the Neoproterozoic than today because of the dimmer Sun, a
larger reservoir of dissolved inorganic carbon should have
damped spatial and depth variations of d13C in Neoproterozoic
oceans. Therefore, the observed isotopic excursions are most
easily explained as secular variations. Their regular occurrence
directly before and after glaciogenic sedimentation suggests that
glaciation was roughly synchronous in many areas, even if not
every isotopic excursion is associated with glaciation. For those
who accepted these basic principles, the arguments for diachro-
nous glaciation receded as the body of Neoproterozoic carbon
isotope data grew.
The first robust palaeomagnetic support for glaciation at sea
level in low palaeolatitudes came from the Elatina Formation in
South Australia (Embleton & Williams 1986). The formation is
a glaciomarine unit comprised of reddish sandstone, siltstone
and diamictite (Lemon & Gostin 1990; Williams et al. 2008),
and a 10-m-thick interval of rhythmically laminated siltstone
has been the subject of repeated palaeomagnetic studies. The
palaeopole determined by Embleton & Williams (1986) implies
a palaeolatitude of c. 58 and, although no field test was performed,
the result is considered reliable because the natural remanent
magnetization is stably carried by detrital haematite. At the time
of the study, the laminations were considered to be annual
varves, and the rhythmic bundles of 12–14 laminae were
thought to represent sunspot cycles (Williams 1981; Williams &
Sonett 1985). Later, they were reinterpreted as tidal rhythmites
(Williams 1989), supporting a glaciomarine origin for the associ-
ated diamictites. Field tests were subsequently carried out using
soft-sediment folds (Schmidt et al. 1991; Schmidt & Willliams
1995) and polarity reversals (Sohl et al. 1999), which confirmed
the primary nature of the remanent magnetization and the palaeo-
latitude of 7.58 or less. A somewhat larger palaeolatitude of c. 158
was obtained in a recent palaeomagnetic study (Raub & Evans
 HISTORY OF NEOPROTEROZOIC GLACIAL GEOLOGY 25

2006) of the cap dolostone (Nuccaleena Formation) above the
Elatina Formation, implying a certain amount of inclination flat-
tening during compaction of the Elatina siltstone, but this hardly
alters the thrust of the result because temperatures are broadly
uniform across the tropics.
Cap carbonates
Arguably the first person to fully appreciate that the basal Edia-
caran (Marinoan) cap carbonate was not only ‘an important strati-
graphic marker’ but also ‘a perplexing and paradoxical
lithostratigraphic unit’ (Table 2.2) was Aitken (1991), a leading
Rocky Mountain carbonate stratigrapher. Concerning the cap
dolostone atop the younger glaciogenic unit (Stelfox Member of
the Icebrook Formation) in the Mackenzie Mountains of NW
Canada, Aitken (1991) stressed that the aphanitic microcrystalline
dolostone was not deposited as fine-grained mud, but as silt and
sand-sized peloids. Ubiquitous subparallel lamination is defined
by oscillatory variation in peloid size. His photographs show
well-sorted, coarse-grained peloids in layers only one or two
peloids deep. Where coarse peloids form the base of a layer,
they commonly rest on facets, apparently resulting from abrasion.
The layers are not laterally continuous, and low-angle cross-
lamination is ubiquitous. The lamination is clearly mechanical in
origin, interspersed locally with microbially stabilized masses of
micropeloidal stromatolite (James et al. 2001).
Aitken (1991) argued that the intrastratal ‘tepees’ for which
the unit was informally named (Eisbacher 1981) are not true peri-
tidal tepees as defined by Assereto & Kendall (1977; see also
Kendall & Warren 1987), which result from layer-parallel expan-
sive growth associated with cementation under alternating vadose
and phreatic conditions. The cap ‘tepees’ are not associated with
breccias or void-filling cements, as are peritidal tepees, and in
plan view their crestlines are always linear and parallel, not poly-
gonal like true tepees (Assereto & Kendall 1977). Bundles of
laminae may exhibit onlap or offlap relations on ‘tepee’ flanks.
These metre-scale trochoidal bedforms were later analysed as
giant wave ripples (Allen & Hoffman 2005).
Large-scale stromatolites, within which are spaced, micrite-
or cement-filled tubes or gutters that maintain a palaeovertical
(‘geoplumb’) orientation irrespective of the inclination of lami-
nations in the host stromatolite, were described from cap dolo-
stones in California (Cloud et al. 1974; Wright et al. 1978),
Namibia (Hegenberger 1987) and later in Alaska, Brazil, Canada
and Mongolia (Fig. 2.3).
Aitken (1991) identified crystal fans and pseudo-stromatolites
in limestone overlying the cap dolostone, correctly interpreting
them as pseudomorphic after aragonite sea-floor cements. His
photograph of ‘blades of calcite’ is from a unique layer of sea-floor
barite (BaSO4) cement, variably calcitized, which occurs in the top
,10 cm of the cap dolostone for .150 km along a strike south-
eastward from the location of Aitken’s (1991) photograph. Thus,
cap dolostones and overlying limestones display a panoply of idio-
syncratic features (Fig. 2.3) that generally occur in a consistent
vertical sequence (Hoffman et al. 1987).
Aitken (1991) offered no genetic explanation for cap carbonates.
Eyles (1993) suggested that they represent detrital rock ‘flour’,
as proposed earlier for carbonate layers within glaciogenic
diamictites (Fairchild 1983). A detrital origin could hardly

Table 2.2. Idiosyncratic sedimentary and early diagenetic features in cap dolostones

Palaeocontinent Cap dolostone (ref.) SCC TPB DGB PEL LAC GWR TBS SFB
p p p p
Amazonia Mirassol d’Oeste (1) – – – –
Arabia Hadash (2) – – – – – – – –
p p p p
Arctic Alaska Nularvik (3) – – – –
p p p p p
Australia Mount Doreen (4) – – –
p p
Australia Nuccaleena (5) – – – – – –
p p p
Australia Cumberland Creek (6) – – – – –
p p
Australia Lower Ranford (7) – – – – – –
Baltica Lower Nyborg (8) – – – – – – – –
p p p p p
Congo Keilberg (9) – – –
Congo Calcaire Rose (10) – – – – – – – –
Congo C1 (11) – – – – – – – –
India Upper Blaini (12) – – – – – – – –
p p p p
Kalahari Dreigratberg (13) – – – –
p p p p p p
Laurentia Ravensthroat (14) – –
p p p p
Laurentia Noonday (15) – – – –
p
Laurentia Lower Canyon (16) – – – – – – –
p p p
Laurentia Lower Dracoisen (17) – – – – –
p p p
Laurentia Cranford (18) – – – – –
p p
Laurentia Hard Luck (19) – – – – – –
p p p p p
Mongolia Ol (20) – – –
p p
Mongolia Baxha (21) – – – – – –
p p p p
South China Lower Doushantuo (22) – – – –
p
Tarim Lower Zhamoketi (23) – – – – – – –
p p
West Africa Oued Djouf (24) – – – – – –
p p p
West Africa Amogjar (25) – – – – –
p p p p p
West Africa Mid Sud-Banboli (26) – – –

Abbreviations: SCC, sheet-crack cements; TPB, tepee breccia; DGB, diagenetic barite; PEL, peloids; LAC, low-angle cross-laminae; GWR, giant wave ripples; TBS, tube
biostrome; SFB, seafloor barite.
References: (1) Nogueira et al. (2003); (2) Allen et al. (2004); (3) Macdonald et al. (2009b); (4) Kennedy (1996); (5) Plummer (1978); (6) Calver & Walter (2000);
(7) Corkoran (2007); (8) Edwards (1984); (9) Hoffman et al. (2007); (10) Cahen & Lepersonne (1981); (11) Cahen (1950); (12) Kaufman et al. (2006); (13) Macdonald
et al. (2011); (14) James et al. (2001); (15) Corsetti & Grotzinger (2005); Corsetti & Kaufman (2005); Wright et al. (1978); (16) Hambrey & Spencer (1987); (17)
Halverson et al. (2004); (18) McCay et al. (2006); (19) F. A. Macdonald, pers. comm.; (20) Macdonald et al. (2009a); (21) F. A. Macdonald & D.S. Jones, pers. comm.;
(22) Jiang et al. (2006); (23) Xiao et al. (2004); (24) Bertrand-Sarfati et al. (1997); (25) Shields et al. (2007); (26) Nédélec et al. (2007).
26 P. F. HOFFMAN
account for the systematic variations in d13C observed in cap
dolostones (Kennedy et al. 1998; Hoffman et al. 2007).
Kaufman et al. (1993) and Grotzinger & Knoll (1995) related
cap carbonates to the overturn of alkalinity-charged, isotopically
depleted, bottom water after a lengthy period of ocean stagnation.
This suggestion requires that primary production be maintained
in the absence of upwellings, which is the overwhelmingly pre-
dominant nutrient flux in the modern ocean. Upwellings would
not be necessary if organic matter did not settle, but then alkalinity
would not build up in the deep nor would a vertical isotopic
gradient develop. This and the physical implausibility of pro-
longed (.10 000 years) ocean stagnation (in the absence of an
ice cover) make the overturn hypothesis unconvincing, despite
its popular appeal.
Kennedy (1996) proposed that cap carbonates represent non-
skeletal analogues of carbonates deposited (in low latitudes)
during times of Quaternary deglaciation according to the ‘coral
reef’ hypothesis (Berger 1982; Opdyke & Walker 1992). This
results from the hypsometry of ocean basins. When sea level is
lowered due to ice-sheet growth, the area of shallow shelves
and inland seas is disproprotionately reduced. These areas are
favoured for carbonate burial because the sediment never
encounters undersaturated deepwaters. When these areas are lost,
all carbonate production occurs in the open ocean and the
alkalinity of deepwaters increases. Upon deglaciation, alkalinity-
rich waters flood the shelves, depositing carbonate and
releasing CO2 to the atmosphere. The process can account for
much of the glacial –interglacial variation in pCO2 (Opdyke &
Walker 1992) and thus provides a positive climate feedback.
This mechanism must have contributed to cap carbonates, but
the estimated average thickness of carbonate produced (Ridgwell
et al. 2003) is only approximately one-tenth of the average
thickness of cap dolostones of 18.5 m (Hoffman et al. 2007,
Table 2.1). The coral reef hypothesis does not explain why cap car-
bonates are only associated with the ultimate deglaciation, and do
not accompany glacial –interglacial cycles within the glacial
period where these are recognized (Leather et al. 2002; Allen
et al. 2004). If cap carbonates were eroded during glacial read-
vances, they would be present as clasts in glacial deposits, which
they are not.
Nearly 70 papers concerning cap carbonates appeared in the
decade after 1997. This reflects the central but contentious role
they have assumed in the ongoing controversy over the nature of
Cryogenian glaciation. Arguably they are the most richly enig-
matic horizons in the entire stratigraphic record.
Causative theories for Neoproterozoic glaciation
Walker et al. (1981) provided the conceptual basis for a self-
reversing global glaciation from the point of the initial ‘white
Earth’ (ice-albedo) instability. They did not speculate on how
the climate ever reached that point because they credited silicate-
weathering feedback with its avoidance. However, a number
of causative theories (not mutually exclusive) were proposed
between 1971 and 1997 to explain Neoproterozoic glaciation
(Table 2.3).
Carbonate burial. Roberts (1971, 1976) was struck by the fact that
the Cryogenian glaciation was preceded by thick successions of
shallow-water carbonate strata in the North Atlantic (Greenland,
Svalbard and the British Isles), western North America and
central Africa. He postulated a period of anomalous carbonate
burial in which atmospheric CO2 was sequestered, producing an
‘anti-greenhouse’ effect. He did not give any reason why this
might have occurred. It would require a rise in the global rate of
silicate weathering relative to the rate of CO2 outgassing (see
below; Continental distribution and Continental break-up).
Table 2.3. Causative theories for Neoproterozoic glaciation
Astronomical theories
1. Solar variation Harland (1964a)
3. Large obliquity Williams (1972, 1975), Jenkins (2000),
Donnadieu et al. (2002)
5. Ice-ring collapses Sheldon (1984)
13. Impact ejecta Bendtsen & Bjerrum (2002), Fawcett &
Boslough (2002)
4. Giant molecular clouds Pavlov et al. (2005)
Oceanographic theories
2. Carbonate burial Roberts (1971)
7. Ocean stagnation Kaufman et al. (1993), Grotzinger & Knoll
(1995)
10. Hypsometric effect Kennedy (1996), Ridgwell et al. (2003)
11. Organic burial Kaufman et al. (1997)
Geodynamic theories
6. Continental Marshall et al. (1988), Worsley & Kidder
distribution (1991), Kirschvink (1992), Schrag et al.
(2002)
9. Continental break-up Eyles (1993), Young (1995a), Donnadieu et al.
(2004a)
15. Basalt weathering Goddéris et al. (2003)
16. True polar wander Li et al. (2004)
Biological theories
8. Biocatalysed Carver & Vardavas (1994), Lenton & Watson
weathering (2004), Kennedy et al. (2006)
12. Methane destruction Pavlov et al. (2000, 2003), Catling et al. (2001),
Claire et al. (2006)
14. Methane substitution Halverson et al. (2002), Schrag et al. (2002)
Large orbital obliquity. The Australian sedimentologist George
E. Williams always had an eye on ‘the big picture’ and possessed
an early appetite for planetary orbital mechanics. Impressed by the
prevalence of varves (i.e. rhythmic lamination, supposedly seaso-
nal) and other seasonal indicators such as polygonal sand wedges
associated with Neoproterozoic glaciogenic strata (Chumakov
1968; Spencer 1971), he proposed that the Earth’s obliquity (i.e.
the angle between the equatorial and ecliptic planes) oscillated
between large (e.g. Neoproterozoic) and small (e.g. Phanerozoic)
values (Williams 1972). When the angle was large, the seasonal
cycle was strong everywhere; when small, the seasons did not
differ greatly in low latitudes (except in strongly monsoonal
areas). In addition, as the angle increased, the meridional insola-
tion gradient declined and actually reversed (i.e. greater annual
insolation at the poles than at the equator) whenever the angle
exceeded 548. Williams (1972, 1974, 1975) inferred that when
the obliquity was very large, low latitudes were subject to glacia-
tion preferentially. As low latitudes cover more area than high lati-
tudes, there should be more ice (and higher planetary albedo)
during periods of low-latitude glaciation than when ice is limited
to the poles. Accordingly, glacial periods occur when obliquity
is large and when there is strong seasonality everywhere (contrary
to the Wegener – Milankovic hypothesis that ice-sheet growth is
favoured by cool summers and low seasonality). Under large obli-
quity, glaciation at the poles is precluded by intense summer-
time insolation.
The large obliquity hypothesis was strengthened by the discov-
ery of 2.5-m-deep sand wedges associated with the Elatina For-
mation (Williams & Tonkin 1985), which, combined with
palaeomagnetic evidence (Embleton & Williams 1986), indicates
strong seasonality at low palaeolatitude (Williams 1993; but see
Maloof et al. 2002). In the Phanerozoic, deep periglacial sand
wedges occur only in middle and high latitudes, where seasonal
forcing is strong (Lachenbruch 1962). With large obliquity, sea-
sonality is strong at all latitudes.
Like any theory, large obliquity was not without difficulties.
Although a large (and chaotic) obliquity could have been imparted
 HISTORY OF NEOPROTEROZOIC GLACIAL GEOLOGY
on the early Earth by stochastic accretion (Laskar & Robutel
1993), obliquity cannot oscillate between large and small angles
because of ‘entrapment’ at small angles by gravitational attraction
between the Moon and the Earth’s equatorial bulge (Laskar et al.
1993). Therefore, the theory was revised such that obliquity was
persistently large until Ediacaran times, when it was rapidly
reduced to small angles before the Cambrian (Williams 1993,
2000). This proved difficult to replicate in models or justify from
orbital theory (Néron de Surgy & Laskar 1997; Williams et al.
1998; Hoffman & Maloof 1999; Pais et al. 1999; Donnadieu
et al. 2002; Levrard & Laskar 2003). Moreover, large obliquity
over most of Precambrian time is at odds with a growing body
of palaeomagnetic evidence that pre-Ediacaran evaporites and
carbonate-dominated successions formed at latitudes inconsistent
with a reversed climatic gradient (Park 1994; Buchan et al.
2001; Evans 2006; Maloof et al. 2006). However, decisive falsifi-
cation of the large obliquity hypothesis, glaciation at high palaeo-
latitude, has yet to be observed.
Ice-ring collapses. Richard P. Sheldon was a sedimentary phos-
phorite specialist with the United States Geological Survey who
developed an ‘outlandish hypothesis’ (Sheldon 1984) for Neopro-
terozoic glaciation following a field excursion in northern Mongo-
lia, where phosphorites, carbonates and glacial marine strata occur
in close stratigraphic proximity. He speculated that ice rings denser
than those of Saturn orbited the Archaean Earth. In the Proterozoic,
sublimation thinned the rings and they became more discrete.
As the Moon pulled away from the Earth, their orbits slowly
decayed and they successively entered the atmosphere. As each
ring approached the Earth, its shadow caused an ice age. Upon
entering the atmosphere and before the next ring approached, the
shadow was lost and greenhouse gases were added, causing a
warm interglacial. Once the final ring had fallen, the tropics
were shadowless for the first time. Ocean overturning driven by
the steepened climatic gradient led to phosphogenesis and
organic diversification. Sheldon (1984) noted that seasonality
would be greatest in glacial times, consistent with Williams’
(1993) observation, because the rings cast shadows only on
the winter hemisphere. Again, this is contrary to the Wegener –
Milankovic theory of ice ages, which posits that ice sheets grow
when summers (not winters) are cool and seasonality is weak
(Köppen & Wegener 1924; Milankovic 1941).
Continental distribution. Atmospheric pCO2 will adjust downwards
in response to an equatorward migration of the continents, because
weathering rates are greatest in the tropics (Marshall et al. 1988;
Worsley & Kidder 1991). Global cooling is therefore consistent
with sedimentological and palaeomagnetic evidence for an
unusual preponderance of low-latitude continents in the Neopro-
terozoic, ‘a situation that has not been encountered at any sub-
sequent time in Earth history’ (Kirschvink 1992). Equatorward
displacement of the continents also changes the planetary
albedo: the tropical ocean is a strong absorber of sunlight, unlike
land areas (minimally vegetated in the Neoproterozoic) and the
fog-bound polar oceans, which are relatively good reflectors
(Kirschvink 1992). Moreover, a positive albedo feedback would
result from the enlargement of tropical land area that would
accompany any glacioeustatic fall in sea level (Kirschvink 1992).
A palaeomagnetist, Kirschvink (1992) accepted the low-latitude
result for the Elatina glaciation (Embleton & Williams 1986)
but rejected the large-obliquity hypothesis because of the co-
occurrence of Neoproterozoic glacial deposits and thick carbonate-
dominated successions. If the meridional climatic gradient were
reversed, the carbonate belts would move from low latitudes to
the poles, ‘where the glaciers (in Williams’ model) should not
encounter them’ (Kirschvink 1992). He was therefore compelled
to assume that if ice sheets reached sea level close to the
equator, higher latitudes must have been glaciated as well, wher-
ever ablation did not exceed precipitation. He envisaged an
27
ocean largely covered by pack ice, perhaps with ‘warm tropical
“puddles” in the sea of ice, shifting slightly from north to south
with the seasons’ (Kirschvink 1992). At the 1989 PPRG meeting
at UCLA (Maugh 1989), he proposed that the ‘white Earth’
disaster had actually occured in the Neoproterozoic, perhaps
repeatedly, and that each pan-glacial period was abruptly termi-
nated when the slow build up of CO2 reached a critical level, as
deduced by Walker et al. (1981). He attributed the deposition
of banded iron-formation during Neoproterozoic glaciation to
ocean stagnation and deepwater anoxia, as previously proposed
by Martin (1964b), and he predicted that the abrupt climate
switches accompanying glaciation and deglaciation should be
represented in widely separate areas by lithologically similar
strata, a result of the global scale of the climatic fluctuations
(Kirschvink 1992). He called the glacial state a ‘Snowball’ Earth
(Maugh 1989) to highlight the central role of planetary albedo in
the phenomenon. It is an evocative description of how the late
Cryogenian Earth might have looked from outer space.
Impact ejecta. Impact ejecta figured in discussions of Neoprotero-
zoic glaciation in two completely different ways. They were pro-
posed as an alternative emplacement mechanism for diamictite,
implying that the low-latitude and carbonate-associated glacial
problem was a chimera (Oberbeck et al. 1993; Rampino 1994).
This was a small reenactment of the pebbly mudflow challenge
of 1957 –1974. Impact ejecta aprons do in fact share many
diagnostic features with glaciogenic diamictites (e.g. unsorted
debris, faceted and striated erratic stones, outsized dropstones).
However, continuous ejecta blankets extend only about one crater-
radius from the crater rim, regardless of crater size (Melosh 1989).
Therefore, an improbably large flux of big impacts (roughly one
Cretaceous –Tertiary sized impact every 100 ka for 200 Ma)
would be required to account for the distribution of Neoproterozoic
diamictites, given the low probability of their exposure at the
Earth’s surface today.
It was later proposed on the basis of modelling experiments
that large impacts could have caused Neoproterozoic glaciation
(Bendtsen & Bjerrum 2002; Fawcett & Boslough 2002). The
models suggest that short-term (,10 year) shielding of sunlight
by ejecta from a Cretaceous –Tertiary sized impact could cause
the ocean surface to freeze over if seawater was as cold as today,
unlike the Cretaceous ocean, which was 8– 128 warmer than
present.
Ocean stagnation. As the patterns of C and Sr isotope variations
associated with Neoproterozoic and Phanerozoic glaciations are
distinct (Kaufman et al. 1993), the cause of glaciation might also
be different. Kaufman et al. (1993) and Grotzinger & Knoll
(1995) attempted to genetically link the isotopic patterns, glacia-
tion and the associated Fe- and Mn-ore formations and ‘cap’ car-
bonates in a conceptual model involving prolonged ocean
stagnation followed by overturn. They hypothesized that when
the ocean was stagnant, organic export and subsequent respiration
in anoxic deep waters would progressively enrich the surface
waters in 13C, while simultaneously creating an isotopically
depleted, bicarbonate-enriched reservoir at depth (Kaufman
et al. 1993; Grotzinger & Knoll 1995). In their model, the attendant
transfer of CO2 from the atmosphere to the deep ocean con-
tributes to global cooling. Ultimately, the growth of sea ice triggers
the formation of cold, saline, deep water, and the meridional
overturning circulation is reestablished. Upwelling of alkalinity-
laden deepwater leads to the precipitation of isotopically depleted
cap carbonates and the simultaneous release of CO2 to the atmos-
phere, which melts the ice (Kaufman et al. 1993; Grotzinger &
Knoll 1995).
The parsimony of the model temporarily outshone its flawed
foundation. Short of covering the ocean with ice, physical
stagnation (as distinct from dynamic stratification) on geological
timescales is implausible because the energy driving the upwelling
28 P. F. HOFFMAN
flux (the rate-limiting step in the overturning circulation) derives
from winds and tides (Wunsch 2002). Winds and tides are not
so easily turned off. Moreover, if the surface ocean was ever
deprived of nutrients upwelled from depth, primary production
would crash, depriving the deep of its alkalinity pump and the iso-
topic gradient of its driver. To its credit, the overturn hypothesis
at least lined up all the key observations in the same viewfinder:
glaciation, isotopes, iron formations and cap carbonates.
Biocatalysed weathering. Carver & Vardavas (1994) constructed a
time-dependent model of the Earth’s mean surface temperature
controlled by the geochemical cycle of carbon (Walker et al.
1981). They parameterized CO2 outgassing as having declined
rapidly before and more slowly after 3.5 Ma. The solar flux rose
almost linearly, and the consumption of CO2 through silicate
weathering was shaped by the rapid growth of 90% of the
present continental crust between 3.5 and 2.0 Ga, and by stepwise
increases in weathering efficiency (i.e. the rate of CO2 consump-
tion for any given pCO2) resulting from biological colonizations
of the land (Schwartzman & Volk 1991). The first colonization
(microbial) is assumed to have occurred before 3.5 Ga. The
second (organisms unspecified) supposedly occurred between 1.2
and 0.7 Ga, and the third in the middle Phanerozoic in response
to the rise of vascular plants. The observed enrichment in 13C
of most pre-Ediacaran carbonates (e.g. Knoll et al. 1986) is cited
in support of the middle (and largest) step in weathering efficiency
(Carver & Vardavas 1994), although the isotopic pattern requires
an increase in fractional organic burial as opposed to overall
carbon burial. Increased Neoproterozoic biocatalysed weathering
receives some support from micropalaeontology (Horodyski &
Knauth 1994), molecular divergence analysis (Heckman et al.
2001) and clay mineralogy (Kennedy et al. 2006), but remains
controversial. The modelled temperature curves (Carver &
Vardavas 1994) feature two relatively cold intervals in Earth
history, early Palaeoproterozoic in response to continental
growth and mid-Neoproterozoic in response to the prescribed
increase in biocatalysed weathering.
Continental break-up. Supercontinents tend to be dry because most
land is far from the ocean. Upon fragmentation, new continental
margins are created and land overall is brought closer to the
source of moisture. As silicate-weathering is catalysed by moisture
as well as by temperature (Walker et al. 1981), continental
break-up should result in lower pCO2 and a colder global
climate. Cooling will be most severe if rifting occurs in the
tropics, and least so for rifting in the polar regions where chemical
weathering rates are low. Continental break-up in tropical Pangaea
was followed by the coolest period (Late Jurassic –Early Cretac-
eous) of the Mesozoic era (Frakes et al. 1992); break-up in the
polar North Atlantic by the warmest period (Early Eocene) of
the Cenozoic (Zachos et al. 2001). This is consistent with Cryogen-
ian cooling, related to the break-up and dispersal of equatorial
Rodinia beginning c. 800 Ma (Li et al. 2008). These ideas have
recently been tested with a simplified atmospheric general circula-
tion model coupled to a model of the geochemical cycle of carbon
(Donnadieu et al. 2004a, b).
It is therefore difficult to accept the argument that Neoprotero-
zoic glaciation could be explained in terms of a close spatial (as
well as temporal) association with rift valleys and rifted margins,
without recourse to extreme climate states (Eyles 1993; Young
1995a; Eyles & Januszczak 2003). [In Reply to a Comment,
Young (1995b) wrote that he did not intend to imply a genetic
relationship, but then his paper should have been titled, ‘Was the
preservation of Neoproterozoic glacial deposits on the margins
of Laurentia related to the fragmentation of two superconti-
nents?’.] The Late Quaternary glacial maxima were arguably as
cold as any time in the Phanerozoic, but moraines on the mount-
ains of the East African rift system do not extend below
c. 3.5 km above sea level (Osmaston 2004). In the Red Sea rift,
only a handful of the highest peaks (.4.0 km above sea level)
on the Ethiopean Dome were glaciated at the Last Quaternary
Maximum (Osmaston 2004). For Neoproterozoic ice sheets to
have reached sea level at the same latitudes implies a drastically
colder climate.
Epilogue: the long road to consensus
Figure 2.2 clearly shows that publications concerning Neoprotero-
zoic glaciation ‘took off’ in the mid-1990s and possibly peaked
in 2007. There is little doubt that the galvanizing factor was the
Snowball Earth hypothesis (Kirschvink 1992; Hoffman &
Schrag 2002). In the first four years after its publication,
Kirschvink (1992) had been cited only three times, and favourably
only once (Klein & Beukes 1993). However, as soon as it was
advocated prominently (Hoffman et al. 1998; Hoffman & Schrag
2000) and in the public arena (Walker 2003), the reaction
was swift.
This is not the place to review all the papers that have appeared
since 1998: more time and more space are needed. Opinions about
the nature of Neoproterozoic glaciation are now at a stalemate.
Most workers agree that one, and possibly two, Cryogenian glacia-
tions occurred simultaneously on virtually all palaeocontinents
(Evans 2000; Halverson 2006; Hoffman & Li 2009; but see
Eyles & Januszczak 2003 for a contrary view). The basic argument
is that if ice sheets reached sea level in the warmest areas (i.e. close
to the palaeoequator and where thick non-skeletal carbonates
accumulated), then colder areas must have been frozen as well.
This argument makes no a priori assumption about correlation,
but correlation follows from the premise.
The stalemate concerns the state of the ocean: was it largely
ice-covered as assumed in the Snowball Earth hypothesis (Kirsch-
vink 1992; Hoffman & Schrag 2002), or substantially ice-free
as posited in the so-called slushball solution (Hyde et al. 2000;
Peltier et al. 2004; but see Bendtsen 2002; Voigt et al. 2011)?
Because of subduction, direct evidence from the ocean floor has
been eliminated, or survives only from post-Cryogenian (Edia-
caran), regional-scale glaciation (Kawai et al. 2008). Those who
favour the snowball model point to its ability to account for syngla-
cial Fe- and Fe –Mn deposits, post-glacial cap carbonates, and iso-
topic evidence for highly elevated pCO2 during and after glaciation
(Bao et al. 2008, 2009). The slushball solution does not account for
these features (Pollard & Kasting 2005). Those who favour the
slushball solution point to evidence that the snowball hypothesis
is incompatible with the fossil record (Knoll 2003; Xiao 2004; Cor-
setti et al. 2006; Moczydlowska 2008) and also with the sedimen-
tology of glacial deposits (e.g. McMechan 2000; Condon et al.
2002; Allen & Étienne 2008). The biotic argument presupposes
a knowledge of the limits to survival of species. The argument
from sedimentology assumes knowledge of when the deposits
were formed because the snowball hypothesis predicts that glacia-
tion was ‘polar’ in character at the beginning and ‘temperate’ or
‘Alaskan’ at the end. Studies of modern ice-sheet stability show
that outlet glaciers are buttressed by ice shelves (Dowdeswell
et al. 2000; De Angelis & Skvarca 2003; Nick et al. 2009): ice-
shelf removal triggers ice-sheet deflation, with potential for
ice-sheet collapse. On a Snowball Earth, the ice shelf is global. Ice-
shelf removal, the first step in snowball deglaciation, should
trigger rapid deflation and collapse of low-latitude ice sheets.
Much of the glacial sediment record on continental margins will
date from this period. These deposits will bear evidence of open
water (e.g. iceberg rafting and dumping, wave ripples) because
open water then existed. Sedimentology is thus a clumsy means
of testing the snowball hypothesis because of uncertainty over
which stage of the glacial cycle is stratigraphically preserved
(Hoffman 2005). Moreover, sediment fluxes cannot be estimated
because chronology is lacking. Even varve chronology (De Geer
1912), from which the duration of the Holocene was estimated
 HISTORY OF NEOPROTEROZOIC GLACIAL GEOLOGY
Agassiz
10
Rejection
Early Martyrs
5 0.4
Bernhardi
MacCulloch
Esmark
Venetz
Playfair
Hutton
Buch
Hall
0 0
1800 1820 1840
Year (AD)
with a fair degree of accuracy, is closed to us because of the rela-
tively low palaeolatitudes of most Neoproterozoic deposits, where
seasonality is weak. The decisive falsifying test is chronometric:
any glaciation that is short-lived (,5 Ma) cannot have been a
snowball glaciation because of the time required to accumulate
enough CO2 to overcome the ‘white Earth’s’ albedo. It is the dur-
ation of the glaciation, not that of particular glacial deposits, that
provides the test. The tightest existing constraint on the duration
of the end-Cryogenian (Marinoan) glaciation is 19.3 + 4.2 Ma
(Condon et al. 2005; Zhang et al. 2008). The older Cryogenian
(Sturtian) glaciation is virtually unconstrained, whereas the mid-
Ediacaran (Gaskiers) glaciation could not have been a snowball
glaciation because of its short duration (Hoffman & Li 2009).
If the snowball hypothesis is not falsified geochronologically,
how and when might a consensus on the nature of Cryogenian
glaciation come about? The history of the Pleistocene glacial
controversy (Fig. 2.4) may offer a preview. Agassiz’s outspoken
advocacy of the glacial theory of Esmark, Venetz and Bernhardi
engendered much excitement, but failed to achieve consensus. In
fact, with few exceptions (James Smith, Charles Maclaren,
Robert Buckland, Robert Chambers and, on-and-off, Charles
Darwin), British geologists soon turned hostile to the glacial
theory and remained so from 1842 until 1860. One can track the
fortunes of the theory through the writings of Charles Lyell
(1841, 1845, 1851, 1852, 1855, 1857, 1863, 1865), or the Anniver-
sary addresses of the President to the Geological Society, London
(see also Woodward 1907; Davies 2007).
Then, in the first half of the 1860s, everything changed. This
was the result of comprehensive regional studies by a new gener-
ation of Scottish geologists led by Ramsay (1860), Jamieson
(1862, 1863, 1865) and the Geikie brothers (1863, 1874) in what
may be termed the ‘Scottish glacial revival’. When they began
their studies, they had assumed the glacial theory was false
because that was the prevailing view. Their own work, systematic
and wide-ranging, left them no choice but to conclude that
Agassiz had been correct after all. From their time onwards,
critics of the glacial theory became the exception rather than
the majority.
One is tempted to think that consensus might be reached faster
in the early 21st century than in the mid-19th century. This is
doubtful. It is clear from the literature that Victorian geologists
actually read each others’ papers. No one could claim that this is
true today.
Will consensus be achieved through some conceptual or techni-
cal breakthough? I doubt it: there have been several already and
consensus is no closer than at the millenium. Was consensus
on Pleistocene glaciation brought about by Jamieson’s (1882)
solution to the submergence problem, which was that each ice
29
1.0
A. Geikie
Fig. 2.4. Bibliographic history of the
Jamieson
Revival 0.8 Pleistocene (‘Newer Pliocene’) glacial
Ramsay
controversy. Number of papers (black) and
books (grey) per year concerning the ‘Great
Northern Drift’. Early Martyrs refer to the
0.6 essentially correct but ignored glacial
theories of Esmark (1824), Dobson (1925),
Venetz (1830) and Bernhardi (1832).
Agassiz refers to the time of tumult
associated with the championing of the
glacial theory by Agassiz (1837, 1840,
0.2
1842). Rejection was a 20-year period of
eclipse of glacial theory. Revival marks the
widespread acceptance of glacial theory
following the studies of Ramsay (1860),
1860 1880 Jamieson (1862, 1863, 1865) and Geikie
(1863).
age was accompanied by submergence of the land, followed by
slow reemergence after the ice disappeared? This problem
(based on marine fossils within tills that are now elevated above
sea level) was central to the interpretation of the Drift (e.g.
Smith 1836). Agassiz had offered no explanation for it, there
being no marine fossils in the Swiss glacial deposits. Geophysicists
attributed ice-age submergence to the gravitational ‘pull’ of a
gigantic polar ice cap on the adjacent seas (Adhémar 1842;
Croll 1875), a theory easily accommodated by those like Lyell
and Murchison who believed that the drift came by way of
iceberg-laden floods of Arctic origin. In contrast, Shaler (1847)
and Jamieson (1865) suggested, and later demonstrated with
geological evidence (Jamieson 1882), that submergence was
caused by deflection of the lithosphere under the load of the ice
sheet (glacioisostasy). However, Jamieson’s (1882) theory of
submergence came after broad acceptance of the glacial theory
and cannot therefore have been its cause.
Consensus on the glacial theory came about because of the over-
whelming weight of evidence of the same kind as that obtained in
skeletal form by Agassiz (1842). If consensus on the Snowball
Earth hypothesis ultimately rests on the weight of evidence of
the kind identified by Kirschvink (1992) (palaeomagnetic evidence
for low-latitude glaciation at sea level, syn-glacial sedimentary
Fe2O3 and MnO2 ore formations, and global evidence of abrupt
climate switching (cap carbonates)) a positive outcome appears
more and more probable in the end because that evidence has
only become stronger.
Must we also wait for two decades or more for consensus
to emerge? Long delays between initial discussion of radical
hypotheses and their broad acceptance is a general phenomenon
in science (e.g. heliocentrism, biological evolution, continental
drift, the big bang), so we must suppose there is a reason for the
delays. The simplest one is that no consensus can emerge until
after the original antagonists, tethered to hasty judgements, have
passed from the scene. Given the lengthening of productive
lifespans, we could be in for a long wait.
Whatever the final outcome concerning the state of the
Cryogenian glacial ocean, it is apparent that a third climate state
needs to be added to the two first recognized by Brooks (1926):
‘non-glacial’ (no ice sheets), ‘glacial’ (polar to temperate ice
sheets) and ‘pan-glacial’ (ice sheets at all latitudes).
As a comparative newcomer to Neoproterozoic glacial studies, I am grateful
for reviews of a draft of this paper by M. Hambrey and two anonymous
reviewers. I also thank Barry Cooper for a preprint of his paper in Earth Sciences
History on early contributions from South Australia. However, I bear sole res-
ponsibility for the views expressed here. This work was supported by the US
National Science Foundation (grant EAR-0417422), the Earth System Evolution
30 P. F. HOFFMAN
Project of the Canadian Institute for Advanced Research (CIFAR) and Harvard
University. This represents a contribution of the IUGS- and UNESCO-funded
IGCP (International Geoscience Programme) Project #512.
References
Adhémar, J. A. 1842. Révolutions de la Mer. Fain et Thunot, Paris.
Agassiz, L. 1837. Discours prononcé à l’ouverture des séances de la
Société Helvétique des Sciences Naturelles, à Neuchâtel, le 24
juillet 1837. Actes de la Société Helvétique des Sciences Naturelles
(22ème session, Neuchâtel, 24 – 26 juillet 1837), 5– 32. (Reprinted
in English: Upon glaciers, moraines, and erratic blocks. Edinburgh
New Philosophical Journal, 24, 364– 383, 1838.)
Agassiz, L. 1840. Études sur les glaciers. Neuchâtel et Soleure. (Re-
printed in English: Studies on glaciers, preceded by the Discourse
of Neuchâtel. Hafner, New York, 1967.)
Agassiz, L. 1842. The glacial theory and its recent progress. New
Edinburgh Philosophical Journal, 33, 217–283.
Aitken, J. D. 1991. The Ice Brook Formation and post-Rapitan, Late
Proterozoic glaciation, Mackenzie Mountains, Northwest Territories.
Geological Survey of Canada, Bulletin, 404, 43.
Allen, P. A. & Hoffman, P. F. 2005. Extreme winds and waves
in the aftermath of a Neoproterozoic glaciation. Nature, 433,
123– 127.
Allen, P. A. & Etienne, J. L. 2008. Sedimentary challenge to Snowball
Earth. Nature Geoscience, 1, 817–825.
Allen, P., Leather, J. & Brasier, M. D. 2004. The Neoproterozoic Fiq
glaciation and its aftermath, Huqf Supergroup of Oman. Basin
Research, 16, 507 – 534.
Alley, R. B., Blankenship, D. D., Rooney, S. T. & Bentley, C. R.
1989. Sedimentation beneath ice shelves – the view from ice
stream B. Marine Geology, 85, 101– 120.
Almeida, F. F. M. 1946. Origem dos minérios de ferro y manganes de
Urucum (Corumbá, Estado Mato Grosso). Boletim de Divisão de
Geologia e Mineralogia, 119, 58.
Anderson, J. B. 1999. Antarctic Marine Geology. Cambridge University
Press, Cambridge.
Anderson, J. B. & Ashley, G. M. (eds) 1991. Glacial Marine Sedimen-
tation: Paleoclimatic Significance. Geological Society of America
Special Paper, 261, Boulder, CO.
Arnaud, E. 2004. Giant cross-beds in the Neoproterozoic Port Askaig
Formation, Scotland: implications for snowball Earth. Sedimentary
Geology, 165, 155–174.
Arnaud, E. 2008. Deformation in the Neoproterozoic Smalfjord
Formation, northern Norway: an indicator of glacial depositional
conditions? Sedimentology, 55, 335– 356.
Assereto, R. L. A. M. & Kendall, C. G. St. C. 1977. Nature, origin
and classification of peritidal tepee structures and related breccias.
Sedimentology, 24, 153–210.
Bao, H., Lyons, J. R. & Zhou, C. 2008. Triple oxygen isotope evidence
for elevated CO2 levels after a Neoproterozoic glaciation. Nature,
452, 504– 506.
Bao, H., Fairchild, I. J., Wynn, P. M. & Spötl, C. 2009. Stretching the
envelope of past surface environments: Neoproterozoic glacial lakes
from Svalbard. Science, 323, 119–122.
Barrett, P. J. (ed.) 1989. Antarctic Cenozoic History from the CIROS-1
Drillhole, McMurdo Sound. DSIR Bulletin, 245, Science Information
Publishing Center, Wellington, New Zealand.
Baud, L. 1933. Le conglomérat argilo-calcareux dans la région de Kayes
et de Bafoulabé et sa position stratigraphique. Compte Rendu, 197,
172– 173.
Beetz, W. 1929. Über das Wahrscheinlich altcambrische oder jungproter-
ozoische Alter der Glazialschichten an der Basis des Kundelungu-
Systems in Katanga und am unteren Kongo. Neues Jahrbuch für
Mineralogie, Geologie und Paläontologie, Beilagebände, Abteiling
B, 61, 61– 82.
Bendtsen, J. 2002. Climate sensitivity to changes in solar insolation in a
simple coupled climate model. Climate Dynamics, 18, 595–609.
Bendtsen, J. & Bjerrum, C. J. 2002. Vulnerability of climate on Earth
to sudden changes in insolation. Geophysical Research Letters, 29,
doi: 10.1029/2002GL014829.
Benn, D. I. & Evans, D. J. A. 1998. Glaciers and Glaciation. Arnold
Publishers, London.
Berger, W. H. 1982. Increase of carbon dioxide in the atmosphere during
deglaciation: The coral reef hypothesis. Naturwissenschaften, 69,
87 – 88.
Berner, R. A. & Maasch, K. A. 1996. Chemical weathering and controls
on atmospheric O2 and CO2: fundamental principles were enunciated
by J. J. Ebelmen in 1845. Geochimica et Cosmochimica Acta, 60,
1633– 1637.
Bertrand-Sarfati, J., Flicoteaux, R., Moussine-Pouchkine, A. &
Aı̈t Kaci Ahmed, A. 1997. Lower Cambrian apatitic stromatolites
and phospharenites related to the glacio-eustatic cratonic rebound
(Sahara, Algeria). Journal of Sedimentary Research, 67, 957– 974.
Bernhardi, A. 1832. Wie kamen die aus dem Norden stammenden
Felsbruchstücke und Geshiebe, welche man in Norddeutschland
und den benachbarten Ländern findet, an ihre gegenwärtigen
Fundorte? (How did the rock fragments and boulders of northern
origin found in northern Germany and neighboring countries get to
their present positions?) Jahrbuch für Mineralogie, Geognosie,
Geologie, und Petrefaktenkunde, 3, 257– 267.
Bestmann, M., Rice, A. H. N., Langenhorst, F., Grasemann, B. &
Heidelbach, F. 2006. Subglacial bedrock welding associated with
glacial earthquakes. Journal of the Geological Society, London,
163, 417– 420.
Biju-Duval, B. & Gariel, O. 1969. Nouvelles observations sur les
phénomènes glaciaires ‘Éocambriens’ de la bordure nord de la
synéclise de Taoudeni, entre le Hank et le Tanezrouft, Sahara occi-
dental. Palaeogeography, Palaeoclimatology, Palaeoecology, 6,
283– 315.
Blackett, P. M. S. 1961. Comparison of ancient climates with the ancient
latitudes deduced from rock magnetic measurements. Proceedings of
the Royal Society of London, Series A, 263, 1 –30.
Blackwelder, E. 1932. An ancient glacial formation in Utah. Journal of
Geology, 40, 289–304.
Blanford, W. T., Blanford, H. F. & Theobald, W. 1859. On the
geological structure and relations of the Talcheer Coal Field, in the
District of Cuttack. Memoirs of the Geological Survey of India, 1,
33 – 89.
Boulton, G. S. 1990. Sedimentary and sea level changes during glacial
cycles and their control on glacimarine facies architecture. In: Dow-
deswell, J. A. & Scourse, J. D. (eds) Glacimarine Evironments:
Processes and Sediments. Geological Society, London, Special Pub-
lications, 53, 15– 52.
Briden, J. C. 1970. Palaeolatitude distribution of precipitated sediments.
In: Runcorn, S. K. (ed.) Palaeogeophysics. Academic Press,
London, 437– 444.
Briden, J. C. & Irving, A. 1964. Palaeolatitude spectra of sedimentary
palaeoclimatic indicators. In: Nairn, A. E. M. (ed.) Problems in
Palaeoclimatology. John Wiley & Sons, New York, 199–224.
Brooks, C. E. P. 1922. The Evolution of Climate. Ernest Benn, London.
Brooks, C. E. P. 1926. Climate Through the Ages. R.V. Coleman,
New York.
Buchan, K. L., Ernst, R. E., Hamilton, M. A., Mertanen, S.,
Pesonen, L. J. & Elming, S.-Å. 2001. Rodinia: the evidence from
integrated palaeomagnetism and U–Pb geochronology. Precambrian
Research, 110, 9– 32.
Budyko, M. I. 1969. The effect of solar radiation variations on the climate
of the Earth. Tellus, 21, 611–619.
Bjørlykke, K. 1967. The Eocambrian Reusch moraine at Bigganjargga
and the geology of Varangerfjord, northern Norway. Norsk Geolo-
giske Undersøkelse, 251, 18 –44.
Cahen, L. 1950. Le Calcaire de Sekelolo, le complexe tillitique et la
dolomie rose C1 dans l’anticlinal de Congo de Kati (Bas-Congo).
Annales de la Musée Royale Congo Belge, Sciences Géologiques,
7, 1 –55.
Cahen, L. 1963. Glaciations anciennes et dérive des continents (Ancient
glaciations and continental drift). Annales de la Société Géologique
de Belgique, 86, 19– 83.
Cahen, L. & Lepersonne, J. 1981. Proterozoic diamictites of Lower
Zaire. In: Hambrey, M. J. & Harland, W. B. (eds) Earth’s Pre-
Pleistocene Glacial Record. Cambridge University Press, Cam-
bridge, 153– 157.
 HISTORY OF NEOPROTEROZOIC GLACIAL GEOLOGY
Caldas, J. 1979. Evidencias de una glaciación Precambriana en la costa
sur del Perú. Segundo Congreso Geológico Chileno, Arica, Vol. J,
29 – 37.
Calver, C. R. & Walter, M. R. 2000. The late Neoproterozoic Grassy
Group of King Island, Tasmania: correlation and palaeogeographic
significance. Precambrian Research, 100, 299– 312.
Carver, J. H. & Vardavas, I. M. 1994. Precambrian glaciations
and the evolution of the atmosphere. Annales Geophysicae, 12,
674– 682.
Catling, D. C., Zahnle, K. J. & McKay, C. P. 2001. Biogenic methane,
hydrogen escape, and the irreversible oxidation of early Earth.
Science, 293, 839– 843.
Cawood, P. A., Nemchin, A. A., Smith, M. & Loewy, S. 2003. Source of
the Dalradian Supergroup constrained by U– Pb dating of detrital
zircon and implications for the East Laurentian margin. Journal of
the Geological Society, London, 160, 231–246.
Chamberlin, T. C. 1898. The ulterior basis of time divisions and the
classification of geological history. Journal of Geology, 6, 449– 462.
Chamberlin, T. C. 1899. An attempt to frame a working hypothesis of
the cause of glacial periods on an atmospheric basis. Journal of
Geology, 7, 545–584.
Charpentier, J. de. 1837. Some conjectures regarding the great revolu-
tions which have so changed the surface of Switzerland, and particu-
larly that of the Canton of Vaud, as to give rise to its present aspect.
Edinburgh New Philosophical Journal, 22, 27 –36.
Chumakov, N. M. 1964. Präkambrische tillit-ähnliche Gesteine der
Sowjetunion. Geologische Rundschau, 54, 83 –102.
Chumakov, N. M. 1968. On the character of the Late Precambrian glacia-
tion of Spitsbergen (in Russian). Doklady Akademiya Nauk USSR,
Geological Series, 180, 1446–1449.
Chumakov, N. M. 1981. Upper Proterozoic glaciogenic rocks and their
stratigraphic significance. Precambrian Research, 15, 373–395.
Chumakov, N. M. 1992. The problems of old glaciations (pre-Pleistocene
glaciogeology in the USSR). Soviet Scientific Reviews, Section G
Geology, 1, 1– 208.
Claire, M. W., Catling, D. C. & Zahnle, K. J. 2006. Biogeochemical
modelling of the rise in atmospheric oxygen. Geobiology, 4,
239– 269.
Cloud, P., Wright, L. A., Williams, E. G., Diehl, P. & Walter, M. R.
1974. Giant stromatolites and associated vertical tubes from the
upper Proterozoic Noonday Dolomite, Death Valley region, eastern
California. Geological Society of America Bulletin, 85, 1869– 1882.
Coleman, A. P. 1907. A Lower Huronian ice age. American Journal of
Science, 23, 187– 192.
Coleman, A. P. 1926. Ice Ages: Recent and Ancient. MacMillan,
New York.
Condon, D. J., Prave, A. R. & Benn, D. I. 2002. Neoproterozoic glacial-
rainout intervals: observations and implications. Geology, 30, 35– 38.
Condon, D., Zhu, M., Bowring, S. A., Wang, W., Yang, A. & Jin, Y.
2005. U– Pb ages from the Neoproterozoic Doushantuo Formation,
China. Science, 308, 95– 98.
Cooper, B. 2009. Snowball Earth: the early contribution from South
Australia. Earth Sciences History, 29, 121– 145.
Corkoran, M. 2007. ‘Cap carbonates’ and Neoproterozoic glacigenic
successions from the Kimberley region, north-west Australia. Sedi-
mentology, 54, 871– 903.
Corsetti, F. A. & Grotzinger, J. P. 2005. Origin and significance of tube
structures in Neoproterozoic post-glacial cap carbonates: example
from Noonday Dolomite, Death Valley, United States. Palaios, 20,
348– 363.
Corsetti, F. A. & Kaufman, A. J. 2005. The relationship between the
Neoproterozoic Noonday Dolomite and the Ibex Formation: new
observations and their bearing on ‘snowball Earth’. Earth-Science
Reviews, 73, 63 –78.
Corsetti, F. A., Olcott, A. N. & Bakermans, C. 2006. The biotic
response to Neoproterozoic snowball Earth. Palaeogeography,
Palaeoclimatology, Palaeoecology, 232, 114– 130.
Crawford, A. R. & Daily, B. 1971. Probable non-synchroneity of Late
Precambrian glaciations. Nature, 230, 111– 112.
Croll, J. 1875. Climate and Time in Their Geological Relations: A Theory
of Secular Changes of the Earth’s Climate. Appleton & Co.,
New York.
31
Crowell, J. C. 1957. Origin of pebbly mudstones. Geological Society of
America Bulletin, 68, 993–1010.
Crowell, J. C. 1964. Climatic significance of sedimentary deposits con-
taining dispersed megaclasts. In: Nairn, A. E. M. (ed.) Problems in
Palaeoclimatology. Interscience, London, 86– 99.
Crowell, J. C. & Frakes, L. A. 1970. Phanerozoic glaciations
and the causes of ice ages. American Journal of Science, 268,
193– 224.
Cunningham, F. F. 1990. James David Forbes, Pioneer Scottish Glaciol-
ogist. Scottish Academic Press, Edinburgh.
Dal, A. 1900. Geologiske iagttagselser omkring Varangerfjorden. Norges
Geologiske Undersøkelse, 28, 1– 16.
David, T. W. E. 1907a. Conditions of climate at different geological
epochs, with special reference to glacial epochs. 10th International
Geological Congress, Mexico City, 1906, Compte Rendu, 437–482.
David, T. E. W. 1907b. Some problems of Australian glaciations. Report
of the Australasian Association for the Advancement of Science,
11, 457– 465.
Davies, G. L. 1968. The tour of the British Isles made by Louis Agassiz in
1840. Annals of Science, 24, 131– 146.
Davies, G. L. H. 2007. Whatever is under the Earth: the Geological
Society of London, 1807– 2007. Geological Society, London.
Davies, K. A. 1939. The glacial sediments of Bunyoro, N.W. Uganda.
Bulletin of the Geological Survey of Uganda, 3, 20 –37.
Davies, K. A. 1940. The glacial series of Bunyoro, north Uganda. 17th
International Geological Congress, Moscow, 1937, 6, 115– 119.
De Angelis, H. & Skvarca, P. 2003. Glacier surge after ice shelf
collapse. Science, 299, 1560– 1562.
De Geer, G. 1912. A geochronology of the last 12,000 years. 11th Inter-
national Geological Congress, Stockholm, 1910, 1, 241– 253.
Deynoux, M. & Trompette, R. 1976. Discussion: Late Precambrian mix-
tites: glacial and/or nonglacial? Dealing especially with the mixtites
of West Africa. American Journal of Science, 276, 1302– 1315.
Dobson, P. 1925. Remarks on Bowlders. American Journal of Science and
Arts, 10, 217.
Donnadieu, Y., Ramstein, G., Fluteau, F., Besse, J. & Meert, J. 2002.
Is high obliquity a plausible cause for Neoproterozoic glaciations?
Geophysical Research Letters, 29, doi: 10.1029/2002GL015902.
Donnadieu, Y., Goddéris, Y., Ramstein, G., Nédélec, A. & Meert, J.
2004a. A ‘snowball Earth’ climate triggered by continental break-up
through changes in runoff. Nature, 428, 303– 306.
Donnadieu, Y., Ramstein, G., Fluteau, F., Roche, D. & Gonopolski,
A. 2004b. The impact of atmospheric and oceanic heat transport on
the sea-ice instability during the Neoproterozoic. Climate Dynamics,
22, 293– 306.
Dorr, J. V. H. 1945. Manganese and iron deposits of Morro do Urucum,
Mato Grosso, Brazil. United States Geological Survey Bulletin,
946-A, 1 –47.
Dott, R. H. Jr. 1961. Squantum ‘tillite’, Massachusetts – evidence of
glaciation or subaqueous mass movements? Geological Society of
America Bulletin, 72, 1289– 1305.
Dow, D. B. 1965. Evidence of a Late Pre-Cambrian glaciation in the Kim-
berley Region of Western Australia. Geological Magazine, 102,
407– 419.
Dowdeswell, J. A. & Scourse, J. D. (eds) 1990. Glacimarine Environ-
ments: Processes and Sediments. Geological Society, London,
Special Publications, 53.
Dowdeswell, J. A., Whittington, J. A., Jennings, A. E., Andrews, J.
T., Mackensen, A. & Marienfield, P. 2000. An origin for lami-
nated glacimarine sediments through sea-ice build-up and supressed
iceberg rafting. Sedimentology, 47, 557– 576.
Drewry, D. 1986. Glacial Geologic Processes. Edward Arnold, London.
Dunn, P. R., Thomson, B. P. & Rankama, K. 1971. Late Pre-Cambrian
glaciation in Australia as a stratigraphic boundary. Nature, 231,
498– 502.
Du Toit, A. L. 1922. The Carboniferous glaciation of South Africa.
Transactions of the Geological Society of South Africa, 24, 188– 227.
Ebelmen, J.-J. 1845. Sur les produits de la décomposition des especes
minérales de la famille silicates. (On the products of the weathering
of silicate minerals.) Annales des Mines, 7, 3– 66.
Ebelmen, J.-J. 1847. Sur la décomposition des roches. (On the weathering
of rocks.) Annales des Mines, 12, 627– 654.
32 P. F. HOFFMAN
Edwards, M. B. 1975. Glacial retreat sedimentation in the Smalfjord
Formation, Late Precambrian, north Norway. Sedimentology, 22,
75 –94.
Edwards, M. B. 1984. Sedimentology of the Upper Proterozoic glacial
record, Vestertana Group, Finnmark, North Norway. Norges Geolo-
giske Undersøkelse Bulletin, 394, 76.
Edwards, M. B. 1997. Discussion of glacial or non-glacial origin of the
Bigganjargga tillite, Finnmark, northern Norway. Geological Maga-
zine, 134, 873– 876.
Eisbacher, G. H. 1981. Sedimentary tectonics and glacial record in the
Windermere Supergroup, Mackenzie Mountains, northwestern
Canada. Geological Survey of Canada Paper, 80-27, 40.
Elverhoi, A. 1984. Glacigenic and associated marine sediments in the
Weddell Sea, Fjords of Spitsbergen and the Barents Sea: a review.
Marine Geology, 57, 53– 88.
Embleton, B. J. J. & Williams, G. E. 1986. Low latitude of deposition for
late Precambrian periglacial varvites in South Australia: implications
for palaeoclimatology. Earth and Planetary Science Letters, 79,
419– 430.
Eriksson, E. 1968. Air– ocean – icecap interactions in relation to climatic
fluctuations and glaciation cycles. Meteorological Monographs, 8,
68 –92.
Esmark, J. 1824. Bidrag til vor jordklodes historie. (Contribution to the
history of our Earth.) Magazin for Naturvidenskaberne, 2(1),
29 –54. (Reprinted in English: Remarks tending to explain the geo-
logical history of the Earth. Edinburgh New Philosophical Journal,
2, 107–121, 1827.)
Evans, D. A. D. 2000. Stratigraphic, geochronological, and paleomagnetic
constraints upon the Neoproterozoic climatic paradox. American
Journal of Science, 300, 347– 433.
Evans, D. A. D. 2006. Proterozoic low orbital obliquity and axial-dipolar
geomagnetic field from evaporite palaeolatitudes. Nature, 444,
51 –55, doi: 10.1038/nature05203.
Eyles, N. 1993. Earth’s glacial record and its tectonic setting. Earth-
Science Reviews, 35, 1– 248.
Eyles, N. & Januszczak, N. 2003. ‘Zipper-rift’: a tectonic model for
Neoproterozoic glaciations during the breakup of Rodinia after 750
Ma. Earth-Science Reviews, 65, 1– 73.
Fairchild, I. J. 1983. Effects of glacial transport and neomorphism on
Precambrian dolomite crystal sizes. Nature, 304, 714– 716.
Fawcett, P. J. & Boslough, M. B. E. 2002. Climatic effects of an
impact-induced equatorial debris ring. Journal of Geophysical
Research, 107(D15), doi: 10.1029/2001JD001230.
Fedden, F. G. S. 1875. On the evidence of ‘ground-ice’ in tropical India,
during the Tálchı́r period. Records of the Geological Survey of India,
8, 16 –18.
Forbes, E. 1846. On the connexion between the distribution of the existing
fauna and flora of the British Isles, and the geological changes which
have affected their area, especially during the epoch of the Northern
Drift. Memoirs of the Geological Survey of Great Britain, 1,
336– 403.
Fourier, J. 1824. Remarques générales sur les températures du globe ter-
restre et des espaces planétaires. Annales de Chimie et de Physique
(2nd series), 27, 136–167. Translated as, General remarks on the
temperature of the terrestrial globe and the planetary spaces. Ameri-
can Journal of Science and Arts, 32, 1– 20, 1837.
Føyn, S. 1937. The Eocambrian series of the Tana district, northern
Norway. Norsk Geologisk Tidsskrift, 17, 65– 164, 4 plates,
1:250,000 scale map.
Frakes, L. A., Francis, J. E. & Syktus, J. I. 1992. Climate Modes of the
Phanerozoic. Cambridge University Press, Cambridge, 274.
Frankel, H. 1987. Jan Hospers and the rise of paleomagnetism. Eos,
Transactions of the American Geophysical Union, 68, 577, 579– 581.
Furon, R. 1933. Observations sur la stratigraphie de l’ouest africain
(Mauritanie et Soudan). Compte Rendu, 196, 1905–1906.
Gaertner, H. R. von. 1943. Bemerkungen über den Tillit von Biggan-
jargga am Varangerfjord. Geologische Rundschau, 34, 226–231.
Garwood, E. J. & Gregory, J. W. 1898. Contribution to the glacial
geology of Spitsbergen. Quarterly Journal of the Geological
Society, London, 54, 197–225.
Geikie, A. 1863. On the phenomena of the glacial drift of Scotland. Trans-
actions of the Geological Society of Glasgow, 1, 190.
Geikie, A. 1865. The Scenery of Scotland, Viewed in Connexion with its
Physical Geology. MacMillan, London.
Geikie, J. 1874. The Great Ice Age. D. Appleton & Co., New York, 545.
Gevers, T. W. & Beetz, W. 1940. Pre-Dwyka glacial periods in southern
Africa. 17th International Geological Congress, Moscow, 1937, 6,
65 – 98.
Gibsher, A. S. & Khomentovsky, V. V. 1990. The section of the Tsagaan
Olum and Bayan Gol Formations of the Vendian – Lower Cambrian
in the Dzabkhan zone of Mongolia. In: Khomentovsky, V. V.,
Gibsher, A. S. & Karlova, G. A. (eds) The Late Precambrian and
Early Paleozoic of Siberia. Institut Geologii I Geofiziki, Sibirskoe
Otdelenie, Akademiya Nauk SSSR, Novosibirsk, 79 –91.
Girdler, R. W. 1964. The palaeomagnetic latitudes of possible ancient
glaciations. In: Nairn, A. E. M. (ed.) Problems in Palaeoclimatology.
Interscience, London, 119–149.
Goddéris, Y., Donnabieu, Y. et al. 2003. The Sturtian ‘snowball’ gla-
ciation: fire and ice. Earth and Planetary Science Letters, 211, 1– 12.
Gold, T. 1955. Instability of the Earth’s axis of rotation. Nature, 175,
526– 529.
Gorin, G. E., Racz, L. G. & Walter, M. R. 1982. Late Precambrian –
Cambrian sediments of Huqf Group, Sultanate of Oman. American
Association of Petroleum Geologists Bulletin, 66, 2609– 2627.
Grotzinger, J. P. & Knoll, A. H. 1995. Anomalous carbonate
precipitates: is the Precambrian the key to the Permian? Palaios, 10,
578– 596.
Halverson, G. P. 2006. A Neoproterozoic chronology. In: Xiao, S. &
Kaufman, A. J. (eds) Neoproterozoic Geobiology and Paleobiology.
Springer, Dordrecht, 231– 271.
Halverson, G. P., Hoffman, P. F., Schrag, D. P. & Kaufman, J. A.
2002. A major perturbation of the carbon cycle before the Ghaub
glaciation (Neoproterozoic) in Namibia: prelude to snowball
Earth? Geophysics, Geochemistry, Geosystems, 3, doi: 10.1029/
2001GC000244.
Halverson, G. P., Maloof, A. C. & Hoffman, P. F. 2004. The Marinoan
glaciation (Neoproterozoic) in northeast Svalbard. Basin Research,
16, 297– 324.
Hambrey, M. J. 1994. Glacial Environments. CRC Press, Boca Raton, FL.
Hambrey, M. J. & Harland, W. B. 1981. Earth’s Pre-Pleistocene
Glacial Record. Cambridge University Press, Cambridge.
Hambrey, M. J. & Harland, W. B. 1985. The Late Proterozoic glacial
era. Palaeogeography, Palaeoclimatology, Palaeoecology, 51,
255– 272.
Hambrey, M. J. & Spencer, A. M. 1987. Late Precambrian glaciation of
central East Greenland. Meddelelser om Grønland, Geoscience, 19,
1 –50.
Hamdi, B. 1992. Late Precambrian glacial deposits in central Iran. 29th
International Geological Congress, Kyoto, 1992, Abstracts, 2, 263.
Harland, W. B. 1964a. Evidence of late Precambrian glaciation and its
significance. In: Nairn, A. E. M. (ed.) Problems in Palaeoclimatol-
ogy. Interscience, London, 119– 149.
Harland, W. B. 1964b. Critical evidence for a great infra-Cambrian
glaciation. Geologische Rundschau, 54, 45 –61.
Harland, W. B. 1983. The Proterozoic glacial record. In: Medaris, L. G.,
Jr., Byers, W. C., Mickelson, D. M. & Shanks, W. C. (eds) Proter-
ozoic Geology. Geological Society of America, Memoir, 161,
279– 288.
Harland, W. B. & Bidgood, E. T. 1959. Palaeomagnetism in some
Norwegian Sparagmites and the Late Pre-Cambrian ice age.
Nature, 184, 1860– 1862.
Harland, W. B. & Rudwick, M. J. S. 1964. The great infra-Cambrian ice
age. Scientific American, August, 42 –49.
Harland, W. B. & Wilson, C. B. 1956. The Hecla Hoek Succession in
Ny Friesland, Spitsbergen. Geological Magazine, 93, 2265– 2286.
Heckman, D. S., Geiser, D. M., Eldell, B. R., Stauffer, R. L., Kardos,
N. L. & Hedges, S. B. 2001. Molecular evidence for the early colo-
nization of land by fungi and plants. Science, 293, 1129–1133.
Hegenberger, W. 1987. Gas escape structures in Precambrian peritidal
carbonate rocks. Communications of the Geological Survey of
Namibia, 3, 49 –55.
Heezen, B. C. & Hollister, C. 1964. Turbidity currents and glaciation.
In: Nairn, A. E. M. (ed.) Problems in Palaeoclimatology. Inter-
science, London, 99– 108.
 HISTORY OF NEOPROTEROZOIC GLACIAL GEOLOGY
Hintze, F. F. 1913. A contribution to the geology of the Wasatch Moun-
tains, Utah. Annals of the New York Academy of Sciences, 23,
85 – 143.
Hitchcock, E. 1841. Diluvium, or drift. In: Final Report on the Geology
of Massachusetts. J.S. & C. Adams, Amherst, MA, 350– 406; post-
script, 3a –11a.
Hoffman, P. F. & Maloof, A. C. 1999. Glaciation: the snowball theory
still holds water. Nature, 397, 384.
Hoffman, P. F. & Schrag, D. P. 2000. Snowball Earth. Scientific Amer-
ican, 282, 68 –75.
Hoffman, P. F. & Schrag, D. P. 2002. The snowball Earth hypothesis:
testing the limits of global change. Terra Nova, 14, 129– 155.
Hoffman, P. F., Kaufman, A. J., Halverson, G. P. & Schrag, D. P.
1998. A Neoproterozoic snowball Earth. Science, 281, 1342–1346.
Hoffman, P. F., Halverson, G. P., Domack, E. W., Husson, J. M.,
Higgins, J. A. & Schrag, D. P. 2007. Are basal Ediacaran (635
Ma) post-glacial ‘cap dolostones’ diachronous? Earth and Planetary
Science Letters, 258, 114– 131.
Holland, T. H. 1908. On the occurrence of striated boulders in the
Blaini Formation of Simla, with a discussion of the geological
age of the beds. Records of the Geological Survey of India, 37,
129– 135.
Holtedahl, O. 1918. Varangerhalvøn: Strøet omkring bunden av Varan-
gerfjord. Norsk Geologiske Undersøkelse, 84, 148– 173.
Hoffman, P. F. 2005. 28th DeBeers Alex. Du Toit Memorial Lecture: on
Cryogenian (Neoproterozoic) ice-sheet dynamics and the limitations
of the glacial sedimentary record. South African Journal of Geology,
108, 557– 576.
Hoffman, P. F. & Li, Z. X. 2009. A palaeogeographic context for Neopro-
terozoic glaciation. Palaeogeography, Palaeoclimatology, Palaeo-
ecology, 277, 158– 172.
Horodyski, R. J. & Kauth, L. P. 1994. Life on land in the Precambrian.
Science, 263, 494– 498.
Howchin, W. 1901. Preliminary note on the existence of glacial beds of
Cambrian age in South Australia. Transactions of the Royal Society
of South Australia, 21, 74 –86.
Howchin, W. 1903. Report on South Australian Glacial Investigation
Committee. Report of the Australasian Association for the Advance-
ment of Science, 9, 194– 200.
Howchin, W. 1908. Glacial beds of Cambrian age in South
Australia. Quarterly Journal of the Geological Society, London, 64,
234– 259.
Howchin, W. 1912. Australian glaciations. Journal of Geology, 20,
193– 227.
Hunt, T. S. 1880. The chemical and geological relations of the atmos-
phere. American Journal of Science, 19, 349– 363.
Hyde, W. T., Crowley, T. J., Baum, S. K. & Peltier, W. R. 2000. Neo-
proterozoic ‘snowball Earth’ simulations with a coupled climate/ice-
sheet model. Nature, 405, 425–429.
Irving, E. 1956. Palaeomagnetic and palaeoclimatological aspects of
polar wandering. Geofisica Pura e Applicada, 33, 23 – 41.
Irving, E. 1957a. Directions of magnetization in the Carboniferous glacial
varves of Australia. Nature, 180, 280– 281.
Irving, E. 1957b. Analysis of the palaeomagnetism of the Torridonian
sandstone series of north-west Scotland. Philosophical Transactions
of the Royal Society, London, Series A, 250, 83 –89.
Irving, E. 1959. Palaeomagnetic pole positions: a survey and analysis.
Geophysical Journal of the Royal Astronomical Society, 2, 51 –79.
Irving, E. 2008. Jan Hospers’s key contributions to geomagnetism. Eos,
Transactions, American Geophysical Union, 89, 457–458.
Irving, E. & Runcorn, S. K. 1957. The origin of the palaeomagnetism of
the Torridonian sandstones of north-west Scotland. Philosophical
Transactions of the Royal Society, London, Series A, 250, 100– 110.
Isotta, C. A., Rocha-Campos, A. C. & Yoshida, R. 1969. Striated pave-
ment of the Upper Pre-Cambrian glaciation in Brazil. Nature, 222,
466– 468.
Jack, R. L. 1913. The Mount Grainger Goldfield. Report of the Geological
Survey of South Australia, 1, 1 –24.
James, N. P., Narbonne, G. M. & Kyser, T. K. 2001. Late Neoprotero-
zoic cap carbonates: Mackenzie Mountains, northwestern Canada:
precipitation and global glacial meltdown. Canadian Journal of
Earth Sciences, 38, 1229–1262.
33
Jamieson, T. F. 1862. On the ice-worn rocks of Scotland. Quarterly
Journal of the Geological Society of London, 18, 164– 184.
Jamieson, T. F. 1863. On the parallel roads of Glen Roy, and their place in
the history of the glacial period. Quarterly Journal of the Geological
Society of London, 19, 235– 259.
Jamieson, T. F. 1865. On the history of the last geological changes in Scot-
land. Quarterly Journal of the Geological Society of London, 21,
161– 203.
Jamieson, T. F. 1882. On the cause of the depression and re-elevation of
the land during the glacial period. Geological Magazine, 9, 400– 407,
457– 466.
Jenkins, G. S. 2000. Global climate model high-obliquity solutions to the
ancient climate puzzles of the faint-young-Sun paradox and low-
latitude Proterozoic glaciation. Journal of Geophysical Research,
105, 7357–7370.
Jensen, P. A. & Wulff-Pedersen, E. 1996. Glacial or non-glacial origin
for the Bigganjargga tillite, Finnmark, northern Norway. Geological
Magazine, 133, 137– 145.
Jiang, G., Sohl, L. E. & Christie-Blick, N. 2003. Neoproterozoic
stratigraphic comparison of the Lesser Himalaya (India) and Yangtse
clock (south China): paleogeographic implications. Geology, 31,
917– 920.
Jiang, G., Kennedy, M. J., Christie-Blick, N., Wu, H. & Zhang, S.
2006. Stratigraphy, sedimentary structures, and textures of the late
Neoproterozoic Doushantuo cap carbonate in South China. Journal
of Sedimentary Research, 76, 978– 995.
Johnston, J. D. 1993. Ice wedge casts in the Dalradian of South Donegal
– evidence for subaerial exposure of the Boulder Bed. Irish Journal
of Earth Sciences, 12, 13– 26.
Katz, H. R. 1961. Late Precambrian to Cambrian stratigraphy in East
Greenland. In: Raasch, G. O. (ed.) Geology of the Arctic, Vol 1.
University of Toronto Press, Toronto, 299– 328.
Kaufman, A. J., Jacobsen, S. B. & Knoll, A. H. 1993. The Vendian
record of Sr and C isotopic variations in seawater: implications for
tectonics and paleoclimate. Earth and Planetary Science Letters,
120, 409–430.
Kaufman, A. J., Knoll, A. H. & Narbonne, G. M. 1997. Isotopes, ice
ages, and terminal Proterozoic Earth history. Proceedings of the
National Academy of Sciences USA, 94, 6600–6605.
Kaufman, A. J., Jiang, G., Christie-Blick, N., Banerjee, D. M. & Rai,
V. 2006. Stable isotope record of the terminal Neoproterozoic Krol
platform in the Lesser Himalayas of northern India. Precambrian
Research, 147, 156–185.
Kawai, T., Windley, B. F., Terabayashi, M., Yamamoto, H., Isozaki,
Y. & Maruyama, S. 2008. Neoproterozoic glaciation in the
mid-oceanic realm: an example from hemi-pelagic mudstones on
Llanddwyn Island, Anglesey, UK. Gondwana Research, 14,
105– 114.
Kendall, C. G. St. C. & Warren, J. 1987. A review of the origin and
setting of tepees and their associated fabrics. Sedimentology, 34,
1007–1027.
Kennedy, M. J. 1996. Stratigraphy, sedimentology, and isotopic geo-
chemistry of Australian Neoproterozoic postglacial cap dolostones:
deglaciation, d13C excursions, and carbonate precipitation. Journal
of Sedimentary Research, 66, 1050– 1064.
Kennedy, M., Droser, M., Mayer, L. M., Pevear, D. & Mrofka, D.
2006. Late Precambrian oxygenation: inception of the clay mineral
factory. Science, 311, 1446–1449.
Kennedy, M. J., Runnegar, B., Prave, A. R., Hoffmann, K.-H. &
Arthur, M. A. 1998. Two or four Neoproterozoic glaciations?
Geology, 26, 1059– 1063.
Kilburn, C., Pitcher, W. S. & Shackleton, R. M. 1965. The stratigra-
phy and origin of the Portaskaig Boulder Bed series (Dalradian). Geo-
logical Journal, 4, 343– 360.
Kiessling, W. 2001. Paleoclimatic significance of Phanerozoic reefs.
Geology, 29, 751– 754.
King, L. H., Rokoengen, K., Fader, G. B. J. & Gunleiksrud, T. 1991.
Till-tongue stratigraphy. Geological Society of America Bulletin,
103, 637–659.
Kirschvink, J. L. 1992. Late Proterozoic low-latitude glaciation: the
snowball Earth. In: Schopf, J. W. & Klein, C. (eds) The Proterozoic
Biosphere. Cambridge University Press, Cambridge, 51 –52.
34 P. F. HOFFMAN
Klein, C. & Beukes, N. J. 1993. Sedimentology and geochemistry of the
glacigenic Late Proterozoic Rapitan iron-formation in Canada. Econ-
omic Geology, 88, 542– 565.
Knoll, A. H. 2003. Life on a Young Planet, the First Three Billion
Years of Evolution on Earth. Princeton University Press, Princeton.
Knoll, A. H. & Walter, M. R. 1992. Latest Proterozoic stratigraphy and
Earth history. Nature, 356, 673–678.
Knoll, A. H., Hayes, J. M., Kaufman, A. J., Swett, K. & Lambert, I. B.
1986. Secular variation in carbon isotope ratios from Upper Protero-
zoic successions of Svalbard and East Greenland. Nature, 321,
831– 838.
Koken, 1907. Indisches Perm und die Permische Eiszeit. (Indian Permian
and the Permian ice age.) Neues Jahrbuch für Mineralogie, Geologie,
und Paläontologie, Festband 1907, 446–545, 19 plates (including
map).
Köppen, W. & Wegener, A. 1924. Die Klimate der geologischen
Vorzeit. (Climates of the geological past.) Gebrüder Borntraeger,
Berlin.
Kuenen, P. H. & Migliorini, C. I. 1950. Turbidity currents as a cause of
graded bedding. Journal of Geology, 58, 91 –127.
Kulling, O. 1934. The Hecla Hoek Formation round Hinlopenstredet.
Geografiska Annaler, 14, 161–253.
Lachenbruch, A. 1962. Mechanics of thermal contraction cracks and
ice-wedge polygons in permafrost. Special Paper, Geological
Society of America, 70, 1 –69.
Laskar, J. & Robutel, P. 1993. The chaotic obliquity of the planets.
Nature, 361, 608–612.
Laskar, J., Joutel, F. & Robutel, P. 1993. Stabilization of the Earth’s
obliquity by the Moon. Nature, 361, 615– 617.
Leather, J., Allen, P. A., Brasier, M. D. & Cozzi, A. 2002. Neoproter-
ozoic snowball Earth under scrutiny: evidence from the Fiq glaciation
of Oman. Geology, 30, 891–894.
Lee, Y. Y. 1936. The Sinian glaciation in the lower Yangtze valley. Bul-
letin of the Geological Society of China, 15, 131–134.
Lee, J. S. & Chao, Y. T. 1924. Geology of the gorge district of the
Yangtze (from Ichang to Tzekuei) with special reference to the
development of the gorges. Bulletin of the Geological Society of
China, 3, 351– 391.
Lee, J. S. & Lee, Y. Y. 1940. Sinian glaciation of China. 17th Inter-
national Geological Congress, Moscow, 1937, Report, 6, 33 – 41.
Lemon, N. M. & Gostin, V. A. 1990. Glacigenic sediments of the late
Proterozoic Elatina Formation and equivalents, Adelaide Geosyn-
cline, South Australia. In: Jago, J. B. & Moore, P. S. (eds) The
Evolution of a Late Precambrian–Early Paleozoic Rift Complex:
the Adelaide Geosyncline. Geological Society of South Australia
Special Publication, Adelaide, 16, 149–163.
Lenk, C., Strother, P. K., Kaye, C. A. & Barghoorn, E. S. 1982. Pre-
cambrian age of the Boston Basin: new evidence from microfossils.
Science, 216, 619–620.
Lenton, T. M. & Watson, A. J. 2004. Biotic enhancement of weathering,
atmospheric oxygen and carbon dioxide in the Neoproterozoic. Geo-
physical Research Letters, 31, L05202, doi:10.1029/2003GL018802.
Levrard, B. & Laskar, J. 2003. Climate friction and the Earth’s obli-
quity. Geophysical Journal International, 154, 970–990.
Li, Z. X., Evans, D. A. D. & Zhang, S. 2004. A 908 spin on Rodinia: poss-
ible causal links between the Neoproterozoic supercontinent, super-
plume, true polar wander and low-latitude glaciation. Earth and
Planetary Science Letters, 220, 409–421.
Li, Z. X., Bogdanova, S. V. et al. 2008. Assembly, configuration, and
break-up history of Rodinia: a synthesis. Precambrian Research,
160, 179– 210.
Lu, S. N., Ma, G. G., Gao, Z. J. & Lin, W. X. 1985. Primary research on
the glaciogenic rocks of the Late Precambrian in China. In: Precam-
brian Geology Committee (eds) Precambrian Geology, No. 1, The
Collected Works on the Late Precambrian Glaciogenic Rocks of
China. Geology Publication House, Beijing, 1– 86.
Lyell, C. 1840. On the geological evidence of the former existence of
glaciers in Forfarshire. Edinburgh New Philosophical Journal,
30, 199–202. Reprinted in Philosophical Magazine (2nd series),
18, 579– 591, 1841.
Lyell, C. 1841. Elements of Geology, 2nd edn. Hilliard, Gray, and Co.,
Boston.
Lyell, C. 1845. Travels in North America, 1841– 2, with Geological
Observations on the United States, Canada, and Nova Scotia
(2 vol.). Wiley & Putnam, New York, 231. (Reprinted in 1978 by
Arno Press, New York.)
Lyell, C. 1851. Manual of Elementary Geology, 3rd edn. John Murray,
London.
Lyell, C. 1852. Manual of Elementary Geology, 4th edn. John Murray,
London.
Lyell, C. 1855. Manual of Elementary Geology, 5th edn. John Murray,
London.
Lyell, C. 1857. Manual of Elementary Geology, 6th edn. John Murray,
London.
Lyell, C. 1863. The Geological Evidence of the Antiquity of Man.
Reprinted unabridged in 2004 by Dover Publications, Mineola, NY.
Lyell, C. 1865. Elements of Geology, 6th enlarged edn. John Murray,
London.
Macdonald, F. A., Jones, D. S. & Schrag, D. P. 2009a. Stratigraphic
and tectonic implications of a newly discovered glacial diamictite-
cap carbonate couplet in southwestern Mongolia. Geology, 37,
123– 126.
Macdonald, F. A., McClelland, W. C., Schrag, D. P. & Macondald,
W. P. 2009b. Neoproterozoic glaciation on a carbonate platform
margin in Arctic Alaska and the origin of the North Slope subterrane.
Geological Society of America Bulletin, 121, 448–473.
Macdonald, F. A., Strauss, J. V., Rose, C. V., Dudás, F. Ö. & Schrag,
D. P. 2011. Stratigraphy of the Port Nolloth Group of Namibia and
South Africa and implications for the age of Neoproterozoic iron for-
mations. American Journal of Science, 310, 862– 888.
Maciel, P. 1959. Tilito Cambriano(?) no Estado de Mato Grosso. Socie-
dad Brasilieras Geologia Boletino, 8, 3– 49.
Maloof, A. C., Kellogg, J. B. & Anders, A. M. 2002. Neoproterozoic
sand wedges: crack formation in frozen soils under diurnal forcing
during a snowball Earth. Earth and Planetary Science Letters, 204,
1 –15.
Maloof, A. C., Halverson, G. P., Kirschvink, J. L., Schrag, D. P.,
Weiss, B. P. & Hoffman, P. F. 2006. Combined paleomagnetic,
isotopic and stratigraphic evidence for true polar wander from the
Neoproterozoic Akademikerbreen Group, Svalbard. Geological
Society of America Bulletin, 118, 1099– 1124.
Marshall, H. G., Walker, J. C. G. & Kuhn, W. R. 1988. Long-term
climate change and the geochemical cycle of carbon. Journal of
Geophysical Research, 93, 791– 801.
Martin, H. 1965a. The Precambrian Geology of South West Africa and
Namaqualand. Precambrian Research Unit Bulletin, 1, University
of Cape Town, South Africa.
Martin, H. 1965b. Beobachtungen zum Problem der jung-
präkambrischen Glazialen Ablagerungen in Südwestafrika
(Observations concerning the problem of the late Precambrian
glacial deposits in South West Africa). Geologische Rundschau, 54,
115– 127.
Martin, H. 1981. The late Palaeozoic Gondwana glaciation. Geologische
Rundschau, 70, 480 – 496.
Martin, H., Porada, H. & Walliser, O. H. 1985. Mixtite deposits of the
Damara sequence, Namibia, problems of interpretation. Palaeogeo-
graphy, Palaeoclimatology, Palaeoecology, 51, 159–196.
Maugh, T. H. II. 1989. Super ice age gave life on Earth growing pains.
Los Angeles Times, September 7, 1, 3, 28.
Mawson, D. 1949a. The Late Precambrian ice age and glacial record of
the Bibliando dome. Journal and Proceedings of the Royal Society
of New South Wales, 82, 150– 174.
Mawson, D. 1949b. The Elatina glaciation: a third recurrence of glacia-
tion evidenced in the Adelaide system. Transactions of the Royal
Society of South Australia, 73, 117–121.
McCabe, C. & Elmore, R. D. 1989. The occurrence and origin of Late
Paleozoic remagnetization in the sedimentary rocks of North
America. Reviews of Geophysics, 27, 471–494.
McCay, G. A., Prave, A. R., Alsop, G. I. & Fallick, A. E. 2006. Glacial
trinity: Neoproterozoic Earth history within the British –Irish Caledo-
nides. Geology, 34, 909–912, doi: 10.1130/G22694A.1
McMechan, M. E. 2000. Vreeland Diamictites – Neoproterozoic glacio-
genic slope deposits, Rocky Mountains, northeast British Columbia.
Bulletin of Canadian Petroleum Geology, 48, 246– 261.
 HISTORY OF NEOPROTEROZOIC GLACIAL GEOLOGY
Melosh, H. J. 1989. Impact Cratering: A Geologic Process. Oxford Uni-
versity Press, New York.
Milankovic, M. 1941. Canon of Insolation and the Ice-Age Problem.
(English translation published by Zavod za Udzbenike i Nastavna
Sredstva, Belgrade, 634, 1998.)
Miller, N. R., Alene, M. et al. 2003. Significance of the Tambien
Group (Tigrai, N. Ethiopia) for Snowball Earth events in the
Arabian-Nubian Shield. Precambrian Research, 121, 263–283.
Moczydlowska, M. 2008. The Ediacaran microbiota and the survival of
Snowball Earth conditions. Precambrian Research, 167, 1 –15.
Molnia, B. F. 1983. Glacial– Marine Sedimentation. Plenum Press,
New York.
Moraes Rego, L. F. de. 1930. Glaciação eopaleozóica no centro do
Brazil (Glaciation in the earliest Palaeozoic of central Brazil).
Anais Academia Brasileira de Ciências, 2, 109–112.
Mu, Y. 1981. Luoquan Tillite of the Sinian System in China. In:
Hambrey, M. J. & Harland, W. B. (eds) Earth’s Pre-Pleistocene
Glacial Record. Cambridge University Press, Cambridge, 402– 413.
Nalivkin, D. V. 1956. Ucheniye o fatsiyak. Geograficheskiye usloviya
obrazovaniya osadkov, 2 (Science of facies. Geographical environ-
ment of sedimentation). Izvestiya Akademia Nauk SSSR, Moscow.
Natland, M. L. & Kuenen, P. H. 1951. Sedimentary history of the
Ventura Basin, California and the action of turbidity currents. In: Tur-
bidity Currents and the Transportation of Coarse Sediments into
Deep Water. Special Publication 2, Society of Economic Paleontolo-
gists and Mineralogists, Tulsa, OK, 76 –107.
Nédélec, A., Affaton, P., France-Lanord, C., Charrière, A. &
Alvaro, J. 2007. Sedimentology and chemostratigraphy of
the Bwipe Neoproterozoic cap dolostones (Ghana, Volta Basin):
a record of microbial activity in a peritidal environment. Comptes
Rendus Geoscience, 339, 223– 239.
Néron de Surgy, O. & Laskar, J 1997. On the long term evolution of the
spin of the Earth. Astronomy and Astrophysics, 318, 975–989.
Newman, R. P. 1995. American instransigence: the rejection of continen-
tal drift in the great debates of the 1920’s. Earth Sciences History, 14,
62 – 83.
Nick, F. M., Vieli, A., Howat, I. M. & Joughin, I. 2009. Large-scale
changes in Greenland outlet glacier dynamics triggered at the termi-
nus. Nature Geoscience, 2, 110– 114.
Nikolaev, J. 1930. The glacial deposits (tillites) of Lower Cambrian age
in the Yenissei Range. Bulletin of the Geological and Prospecting
Service of the U.S.S.R., 49(7), 1 – 15.
Nogueira, A. C. R., Riccomini, C., Sial, A. N., Moura, C. A. V. & Fair-
child, T. R. 2003. Soft-sediment deformation at the base of the Neo-
proterozoic Puga cap carbonate (southwestern Amazon craton,
Brazil): confirmation of rapid icehouse to greenhouse transition in
snowball Earth. Geology, 31, 613– 616.
Norin, E. 1937. Geology of the western Quruq Tagh, eastern Tien Shan.
Reports of the Sino-Swedish Expedition III. Geology. Bokförlags
Aktiebolaget Thule, Stockholm.
North, G. R. 1990. Multiple solutions in energy balance climate models.
Palaeogeography, Palaeoecology, Palaeoclimatology (Global and
Planetary Change Section), 82, 225– 235.
Oberbeck, V. R., Marshall, J. R. & Aggarwal, H. 1993. Impacts, til-
lites, and the breakup of Gondwanaland. Journal of Geology, 101,
1 – 19.
Opdyke, B. N. & Walker, J. C. G. 1992. Return of the coral reef hypoth-
esis: basin to shelf partitioning of CaCO3 and its effect on atmos-
pheric CO2. Geology, 20, 733–736.
Opdyke, B. N. & Wilkinson, B. H. 1990. Palaeolatitude distribution of
Phanerozoic marine ooids and cements. Palaeogeography, Palaeocli-
matology, Palaeoecology, 78, 135–148.
Opdyke, N. D. 1962. Palaeoclimatology and continental drift. In:
Runcorn, S. K. (ed.) Continental Drift. Academic Press,
New York, 41– 65.
Oreskes, N. 1999. The Rejection of Continental Drift. Oxford University
Press, New York.
Osmaston, H. 2004. Quaternary glaciations in the East African moun-
tains. In: Ehlers, J. & Gibbard, P. L. (eds) Quaternary Glaciations
– Extent and Chronology, Part III. Elsevier, Amsterdam, 139–150.
Pais, M. A., Le Mouël, J. L., Lambeck, K. & Poirier, J. P. 1999. Late
Precambrian paradoxical glaciation and obliquity of the Earth – a
35
discussion of dynamical constraints. Earth and Planetary Sciences,
174, 155–171.
Park, J. K. 1994. Palaeomagnetic constraints on the position of Laurentia
from middle Neoproterozoic to Early Cambrian times. Precambrian
Research, 69, 95– 112.
Pavlov, A. A., Kasting, J. F., Brown, L. L., Rages, K. A. & Freedman,
R. 2000. Greenhouse warming by CH4 in the atmosphere of early
Earth. Journal of Geophysical Research, 105, 11 981– 11 990.
Pavlov, A. A., Hurtgen, M. T., Kasting, J. F. & Arthur, M. A. 2003.
Methane-rich Proterozoic atmosphere? Geology, 31, 87– 90.
Pavlov, A. A., Toon, O. B., Pavlov, A. K., Bally, J. & Pollard, D.
2005. Passing through a giant molecular cloud: ‘snowball’ glaciations
produced by interstellar dust. Geophysical Research Letters, 32,
L03705, doi: 10.1029/2004GL021890.
Peltier, W. R., Tarasov, L., Vettoretti, G. & Solheim, L. P. 2004.
Climate dynamics in deep time: modeling the ‘snowball bifurcation’
and assessing the plausibility of its occurrence. In: Jenkins, G. S.,
McMenamin, M. A. S., McKey, C. P. & Sohl, L. (eds) The
Extreme Proterozoic: Geology, Geochemistry, and Climate. Geophy-
sical Monograph, American Geophysical Union, Washington, DC,
146, 107–124.
Plummer, P. S. 1978. Note on the palaeoenvironmental significance of the
Nuccaleena Formation (upper Precambrian), central Flinders Ranges,
South Australia. Journal of the Geological Society of Australia, 25,
395– 402.
Pollard, D. & Kasting, J. F. 2005. Snowball Earth: a thin-ice solution
with flowing glaciers. Journal of Geophysical Research, 110,
C07010, doi: 10.1029/2004JC002525.
Powell, R. D. 1984. Glacimarine processes and inductive lithofacies
modelling of ice shelf and tidewater glacier sediments based on
Quaternary examples. Marine Geology, 57, 1– 52.
Powell, R. D. 1990. Glacimarine processes at grounding-line fans and
their growth to ice-contact deltas. In: Dowdeswell, J. A. &
Scourse, J. D. (eds) Glacimarine Environments: Processes and Sedi-
ments. Geological Society, London, Special Publication, 53, 53 –73.
Powell, R. D. & Domack, E. W. 1995. Modern glaciomarine environ-
ments. In: Menzies, J. (ed.) Glacial Environments, Vol. 1. Modern
Glacial Environments: Processes, Dynamics and Sediments.
Butterworth-Heinemann, Oxford, 445– 486.
Preiss, W. V. 1987. The Adelaide Geosyncline: Late Proterozoic stratigra-
phy, sedimentation, palaeontology and tectonics. Geological Survey
of South Australia Bulletin, 53, 438.
Rampino, M. R. 1994. Tillites, diamictites, and the ballistic ejecta of large
impacts. Journal of Geology, 102, 439– 456.
Ramsay, A. C. 1855. Angular, subangular, polished and striated fragments
and boulders in the Permian breccia of Shropshire, Worcestershire,
etc.; and on the probable existence of glaciers and icebergs in the
Permian epoch. Quarterly Journal of the Geological Society of
London, 11, 185– 205.
Ramsay, A. C. 1860. The Old Glaciers of Switzerland and North Wales.
Longman, Green, Longman & Roberts, London.
Rankama, K. 1973. The Late Precambrian glaciation, with particular
reference to the Southern Hemisphere. Journal and Proceedings,
Royal Society of New South Wales, 106, 89 –97.
Raub, T. D. & Evans, D. A. D. 2006. Magnetic reversals in basal
Ediacaran cap carbonates: a critical review. Eos, Transactions of
the American Geophysical Union, Joint Assembly Supplement, 87,
Abstract GP41– 02.
Reading, H. G. & Walker, R. G. 1966. Sedimentation of Eocambrian
tillites and associated sediments in Finnmark, northern Norway.
Palaeogeography, Palaeoclimatology, Palaeoecology, 2, 177– 212.
Reusch, H. 1891. Skuringmærker og morængrus eftervist i Finnmarken
fra en periode meget ældre end ‘istiden’ (Glacial striae and boulder-
clay in Norwegian Lapponie from a period much older than the last
ice age). Norges Geologiske Undersøkelse, 1, 78 –85 (English
summary, 97– 100).
Rice, A. H. N. & Hofmann, C.-C. 2000. Evidence for a glacial origin of
Neoproterozoic III striations at Oaibaccannjar’ga, Finnmark, north-
ern Norway. Geological Magazine, 137, 355– 366.
Ridgwell, A. J., Kennedy, M. J. & Caldeira, K. 2003. Carbonate depo-
sition, climate stability, and Neoproterozoic ice ages. Science, 302,
859– 862.
36 P. F. HOFFMAN
Robert, M. 1940. La glaciation du Kundelungu au Katanga (Congo
Belge). 17th International Geological Congress, Moscow, 1937,
Report 6, 99 – 113.
Roberts, J. D. 1971. Late Precambrian glaciation: an anti-greenhouse
effect? Nature, 234, 216.
Roberts, J. D. 1976. Late Precambrian dolomites, Vendian glaciation, and
synchroneity of Vendian glaciations. Journal of Geology, 84, 47 – 63.
Rodgers, J. 1957. The distribution of marine carbonate sediments: a
review. In: Le Blanc, R. J. & Breeding, J. G. (eds) Regional
Aspects of Carbonate Deposition. Society of Economic Paleontolio-
gists and Mineralogists (SEPM) Special Publication No. 5, 2 – 14,
Tulsa, Oklahoma.
Rogers, A. W. 1915. The geology of part of Namaqualand. Transactions
of the Geological Society of South Africa, 18, 72 –101, 14 plates.
Rosendahl, H. 1931. Bidrag til Varangernesets geologi. Norsk Geologisk
Tidsskrift, 25, 327– 349.
Rosendahl, H. 1945. Prekambrium-Eokambrium i Finnmark. Norsk
Geologisk Tidsskrift, 25, 327–349.
Rudwick, M. J. S. 1964. The infra-Cambrian glaciation and the origin of
the Cambrian fauna. In: Nairn, A. E. M. (ed.) Problems in Palaeo-
climatology. Interscience, London, 150–155, 184– 185.
Runcorn, S. K. 1961. Climatic change through geological time in light of
the palaeomagnetic evidence for polar wandering and continental
drift. Quarterly Journal of the Royal Meteorological Society, 87,
282– 313.
Sagan, C. & Mullen, G. 1972. Earth and Mars: evolution of atmospheres
and surface temperatures. Science 177, 52 –26.
Sayles, R. W. 1914. The Squantum Tillite. Bulletin of the Museum of
Comparative Zoology at Harvard College, 56, 141– 175, 12 plates.
Sayles, R. W. & LaForge, L. 1910. The glacial origin of the Roxbury
Conglomerate. Science, 32, 723–724.
Schermerhorn, L. J. G. 1974. Late Precambrian mixtites: glacial and/or
non-glacial? American Journal of Science, 274, 673– 824.
Schermerhorn, L. J. G. 1977. Late Precambrian dolomites, Vendian
glaciation, and synchroneity of Vendian glaciations: a discussion.
Journal of Geology, 85, 247– 250.
Schermerhorn, L. J. G. & Stanton, W. I. 1963. Tilloids in the West
Congo geosyncline. Quarterly Journal of the Geological Society of
London, 119, 201–241.
Schimper, K. 1837. Über die Eiszeit. (On the Ice Age.) Actes Société
Helvetique des Sciences Naturelles, Neuchâtel, 38– 51.
Schmidt, P. W. & Williams, G. E. 1995. The Neoproterozoic climatic
paradox: equatorial palaeolatitude for Marinoan glaciation near sea
level in South Australia. Earth and Planetary Science Letters, 134,
107– 124.
Schmidt, P. W., Williams, G. E. & Embleton, B. J. J. 1991. Low palaeo-
latitude of Late Proterozoic glaciation: early timing of remnance
in Haematite of the Elatina Formation, South Australia. Earth and
Planetary Science Letters, 105, 355–367.
Schiøtz, O. E. 1898. Om Dr Reusch’s präglaciale skuringdmerker. Nyt
Magazin Naturvissenschaften, 36, 1 –10.
Schrag, D. P., Berner, R. A., Hoffman, P. F. & Halverson, G. P. 2002.
On the initiation of a snowball Earth. Geophysics, Geochemistry,
Geosystems, 3, doi: 10.1029/2001GC000219.
Schwartzman, D. W. & Volk, T. 1991. Biotic enhancement of weather-
ing and surface temperatures since the origin of life. Palaeogeogra-
phy, Palaeoecology, Palaeoclimatology, 90, 357– 374.
Schwarzbach, M. 1964a. Criteria for the recognition of ancient glacia-
tions. In: Nairn, A. E. M. (ed.) Problems in Palaeoclimatology.
Interscience, London, 81 –85.
Schwarzbach, M. 1964b. Climates of the Past. D. Van Nostrand Co.,
London. (English translation of the 2nd revised edn of Das Clima
der Vorzeit. Ferdinand Enke Verlag, Stuttgart, 1961.)
Sellers, W. D. 1969. A global climatic model based on the energy
balance of the Earth-atmosphere system. Journal of Applied Meteor-
ology, 8, 392–400.
Sellers, W. D. 1990. The genesis of energy balance modeling and
the cool Sun paradox. Palaeogeography, Palaeoecology, Palaeocli-
matology (Global and Planetary Change Section), 82, 217–224.
Shaler, N. S. 1847. Preliminary report on the secent changes of level
on the coast of Maine: with reference to their origin and relation to
other similar changes. Memoirs of the Boston Society of Natural
History, 2, 321–323, 335– 340.
Sheldon, R. P. 1984. Ice-ring origin of the Earth’s atmosphere and
hydrosphere and late Proterozoic –Cambrian phosphogenesis.
Special Publication, Geological Survey of India, 17, 17 – 21.
Shields, G. A., Deynoux, M., Strauss, H., Paquet, H. & Nahon, D.
2007. Barite-bearing cap dolostone of the Taoudéni Basin, northwest
Africa: sedimentary and isotopic evidence for methane seepage after
a Neoproterozoic glaciation. Precambrian Research, 154, 209– 235.
Siever, R. 1968. Sedimentological consequences of a steady-state
ocean-atmosphere. Sedimentology, 11, 5 –29.
Smith, J. 1836. On indications of changes in the relative levels of sea and
land in the West of Scotland. Proceedings of the Geological Society,
London, 2, 427–429.
Smith, J. 1839. On the climate of the newer pliocene tertiary period.
Proceedings of the Geological Society of London, 3, 118– 119.
Sohl, L. E., Christie-Blick, N. & Kent, D. V. 1999. Paleomagnetic
polarity reversals in Marinoan (ca 600 Ma) glacial deposits of
Australia: implications for the duration of low-latitude glaciation in
Neoproterozoic time. Geological Society of America Bulletin, 111,
1120– 1139.
Spalletti, L. & Del Valle, A. 1984. Las diamictitas del sector oriental
de Tandilia: characteres sedimentológicos y origen. Revista de la
Asociación Geológica Argentina, 39, 188– 206.
Spencer, A. M. 1971. Late Pre-Cambrian glaciation in Scotland. Geologi-
cal Society of London Memoir, 6, 99.
Spjeldnaes, N. 1964. The Eocambrian glaciation in Norway. Geologische
Rundschau, 54, 24 –45.
Sprigg, R. C. 1986. The Adelaide Geosyncline: a century of controversy.
Earth Sciences History, 5, 66– 83.
Strahan, A. 1897. On glacial phenomena of Palaeozoic age in the Varan-
ger Fiord. Quarterly Journal of the Geological Society, London, 53,
137– 146, Discussion, 153–156.
Sugden, D. E. & John, B. S. 1976. Glaciers and Landscape: A Geomor-
phological Approach. Edward Arnold, London.
Sutherland, P. C. 1870. Notes on an ancient boulder-clay of
Natal. Quarterly Journal of the Geological Society of London, 26,
514– 517.
Thompson, M. D. & Bowring, S. A. 2000. Age of the Squantum ‘tillite’,
Boston Basin, Massachusetts: U– Pb zircon constraints on terminal
Neoproterozoic glaciation. American Journal of Science, 300,
630– 655.
Thomson, J. 1871. On the stratified rocks of Islay. Report of the 41st
Meeting of the British Association for the Advancement of Science,
Edinburgh. John Murray, London, 110– 111.
Thomson, J. 1877. On the geology of the island of Islay. Transactions of
the Geological Society of Glasgow, 5, 200–222.
Tyndall, J. 1861. On the absorbtion and radiation of heat by gases and
vapours, and on the physical connexion of radiation, absorbtion,
and conduction – the Bakerian Lecture. Philosophical Magazine
(series 4), 22, 169– 194, 273–285.
Tyndall, J. 1863. On the radiation through the Earth’s atmosphere. Phi-
losophical Magazine (series 4), 25, 200–206.
Venetz, I. 1830. Sur l’extension qu’il présume que les glaciers avaient
autrefois, et sur leur retraits dans leurs limites actuelles. Actes
Société Helvetique des Sciences Naturelles, Lausanne, 31.
Voigt, A., Abbot, D. S., Pierrehumbert, R. T. & Marotzke, J. 2011.
Initiation of a Marinoan Snowball Earth in a state-of-the-art atmos-
phere-ocean general circulation model. Climate of the Past, 7, 1– 15.
Walde, D. H. G., Gierth, E. & Leonardos, O. H. 1981. Stratigraphy and
mineralogy of the manganese ores of Urucum, Mato Grosso, Brazil.
Geologische Rundschau, 70, 1077– 1085.
Walker, G. 2003. Snowball Earth: The Story of the Great Global Cata-
strophe that Spawned Life As We Know It. Crown Publishers,
New York, 269.
Walker, J. C. G., Hays, P. B. & Kasting, J. F. 1981. A negative feedback
mechanism for the long-term stabilization of Earth’s surface tempera-
ture. Journal of Geophysical Research, 86(C10), 9776– 9782.
Wang, Y., Lu, S. & Gao, Z. 1981. Sinian tillites of China. In: Hambrey,
M. J. & Harland, W. B. (eds) Earth’s Pre-Pleistocene Glacial
Record. Cambridge University Press, Cambridge, 386–401.
 HISTORY OF NEOPROTEROZOIC GLACIAL GEOLOGY
Wegener, A. 1912. Die Entstehung der Kontinente. Petermanns Geogra-
phische Mitteilungen, 58, 185–195, 253– 256, 305–309. (Reprinted
in English translation: The origins of the continents. Journal of
Geodynamics, 32, 29– 63, 2001.)
Wegener, A. 1929. Die Enstehung der Kontinente und Ozeane, 4th edn.
Friedrich Vieweg & Sohn, Braunschweig. (Reprinted in English
translation by John Biram: The Origin of Continents and Oceans.
Dover Publications, New York, 246, 1966.)
Wegmann, C. E., Dangeard, L. & Graindor, M. J. 1950. Sue quelques
caractères remarquables de la formation pré-cambrienne connue
sous le nom de Poudinage de Granville. Compte Rendus, 230,
979– 980.
Wetherald, R. T. & Manabe, S. 1975. The effects of changing the Solar
constant on the climate of a general circulation model. Journal of the
Atmospheric Sciences, 32, 2044–2059.
Whitten, G. F. 1970. The investigation and exploitation of the Razorback
Ridge iron deposit. Geological Survey of South Australia Reports of
Investigations, 33, 165.
Williams, D. M., Kasting, J. F. & Frakes, L. A. 1998. Low-latitude
glaciation and rapid changes in the Earth’s obliquity explained by
obliquity – oblateness feedback. Nature, 396, 453–455.
Williams, G. E. 1972. Geological evidence relating to the origin
and secular rotation of the Solar system. Modern Geology, 3,
165– 181.
Williams, G. E. 1974. Discussion of Late Precambrian glacial climate and
the Earth’s obliquity. Journal of the Geological Society, London, 130,
599– 601.
Williams, G. E. 1975. Late Precambrian glacial climate and the Earth’s
obliquity. Geological Magazine, 112, 441–544.
Williams, G. E. 1979. Sedimentology, stable-isotope geochemistry and
palaeoenvironment of dolostones capping late Precambrian glacial
sequences in Australia. Journal of the Geological Society of Austra-
lia, 26, 377–386.
Williams, G. E. 1981. Sunspot periods in the late Precambrian glacial
climate and solar – planetary relations. Nature, 291, 624–628.
Williams, G. E. 1989. Late Precambrian tidal rhythmites in South Austra-
lia and the history of the Earth’s rotation. Journal of the Geological
Society, London, 146, 97 –111.
Williams, G. E. 1993. History of the Earth’s obliquity. Earth-Science
Reviews, 34, 1 –45.
Williams, G. E. 2000. Geological constraints on the Precambrian history
of Earth’s rotation and the Moon’s orbit. Reviews of Geophysics, 38,
37 – 59.
Williams, G. E. & Sonett, C. P. 1985. Solar signature in sedimentary
cycles from the late Precambrian Elatina Formation, Australia.
Nature, 318, 523– 527.
Williams, G. E. & Tonkin, D. G. 1985. Periglacial structures and palaeo-
climatic significance of a late Precambrian block field in the Cattle
Grid copper mine, Mount Gunson, South Australia. Australian
Journal of Earth Sciences, 32, 287–300.
Williams, G. E., Gostin, V. A., McKirdy, D. M. & Preiss, W. V. 2008.
The Elatina glaciation, late Cryogenian (Marinoan Epoch), South
Australia: sedimentary facies and palaeoenvironments. Precambrian
Research, 163, 307– 331.
Willis, B. 1904. Geological research in eastern Asia. Carnegie Institution
of Washington, Yearbook, 3, Washington, DC, 275–291.
37
Wilson, C. B. & Harland, W. B. 1964. The Polarisbreen Series and other
evidences of late pre-Cambrian ice ages in Spitsbergen. Geological
Magazine, 101, 198– 219.
Wilson, J. T. 1966. Did the Atlantic close and then reopen? Nature, 211,
676– 681.
Witzke, B. J. 1990. Palaeoclimatic constraints for Palaeozoic palaeolati-
tudes of Laurentia and Euramerica. In: McKerrow, W. S. &
Scotese, C. R. (eds) Palaeozoic Palaeogeography and Biogeogra-
phy. Geological Society of London Memoirs, 12, 57 –73.
Woodward, H. B. 1907. The History of the Geological Society of
London. Geological Society, London, 336. Reprinted in 1978
by Arno Press, New York.
Woodward, H. P. 1884. Report on the range to the east of Farina.
Parliamentary Papers for 1884. Government of South Australia,
40, 1– 5.
Worsley, T. R. & Kidder, D. L. 1991. First-order coupling of paleogeo-
graphy and CO2, with global surface temperature and its latitudinal
contrast. Geology, 19, 1161– 1164.
Wright, L., Williams, E. G. & Cloud, P. 1978. Algal and cryptalgal
structures and platform environments of the late pre-Phanerozoic
Noonday Dolomite, eastern California. Geological Society of
America Bulletin, 89, 321–333.
Wunsch, C. 2002. What is the thermohaline circulation? Science, 298,
1179–1180.
Xiao, S. 2004. Neoproterozoic glaciations and the fossil record. In:
Jenkins, G. S., McMenamin, M. A. S., McKey, C. P. & Sohl, L.
(eds) The Extreme Proterozoic: Geology, Geochemistry, and
Climate. Geophysical Monograph, American Geophysical Union,
Washington, DC, 146, 199–214.
Xiao, S., Bao, H. et al. 2004. The Neoproterozoic Quruqtagh Group in
eastern Chinese Tianshan: evidence for a post-Marinoan glaciation.
Precambrian Research, 130, 1– 26.
Yeo, G. M. 1981. The Late Proterozoic Rapitan glaciation in the northern
Cordillera. In: Campbel, F. H. A. (ed.) Proterozoic Basins of Canada.
Geological Survey of Canada Paper, 81– 10, 25 –46.
Young, G. M. 1976. Iron-formation and glaciogenic rocks of the Rapitan
Group, Northwest Territories, Canada. Precambrian Research, 3,
137– 158.
Young, G. M. 1995a. Are Neoproterozoic glacial deposits preserved on
the margins of Laurentia related to the fragmentation of two super-
continents? Geology, 23, 153–156.
Young, G. M. 1995b. Reply: Are Neoproterozoic glacial deposits pre-
served on the margins of Laurentia related to the fragmentation of
two supercontinents? Geology, 23, 1054– 1055.
Zachos, J., Pagani, M., Sloan, L. & Thomas, E. 2001. Trends, rhythms,
and aberrations in global climate 65 Ma to present. Science, 292,
686– 693.
Zhang, S., Jiang, G. & Han, Y. 2008. The age of the Nantuo
Formation and Nantuo glaciation in South China. Terra Nova, 20,
289– 294.
Ziegler, P. A. 1960. Frühpaläozoische Tillite in östlichen Yukon-
Territorium (Kanada). Eclogae Geologicae Helveticae, 52, 735– 741.
Ziegler, A. M., Hulver, M. L., Lottes, A. L. & Schmachtenberg, W.
F. 1984. Uniformitarianism and paleoclimates: inferences from the
distribution of carbonate rocks. In: Brenchley, P. J. (ed.) Fossils
and Climate. John Wiley & Sons, New York, 3 –25.
 Chapter 3

Recognition of glacial influence in Neoproterozoic sedimentary successions

EMMANUELLE ARNAUD1* & JAMES L. ETIENNE2

1
School of Environmental Sciences, University of Guelph, Guelph, Ontario, N1G 2W1, Canada
2
Neftex Petroleum Consultants Ltd, 97 Milton Park, Abingdon, Oxfordshire OX14 4RY, UK
*Corresponding author (e-mail: earnaud@uoguelph.ca)

Abstract: This chapter provides an overview and key references of glacial processes and resulting sedimentary products in subglacial,
terrestrial proglacial and glaciomarine or glaciolacustrine settings. These settings are characterized by a wide variety of processes ranging
from subglacial lodgement and deformation, ice-push and sediment remobilization, which in turn result in a wide range of products such
as diamictite, conglomerate, sandstone, siltstone and mudstone. The sedimentary record of proglacial settings exhibits the most lateral
and vertical variability due to the dynamic nature of ice margins and the most direct record of climatic fluctuations. Many Neoproterozoic
successions, however, preserve glaciomarine deposits that can provide a more continuous and high-resolution (though indirect) record of
change. This chapter will enable the reader to identify features that may be used to infer a glacial influence on the formation of ancient
deposits. The chapter also outlines some of the important issues that require consideration when evaluating palaeoclimatic models for
Neoproterozoic sedimentary successions. These include the equivocal significance of most commonly used proxies such as occurrence
of diamictite, outsized clasts in laminated sediments, clast characteristics, lithostratigraphic trends and sequence boundaries. Careful
analysis of multiple lines of sedimentary evidence, together with other proxies of climatic changes, can yield meaningful reconstructions
and provide a basis for testing palaeoclimate models for this time period. A summary table outlining the characteristics of diamictite with
different depositional origins is also included in order to assist with the interpretation of the Neoproterozoic sedimentary record.

The ability to recognize glacial deposits, or a glacial influence on
sedimentation, has long been a subject of debate (Schermerhorn
1974; Deynoux & Trompette 1976; Jensen & Wulff-Pedersen
1996), and yet is critical for the accurate reconstruction of
palaeoenvironmental conditions during past ice ages. This
chapter summarizes the historical development of the terminology
and approach used in identifying deposits of glacial origin, the
characteristics of glacial environments in terms of processes, the
resulting sedimentary facies and stratal geometries, as well as
some key points to consider when reconstructing palaeoenviron-
mental conditions in ancient successions. The chapter is far from
exhaustive, and many others have previously covered these
topics in more detail elsewhere. Rather, it is meant to provide an
overview with key references for workers who focus on other
aspects of Neoproterozoic geology and may not necessarily be
familiar with glacial deposits. As the focus of this book is on
Neoproterozoic ice ages, characteristics related to micro- or
macrofauna and biogenic indicators associated with glacial set-
tings are not discussed. Similarly, geomorphic features are only
briefly mentioned, as most Neoproterozoic successions do not
have the prerequisite outcrop exposure or preservation potential
of landforms found in more recent glacial settings.
Historical development
The terminology used to refer to glacial deposits has evolved over
the years, with an increasing appreciation of the processes occur-
ring in glaciated basins and a desire to refine palaeoenvironmental
interpretations to include more information about the severity of
climatic conditions, the nature of the ice mass, the record of associ-
ated sea-level changes and depositional setting (Flint et al. 1960;
Carey & Ahmad 1961; Dreimanis 1978; Hambrey & Harland
1978; Boulton & Deynoux 1981; Boulton 1990; Eyles 1993;
Hambrey & Glasser 2003). The criteria used to identify glacial
deposits in ancient sedimentary successions has also evolved
with this refined understanding of processes and a recognition of
the non-uniqueness of many of the sedimentary characteristics
that were once thought diagnostic of glacial deposits (Crowell
1964; Harland 1964; Harland et al. 1966; Schermerhorn 1974;
Flint 1975; Hambrey & Harland 1978; Boulton & Deynoux 1981;
Dreimanis & Schluchter 1985; Chumakov 1992; Eyles 1993;
Crowell 1999). This improved understanding has occurred as a
result of the emergence of modern sedimentological techniques,
which focus on process and products, and the application of
facies and basin analysis (Reading 1978), which have enhanced
our ability to reconstruct palaeoenvironmental conditions from
ancient sedimentary successions (see Eyles & Januszczak 2004
for a historical review).
For the purpose of this book, we encouraged authors describing
deposits from around the world to use the non-genetic term diamic-
tite to describe poorly sorted materials that contain a mixture of
gravel-, sand- and mud-sized particles (Fig. 3.1). Terms such as
‘tillite’ were to be reserved for the interpretation section where
poorly sorted materials could be shown to have been deposited
directly by ice, without significant subsequent disaggregation
and flow. In describing reworked glacial deposits, the interpretive
term ‘glaciogenic’ was recommended as a modifier (e.g. glacio-
genic debris flow). Authors were also encouraged to use ‘outsized
clast’ or ‘lonestone’ rather than the term ‘dropstone’ in their
descriptions, leaving the latter word to the discussion where a
dropstone origin could be interpreted based on deformation of
underlying laminae and onlapping overlying laminae. The use of
terms ‘varvite’ or ‘varve’ to describe laminated fine-grained sedi-
ment in glaciogenic successions was discouraged, considering the
lack of radiometric control required to confirm annual depositional
cycles. These are best described as ‘rhythmically laminated’ when
a cyclical nature can be identified or simply ‘laminated’ when the
stacking pattern appears random. Readers are referred to Hambrey
& Glasser (2003) for an up-to-date discussion of terminology
related to glacial sediments and Evans et al. (2006) for a more
specific discussion of subglacial till classification.
Characteristics of glacial environments
Deposition occurs within a number of different settings in gla-
ciated basins, including subglacial, proglacial, glaciolacustrine
and glaciomarine environments. Each of these depositional set-
tings has characteristic processes, deposits and landforms

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 39– 50. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.3
40 E. ARNAUD & J. L. ETIENNE
Fig. 3.1. Classification of poorly sorted sediments according to Hambrey & Glasser (2003). Reproduced with permission from Springer.
(Menzies 1995, 1996; Benn & Evans 1998, 2010; Evans 2005).
Sedimentary analysis of Neoproterozoic successions can thus
provide much-needed palaeoenvironmental information when
this variability is taken into consideration. In contrast to more
recent glacial deposits, landforms tend not to be preserved or are
poorly exposed in ancient successions, although there are excep-
tions. An understanding of the processes, together with the result-
ing deposits and their three-dimensional distribution in these
settings, is thus critical for palaeoenvironmental reconstruction.
Glaciomarine successions tend to be the most common in the
Neoproterozoic geological record, as marine basins provide
accommodation space away from subsequent ice advances and
erosion, allowing preservation of thick packages of glaciogenic
sediments (Carey & Ahmad 1961; Nystuen 1985; Eyles 1993).
Although these settings tend to provide the longest and highest res-
olution records of Neoproterozoic climate change, the depocentres
tend to be dislocated from the glaciated shelves, which acted as
the sediment supply pathway, complicating efforts to reconstruct
the distribution of palaeo-ice masses. An appreciation of other
glacial settings is important as some successions will record con-
ditions at the basin margin, close to land-based glaciers (Rieu
et al. 2006; Arnaud 2008). Basin margin successions such as
these are particularly useful in terms of palaeoclimate reconstruc-
tion as they can provide information regarding the nature of the ice
mass (stagnant, active or fast-moving) and allow better control on
the relative timing of the changes associated with that ice margin.
They may also provide a better sense of the temporal relationship
between glacial conditions and palaeoenvironmental perturbations
inferred from geochemical data that have recently been so central
to Neoproterozoic research.
A discussion of the various settings in glaciated basins follows
with particular emphasis on the nature of these environments,
the processes that operate within them and the resulting sedimen-
tary products. References provided herein are by no means exhaus-
tive, but were selected as they focus on sedimentary characteristics
and emphasize their connection to processes, thus facilitating the
reconstruction of palaeoenvironmental conditions from the sedi-
mentary record. Readers are referred to the ‘glaciers online’ site
of Alean & Hambrey (2008) as well as Eyles (1993), Crowell
(1999), Hambrey & Glasser (2003) and Etienne et al. (2007) for
helpful photographs and additional general discussions of criteria
used in the recognition of ancient glacial deposits.
Subglacial settings
Conditions in subglacial settings are primarily controlled by the
relationship between pressure and temperature, which ultimately
controls the transition between solid and liquid phases and the
alternation between melting and regelation in the basal traction
zone. In other instances, ice will melt and refreeze in response to
broader temperature fluctuations. These subglacial conditions in
turn control the ability of the ice to erode, transport and deposit
debris. The thermal characteristics of ice masses thus have a
direct impact upon the accumulation of subglacial debris, with
polythermal (variable distribution of warm and cold-based ice)
and temperate glaciers (basal ice close to pressure melting point)
accumulating the most significant volumes of debris. Although
cold-based glaciers have been shown to entrain, transport and
deposit sediment, they generally accumulate significantly less
debris. In some cases, subglacial lakes may form, as shown by
recent inventories of subglacial lakes in Antarctica (Siegert et al.
2005). Satellite data have demonstrated significant ice elevation
changes associated with these subglacial lakes, and these have
been attributed to major discharge events and reorganization of
subglacial meltwater (Wingham et al. 2006; Fricker et al. 2007).
Subglacial processes that are most relevant for palaeoenviron-
mental reconstructions include lodgement or plastering of debris
onto the substrate, deformation of soft-sediment substrates, hydro-
fracturing of substrate (Boulton & Caban 1995), plucking of
angular material from hardrock substrates as a result of freeze
and thaw, abrasion of both the debris within the ice and the under-
lying substrate, and meltwater erosion. Meltwater processes are
particularly significant in temperate glacial settings. The meltwater
can be found in subglacial distributary channels, which comprise
part of the drainage system of the glacier, or as a thin film at the
base of the glacier, which helps facilitate movement.
These processes are responsible for subglacial deposits such as
lodgement and deformation tills (Boyce & Eyles 2000; Evans et al.
2006), commonly characterized by diamicton (the unlithified
equivalent of diamictite), in addition to gravel and sand deposited
in subglacial meltwater tunnels (Fig. 3.2; Table 3.1). Whereas
deposition related to meltout or undermelting of ice in a subglacial
setting is theoretically possible, studies have shown that it is very
difficult to distinguish these deposits from other subglacial depos-
its considering the high likelihood of subsequent overprinting by
 RECOGNITION OF GLACIAL DEPOSITS 41

Fig. 3.2. Photographs of typical glacial sedimentary indicators in Neoproterozoic successions. (a) Mud-rich diamictite with extra-basinal clasts; (b) outsized clast
showing clear piercing and onlap relationship typical of dropstones; (c) striated pavement, Varangerfjorden, Norway; (d) striated clast; (e) bullet-shaped clast; (f) faceted
clast. It should be noted that none of these features are exclusive to glacial conditions, so careful analysis of other features is necessary to confirm a glacial origin. See text
for discussion.

deformation and lodgement processes (Evans et al. 2006). Subgla-
cial processes also result in striated pavements, and clasts that
show evidence of transport and erosion in the basal traction zone
such as striations, faceting, chattermarks and a bullet shape (also
referred to as stoss and lee clasts, flat-iron clasts or bullet-shaped
boulders; Fig. 3.2, Table 3.1). Care must be taken to distinguish
between striations arising from subglacial erosion and those that
are related to slickenside development associated with faulting
or pressure solution. Diamictite units may exhibit a gradational
change from deformed and sheared at the base through chaotically
stratified to massive at the top, indicative of the increase in stress
with proximity to the ice – sediment interface (Boulton & Hind-
marsh 1987; Hart & Boulton 1991; Boulton 1996; Benn &
Evans 1996). Other sedimentary characteristics that can be indic-
ative of subglacial settings are boulder pavements (boulders
preferentially aligned one behind the other) and soft-sediment
deformation features such as boudinage, tectonic lamination,
rooted recumbent folding and inclusions of underlying material
that exhibit shear stress in a preferred direction (Table 3.1; Hart
& Boulton 1991; Hart & Roberts 1994; McCarroll & Rijsdijk
2003; Roberts & Hart 2005; Arnaud 2008). Clastic dykes may
also develop in areas where a frozen substrate and the overlying
ice mass result in high pore water pressures (Boulton & Caban
1995; Le Heron & Etienne 2005). Preferred orientation of the
a-axes of clasts (fabric) has also been used to discriminate
between various depositional origins of diamictite (Domack &
Lawson 1985; Hart 1994; Benn & Evans 1996); however, others
have argued that clast fabric is not always indicative of a particular
depositional setting and that other indicators of glacial conditions
are needed to confirm interpretations (Bennett et al. 1999;
Hambrey & McKelvey 2000). Preferred clast orientations can be
very useful in determining direction of transport, which may in
turn facilitate palaeoenvironmental discrimination (Table 3.1).
Clast fabric analyses are somewhat more difficult for Neoprotero-
zoic successions that have undergone multiple phases of post-
depositional deformation. Finally, it is difficult to say what
sedimentary expression subglacial lakes will have in the rock
record and how their existence may be inferred without the
 42
Table 3.1. Characteristics of diamictite units of different origin

Types of diamictite Textural features Compositional Sedimentary Deformation Basal contact Geometry Associated deposits Useful references
features structures

Lodgment & Poorly sorted Mixed provenance Massive or Glaciotectonic Sharp Thin (,m to m) Well bedded conglomerate Benn & Evans (1996)
deformation tillite* Boulder pavements (intra- and chaotically deformation such as Erosional or and tabular and sandstone, commonly Boulton (1972)
Evidence of glacial extrabasinal clasts) stratified shear structures (rooted deformed Laterally cross-stratified and Boyce & Eyles
transport on particle Compositional Gradational upward folds, boudinage, Basal shear plane extensive exhibiting cut and fill (2000)
surface† immaturity change from highly attenuated bedding, Grooves or (hundreds of geometry (glaciofluvial McCarroll & Rijsdijk
Textural immaturity Inclusions/ intraclasts deformed to tectonic laminae), deformation in km2) outwash) (2003)
Wide range of stone or rafts of other massive nappes, imbricate sub-till Discontinuous if Thick deposits of windblown Hart & Roberts
roundness sediment types that Preferred orientation structures, thrust or sediments close to ice silt (loess), sandstone (1994)

E. ARNAUD & J. L. ETIENNE

Micromorphological exhibit ductile of shear structures reverse faulting, shear margin wedges, convoluted Menzies (2000)
characteristics deformation planes and decollement bedding from Evans et al. (2006)
associated with high surfaces. cryoturbation and in situ
stress conditions Clastic dykes breccia‡ (periglacial)
Strong a-axis fabrics Laminated or massive
(lodgement) mudstone with/without
outsized clasts (lacustrine)
Meltout tillite § Poorly sorted Mixed provenance Crudely stratified Extensional faulting from Conformable Rapid lateral Well bedded conglomerate Boulton (1972)
Localized sorting from (intra- and (discontinuous melting of buried ice Sharp facies changes and sandstone, commonly Lawson (1982)
meltwater reworking extrabasinal clasts) layers of sorted Convoluted bedding and Undisturbed cross-stratified and McDonald & Shilts
Evidence of glacial Compositional sediments folding substrate exhibiting cut and fill (1975)
transport on particle immaturity or draped layers geometry (glaciofluvial Evans et al. (2006)
surface such as Soft sediment or soft over large clasts) outwash)
striations bedrock inclusions Laminated or massive
Textural immaturity or intraclasts that mudstone with or without
Wide range of stone exhibit no outsized clasts from
roundness deformation ponded water or lake
deposits
‘Rainout’ deposits Poorly sorted Mixed provenance Massive to crudely Grounding ice berg Gradational Variable Lodgement till from ice Eyles et al. (1985)
Evidence of glacial (intra- and stratified structures and ice-keel thickness advance onto shelf Thomas & Connell
transport on particle extrabasinal clasts) Discontinuous furrows Tabular to Sediment gravity flow (1985)
surface† Compositional stringers of sand blanket-like deposits
Textural immaturity immaturity and gravel (boulder geometry, Laminated mudstone with/
Wide range of stone Clast clusters from pavements) from draping without outsized clasts
roundness iceberg dumping winnowing underlying Subaquatic fan deposits in
Diamict pellets from Variable clast content topography proximal settings
frozen debris on from changing Laterally Pelagic clastic or carbonate
icebergs sedimentation rate continuous sediments
Dropstones or supply of (distal) to rapid
ice-rafted debris lateral facies
changes
(proximal)
Terrestrial debris flowsPoorly sorted Local (non-glacial) or Massive to Flow nose Erosional with Extent (m– km) Lodgement or meltout tills Lawson (1982)
Evidence of glacial mixed (glaciogenic chaotically Basal shearing structures rip-up clasts Benvenuti & Martini
transport on particle source) provenance stratified Convoluted bedding or (2002)
surface* if glaciogenic Compositional folding Colella & Prior
source maturity depending (1990)
Textural maturity on source
depending on source Inclusions or rafts of
Wide range of stone other sediment
roundness depending types that exhibit
on source ductile deformation
Subaqueous debris flows Poorly sorted Local (non-glacial) or Massive to Flow noses Erosional with Moderate to thick Lodgement tills Nardin et al. (1979),
Evidence of glacial mixed (glaciogenic chaotically Basal shear structures rip-up clasts Tabular to ‘Rainout’ deposits Visser (1983),
transport on particle source) provenance stratified Folding and convoluted Conformable lenticular Laminated mudstone with/ Prior et al. (1984),
surface† Compositional Graded bedding bedding when Channelized without outsized clasts Postma et al.
Textural immaturity maturity depending Coarse-tail grading Load casts, water escape hydroplaning Extent (m to 10s Subaquatic fan deposits (1988), Hart &
Wide range of stone on source structures has occurred of km) Pelagic clastic or carbonate Roberts (1994),
roundness Inclusions or rafts of sediments Mohrig et al.

RECOGNITION OF GLACIAL DEPOSITS

other sediment Other sediment gravity flow (1999), Eyles &
types that exhibit deposits (e.g. turbidites) Eyles (2000),
ductile Hambrey &
deformation, some McKelvey (2000),
can be very large Laberg & Vorren
(10s of m in size) (2000), Mulder &
Alexander (2001),
McCarroll &
Rijsdijk (2003)

Note: No one of these features are sufficient for identification of a glacial or glacially influenced origin and any one of the features may be absent.
*We combine these two types of till, as lodgement and deformation processes will often both affect individual deposits (see discussion in Evans et al. 2006); Evans et al. (2006) argued that meltout till generated in subglacial
settings could similarly not be easily distinguished).
†
Evidence of glacial transport: striations, chattermarks, polish, grooves, facets, bullet shape (also referred to as ‘stoss and lee’ or ‘flat iron’ clasts).
‡
These sedimentary characteristics are not unique to periglacial settings; their origin may be difficult to confirm even with careful stratigraphic and sedimentological consideration.
§
Although meltout of debris is theoretically possible in subglacial settings, their identification in the rock record seems unlikely (Evans et al. 2006). Here, we only consider meltout tillite in the context of proglacial terrestrial settings
(see text for discussion).

43
44 E. ARNAUD & J. L. ETIENNE
benefit of the geomorphic context that is available in modern envi-
ronments. Studies to date have been based on remote sensing data,
so very little is known about the types of deposits within these
lakes and their longer-term preservation potential. Their unequivo-
cal identification in the rock record would appear difficult, as
deposits would look very similar to glaciomarine successions
recording advances and recessions of ice. In Neoproterozoic suc-
cessions, biota are not available to distinguish between the lacus-
trine and marine settings.
In terms of three-dimensional geometry, lodgement and defor-
mation tills tend to be tabular, relatively thin (on the scale of less
than a metre to several metres) and can cover extensive areas (typi-
cally hundreds of km2). Basal contacts can be erosional or show
evidence of deformation and incorporation of materials from the
underlying substrate. Conglomerate and sandstone from subglacial
tunnels will have sharp outer contacts and a pipe-like geometry
in cross-section. The nature of the associated sediments will
depend on whether or not this ice mass overrode land or a
marine shelf. Terrestrial subglacial deposits may be associated
with coarse-grained glaciofluvial outwash sediments and other
terrestrial proglacial sediments such as glaciolacustrine facies
(Boulton 1972), whereas glaciomarine subglacial deposits will
be associated with subaqueous proglacial sediments, tidal rhyth-
mites, calcareous, biogenic or clastic fine-grained sediments
deposited by suspension settling and/or a range of fine- to coarse-
grained sediment gravity flow deposits (e.g. turbidites and hyper-
concentrated flows; Domack 1988; Cowan & Powell 1990;
Anderson & Ashley 1991; Fairchild 1993; Dowdeswell et al.
1996, 1998; Anderson 1999; Taylor et al. 2002).
Terrestrial proglacial settings
Conditions in proglacial settings are primarily controlled by the
proximity to the ice margin and whether or not the ice margin is
stagnant, actively advancing or surging. Proglacial processes
include in situ meltout of debris, remobilization of debris down-
slope, reworking by meltwater and melting of buried stagnant
ice (Boulton 1972; Lawson 1982). These processes typically
result in highly irregular topography, ponding of meltwater and
development of proglacial lakes. The entire proglacial succession
may be deformed and cannibalized by subsequent ice advances
that bulldoze, thrust and re-work the deposits. Such events are
known from modern environments (Huddart & Hambrey 1996)
and the ancient glacial geological record (Le Heron et al. 2005).
Stagnating ice margins will often have multiple recessional mor-
aines and abundant supraglacial debris with buried ice. In contrast,
active ice margins will be characterized by multiple push moraines
and ridges associated with the fluctuating ice margin (Boulton
et al. 1999; Bennett 2001; Bennett et al. 2004).
Highly variable depositional conditions in terrestrial proglacial
environments (ponding, meltwater flow, bulldozing and remobi-
lization of sediments) tend to result in complex lithofacies assem-
blages with a high degree of lateral variability. These assemblages
may include diamictite (derived from meltout, subaerial debris
flow and bulldozing and deformation of glaciogenic debris;
Table 3.1), conglomerate, sandstone, siltstone and mudstone
(meltwater processes) and laminated mudstone and siltstone
(ponding) in these settings. Clasts may show evidence of recent
transport in the basal traction zone such as faceting, striations,
chattermarks and bullet-shape, but surface features rarely survive
fluvial transport. Deformation structures ranging in scale from
centimetres to hundreds of metres can be quite prevalent in these
proglacial settings (Benn & Evans 1996; Phillips et al. 2002).
Extensional faulting and ductile folding of sediments associated
with in situ melting of ice can be found in all facies types. In
addition, reverse or thrust faults, shear planes and low-angle décol-
lement surfaces, nappes and complex folding and shear structures
can be found resulting from bulldozing by ice.
The wide range in depositional conditions associated with
terrestrial proglacial environments often leads to an inherently
complex distribution of sediment types and three-dimensional geo-
metries that vary from irregular to channelized and lenticular with
common intraformational unconformities and sudden bed termin-
ation. Basal contacts are variable from sharp to irregular and
deformed or erosional to conformable and draping over irregular
topography. Associated sediments are likely to be subglacial and
glaciofluvial outwash deposits.
Glaciolacustrine and glaciomarine settings
Conditions in glacially influenced water bodies are primarily con-
trolled by their proximity to the ice margin, whether or not the
margin terminates in the water body, the nature of the ice margin
(i.e. tidewater glacier or floating ice shelf), the thermal regime of
the ice mass (e.g. temperate v. polar) and the characteristics of
the water body (lake v. ocean; deep sea, shelf, fjord or shoreline;
Smith & Ashley 1985; Eyles et al. 1985; Powell & Elverhoi
1989; Dowdeswell & Scourse 1990; Anderson & Ashley 1991;
Dowdeswell et al. 1998; Evans & Pudsey 2002). Lakes that are
dammed by ice or sediment (such as moraines) are also regulated
by dam stability (Tweed & Russell 1999). Glacial processes dom-
inate in settings that are close to the ice margin (,kms), whereas
normal lacustrine/marine processes dominate in settings that are
distal to the ice margin (.tens of km). Where ice terminates in a
water body (ice-contact lakes or tidewater glacier marine settings),
depositional settings are affected by melt-out of subglacial and
englacial debris, slumping off the ice margin, iceberg calving,
efflux from subglacial and englacial meltwater conduits and sedi-
ment input from surface meltwater drainage. These processes
supply sediment to grounding line fan complexes and more
distal areas. In contrast, water bodies fed by glacial meltwater
are primarily influenced by diurnal and seasonal fluctuations in
meltwater discharge and seasonal ice cover. Meltwater processes
are very important in settings influenced by temperate glaciers
and limited to non-existent in polar settings. The configuration
of the basin will influence the types of non-glacial processes
that also contribute to the distribution and nature of deposits
found in these settings. For example, lakes may experience
thermal stratification and seasonal mixing, whereas deposition in
oceans will be affected by tides and salinity. Sediment gravity
flows resulting from high sedimentation rates are common in
most water bodies but may be more prevalent in steep-sided
lakes or fjords. However, it is important to note that subaqueous
glaciogenic debris flows may have run-out distances of hundreds
of kilometres, even on shallow slopes (Dowdeswell et al. 1996,
1998; Taylor et al. 2002). Oscillatory waves will affect shallow
water settings, while storm waves will affect both shallow and
shelf settings.
Processes include rapid deposition of coarse-grained sediments
at the mouth of subglacial tunnels, density underflows and
suspension-settling of fine-grained sediments related to buoyant
meltwater plumes and sub-ice shelf melting, winnowing or
reworking by sub-ice shelf or open water currents as well as bull-
dozing, lodgement and deformation of sediments related to ice
advances (Lønne 1995; Ó Cofaigh & Dowdeswell 2001; Evans
et al. 2005; McKay et al. 2008; Naish et al. 2009). Remobilization
of debris by gravity flows is common in these settings and results
from high sedimentation rates, oversteepening along the edges of
grounding-line fans, and normal sediment-transfer processes at
the shelf-slope break or along steep-sided fjords. Gravity flows
may also be initiated directly by subaqueous iceberg calving, or
as a result of fluctuating water depths associated with calving or
abrupt lake drainage events. Icebergs can have an impact on
both proximal and distal settings through iceberg roll-over and
dumping, meltout of entrained debris and scouring and bulldozing
when grounded in shallow water (Thomas & Connell 1985; Gilbert
 RECOGNITION OF GLACIAL DEPOSITS
1990; Woodworth-Lynas & Guigne 1990). Deposition of cold-
water carbonates is also common in glacially influenced polar
settings. Ikaite (hydrated calcium carbonate) is one such mineral
that forms in near-freezing water temperatures. Unstable at
higher temperatures, ikaite has poor preservation potential and
typically undergoes replacement by calcite. Where the original
crystal structure is preserved, these pseudomorphs are called
glendonites and are an important palaeoenvironmental indicator
of cold-water conditions (James et al. 2005; Domack et al. 2007;
Halverson 2011).
Processes such as those identified above often lead to a predomi-
nance of diamictite and mudstone in glacially influenced lacus-
trine/marine settings, although conglomerate and sandstone can
also be found associated with grounding-line fans and subglacial
tunnel outflow as well as stratified facies related to sub-ice shelf
sedimentation or reworking by sub-ice shelf or open water currents
(Eyles et al. 1985; Lønne 1995; Nemec et al. 1999; Barrett 2007;
McKay et al. 2008). In open water conditions where icebergs are
calving or in sub ice-shelf settings, suspension settling of fine-
grained sediment and rainout of ice-rafted debris can lead to the
formation of diamictite (Table 3.1) or mudstone with outsized
clasts. Where outsized clasts rafted from icebergs deform under-
lying laminated sediments (truncating laminae, loading sediment
to develop compaction-related folds and/or water escape struc-
tures) they may be referred to as dropstones (Fig. 3.2; Thomas &
Connell 1985). Material of pebble size or greater is generally con-
sidered as a good proxy for ice-rafting, although some care is
needed in interpretation (see discussion below). In some cases,
frozen masses of ice-rafted debris are released from icebergs into
water bodies and result in till pellets or inclusions of diamict
within either fine-grained sediments or a diamict with slightly
different characteristics (Gilbert 1990). These can be distinguished
from intraclasts resulting from erosion at the base of sediment
gravity flows by closely examining the type of materials and
their relationship to underlying materials, the nature of the outer
boundary of these inclusions (sharp and irregular for erosional
intraclast v. diffuse for till pellets) as well as the basal contact of
the unit in which they are found. Ice advance onto the continental
shelf can also result in the deposition of lodgement or deformation
tillite (Table 3.1), as demonstrated by overconsolidated diamicton
displaying shear structures, clastic dykes and associated subglacial
landforms such as grooves, lineations and drumlins in recent con-
tinental shelf sediments of Antarctica (Camerlenghi et al. 2001;
Ó Cofaigh et al. 2002; Evans et al. 2005; Naish et al. 2009). Dia-
mictite may of course also form as a result of sediment gravity
flows (Table 3.1; Wright & Anderson 1980; Eyles & Eyles 2000;
Mulder & Alexander 2001). Rhythmically laminated mudstone
may result from tidal pumping in ice-proximal marine settings
(Smith et al. 1990) or from seasonal discharge fluctuations in
non-ice contact lakes (Leonard 1986). The latter are referred to
as varves in more recent deposits, but the use of this term is dis-
couraged in Neoproterozoic successions as a seasonal control
cannot be clearly demonstrated due to limited geochronological
constraints. Although Evans & Pudsey (2002) suggested that ice-
shelf deposits could be distinguished from tidewater glacier depos-
its by the absence of meltwater-derived facies, similar stratified
sediments have been reported from ice-shelf settings in Antarctica
and attributed to grounding line zones (Evans et al. 2005; McKay
et al. 2008; Naish et al. 2009), making it unlikely that the two will
be easy to distinguish in ancient glaciogenic successions in the
absence of biogenic facies, which are sensitive to ice cover, or con-
textual knowledge of the ice margin available in more recent gla-
ciomarine environments.
Laminated or massive fine-grained sediments with rare clasts
dominate in ice distal settings where typical marine processes
dominate and suspension settling from icebergs provide a marginal
contribution to sedimentation. Boulder pavements may develop on
marine shelves as a result of winnowing of diamict and overriding
by ice during sea-level lowstands (Eyles 1994). Deformation may
45
be found in ice-proximal grounding line fan deposits as a result of
slumping and bulldozing by ice, in remobilized glaciogenic debris-
flow deposits along the continental slope and in other deposits
where iceberg grounding has occurred (Table 3.1; Woodworth-
Lynas & Guigne 1990; Hart & Roberts 1994).
The record of climatic changes over time in glacial
environments
Ancient glaciogenic sedimentary successions contain a record of
fluctuating climatic conditions over time (Boulton 1990). Ice
ages typically consist of interstadials (relatively warmer) and sta-
dials (relatively colder) periods, which are recorded in the partial
advance and recession of ice and changes in the lateral and vertical
distribution of sediments associated with these fluctuating environ-
ments. Such fluctuations have been documented in several Neopro-
terozoic successions (Allen et al. 2004; Arnaud & Eyles 2006), as
well as in more recent glacial successions (Domack 1983; Barrett
2007; Naish et al. 2009). Changes from glacial to non-glacial
periods will similarly be recorded at the margins of marine
basins. Here, glacioisostacy (crustal movements associated with
ice loading and unloading) and glacioeustasy (changes in global
ocean volume associated with changes in terrestrial ice volume)
will combine with local tectonics and sediment supply to influence
relative sea level and ultimately the nature of stratal geometries
(Ravnås & Steel 1998; Powell & Cooper 2002; Allen 2007).
Glacio-eustatic variability can lead to significant correlative
global cycles of deposition.
Glaciations affect various parts of the world and can be
expressed quite differently depending on the thermal character-
istics of the ice mass mentioned above. These characteristics can
vary spatially and temporally. Today, temperate glaciers (i.e.
warm ice close to melting point) are found in Alaska, polythermal
glaciers (mixture of warm and cold ice) in Greenland and polar gla-
ciers (cold ice well below melting point) in Antarctica. However,
these thermal characteristics can change over time, as shown by the
sedimentary record of the Ross Ice Shelf, which shows temperate
conditions in the past compared to the polar conditions of today
(Naish et al. 2009).
Reconstructing Neoproterozoic palaeoenvironmental
conditions: key considerations
The sedimentary record is at times difficult to interpret as many, if
not most, sedimentary characteristics are not unique to specific
environments. The following discussion highlights key points to
consider when reconstructing palaeoenvironmental conditions
from Neoproterozoic successions, focusing on indicators typically
used to infer glacial conditions.
The palaeoclimatic significance of diamictite
Diamictite can result from both glacial and non-glacial processes
(Table 3.1). In glaciated settings, diamictite results from lodge-
ment and deformation of subglacial sediments, in situ meltout of
ice and release of poorly sorted basal, englacial or supraglacial
debris, slumping and reworking of glaciogenic debris in proglacial,
glaciolacustrine or glaciomarine settings and from ‘rainout’ of
fine-grained sediment and ice-rafted debris. In non-glaciated set-
tings, diamictite can form as a result of terrestrial or subaqueous
debris flows (hyperconcentrated flows) or as a result of impact
ejecta (Fisher 1971; Nardin et al. 1979; Rampino 1994; Martins-
Neto 1996; Sohn 2000; Weaver et al. 2000; Mulder & Alexander
2001; Benvenuti & Martini 2002). Studies of modern glaciated set-
tings and of sediment gravity flows and their deposits have greatly
improved our ability to link depositional processes with their
46 E. ARNAUD & J. L. ETIENNE
sedimentary products. However, careful consideration of sedi-
mentary characteristics, geometry, as well as lateral and vertical
distribution of diamictite and associated deposits, is needed to dis-
tinguish between these various depositional origins (Arnaud &
Eyles 2006; Table 3.1).
For example, a diamictite formed as a result of ‘rainout’ may
have a gradational basal contact, whereas those related to sedi-
ment gravity flows or subglacial lodgement tend to have sharp con-
formable or erosional basal contacts (Table 3.1). In a marine
setting, diamictite showing draped upper contact may be attributed
to a sediment gravity flow rather than subglacial processes.
Diamictite formed as a result of subglacial lodgement may
exhibit upwardly increasing deformation (Benn & Evans 1996),
in contrast to deposits of sediment gravity flows where the
maximum deformation will occur at the base of the deposit
(Nardin et al. 1979; Mulder & Alexander 2001). Associated
facies play a particularly key role in distinguishing between terres-
trial or marine settings in these ancient successions where faunal
indicators are lacking.
Careful analysis is particularly important in considering
ancient glaciogenic successions, as many of them were deposited
in tectonically active settings where diamictite may record tectoni-
cally induced sediment instability (Krzyszkowski 1993; Tanaka &
Maejima 1995) rather than renewed glacial conditions. Sediment–
landform associations may not be available in these basins to better
constrain a glacial influence.
Even with the most careful examination, the depositional origin of
diamictite will sometimes be difficult to determine unequivocally as
indicators are sometimes absent or obscured by post- depositional
tectonism or weathering. This is particularly significant considering
that Neoproterozoic deposits will have been subjected to a long
history of post-depositional modification with sometimes very
limited and poor preservation in structurally complex settings.
Weathering is problematic when looking for surface markings on
carbonate clasts (e.g. striations, faceting or bullet-shape) within Neo-
proterozoic diamictite, a common problem considering that many
Neoproterozoic glaciogenic successions overlie carbonate strata
and have incorporated these softer lithologies.
In other situations, interpretation will be confounded by the
fact that the poorly sorted debris has been affected by a number
of different processes, and therefore retains a composite sedi-
mentary signature of these processes. In their comprehensive
review of subglacial processes and products, Evans et al. (2006)
suggested that it will be difficult to distinguish between different
types of tillites in ancient glacial deposits, as diamictite will
ultimately record a composite of these subglacial processes and
the dynamic nature of subglacial systems as they change over
time and space. In another example, glaciomarine deposits could
be subsequently over-ridden by ice, preserving some character-
istics of both glaciomarine and subglacial processes and thus
making it very difficult to identify whether these represent open
glaciomarine or subglacial conditions. Another equally difficult
situation is where large-scale controls on the sedimentary suc-
cession can leave similar signatures in the rock record such as
diamictites formed in offshore Alaska during the late Miocene
(Eyles et al. 1991), which were affected by both climatic and
tectonic controls and Neoproterozoic volcaniclastic deposits in
the Arabian –Nubian shield (Miller et al. 2009), which may be
related to glaciogenic and/or volcaniclastic processes. Debates
over the palaeoclimatic significance of the Neoproterozoic
diamictite in Namibia, for example, revolve in part around the
relative importance of tectonic and climatic controls on sedimen-
tation and the different processes ascribed to these deposits.
It is important to acknowledge these uncertainties where rel-
evant, because very different palaeoenvironmental conditions are
implied depending on the depositional process responsible for
the formation of diamictite. In places where it is possible, careful
consideration of the sedimentary characteristics and spatial geome-
try of the diamictite and its associated deposits can greatly refine
our depositional models by distinguishing between different
types of depositional process, glacial setting (e.g. terrestrial
proglacial v. glaciomarine/glaciolacustrine) and thermal regime,
which can in turn contribute much needed constraints on evaluating
palaeoclimate models that have been proposed for this time period
(Condon et al. 2002; Leather et al. 2002; Allen et al. 2004; Arnaud
& Eyles 2006; Rieu et al. 2006; Allen & Etienne 2008).
The palaeoclimatic significance of outsized clasts in
bedded sediments
Outsized clasts in laminated sediments have often been interpreted
as ice-rafted debris or ‘dropstones’ and a clear indication of glacial
conditions. Piercing or disturbance of underlying laminae and
draping of sediment on top of the outsized clasts are probably
the best criteria to imply that the clast was dropped from floating
ice (Thomas & Connell 1985). Without this deformation and
onlap relationship, the outsized clast may in fact be an outrunner
in a succession of turbidites and other sediment gravity flows
(Postma et al. 1988) or a clast surrounded by lamination resulting
from post-depositional compaction or deformation.
Once a ‘dropstone’ origin has been established, care must still
be taken in terms of inferring glacial conditions from the presence
of these outsized clasts. Ice-rafted debris may be derived from
seasonal ice unrelated to any ice margin, proximal icebergs or
far-travelled icebergs (Gilbert 1990). Distinction of seasonal ice,
proximal icebergs and far-travelled icebergs is very unlikely in
ancient successions. It is therefore important to recognize that
the occurrence of dropstones can have varying palaeoclimatic
significance (cold climate if resulting from seasonal ice, locally gla-
ciated or glaciated some distance away if resulting from icebergs).
This is an important consideration when using the sedimentary
record to test competing palaeoclimate models for this time period.
The palaeoclimatic significance of clast characteristics
Clast characteristics suggesting transport in the basal traction zone of
glaciers are also often used to infer glacial conditions. These include
the presence of striations, chattermarks, faceting, bullet-shaped
boulders or flat iron clasts and extra-basinal provenance. It is impor-
tant to note however, that although these confirm a glaciogenic influ-
ence on transport history, contextual analysis is required to ensure
that the clasts have not been re-worked into younger deposits
lacking a direct glacial influence. Although it is true that such
surface markings are unlikely to survive with significant transport
in meltwater glaciofluvial systems, many glaciomarine deposits
(which tend to be the most common in ancient successions) may
not have undergone significant fluvial reworking. This is particularly
relevant when considering the amount of time elapsed between the
deposition of a diamictite thought to record glacial conditions and
its associated ‘cap’ carbonate thought to record warmer post-glacial
conditions in Neoproterozoic successions.
The palaeoclimatic significance of stratigraphic trends and
sequence boundaries
Many studies of glacially influenced successions have equated
alternating beds of diamictite and other sediments as the product
of multiple advances and recessional phases of ice margins and
multiple diamictite-bearing intervals of successions separated by
thick sections of non-diamictite bearing sediments as representing
discrete ice ages. Although this may turn out to be true (Naish et al.
2009), such interpretations must be made with caution considering
the various depositional origins that can be ascribed to diamictite,
the possibility that the succession at times is too distal to the ice
margin to preserve any clear indicator of glacial conditions (e.g.
relatively high-order transgressive or highstand systems tracts
 RECOGNITION OF GLACIAL DEPOSITS
characterized by ice-distal mudstones in glaciomarine settings
with limited or no icebergs), and the lack of geochronological
control to constrain the amount of time represented by those
thick non-diamictite bearing successions.
The lateral and vertical heterogeneity of lithofacies within
glacially influenced successions can also be highly complex. In
both terrestrial and shallow marine (neritic) environments,
regional sequence boundaries may be developed as a result of sub-
glacial erosion either directly by ice via processes of regelation,
abrasion from subglacially entrained debris, or by meltwater drai-
nage under normal or elevated hydrostatic pressures. Multiple
glacial advance-recessional cycles across marine shelf areas may
result in stacked sequence boundaries, with lowstand packages
of sediment in slope to bathyal environments providing the only
continuous record of glacial – interglacial cyclicity. A cryptic
shallow marine record provided by multistorey stacking of sub-
glacial drainage conduits such as tunnel valleys may also be
preserved, although these are not well documented from Neopro-
terozoic successions (see Ordovician examples in Le Heron
et al. 2009). More localized disconformities may develop in
marine environments as a result of iceberg keel-scour and pro-
cesses of submarine mass-wasting related to high glaciomarine
sedimentation rates or local tectonic activity. In terrestrial progla-
cial environments, the suite of erosion processes associated with
braided fluvial systems, glaciotectonic deformation, and incision
generated by glacial lake outburst floods or jökulhlaups resulting
from subglacial eruptions can all lead to local or regional sequence
boundary development. This picture can become even more com-
plicated where thin cold-based ice occurs in accumulation areas
and pre-existing sediments such as periglacial deposits can be
preserved with little or no modification, effectively leading to a
correlative surface lacking erosion.
In addition, as ancient glaciogenic successions tend to be prefer-
entially preserved in basins that were tectonically active, the evol-
ution of the basin must be well constrained in order to assess the
impacts on accommodation space, sediment supply, gross-scale
stratigraphic architecture and characteristics of sequence bound-
aries (Allen 2007).
The points above highlight some of the considerations when
applying a lithostratigraphic or sequence stratigraphic analysis
within glacially influenced successions.
Conclusions
Criteria for the recognition of glacial deposits in Neoproterozoic
sedimentary successions have developed over the last 50 years
with an increasing appreciation for the different kinds of settings
and processes that occur within glaciated basins. The analysis of
modern glacial settings and processes has allowed us to better
recognize the complexity of glaciated basins and the various
challenges in confirming a glacial influence on deposition. This
includes a better appreciation of the processes occurring in glacio-
marine settings as well as the difficulty in distinguishing the nature
of floating ice producing dropstones (local iceberg, distant iceberg
or sea ice), the depositional origin of diamictite (glacial, glacially
influenced or non-glacial) and the controls on large-scale
stratigraphic trends in facies successions (alternating lithofacies
assemblages associated with autocyclicity or eustatic trends of
transgression and regression).
Poorly sorted deposits (diamictite), outsized clasts in laminated
lithofacies (dropstones), high lateral facies variability in proximal
settings and the presence of clasts that show evidence of subglacial
transport are common features in glacial settings and these remain
common indicators used in the study of ancient glaciogenic succes-
sions. However, recent studies have shown that such sedimentary
characteristics can have widely different climatic implications
such that an understanding of the processes and products of
glacial settings and sediment gravity flows as well as careful
47
evaluation of sedimentary characteristics, geometry of deposits,
deformation styles and associated facies are needed to identify a
glaciogenic source and reconstruct the extent and nature of
glacial conditions at the time of deposition.
E. Arnaud’s research in glacial sedimentology is funded by the Natural Sciences
and Engineering Research Council. J. L. Etienne’s work follows research experi-
ence on modern, Pleistocene and Neoproterozoic glacial deposits funded under
NERC Studentship NER/S/A/2000/03690/ and Schweizerischer National-
Fonds Grant 103502. E. Arnaud thanks Carolyn Eyles (McMaster University).
J.L. Etienne thanks Mike Hambrey, Neil Glasser (University of Aberystwyth),
Dan Le Heron (Royal Holloway) and Philip Allen (Imperial College London).
Constructive reviews from David Evans and Eugene Domack greatly improved
the manuscript and are much appreciated. This work represents a contribution
of the IUGS- and UNESCO-funded IGCP (International Geoscience Programme)
Project #512.
References
Alean, J. & Hambrey, M. J. 2008. ‘Earth’s glacial record’ in Glaciers
Online. (Available at http://www.swisseduc.ch/glaciers/earth_icy_
planet/glaciers15-en.html.)
Allen, P. A. 2007. The Huqf Supergroup of Oman: basin development
and context for Neoproterozoic glaciation. Earth-Science Reviews,
84, 139– 185.
Allen, P. A. & Etienne, J. L. 2008. Sedimentary challenge to snowball
Earth. Nature Geoscience, 1, 817– 825.
Allen, P. A., Leather, J. & Brasier, M. D. 2004. The Neoproterozoic
Fiq glaciation and its aftermath, Huqf Supergroup of Oman. Basin
Research, 16, 507– 534.
Anderson, J. B. 1999. Antarctic Marine Geology. Cambridge University
Press, Cambridge.
Anderson, J. B. & Ashley, G. M. (eds) 1991. Glacial marine sedimen-
tation; paleoclimatic significance. Geological Society of America
Special Paper, 261, 232.
Arnaud, E. 2008. Deformation in the Neoproterozoic Smalfjord
Formation, northern Norway: an indicator of glacial depositional
conditions? Sedimentology, 55, 335–356.
Arnaud, E. & Eyles, C. H. 2006. Neoproterozoic environmental change
recorded in the Port Askaig Formation, Scotland: climatic vs tectonic
controls. Sedimentary Geology, 183, 99 –124.
Barrett, P. J. 2007. Cenozoic climate and sea level history from glaci-
marine strata off the Victoria Land coast, Cape Roberts Project,
Antarctica. In: Hambrey, M. J., Christoffersen, P., Glasser, N.
F. & Hubbard, B. (eds) Glacial Sedimentary Processes and Pro-
ducts. International Association of Sedimentologists Special Paper,
39, 259– 287.
Benn, D. I. & Evans, D. J. A. 1996. The interpretation and classification
of subglacially-deformed materials. Quaternary Science Reviews, 15,
23– 52.
Benn, D. I. & Evans, D. J. A. 1998. Glaciers and Glaciation. Arnold
Publishers, London.
Benn, D. I. & Evans, D. J. A. 2010. Glaciers and Glaciation, 2nd edn.
Hodder, London.
Bennett, M. R. 2001. The morphology, structural evolution and signifi-
cance of push moraines. Earth Science Reviews, 53, 197–236.
Bennett, M. R., Waller, R. I., Glasser, N. F., Hambrey, M. J. &
Huddart, D. 1999. Clast fabrics: genetic fingerprint or wishful
thinking? Journal of Quaternary Science, 14, 125–135.
Bennett, M. R., Huddart, D. et al. 2004. Sedimentary and tectonic
architecture of a large push moraine: a case study from Hagafjellsjo-
kull-Eystri, Iceland. Sedimentary Geology, 172, 269– 292.
Benvenuti, M. & Martini, I. P. 2002. Analysis of terrestrial hypercon-
centrated flows and their deposits. Special Publications International
Association of Sedimentologists, 32, 167– 193.
Boulton, G. S. 1972. Modern arctic glaciers as depositional models for
former ice sheets. Journal of the Geological Society, London, 128,
361– 393.
Boulton, G. S. 1990. Sedimentary and sea-level changes during
glacial cycles and their control on glacimarine facies architecture.
In: Dowdeswell, J. A. & Scourse, J. D. (eds) Glacimarine
48 E. ARNAUD & J. L. ETIENNE
Environments: Processes and Sediments. Geological Society,
London, Special Publications, 53, 15– 53.
Boulton, G. S. 1996. Theory of glacial erosion, transport and deposition
as a consequence of subglacial sediment deformation. Journal of
Glaciology, 140, 43 –62.
Boulton, G. S. & Caban, P. 1995. Groundwater flow beneath ice sheets:
Part II — its impact on glacier tectonic structure and moraine for-
mation. Quaternary Science Reviews, 14, 563– 587.
Boulton, G. S. & Deynoux, M. 1981. Sedimentation in glacial environ-
ments and the identification of tills and tillites in ancient sedimentary
sequences. Precambrian Research, 15, 397– 422.
Boulton, G. S. & Hindmarsh, R. C. A. 1987. Sediment deformation
beneath glaciers: rheology and sedimentological consequences.
Journal of Sedimentary Research, 92, 9059–9082.
Boulton, G. S., Van Der Meer, J. J. M., Beets, D. J., Hart, J. K. &
Ruegg, G. H. J. 1999. The sedimentary and structural evolution of
a recent push moraine complex: Holmstrombreen, Spitsbergen.
Quaternary Science Reviews, 18, 339–371.
Boyce, J. I. & Eyles, N. 2000. Architectural element analysis applied to
glacial deposits: internal geometry of a late Pleistocene till sheet,
Ontario, Canada. Geological Society of America Bulletin, 112,
98 –118.
Camerlenghi, A., Domack, E. et al. 2001. Glacial morphology and
post glacial contourites in northern Prince Gustav Channel (NW
Weddell Sea, Antartica). Marine Geophysical Researches, 22,
417– 443.
Carey, S. W. & Ahmad, N. 1961. Glacial marine sedimentation. In:
Raasch, G. O. (ed.) Geology of the Arctic. Volume II. University
of Toronto Press, Toronto, 865– 894.
Chumakov, N. M. 1992. The Problems of Old Glaciations (Pre-
Pleistocene Glaciogeology in the USSR). Harwood Academic Pub-
lishers, New York.
Colella, A. & Prior, D. B. (eds) 1990. Coarse-Grained Deltas. Inter-
national Association of Sedimentologists Special Publication, 10.
Blackwell Scientific Publications, Oxford.
Condon, D. J., Prave, A. R. & Benn, D. J. 2002. Neoproterozoic glacial-
rainout intervals: Observations and implications. Geology, 30,
35 –38.
Cowan, E. A. & Powell, R. D. 1990. Suspended sediment transport and
deposition of cyclically interlaminated sediment in a temperate
glacial fjord, Alaska, U.S.A. In: Dowdeswell, J. A. & Scourse,
J. D. (eds) Glacimarine Environments: Processes and Sediments.
Geological Society, London, Special Publications, 53, 75– 89.
Crowell, J. C. 1964. Climatic significance of sedimentary deposits con-
taining dispersed megaclasts. In: Nairn, A. E. M. (ed.) Problems in
Palaeoclimatology. Interscience Publishers, London, 86 – 99.
Crowell, J. C. 1999. Pre-Mesozoic Ice Ages: their Bearing on Under-
standing the Climate System, 192. Geological Society of America,
Memoir, 106.
Deynoux, M. & Trompette, R. 1976. Late Precambrian mixtites: glacial
and/or non-glacial? Dealing especially with the mixtites of West
Africa. American Journal of Science, 276, 1302–1324.
Domack, E. W. 1983. Facies of late Pleistocene glacial-marine sediments
on Whidbey Island, Washington: an istostatic glacial marine
sequence. In: Molnia, B. F. (ed.) Glacial-Marine Sedimentation.
Plenum, New York, 535–570.
Domack, E. W. 1988. Biogenic facies in the Antarctic glaciomarine
environment: basis for a polar glaciomarine summary. Palaeogeogra-
phy, Palaeoclimatology, Palaeoecology, 63, 357– 372.
Domack, E. W. & Lawson, D. E. 1985. Pebble fabric in an ice-rafted
diamicton. Journal of Geology, 93, 577– 591.
Domack, E. W., Halverson, G. P., Willmott, V., Leventer, A.,
Brachfield, S. & Ishman, S. 2007. Spatial and temporal distri-
bution of ikaite crystals in Antarctic glacial marine sediments.
United States Geological Survey and The National Academies,
USGS OF-2007-1047, Extended Abstract 015, 5.
Dowdeswell, J. A. & Scourse, J. D. (eds) 1990. Glaciomarine Environ-
ments: Processes and Sediments. Geological Society, London,
Special Publications, 53.
Dowdeswell, J. A., Elverhøi, A. & Spielhagen, R. 1998. Glacimarine
sedimentary processes and facies on the polar North Atlantic
Margins. Quaternary Science Reviews, 17, 243–272.
Dowdeswell, J. A., Kenyon, N. H., Elverhøi, A., Laberg, J. S.,
Hollender, F.-J., Mienert, J. & Siegert, M. J. 1996. Large-scale
sedimentation on the glacier-influenced Polar North Atlantic margins:
Long-range side-scan sonar evidence. Geophysical Research Letters,
23, 3535– 3538.
Dreimanis, A. 1978. The problems of waterlain tills. In: Schluchter, C.
(ed.) Moraines and Varves. A. A. Balkema Publishers, Rotterdam,
167– 177.
Dreimanis, A. & Schluchter, C. 1985. Field criteria for the recognition
of till and tillite. Palaeogeography, Palaeoclimatology, Palaeoecol-
ogy, 51, 7– 14.
Etienne, J. L., Allen, P. A., Rieu, R. & Le Guerroué, E. 2007.
Neoproterozoic glaciated basins: a critical review of the Snowball
Earth hypothesis by comparison with Phanerozoic basins. In:
Hambrey, M. J., Christoffersen, P., Glasser, N. F. & Hubbard,
B. (eds) Glacial Sedimentary Processes and Products. International
Association of Sedimentologists Special Publication, 39, 343–399.
Evans, D. J. A. (ed.) 2005. Glacial Landsystems. Oxford University Press.
Evans, J. & Pudsey, C. J. 2002. Sedimentation associated with Antarctic
Peninsula ice shelves: implications for palaeoenvironmental recon-
structions of glaciomarine sediments. Journal of the Geological
Society, London, 159, 233–237.
Evans, D. J. A., Phillips, E. R., Hiemstra, J. F. & Auton, C. A. 2006.
Subglacial till: Formation, sedimentary characteristics and classifi-
cation. Earth Science Reviews, 78, 115– 176.
Evans, J., Pudsey, C. J., O’Cofaigh, C., Morris, P. & Domack, E. 2005.
Late Quaternary glacial history, flow dynamics and sedimenta-
tion along the eastern margin of the Antarctic Peninsula Ice Sheet.
Quaternary Science Reviews, 24, 741– 774.
Eyles, C. H. 1994. Intertidal boulder pavements in the northeastern Gulf
of Alaska and their geological significance. Sedimentary Geology,
88, 161– 173.
Eyles, N. 1993. Earth’s glacial record and its tectonic setting. Earth-
Science Reviews, 35, 1 –248.
Eyles, C. H. & Eyles, N. 2000. Subaqueous mass flow origin for Lower
Permian diamictites and associated facies of the Grant Group, Barb-
wire Terrace, Canning Basin, Western Australia. Sedimentology, 47,
343– 356.
Eyles, N. & Januszczak, N. 2004. ‘Zipper-rift’: a tectonic model for
Neoproterozoic glaciations during the breakup of Rodina after 750
Ma. Earth-Science Reviews, 65, 1– 73.
Eyles, C. H., Eyles, N. & Lagoe, M. B. 1991. The Yakataga Formation: a
six million year record of temperate glacial marine sedimentation
in the Gulf of Alaska. In: Anderson, J. B. & Ashley, G. M. (eds)
Glacial Marine Sedimentation: Paleoclimatic Significance. Geologi-
cal Society of America Special Paper, 261, 159– 180.
Eyles, C. H., Eyles, N. & Miall, A. D. 1985. Models of glaciomarine
sedimentation and their application to the interpretation of ancient
glacial sequences. Palaeogeography, Palaeoclimatology, Palaeo-
ecology, 51, 15 –84.
Fairchild, I. 1993. Balmy shores and icy wastes: The paradox of carbon-
ates associated with glacial deposits in Neoproterozoic times. Sedi-
mentology Review, 1, 1– 16.
Fisher, R. V. 1971. Features of coarse-grained, high concentration
fluids and their deposits. Journal of Sedimentary Petrology, 41,
916– 927.
Flint, R. F. 1975. Features other than diamicts as evidence of ancient gla-
ciations. In: Wright, A. E. & Morsley, F. (eds) Ice Ages: Ancient
and Modern. Seelhouse Press, Liverpool, 121–135.
Flint, R. F., Sanders, J. E. & Rodgers, J. 1960. Diamictite, a substitute
term for symmictite. Bulletin of the Geological Society of America,
71, 1809– 1810.
Fricker, H. A., Scambos, T., Bindschadler, R. & Padman, L. 2007. An
active subglacial water system in West Antarctica mapped from
space. Science, 315, 1544–1548.
Gilbert, R. 1990. Rafting in glacimarine environments. In: Dowdes-
well, J. A. & Scourse, J. D. (eds) Glacimarine Environments:
Processes and Sediments. Geological Society, London, Special Pub-
lications, 53, 105– 121.
Halverson, G. P. 2011. Glacial sediments and associated strata of the
Polarisbreen Group, Northeastern Svalbard. In: Arnaud, E.,
Halverson, G. P. & Shields-Zhou, G. (eds) The Geological
 RECOGNITION OF GLACIAL DEPOSITS
Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 571–579.
Hambrey, M. J. 1994. Glacial Environments. UCL Press, London.
Hambrey, M. J. & Glasser, N. F. 2003. Glacial sediments: processes,
environments and facies. In: Middleton, G. V. (ed.) Encyclopedia
of Sediments and Sedimentary Rocks. Kluwer Academic Publishers,
Dordrecht, 316–331.
Hambrey, M. J. & Harland, W. B. 1978. Analysis of Pre-Pleistocene
glacigenic rocks: aims and problems. In: Schluchter, C. (ed.)
Moraines and Varves. A. A. Balkema Publishers, Rotterdam,
271– 275.
Hambrey, M. J. & McKelvey, B. 2000. Neogene fjordal sedimentation
on the western margin of the Lambert Graben, East Antarctica. Sedi-
mentology, 47, 577– 607.
Harland, W. B. 1964. Critical evidence for a great infra-cambrian glacia-
tion. Geologisches Rundschau, 54, 45– 61.
Harland, W. B., Herod, K. N. & Krinsley, D. H. 1966. The definition
and identification of tills and tillites. Earth-Science Reviews, 2,
225– 256.
Hart, J. K. 1994. Till fabric associated with deformable beds. Earth
Surface Processes and Landforms, 19, 15 – 32.
Hart, J. K. & Boulton, G. S. 1991. The interrelation of glaciotectonic
and glaciodepositional processes within the glacial environment.
Quaternary Science Reviews, 10, 335– 350.
Hart, J. K. & Roberts, D. H. 1994. Criteria to distinguish between
subglacial glaciotectonic and glaciomarine sedimentation, I.
Deformation styles and sedimentology. Sedimentary Geology, 91,
191– 213.
Huddart, D. & Hambrey, M. J. 1996. Sedimentary and tectonic develop-
ment of a high-arctic, thrust-moraine complex: Comfortlessbreen,
Svalbard. Boreas, 25, 227– 243.
James, N. P., Narbonne, G. M., Dalrymple, R. W. & Kyser, C. 2005.
Glendonites in Neoproterozoic low-latitude interglacial sedimentary
rocks, northwest Canada: insights on the Cryogenian ocean and
Precambrian cold carbonates. Geology, 33, 9 – 12.
Jensen, P. A. & Wulff-Pedersen, E. 1996. Glacial or non-glacial origin
of the Bigganjargga tillite, Finnmark, northern Norway. Geological
Magazine, 133, 137– 145.
Krzyszkowski, D. 1993. Pleistocene glaciolacustrine sedimentation in
a tectonically-active zone, Kleszczow Graben, central Poland.
Sedimentology, 40, 623– 644.
Laberg, J. S. & Vorren, T. O. 2000. Flow behaviour of the submarine
glacigenic debris flows on the Bear Island Trough Mouth Fan,
western Barents Sea. Sedimentology, 47, 1105–1117.
Lawson, D. E. 1982. Mobilization, movement and deposition of active
subaerial sediment flows, Matanuska Glacier, Alaska. Journal of
Geology, 90, 279–300.
Leather, J., Allen, P. A., Brasier, M. D. & Cozzi, A. 2002. Neoproter-
ozoic snowball Earth under scrutiny: evidence from the Fiq glaciation
of Oman. Geology, 30, 891– 894.
Le Heron, D. P. & Etienne, J. L. 2005. A complex subglacial clastic
dyke swarm, Sólheimajökull, southern Iceland. Sedimentary
Geology, 181, 25 –37.
Le Heron, D. P., Craig, J. & Etienne, J. L. 2009. Ancient glaciations and
hydrocarbon accumulations in North Africa and the Middle East.
Earth Science Reviews, 93, 47 –76.
Le Heron, D. P., Sutcliffe, O. E., Whittington, R. J. & Craig, J.
2005. The origins of glacially-related soft-sediment deformation
structures in Upper Ordovician glaciogenic rocks: implication for
ice sheet dynamics. Palaeogeography, Palaeoclimatology, Palaeo-
ecology, 218, 75 – 103.
Leonard, E. M. 1986. Varve studies at Hector Lake, Alberta, Canada, and
the relationship between glacial activity and sedimentation. Quatern-
ary Research, 25, 199– 214.
Lønne, I. 1995. Sedimentary facies and depositional architecture of ice-
contact glaciomarine systems. Sedimentary Geology, 98, 13 –43.
Martins-Neto, M. A. 1996. Lacustrine fan-deltaic sedimentation in a
Proterozoic rift basin: the Sopa-Brumadinho Tectonosequence,
southeastern Brazil. Sedimentary Geology, 106, 65 –96.
McCarroll, D. & Rijsdijk, K. F. 2003. Deformation styles as a key for
interpreting glacial depositional environments. Journal of Quatern-
ary Science, 18, 473–489.
49
McDonald, B. C. & Shilts, W. W. 1975. Interpretation of faults in gla-
ciofluvial sediments. In: Jopling, A. V. & McDonald, B. C. (eds)
Glaciofluvial and Glaciolacustrine Sedimentation. SEPM Special
Publication, 123–131.
McKay, R. M., Dunbar, G. B., Naish, T. R., Barrett, P. J., Carter, L.
& Harper, M. 2008. Retreat history of the Ross Ice Sheet (Shelf)
since the last glacial maximum from deep-basin sediment cores
around Ross Island. Palaeogeography, Palaeoclimatology, Palaeo-
ecology, 260, 245– 261.
Menzies, J. (ed.) 1995. Modern Glacial Environments: Processes,
Dynamics and Sediments. Butterworth-Heinemann, Oxford.
Menzies, J. (ed.) 1996. Past Glacial Environments: Sediments, Forms and
Techniques. Butterworth-Heinemann, Oxford.
Menzies, J. 2000. Microstructures within diamictites of the Lower Gow-
ganda Formation (Huronian) near Elliot Lake, Ontario. Journal of
Sedimentary Research, 35, 210– 216.
Miller, N. R., Stern, R. J., Avigad, D., Beyth, M. & Schilman, B.
2009. Cryogenian slate-carbonate sequences of the Tambien Group,
northern Ethiopia (I): pre-sturtian chemostratigraphy and corre-
lations. Precambrian Research, 170, 129– 156.
Mohrig, D., Elverhoi, A. & Parker, G. 1999. Experiments on the rela-
tive mobility of muddy subaqueous and subaerial debris flows and
their capacity to remobilize antecedent deposits. Marine Geology,
154, 117–129.
Moncrieff, A. M. 1989. Classification of poorly sorted sedimentary
rocks. Sedimentary Geology, 65, 191– 194.
Mulder, T. & Alexander, J. 2001. The physical character of subaqueous
sedimentary density flows and their deposits. Sedimentology, 48,
269– 299.
Naish, T., Powell, R. et al. 2009. Obliquity-paced Pliocene West
Antarctic ice sheet oscillations. Nature, 458, 322–328.
Nardin, T. R., Hein, F. J., Gorsline, D. S. & Edwards, B. D. 1979.
A review of mass movement processes, sediment and acoustic
characteristics and contrasts in slope and base of slope
systems versus canyon-fan basin floor systems. In: Doyle, L. J. &
Pilkey, O. H. (eds) Geology of Continental Slopes. Society
of Economic Paleontologists and Mineralogists, Special Publication,
61– 73.
Nemec, W., Lønne, I. & Blikra, L. H. 1999. The Kregnes moraine in
Gauldalen, west central Norway: anatomy of a Younger Dryas pro-
glacial delta in a palaeofjord basin. Boreas, 28, 254–476.
Nystuen, J. P. 1985. Facies and preservation of glaciogenic sequences
from the Varanger ice age in Scandinavia and other parts of the
North Atlantic Region. Palaeogeography, Palaeoclimatology,
Palaeoecology, 51, 209–229.
´ Cofaigh, C. & Dowdeswell, J. A. 2001. Laminated sediments in gla-
O
cimarine environments: diagnostic criteria for their interpretation.
Quaternary Science Reviews, 20, 1411– 1436.
´ Cofaigh, C., Pudsey, C. J., Dowdeswell, J. A. & Morris, P. 2002.
O
Evolution of subglacial bedforms along a paleo-ice stream, Antarctic
Peninsula continental shelf. Geophysical Research Letters, 29,
1199–1203.
Phillips, E. R., Evans, D. A. J. & Auton, C. A. 2002. Polyphase
deformation at an oscillating ice margin following the Loch
Lomond Readvance, central Scotland, U.K. Sedimentary Geology,
149, 157–182.
Postma, G., Nemec, W. & Kleinspehn, K. L. 1988. Large floating clasts
in turbidites: a mechanism for their emplacement. Sedimentary
Geology, 58, 47 –61.
Powell, R. D. & Cooper, J. M. 2002. A glacial sequence stratigraphic
model for temperate glaciated continental shelves. In: Dowdeswell,
J. A. & O ´ Cofaigh, C. (eds) Glacier-Influenced Sedimentation on
High Latitude Continental Margins. Geological Society, London,
Special Publications, 203, 215–244.
Powell, R. D. & Elverhoi, A. (eds) 1989. Modern glacimarine environ-
ments: Glacial and marine controls of modern lithofacies and bio-
facies. Marine Geology, 85, 416.
Prior, D. B., Bornhold, B. D. & Johns, M. W. 1984. Depositional
characteristics of a submarine debris flow. Journal of Geology, 92,
707– 727.
Rampino, M. R. 1994. Tillites, diamictites and ballistic ejecta of large
impacts. The Journal of Geology, 102, 439–456.
50 E. ARNAUD & J. L. ETIENNE
Ravnås, R. & Steel, R. J. 1998. Architecture of marine rift-basin succes-
sions. American Association of Petroleum Geologists Bulletin, 82,
110– 146.
Reading, H. G. 1978. Sedimentary Environments and Facies, 1st edn.
Blackwell Scientific Publications, Oxford.
Rieu, R., Allen, P. A., Etienne, J. L., Cozzi, A. & Weichert, U. 2006.
A Neoproterozoic glacially influenced basin margin succession and
‘atypical’ cap carbonate associated with bedrock paleovalleys,
Mirbat area, southern Oman. Basin Research, 18, 471– 496.
Roberts, D. H. & Hart, J. K. 2005. The deforming bed characteristics of
a stratified till assemblage in north East Anglia, U.K.: investigating
controls on sediment rheology and strain signatures. Quaternary
Science Reviews, 24, 123–140.
Schermerhorn, L. J. G. 1974. Late Precambrian mixtites: glacial and/or
non-glacial. American Journal of Science, 274, 673–824.
Siegert, M. J., Carter, S., Tabacco, I., Popov, S. & Blankenship, D. D.
2005. A revised inventory of Antarctic subglacial lakes. Antarctic
Science, 17, 453–460.
Smith, N. D. & Ashley, G. M. 1985. Proglacial lacustrine environments.
In: Ashley, G. M., Shaw, J. & Smith, N. D. (eds) Glacial Sedimen-
tary Environments. SEPM Short Course Notes, 16, 135– 216.
Smith, N. D., Phillips, A. C. & Powell, R. D. 1990. Tidal drawdown:
A mechanism for producing cyclic sediment laminations in glacio-
marine deltas. Geology, 18, 10 –13.
Sohn, Y. K. 2000. Depositional process of submarine debris flows in the
Miocene fan deltas, Pohang Basin, S.E. Korea, with special reference
to flow transformation. Journal of Sedimentary Research, 70,
491– 503.
Tanaka, J. & Maejima, W. 1995. Fan-delta sedimentation on the basin
margin slope of the Cretaceous, strike-slip Izumi Basin, southwestern
Japan. Sedimentary Geology, 98, 205– 213.
Taylor, J., Dowdeswell, J. A., Kenyon, N. H. & O’Cofaigh, C. 2002.
Late Quaternary architecture of trough-mouth fans: debris flows and
suspended sediments on the Norwegian margin. In: Dowdeswell,
J. A. & O ´ Cofaigh, C. (eds) Glacier-Influenced Sedimentation on
High Latitude Continental Margins. Geological Society, London,
Special Publications, 203, 55– 71.
Thomas, G. S. P. & Connell, R. J. 1985. Iceberg drop, dump and ground-
ing structures from Pleistocene glacio-lacustrine sediments, Scotland.
Journal of Sedimentary Petrology, 55, 243– 249.
Tweed, F. S. & Russell, A. J. 1999. Controls on the formation and
sudden drainage of glacier-impounded lakes: implications for
jökulhaup characteristics. Progress in Physical Geography, 23,
79 – 110.
Visser, J. N. J. 1983. The problems of recognizing ancient subaqueous
debris flow deposits in glacial sequences. Transactions of the Geo-
logical Society of South Africa, 86, 127– 135.
Weaver, P. P. E., Wynn, R. B., Kenyon, N. H. & Evans, J. 2000. Con-
tinental margin sedimentation, with special reference to the north-east
Atlantic margin. Sedimentology, 47(Suppl. 1), 239– 256.
Wingham, D. J., Siegert, M. J., Sheperd, A. & Muir, A. S. 2006. Rapid
discharge connects Antarctic subglacial lakes. Nature, 440,
1033– 1036.
Woodworth-Lynas, C. M. T. & Guigne, J. Y. 1990. Iceberg scours in
the geological record: examples from glacial Lake Agassiz. In:
Dowdeswell, J. A. & Scourse, J. D. (eds) Glaciomarine Environ-
ments: Processes and Sediments. Geological Society, London,
Special Publications, 53, 217– 223.
Wright, R. & Anderson, J. B. 1980. The importance of sediment gravity
flow to sediment transport and sorting in a glacial marine environ-
ment: Eastern Weddell Sea, Antarctica. Geological Society of
America Bulletin, 93, 951–963.
 Chapter 4

Chemostratigraphy and the Neoproterozoic glaciations

GALEN P. HALVERSON1,3* & GRAHAM SHIELDS-ZHOU2

1
School of Earth and Environmental Sciences, The University of Adelaide, North Terrace, Adelaide, SA 5005, Australia
2
Department of Earth Sciences, University College London, Gower Street, London, WD1E 6BT, UK
3
Present address: Department of Earth and Planetary Sciences, McGill University, 3450 University Street, Montreal,
QC, H3A 2A7, Canada
*Corresponding author (e-mail: galen.halverson@mcgill.ca)

Abstract: Although the pre-glacial Proterozoic isotopic record is poorly constrained, it is apparent that the chemical and isotopic
composition of the oceans began to change during the early to mid-Neoproterozoic and experienced considerable fluctuations alongside
climatic instability during much of the subsequent Cryogenian and Ediacaran periods. The earliest known large negative d13C excursion
appears to post-date 811 Ma and fluctuations became progressively more extreme, culminating in the late-Ediacaran ‘Shuram– Wonoka’
anomaly. The negative excursions are commonly associated with pre-glacial and post-glacial times, while extremely high d13C values are
characteristic of strata between glaciations. The precise causal mechanism for these excursions is subject to debate. Seawater 87Sr/86Sr
rose during the Neoproterozoic, with abrupt increases following deglaciation consistent with enhanced weathering rates. Reported
marine sulphate and pyrite d34S data exhibit marked variation through this interval, although the changes are not always consistent
within or between sedimentary successions of equivalent age. Iron-speciation studies indicate that much of this variation was caused
by fluctuating and low sulphate concentrations in seawater, which at times led to the build-up of ferruginous conditions in the
ocean. The application of chemostratigraphy to understanding and correlating the Neoproterozoic glaciations evokes considerable
controversy, and many questions persist regarding the reliability and calibration of the d13C, 87Sr/86Sr and d34S record. Nevertheless,
the individual glaciations appear to be characterized by distinct combined chemostratigraphic signatures, in large part due to the generally
increasing strontium isotope composition of seawater through the Neoproterozoic Era.

Chemical stratigraphy has enjoyed widespread application to the
study of the Neoproterozoic sedimentary record, in particular
with regard to the number, correlation, causes and consequences
of glaciations during this era. Owing to the scarcity of biostratigra-
phically useful fossils, it has become the tool of choice for global
stratigraphic correlation of Neoproterozoic strata (Kaufman &
Knoll 1995; Walter et al. 2000). In their broad survey of carbon-
ates, Schidlowski et al. (1975) pioneered the application of
carbon-isotope stratigraphy to the Precambrian sedimentary
record. Williams (1979) subsequently documented negative d13C
signatures associated with post-glacial carbonates in Australia.
Knoll et al. (1986) later focused attention on the Neoproterozoic
Era with coupled carbonate (d13Ccarb) and total organic carbon
(d13Corg) carbon-isotope records from carbonate-dominated suc-
cessions in Svalbard and East Greenland that showed large and
in-phase variations ostensibly linked to glacial episodes. This
study complemented contemporaneous investigations that simi-
larly showed large fluctuations in the d13C composition of
ancient marine carbonates spanning the Precambrian –Cambrian
boundary (Hsu et al. 1985; Tucker 1986; Magaritz et al. 1986).
These studies bolstered the argument that ancient carbonates and
sedimentary organic matter could be used as proxies for the
carbon-isotopic composition of the seawater in which they
formed. Inasmuch, they opened the door to carbon-isotope che-
mostratigraphy as a commonplace tool for basin- and global-scale
correlation and as a measure of the behaviour of the exogenic
carbon cycle, in particular as it related to extreme climate vari-
ability and biospheric change (e.g. Derry et al. 1992; Kaufman
& Knoll 1995).
Notwithstanding challenges to the integrity of carbon-isotope
signatures as proxies for seawater composition (e.g. Frimmel
2009; Knauth & Kennedy 2009; Derry 2010) and poor radiometric
age control on many Neoproterozoic successions, carbon-isotope
stratigraphy is a powerful tool for establishing a chronological
framework for this interval, during which extensive carbonates
were deposited but for which biostratigraphy is of limited (but
increasing) use (Knoll & Walter 1992; Knoll 2000). Multiple nega-
tive carbon-isotope excursions and intervening intervals of sus-
tained high d13Ccarb appear to correlate globally (e.g. Kaufman
et al. 1997; Halverson et al. 2005). However, because there are
also multiple glaciations, each apparently associated with negative
carbon-isotope anomalies, these excursions are non-unique and
robust correlations, in the absence of firm radiometric ages,
require additional data.
Like carbon isotopes, sulphur-isotope data on both pyrite and
sulphate show impressive variability spanning the Neoproterozoic
glaciations (Gorjan et al. 2000; Hurtgen et al. 2002). However, the
strontium-isotope record is the most useful in discriminating
between the various negative carbon-isotope anomalies because
87
Sr/86Sr increases throughout most of the Neoproterozoic
(Shields 1999; Melezhik et al. 2001; Halverson et al. 2007) such
that each of the prominent carbon-isotope anomalies appears to
be associated with distinct strontium-isotope signatures.
Many other chemostratigraphic methods have been applied to
the Neoproterozoic sedimentary record. Iron speciation data
have revealed remarkable fluctuations in the redox state of
seawater throughout the Neoproterozoic, in particular spanning
glaciations and leading up to the first appearance of the Ediacaran
fauna (Canfield et al. 2007, 2008). Post-glacial carbonates have
also proven to be exceptional archives of the highly unusual
environmental conditions during recovery from the Cryogenian
glaciations, preserving in some cases extraordinary geochemical
anomalies in non-traditional geochemical proxies, such as
calcium and boron isotopes (Kasemann et al. 2005) and D17O
(Bao et al. 2008). For a more extensive review of these and
other chemostratigraphic methods applied to the Neoproterozoic
sedimentary record, the reader is referred to Halverson et al.
(2010). A detailed account of the application of the chemical
index of alteration (CIA) to the Neoproterozoic glacial record is
presented in this volume by Bahlburg & Dobrzinski (2011), and
chemostratigraphic and geochronological data from individual
sedimentary successions are summarized in reviews throughout

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 51– 66. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.4
52 G. P. HALVERSON & G. SHIELDS-ZHOU
this volume. The aim of this chapter is to provide a brief review of
the carbon-, sulphur- and strontium-isotope systems and the iron-
speciation proxy and a synthesis of published data bracketing the
Neoproterozoic glaciations, with a focus on the structure of these
records and their temporal relation to glaciation.
Carbon isotopes
Under normal conditions and at the scale of resolution appropriate
in Precambrian chemostratigraphy, the carbon-isotope compo-
sitions of unaltered marine carbonates (d13Ccarb) precipitated in
equilibrium with seawater closely approximate the composition
of the total dissolved inorganic carbon (DIC) pool (Hayes et al.
1999). If it is assumed that the carbonates retain their primary
signatures after burial, then carbon-isotope chemostratigraphy
has diverse applications, such as establishing global correlations
(e.g. Kaufman et al. 1993, 1997), regional studies of basin archi-
tecture and carbonate sequence stratigraphy (Halverson et al.
2002; Cozzi et al. 2004; Jiang et al. 2007), modelling of the
global carbon cycle (Derry et al. 1992), and as a constraint on mech-
anisms for the causes of Neoproterozoic glaciation (Kaufman et al.
1997; Halverson et al. 2002; Schrag et al. 2002) and post-glacial
carbonate deposition (Kaufman et al. 1997; Hoffman et al. 1998;
Kennedy et al. 2001; Higgins & Schrag 2003; Shields 2005).
Debate continues as to whether Neoproterozoic marine carbonates
faithfully preserve their primary depositional signatures, particu-
larly where they are highly 13C-depleted (Bristow & Kennedy
2008; Frimmel 2009; Knauth & Kennedy 2009; Derry 2010).
Nevertheless, the general consensus is that, by and large, ancient
marine carbonates do preserve seawater compositions, so secular
trends in d13Ccarb can be interpreted in terms of inputs of carbon
to and outputs from the global ocean (Holser 1997).
The record of the carbon-isotope composition of sedimentary
organic matter (d13Corg), while much less applied in the Neoproter-
ozoic, is often used where carbonates are not sufficiently abundant
to construct a d13Ccarb profile. The analysis of d13Corg has recently
surged due to the hypothesis that some of the extreme negative
d13Ccarb anomalies might be the result of partial oxidation of an
enormous reactive organic carbon reservoir in the deep ocean
(Rothman et al. 2003). Organic carbon data are intrinsically
more difficult to acquire than inorganic carbon data, and the
connection between extant kerogen (or total organic carbon,
TOC) and original biomass is inevitably blurred by diagenesis,
thermal alteration of the organic matter (Hayes et al. 1983;
Kaufman et al. 1991; Hayes et al. 1999) and input of detrital
kerogen. Nevertheless, early stratigraphically constrained Neopro-
terozoic d13Corg data sets (e.g. Knoll et al. 1986; Kaufman et al.
1997) showed broad correlation with contemporaneous d13Ccarb,
suggesting that it could be used in parallel with or in place of
(where primary carbonates are unavailable) inorganic carbon
data as a proxy for seawater compositions, albeit with the caveat
that the net fractionation between the original DIC reservoir and
extracted kerogen is inherently variable. A compilation of
organic carbon data by Hayes et al. (1999) suggested an average
difference between contemporaneous carbonates and organic
matter (1TOC) of c. 30‰ in the Neoproterozoic, but with large fluc-
tuations (+10‰) associated with d13Ccarb anomalies, in particular
those spanning glaciations.
More recent and detailed data sets from Ediacaran-aged sedi-
ments have revealed sustained intervals where d13Ccarb and
d13Corg are decoupled or even anti-correlated (Calver 2000; Fike
et al. 2006; McFadden et al. 2008). Swanson-Hysell et al.
(2010) recently argued, based on new, paired data from
Australia, that decoupling of these two proxies began sometime
in the Cryogenian. Decoupling between these two proxies is a
prediction of the Rothman et al. (2003) hypothesis, because the
implied oceanic organic-carbon reservoir would be much larger
than the DIC reservoir. While questions remain as to the
plausibility of sufficient oxidants being available to generate the
accompanying highly negative d13Ccarb anomalies (Bristow &
Kennedy 2008), the undeniable lack of correlation in certain data
sets requires a mechanistic explanation.
Sulphur isotopes
Sedimentary sulphur-isotope ratios are measured on sedimentary
sulphates or sulphides. Sulphur-isotope data on sulphates can be
recovered, with varying degrees of reliability, from evaporites,
barite, phosphorites and carbonates (as carbonate-associated sul-
phate, CAS). Sulphate-bearing marine minerals commonly do
not form in isotopic equilibrium with seawater, so additional argu-
ments need to be marshalled to support the presumption that they
record seawater sulphate isotopic compositions. The d34S values of
marine evaporite minerals may be higher or lower than contem-
poraneous seawater depending on the extent of basin restriction
and connection to meteoric influence. Stratiform barite d34S
values are also prone to deviate from seawater, although lowermost
values are generally close approximations to ocean composition
(Shields 2007). Phosphorite can also show higher d34S values
than seawater due to Rayleigh fractionation during bacterial sul-
phate reduction (e.g. Hough et al. 2006), and the same may also
be true of some diagenetic carbonate rocks (e.g. Marenco et al.
2008a). An additional problem with carbonate- (and phosphate-)
associated sulphate sulphur-isotopic analyses derives from the
inadvertent incorporation of contaminant sulphate during the dis-
solution process before analysis (Marenco et al. 2008b).
Unlike marine sulphate minerals, sulphides (typically pyrite)
record the net fractionation imparted during bacterial sulphate
reduction (BSR), plus additional fractionation effects contributed
by disproportionation reactions during oxidative recycling of sul-
phide (Canfield & Teske 1996; Detmers et al. 2001). Sulphate
and pyrite d34S trends broadly mirror one another, but pyrite is gen-
erally much more variable, attributable to the fact that reduction,
oxidation and disproportionation reactions commonly occur
during early diagenesis, where distinct local effects, such as pore
water sulphate concentrations, availability of appropriate sub-
strates for BSR, and physiology of the BSR population (Detmers
et al. 2001) strongly influence the final sulphur-isotope signature.
Sulphur isotopes have been widely applied to problems in Pre-
cambrian Earth history. For example, a gradual increase in the
spread of d34Spyrite in shales (d34Ssulphate – d34Spyrite isotopic dis-
crimination) broadly coincides with separate geological and geo-
chemical evidence for a Neoproterozoic rise in the oxygenation
of the Earth’s environment (Canfield & Teske 1996; Fike et al.
2006). Originally, this isotopic shift was linked to the evolutionary
radiation of sulphide-oxidizing bacteria (Canfield & Teske 1996),
but recent evidence puts this evolutionary event much earlier
(Johnston et al. 2005). Instead, this isotopic shift could have
been caused by a rise in the importance of sulphide oxidation
and disproportionation reactions during early diagenesis, which
may be indirectly linked to oxygenation through seafloor redox
changes. The d34S pyrite compilation is also the basis for the
model that the global deep ocean became dominantly euxinic,
rather than oxygenated, as massive banded iron-formation depo-
sition came to a close in the late Palaeoproterozoic (Poulton
et al. 2004). The Neoproterozoic is characterized by extraordinary
fluctuations in both d34Spyrite and d34Ssulphate in the Neoprotero-
zoic, which are closely coupled to glacial events (Halverson &
Hurtgen 2007). However, the extent to which these variations
truly reflect global seawater sulphate compositions and environ-
mental conditions is poorly established.
Highly precise measurement of all four sulphur isotopes
(32S, 33S, 34S, 36S) has shown resolvable and consistent variations
in the parameters D33S and D36S, which can reflect both mass-
independent (MIF) and mass-dependent fractionation (MDF) pro-
cesses. MIF of sulphur isotopes has been applied to reconstructing
 CHEMOSTRATIGRAPHY AND THE NEOPROTEROZOIC GLACIATIONS
the evolution of atmospheric oxygen levels (Farquhar et al. 2000),
but has not been systematically applied to the Neoproterozoic,
when atmospheric O2 concentrations likely exceeded the low
level (,1025 present levels) required to suppress MIF fraction-
ation in the atmosphere (Pavlov & Kasting 2002). On the other
hand, variations in D33S and D36S resulting from MDF record
information about specific sulphur transformations and pathways
that cannot be gleaned from traditional measurements of
2
S/4S alone (Johnston et al. 2005; Ono et al. 2006). Although
not reviewed here, the technique of measuring quadruple sulphur
isotopes promises to yield significant insight into behaviour of
the sulphur cycle leading up to and following the Neoproterozoic
glaciations.
Iron speciation
The impressive range of d34Spyrite and d34Ssulphate values that are
typical of the Neoproterozoic are commonly interpreted in terms
of a low sulphate ocean (Halverson & Hurtgen 2007). This
interpretation is consistent with the observation that seawater
during the Neoproterozoic was occasionally rich in dissolved
ferrous iron, based on the occurrence of iron formation. Iron-
speciation studies shed further light on the redox state of the
oceans and are being carried out increasingly on Neoproterozoic
successions (Canfield et al. 2008). Redox conditions in the water
column affect the speciation of iron in marine sediments. The
highly reactive iron pool (FeHR) is that which is geochemically
available during diagenesis and includes iron bound in carbonates,
iron oxides and oxyhydroxides, and iron sulphides (Canfield
1989). Sediments deposited under an oxic water column tend to
contain less reactive iron as a fraction of total sedimentary iron
(FeHR/FeT) than those deposited under an anoxic water column,
with a proposed cut-off at 0.38 (Raiswell et al. 1988; Lyons &
Severmann 2006). Canfield et al. (2008) further proposed that
the measure of pyrite-bound iron v. total reactive iron (FePy/
FeHR) can be used to distinguish iron-rich anoxic v. euxinic
water columns, with sediments exhibiting FePy/FeHR . 0.8
being classified as euxinic.
Neoproterozoic sediments deposited beneath a storm wave
base typically exhibit high FeHR/FeT ratios (Fig. 4.1), indicating
that anoxia was common in the deeper marine environment.
However, sulphidic conditions are so far only indicated during
the pre-glacial Neoproterozoic and early Cambrian (Canfield
et al. 2008), suggesting ferruginous rather than sulphidic
(euxinic) anoxia during the interval of climatic fluctuations.
Because ferruginous conditions can only arise when the molar
flux of FeHR to the deep ocean is greater than half the flux of
sulphide, their reappearance during the Neoproterozoic after an
absence of more than a billion years is consistent with a low-
sulphate ocean reservoir.
Strontium isotopes
Over timescales of .106 years, the 87Sr/86Sr ratio of seawater
records the relative strontium fluxes from continental weathering
and hydrothermal input, superimposed on the long-term increase in
87
Sr/86Sr as a result of radioactive decay of 87Rb (Edmond 1992).
As such, marine 87Sr/86Sr is a useful measure of evolving tectonic
and long-term climatic regimes. Nevertheless, the ensemble of
driving mechanisms for the variations in seawater 87Sr/86Sr
spanning the Neoproterozoic remains controversial. It is doubtful
that seawater 87Sr/86Sr can be simplified to a simple balance
between the hydrothermal and continental inputs, but rather
must also be influenced by continental configuration and CO2 out-
gassing rates (Halverson et al. 2007; Shields 2007).
Strontium-isotope stratigraphy is based on the presumption that
seawater was always isotopically homogeneous with respect to Sr
53
because of its long ocean residence time. Although isotopic analy-
sis is now standard procedure, samples need to be carefully
screened because of the risk of contaminating the sample with
diagenetically altered or detrital materials. This risk is reduced
for Phanerozoic studies for which microscopically well-preserved
low-Mg calcite tests (e.g. foraminifera, brachiopods and belem-
nites) can be analysed. However, for much of the Precambrian, it
is necessary to identify least altered components of the bulk car-
bonate rock (e.g. Derry et al. 1992; Kaufman et al. 1993).
Strontium isotope analysis comprises the following six steps, of
which the first four vary considerably between research groups: (i)
sample selection using petrographic and geochemical criteria; (ii)
physical extraction from the host sample; (iii) chemical pre-
treatment to remove contaminant Sr ions; (iv) sample dissolution;
(v) chemical separation of the element Sr; (vi) mass spectrometric
analysis. Studies show that least altered ratios can most easily be
obtained by targeted microsampling of fine-grained calcite
(micrite), which is free of later recrystallization and dolomitiza-
tion, although early diagenetic cements may also be reliable
tracers of ocean composition (e.g. Kaufman et al. 1993; Fairchild
et al. 2000). Chemical pre-leaching helps remove contaminant Rb
and radiogenic Sr (Gorokhov et al. 1995), while subsequent
incomplete dissolution in weak acids such as acetic acid has
been shown to limit inadvertent dissolution of clay mineral-bound
Sr (Bailey et al. 2000).
The Neoproterozoic strontium-isotope record is typically
reconstructed from calcite that has relatively high strontium and
low rubidium concentrations or otherwise exhibits evidence of
limited exchange with diagenetic fluids, as measured, for
example, in Mn/Sr ratios (Brand & Veizer 1981; Jacobsen &
Kaufman 1999). Although carbonate originally precipitated as
aragonite is intrinsically more suitable due to its high original Sr
content (Kulp et al. 1952), the abundance of metastable carbonate
precipitates in the Precambrian indicates that some apparently
pristine components could be the result of diagenetic replacement
(Fairchild et al. 2000). Because alteration usually increases the
87
Sr/86Sr of carbonates (Banner & Hanson 1990), the lowest
measured values within a suite of rocks are typically inferred to
most closely approximate the composition of the seawater from
which it precipitated. However, this ought not merely to be
assumed, as post-depositional fluids are not always more radio-
genic than coeval seawater. With this in mind, least-altered ratios
are best extrapolated from diagenetic trends, for example by
plotting Mn/Sr, Mg/Ca, 1/Sr or d18O against measured 87Sr/86Sr.
Recent compilations of the 87Sr/86Sr evolution of Neoprotero-
zoic oceans share the common, dominant feature of ratios rising
from as low as c. 0.7055 in the early Neoproterozoic to as high
as 0.7085 in the Ediacaran Period (Jacobsen & Kaufman 1999;
Walter et al. 2000; Melezhik et al. 2001; Halverson et al. 2007a;
Shields 2007). This feature alone makes strontium-isotope
chemostratigraphy a key tool in establishing correlations and rela-
tive age in otherwise poorly radiometrically dated Neoproterozoic
successions. Additional detail in the strontium-isotope record
remains controversial due to the relatively poor resolution of the
Neoproterozoic record and variable quality of samples used in
compilations (Melezhik et al. 2001). However, an apparently
reproducible finer-scale structure in the record is emerging with
the promise of improving the Neoproterozoic chronology and the
link between coupled tectonic– climatic processes. Importantly,
and as will be discussed further below, the seawater strontium-
isotope record appears to be distinct, spanning each of the Neo-
proterozoic glacial epochs (Table 4.1).
The pre-glacial Neoproterozoic
The most characteristic features of the Neoproterozoic chemostra-
tigraphic record are the generally high d13Ccarb values that prevail
throughout most of the era and the episodic negative anomalies
54 G. P. HALVERSON & G. SHIELDS-ZHOU

Fig. 4.1. Working chemostratigraphic compilations of d13Ccarb, 87Sr/86Sr, d34Ssulphate, d34Spyrite and iron-speciation (FeHR/FeT) data for the Neoproterozoic (modified
after Halverson et al. 2005, 2007a, 2010; Halverson 2006; Canfield et al. 2008). This compilation does not include all available high-quality data due to the difficulty
of integrating data that have poor and variable age constraints (see text for discussion). The names shown vertically in the top of the figure are the informal names of the
principal negative anomalies, as well as the distinct positive shift in d13Ccarb recorded in the lower Little Dal Group (Halverson 2006) and the ‘Keele Peak’ positive
anomaly (Kaufman et al. 2007). Note that there are conflicting sulphur-isotope data, in particular for the late Ediacaran Period (e.g. Ries et al. 2009). A low of 87Sr/86Sr in
the earliest Neoproterozoic (dashed oval) is based on data from Gorokhov et al. (1995). Iron-speciation data (FeHR/FeT) are from Canfield et al. (2008), replotted as
five-point running averages, and from Nagy et al. (2009). Open boxes and diamonds are additional d34Ssulphate and iron-speciation data from the Chuar Group as published
in Johnston et al. (2010). Principle U–Pb geochronological constraints on the d13Ccarb record are shown in triangles at the top of the compilation: (1) an 811.5 + 0.3 Ma
tuff in the Fifteenmile Group, Yukon Territory (Macdonald et al. 2010b); (2) a 760 + 1 Ma tuff in the Ombombo Subgroup, NW Namibia (Halverson et al. 2005); (3) a
716.5 + 0.2 Ma tuff in the lower Rapitan Group, Yukon Territory (Macdonald et al. 2010b); (4) an ash bed in the Ghaub Formation in northern Namibia dated at
635 + 0.6 Ma (Hofmann et al. 2005) and an ash bed in the lower Duoshantuo Formation in South China dated at 635.2 + 0.6 Ma (Condon et al. 2005) bracket the
age of the Cryogenian –Ediacaran boundary; (5) a 542.0 + 0.6 Ma ash bed at the Precambrian –Cambrian boundary in Oman (Amthor et al. 2003); and several
slightly older ages (overlapping triangles) from the latest Ediacaran of the Nama Group, S Namibia (Grotzinger et al. 1995). Shaded boxes and lines show the inferred
range of glacial events, with the gradational shading in the middle Cryogenian box reflecting the wide range of ages for middle Cryogenian glaciation (e.g. Allen &
Etienne 2008).
 CHEMOSTRATIGRAPHY AND THE NEOPROTEROZOIC GLACIATIONS
that punctuate it (Knoll et al. 1986; Kaufman & Knoll 1995;
Shields & Veizer 2002). These basic observations raise several
questions. First, when did the average values near 0‰ typical of
the Mesoproterozoic (Brasier & Lindsay 1995) give way to the
average values of þ5‰ (Halverson et al. 2005) characteristic of
the Neoproterozoic? Second, did the increased variability in
d13Ccarb accompany the higher average values and, if not, when
did this begin? Finally, what drove these fundamental shifts in
the behaviour of the exogenic carbon cycle?
These questions are difficult to answer because of the poor
carbon-isotope coverage and radiometric ages spanning the
Mesoproterozoic– Neoproterozoic transition. Bartley et al.
(2001) and Frank et al. (2003) argued that the increase in the
amplitude and frequency of fluctuations in d13Ccarb began some-
time in the middle to late Mesoproterozoic, based on data from
Siberia and NE Canada, respectively. Halverson (2006) argued
that sustained d13Ccarb  5‰ did not start until the early Neo-
proterozoic based on data from the Platform Assemblage of the
Little Dal Group in NW Canada, where the lowermost data show
a distinct ramp from 0 to 5‰ (Fig. 4.1), followed by generally
high (5‰) values up to the onset of the Bitter Springs anomaly
(see below). However, the age and correlation of the lowermost
Little Dal Group and its equivalence to other rocks within Succes-
sion B in NW Canada is not known and only constrained to be
between c. 1004 (Rainbird et al. 1996) and 811 Ma (Macdonald
et al. 2010b). It is therefore unclear whether the lower Little Dal
positive d13C shift (Fig. 4.1) truly does record the first highly
13
C-enriched seawater in the Neoproterozoic, or what the precise
age of this event might be.
Much better constrained now is the timing of presumably the
first major negative d13Ccarb excursion in the Neoproterozoic: the
Bitter Springs anomaly (Table 4.1; Halverson et al. 2005).
Although older negative d13Ccarb excursions have been reported
(Bartley et al. 2001; Shields 2007), these are generally from
poorly dated successions and may be the result of alteration. For
example, 13C-depleted I6 unit rocks of the Atar Group, Mauritania
(Shields 2007), although now known to be c. 1.1 Ga in age
(Rooney et al. 2010), are likely to have undergone significant iso-
topic alteration (Fairchild et al. 1990). The Bitter Springs negative
anomaly was first fully documented in the Bitter Springs
Formation in the Amadeus basin of central Australia (Hill et al.
2000). It was subsequently recorded in more detail in Svalbard
(Halverson et al. 2005), where it is clearly defined by reciprocal
d13C shifts of .8‰ that perfectly coincide with rare exposure sur-
faces in the Akademikerbreen Group. The anomaly also appears to
occur in the upper Little Dal Group (Upper Carbonate Formation)
in the Mackenzie Mountains (Halverson 2006). These global cor-
relations are reinforced by additional geochemical data (Halverson
et al. 2007a), namely 87Sr/86Sr (see below) and an updated and
high-resolution d13Ccarb record through the Bitter Springs For-
mation (Swanson-Hysell et al. 2010), as well as the coincidence
of the shifts with prominent sequence boundaries. Both features
have been interpreted to relate to large true polar wander episodes
(Maloof et al. 2006).
The Bitter Springs anomaly (Table 4.1; Fig. 4.1) has now been
reproduced across northwestern Canada: within the middle
Lower Tindir Group (Tatunduk Inlier), upper Fifteenmile Group
(Coal Creek Inlier), and the Wynniatt Formation of the Shaler
Supergroup (Victoria Island; Macdonald et al. 2010b; Jones
et al. 2010), although in this latter case only the negative shift is
preserved and this is recorded in sediments intruded by mafic
sills, which means it may be an overprint related to contact meta-
morphism rather than a primary signal. A likely correlative has
also been documented in the Tambien Group of northern Ethiopia
(Alene et al. 2006; Miller et al. 2009). Macdonald et al. (2010b)
produced a precise U – Pb zircon age of 811.51 + 0.25 Ma on a
tuff c. 50 m beneath the negative shift presumed to define the
base of the anomaly in upper Fifteenmile Group, thus constrain-
ing the onset of the Bitter Springs anomaly to c. 810 Ma and
55
providing a more robust basis for global correlations. At the
upper boundary of the Bitter Springs anomaly, d13Ccarb shifts
back to values exceeding 5‰.
The d13Ccarb of seawater at the time of the putative and possibly
regional Kaigas (Bayisi) glaciation at c. 750– 740 Ma (Frimmel
et al. 1996; Xu et al. 2009) is not well known. Negative d13C
values occur in the lower Wallekraal Formation, which overlies
the Kaigas Formation (Fölling & Frimmel 2002) but no large
d13C anomaly occurs between the Bitter Springs anomaly and
the Islay anomaly in Svalbard (Fig. 4.1). Instead, d13Ccarb values
remain generally high until the onset of the Islay anomaly (Prave
et al. 2009), which appears to precede the first major Cryogenian
glaciation (Fig. 4.1) and is discussed in more detail in the
following section.
The strontium-isotope record through the early Neoproterozoic
shown in Figure 4.1 is largely derived from data from Svalbard and
northwestern Canada, even though other presumably primary
values have been obtained from other successions such as the
Bitter Springs Formation. As noted by Bartley et al. (2001),
seawater 87Sr/86Sr values reach a nadir around the Mesoproter-
ozoic –Neoproterozoic transition, presumably related to the final
assembly of Rodinia. The least radiogenic latest Mesoproterozoic
87
Sr/86Sr values reported by Bartley et al. (2001) hover around
0.7055, which is virtually the same as a value of 0.7056 from
limestones of the Atar Group, Taoudeni Basin (Veizer et al.
1983), now dated at 1.1 Ga (Rooney et al. 2010) and the lowest
values in the lower Little Dal Group (Halverson et al. 2007a).
However, Gorokhov et al. (1995) reported values from pristine
limestones of the presumed early Neoproterozoic Burovaya For-
mation (Turukhansk uplift, NW Siberia) as low as 0.7052. Thus,
0.7055 is a plausible approximation for seawater 87Sr/86Sr at the
Mesoproterozoic –Neoproterozoic boundary, but probably does
not represent the nadir before the protracted Neoproterozoic rise
in 87Sr/86Sr.
Strontium-isotope ratios had risen to 0.7063 by the onset of the
Bitter Springs anomaly (Fig. 4.1). Although a more discreet and
smaller shift to slightly higher ratios accompanies the negative
d13C shift at the base of the anomaly, 87Sr/86Sr returns
to c. 0.7063 at the end of the Bitter Springs anomaly, before appar-
ently increasing to as high as 0.7070 in the Backlundtoppen
Formation in Svalbard (prior to the Islay anomaly). No intervening
data from this transition have been documented. As will be
discussed below, this middle Neoproterozoic 87Sr/86Sr peak is fol-
lowed by a temporary return to values as low as 0.7063.
The sulphur-isotope record for the early Neoproterozoic is rela-
tively sparse, despite the fact that voluminous evaporites were
deposited in Australia and NW Canada c. 830 to 800 Ma (Evans
2006). Indeed, the compilation in Figure 4.1 does not include any
data older than c. 820 Ma, although this lacuna is in part an artefact
of poor control on ages and correlations, which limit the data that
can be used. d34Ssulphate from evaporites in the Gillen Member of
the Bitter Springs Formation (i.e. before the Bitter Springs
anomaly) ranges from 15 to 21‰ (Gorjan et al. 2000; Hill et al.
2000). From the available data, it appears that early Neoproterozoic
seawater d34Ssulphate values do not depart greatly from the Gillen
Member values (Fig. 4.1), averaging c. 20‰ through the middle
Neoproterozoic, close to average values for the Cretaceous and
Cenozoic (Paytan et al. 1998, 2004; Kampschulte & Strauss
2004), but distinct from the highly 34S-enriched values that
characterize the early and late Ediacaran and early Palaeozoic
(Shields et al. 2004; Kampschulte & Strauss 2004; Halverson &
Hurtgen 2007).
Although most d34Ssulphate data of the composite curve for this
time interval are derived from CAS measurements on Svalbard
samples (Halverson et al. 2010), available data from evaporites
spanning the Bitter Springs anomaly in the Amadeus Basin
(Gorjan et al. 2000; Hill et al. 2000) yield similar values. Pyrite
data in this record are also sparse and similarly derived mainly
from Svalbard (Halverson et al. 2010) and the Bitter Springs
56 G. P. HALVERSON & G. SHIELDS-ZHOU

Table 4.1. Summary of the d13Ccarb and 87Sr/86Sr signatures associated with the major Neoproterozoic negative carbon isotope anomalies (Fig. 4.1) and an abridged
compilation of locations where they are well preserved (for the Maieberg and Rasthof anomalies, only key locations where the anomalies are preserved in great detail
are given). PB, Paraguai Belt; UTG, Upper Tindir Group

Shuram-Wonoka anomaly
d13Ccarb Minimum: , –10‰; protracted recovery
87
Sr/86Sr 0.7081– 0.7084

Region Unit Reference

S Namibia Nama Gp Workman et al. (2002)
Oman Shuram/Kufai Fm Burns & Matter (1993), Le Guerroué et al. (2006)
S China Doushantuo Fm Jiang et al. (2007), McFadden et al. (2008)
S Australia Wonoka Fm Pell et al. (1993), Calver (2000)
N Norway Nyborg Fm Halverson et al. (2005)
SW USA Johnnie Fm Corsetti & Kaufman (2003)
SE Siberia Nikol’skaya/Chenchinskaya Pokrovskii et al. (2006), Melezhik et al. (2009)

Maieberg anomaly
d13Ccarb Minimum: –4 to – 6‰, gradual to abrupt return to near 0‰ values
87
Sr/86Sr 0.7072 at base, increases upsection to 0.7078– 0.7080

Region Unit Reference

NW Namibia Maieberg Fm Hoffman et al. (1998)

S Namibia Bloeddrif Mb Fölling & Frimmel (2002)
Oman Hadash/Masirah Bay Le Guerroué et al. (2006)
NW China Tureeken Fm Xiao et al. (2004)
S China Doushantuo Fm Jiang et al. (2007), McFadden et al. (2008) and others
SW Mongolia Oi Mb Macdonald et al. (2009a)
E Svalbard Dracoisen Fm Kaufman et al. (1997), Halverson et al. (2004)
Scotland Cranford Ls McCay et al. (2006)
NE Alaska Katakturuk Dolomite (K2) Macdonald et al. (2009b)
SW USA Noonday Dolomite Prave (1999)
Brazil (PB) Araras Fm Nogueira et al. (2007)
SE Siberia Barakunskaya Fm Pokrovskii et al. (2006), Melezhik et al. (2009)

Trezona anomaly
d13Ccarb Minimum: –6 to – 8‰, increases before glaciation
87
Sr/86Sr 0.7072– 0.7074

Region Unit Reference

NW Namibia Ombaatjie Fm Halverson et al. (2002)

South Australia Trezona Fm McKirdy et al. (2001)
N Norway Grasdal Fm Halverson et al. (2005)
Scotland Craignish/Ardrishaig Prave et al. (2009)
NE Alaska Katakturuk Dolomite (K1) Macdonald et al. (2009b)
Mackenzie Mts Keele Fm Halverson et al. (2005)

Rasthof anomaly
d13Ccarb Minimum: –4 to 0‰, rises rapidly
87
Sr/86Sr 0.7067– 0.7069 at base, increases up section to 0.7072– 0.7073

Region Unit Reference

NW Namibia Rashtof Fm Hoffman et al. (1998), Yoshioka et al. (2003)

SW Mongolia Base Tayshir Mb Shields et al. (1997), Macdonald et al. (2009a)
S Australia Tapley Hill Fm McKirdy et al. (2001)
Scotland Bonahaven Fm* Brasier & Shields (2000), McCay et al. (2006), Prave et al. (2009)
NE Alaska Katakturuk Dolomite (K1) Macdonald et al. (2009b)
Mackenzie Mts Twitya Fm. Hoffman & Schrag (2002)
SW USA base Beck Spring Fm Corsetti & Kaufman (2003)

Islay anomaly
d13Ccarb Minimum: –5 to – 9‰, return to .0‰ before glaciation
87
Sr/86Sr 0.7065– 0.7067, declines upsection to 0.7064

Region Unit Reference

Tasmania Black River Fm Calver (1998)

E Svalbard Elbobreen Fm Halverson et al. (2004)

(Continued)
 CHEMOSTRATIGRAPHY AND THE NEOPROTEROZOIC GLACIATIONS 57

Table 4.1. Continued

Scotland Islay/Lossit Fm Brasier & Shields (2000), Prave et al. (2009), McCay et al.
(2006), Sawaki et al. (2010b)
E Greenland Bed Group 19 Fairchild et al. (2000)
Victoria Island Killian Formation Macdonald et al. (2010b)
Mackenzie Mts Copper Cap Fm Halverson (2006)
Alaska-Yukon Upper Dolomite (LTG) Macdonald et al. (2010a)
SW USA Upper Beck Spring Fm Corsetti & Kaufman (2003)

Bitter Springs anomaly

d13Ccarb Minimum: –5 to 0‰, bracketed by sharp positive and negative shift
87
Sr/86Sr 0.7063–0.7065

Region Unit Reference

N Ethiopia Tambien Group Alene et al. (2006)

Central Australia Bitter Springs Formation Hill et al. (2000)
Western Australia Hussar Formation Hill (2005)
E Svalbard Grusdievbreen/Svanbergfjellet Halverson et al. (2005)
Scotland Ballachulish Sbgp Prave et al. (2009)
Mackenzie Mts Upper Carbonate Fm Halverson (2006)
Yukon Upper Fifteenmile Gp (PF1) Macdonald et al. (2010b)
Alaska-Yukon Upper Shale (LTG) Macdonald et al. (2010a)

*Prave et al. (2009) argued that the post-Port Askaig negative d13C anomaly is missing, in which case the Bonahaven anomaly, which is stratigraphically above the
glacigenic Port Askaig Formation, might be equivalent to the Tayshir anomaly in Mongolia.

Formation (Hill et al. 2000). The most salient pattern in the d34S
data is a coupled increase in d34Ssulphate and decrease in d34Spyrite
coincident with the Bitter Springs anomaly (Fig. 4.1), resulting
in a transient increase in D34S. Other constraints on the d34Ssulphate
of early Neoproterozoic seawater come from studies on succes-
sions from Arctic Canada. d34Ssulphate values there range from 14
to 18‰ in the pre-Bitter Springs anomaly (.810 Ma) Gypsum
Formation of the Little Dal Group (Turner 2009), while similar
values (16 – 19‰) have been reported from the post-Bitter
Springs anomaly (,780 Ma) Redstone River Formation of the
overlying Coates Lake Group (Strauss 1993).
The sulphur-isotope record throughout the remainder of the
early –middle Neoproterozoic is thin, but suggests a return
to reduced D34S resulting from higher average d34Spyrite sometime
prior to the middle Cryogenian glaciation.
Spanning the Cryogenian glaciations
Significant discussion remains as to the number, timing and dur-
ation of the Cryogenian glaciations, and the base of the Cryogenian
period has yet to be formally defined. Two particular persistent
sources of controversy and confusion are the possible occurrence
of an early (c. 745–750 Ma; Borg et al. 2003) ‘Kaigas’ glaciation
(Frimmel et al. 1996) that may not have been global in extent and
the seemingly contradictory age constraints on the so-called ‘Stur-
tian’ or early Cryogenian glaciation. There is no incontrovertible
evidence that the Kaigas diamictite in the Gariep belt of southern
Namibia (see Macdonald et al. (2011) for a stratigraphic and
sedimentological reappraisal of the Kaigas diamictites) and other
older diamictites in the Katanga Supergroup in Zambia (Key
et al. 2001) and NW China (Xu et al. 2009) are glacial in origin
(Hoffman & Li 2009). Chemostratigraphic data spanning these
isolated diamictites are also sparse. Therefore, although the
putative Kaigas glaciation is plotted in Figure 4.1, we do not
review the chemostratigraphic framework for this possible middle-
Neoproterozoic glaciation. As for the sticky question of the age of
early Cryogenian glaciation (i.e. glaciation between the older
Kaigas event and the end-Cryogenian ‘Marinoan’ event), reviewed
recently by Hoffman & Li (2009), we acknowledge that it is
impossible to discriminate at this point between the model that
there were multiple middle Cryogenian glaciations or a single,
long-lived glacial epoch that spanned from c. 716 (Macdonald
et al. 2010b) to 660 Ma (Fanning & Link 2008).
These deficiencies in the Cryogenian chronology are exacer-
bated by presumably larges hiatuses in the records on continental
shelves (from whence most of the chemostratigraphic records are
derived) resulting from glacioeustatic fall in sea level and glacial
erosion. These complications in the Cryogenian record make it
impossible to produce an accurate chemostratigraphic record for
the Cryogenian period and the compilation in Figure 4.1 can
only be regarded as a work in progress. Nevertheless, it is
notable that independent compilations emerging from widely sep-
arate sedimentary basins do seem to be converging on a similar
overall chemostratigraphic structure for the interval spanning the
Cryogenian glaciations (e.g. Halverson et al. 2005; Prave et al.
2009; Macdonald et al. 2010b).
The negative d13Ccarb anomalies that coincide with deglaciation
and are preserved in transgressive –regressive sequences overly-
ing glacial deposits and their equivalent glacial unconformities
(i.e. cap carbonate sequences; Hoffman & Schrag 2002) are well
documented (Table 4.1; Williams 1979; Kaufman et al. 1997;
Kennedy et al. 1998; Halverson et al. 2005; Hoffman et al.
2007). Therefore, only the most important features are summarized
here. Importantly, no more than two distinct, post-glacial negative
d13Ccarb anomalies linked to Cryogenian glaciation are known to
occur in a single succession (Kennedy et al. 1998). Where there
are two such post-glacial anomalies within one succession or,
where there is just one of known age, consistent distinctions can
be drawn between the two anomalies (Table 4.1; Kennedy et al.
1998; Hoffman & Schrag 2002).
As a general rule, the older of the two post-glacial sequences
lacks the basal transgressive portion, meaning the maximum flood-
ing surface is the base of the sequence (Hoffman & Schrag 2002;
Hoffman et al. 2011). d13Ccarb typically begins negative and trends
positive, returning to values .0‰ within a few metres (Kennedy
et al. 1998; Halverson et al. 2005). For example, in the Rasthof
Formation of NW Namibia, d13Ccarb begins at around –4.3‰
(herein referred to as the Rasthof anomaly) and crosses over
abruptly to positive values c. 10 m up-section, at the contact
58 G. P. HALVERSON & G. SHIELDS-ZHOU
between a basal rhythmite member and an overlying chaotic
microbialite member (Hoffman et al. 1998; Yoshioka et al.
2003). The thickness of this anomaly in Namibia is in large part
a function of the amount of allodapic dolomite within the lower
member (Halverson et al. 2005). Only in rare instances is a nega-
tive trend preserved in the lowermost part of the sequence (e.g. in
the presumed Rasthof-equivalent basal Tayshir Member of the
Tsagaan Oloom Formation, Mongolia; Macdonald et al. 2009a).
In contrast, the basal transgressive portion of the sequence over-
lying the end-Cryogenian glacials and defining the base of the
Ediacaran Period (Knoll et al. 2006) commonly preserves an
upward negative trend, with values typically in the range of –2
to –4‰ (Hoffman & Schrag 2002; Hoffman et al. 2007; Rose &
Maloof 2010). In the carbonate-dominated post-glacial platformal
sequence in NW Namibia (Maieberg Formation) where the Maie-
berg anomaly has been most extensively studied, d13Ccarb con-
tinues to decline to nearly –6‰ at the maximum flooding
surface before gradually returning to higher values over hundreds
of metres of the section (Hoffman et al. 1998). However, this
apparent protracted recovery is not typical of the basal Ediacaran
negative d13Ccarb anomaly, and in many other carbonate-
dominated locations, including the slope equivalents of the
Maieberg Formation in Namibia (Halverson et al. 2005), the
Doushantuo Formation of South China (e.g. McFadden et al.
2008), and the Barakunskaya Formation in southern Siberia (Pok-
rovskii et al. 2006), the anomaly is relatively condensed, spanning
only several metres to a few tens of metres. Thus, relative sedimen-
tation rates must be considered when correlating chemostrati-
graphic signals.
Large negative d13Ccarb anomalies also precede two Cryogenian
glaciations (Table 4.1; Fig. 4.2). The so-called Trezona anomaly,
named after the Trezona Formation in South Australia where
the anomaly was first fully documented (McKirdy et al. 2001),
occurs prior to the end-Cryogenian (i.e. the ‘Marinoan’ or
‘Elatina’ glaciation in South Australia) glaciation and following
the so-called Keele Peak (Kaufman et al. 1997), where d13Ccarb
reaches values þ9‰ (Figs 4.1 & 4.2). This negative anomaly
is best documented in the Ombaatjie Formation in the Otavi
Group, NW Namibia, where it is defined by a gradual, facies-
independent decline from c. þ7 to –5‰ through up to 50 m of
dominantly shallow-water carbonates, followed by a positive
shift of a few ‰ (Halverson et al. 2002). The extent of truncation
of the anomaly can be used as a general indicator of the depth of
erosion on the glacial surface.
A negative anomaly of similar magnitude, referred to as the
Islay anomaly (Prave et al. 2009) after the formation in Scotland
(Figs 4.1 & 4.2) where it was first documented (Brasier &
Shields 2000; McCay et al. 2006; Sawaki et al. 2010a), occurs
prior to the older Cryogenian glaciation in several successions.
In NE Svalbard, this anomaly is virtually identical in shape, mag-
nitude and stratigraphic context to the Trezona anomaly in
Namibia (Fig. 4.2) and is similarly variably erosionally truncated
(Halverson et al. 2004). However, in Scotland, d13C returns to
positive values prior to glaciation (Prave et al. 2009). In NW
Canada and East Greenland, the presumably (but not definitively)
equivalent anomalies in the Coates Lake Group (Halverson 2006)
and Bed Groups 19–20 (Fairchild et al. 2000), respectively,
feature an even more prominent positive shift, possibly providing
an important distinction from the Trezona anomaly.
A potentially different late Cryogenian negative d13Ccarb
anomaly of similar magnitude has been found in the Tayshir
Member of the lower Tsagaan Oloom Formation in Mongolia
(Macdonald et al. 2009a). This anomaly has been tentatively cor-
related with the Bonahaven anomaly in Scotland (Prave et al.
2009) and an anomaly of much smaller magnitude in the Gruis For-
mation of NW Namibia, between the Rasthof and Trezona
anomalies (Fig. 4.2). A similar large negative anomaly in this stra-
tigraphic position has not been documented elsewhere. Therefore,
it remains plausible that the anomaly in the Tayshir Member is
equivalent to the Trezona anomaly and the return to positive
d13Ccarb values above the anomaly has been removed by erosion
as it was in Namibia (Macdonald et al. 2009a).
Although the homologous pattern of the Islay and Trezona
anomalies suggests a similar origin, it diminishes the utility of
these anomalies for correlation purposes. Fortunately, the
anomalies are clearly distinguished from one another chemostrati-
graphically by their strontium-isotope signatures. Whereas
87
Sr/86Sr ratios of 0.7072–0.7073 coincide with the Trezona
anomaly (Halverson et al. 2007a), 87Sr/86Sr through the older
Islay anomaly distinctly declines from c. 0.7068 to 0.7064
(Fig. 4.2; Table 4.1), with the lowest values in the Islay Formation
corresponding to the return towards positive d13Ccarb values
(Sawaki et al. 2010a). Similar 87Sr/86Sr values have been reported
from the likely equivalent interval in the Coates Lake Group
(0.7066; Halverson et al. 2007a) and Bed Group 20 (0.7063;
Fairchild et al. 2000). Thus, it appears that the onset of the
middle-Cryogenian glaciation is preceded by a 0.0005 decline in
the 87Sr/86Sr of seawater following the c. 765 Ma peak of
c. 0.7070 (Fig. 4.1) documented in the Backlundtoppen Form-
ation. In contrast to this negative slope leading up to the middle
Cryogenian glaciation, the 87Sr/86Sr record in the following
interglacial interval has a positive slope (Fig. 4.1), with a smooth
and seemingly rapid increase in 87Sr/86Sr from 0.7067 to 0.7073
recorded in the Tsagaan Oloom Formation in Mongolia (Shields
et al. 1997).
Strontium-rich aragonite and barite cements of the Hayhook
Formation in the upper part of the transgressive portion of the
post-Ice Brook (base Ediacaran) sequence in the Mackenzie
Mountains, NW Canada, have 87Sr/86Sr ratios of c. 0.70715
(Halverson et al. 2007a; Hoffman & Halverson 2011). These
values are effectively indistinguishable from a ratio of 0.70718
measured within the pre-glacial Keele Peak in the same section
(Halverson et al. 2007a) and therefore indicate no net change
in 87Sr/87Sr spanning the end-Cryogenian glaciation.
The sulphur-isotope record spanning the Cryogenian glaciations
is almost entirely based on pyrite and CAS (there being few eva-
porites during this time span) and is intensely variable (Fig. 4.1).
This variability reflects a combination of low seawater sulphate
concentrations and strongly fluctuating environmental conditions
tied to global glaciation (Hurtgen et al. 2002), as well as possible
post-depositional effects. Sulphur-isotope data (Fig. 4.1) from the
Chuar Group in southwestern USA show a return to d34Spyrite
values clustering around 0‰ and low d34S values around
750 Ma, coinciding with at least local development of euxinic
deep waters as determined from iron-speciation data on mid-shelf
black shales (Johnston et al. 2010). Sulphur-isotope data are sparse
for the subsequent interval leading up to the middle Cryogenian
glaciation (i.e. the c. 717 Ma Rapitan glaciation; Macdonald
et al. 2010b). However, abundant pyrite and CAS data have
been produced from the interglacial interval, in particular from
Namibia (Hurtgen et al. 2002; Gorjan et al. 2003) and Australia
(Gorjan et al. 2000; McKirdy et al. 2001). Whereas d34Spyrite
data from within the Sturt and equivalent glacial units in Australia
range from 0 to –20‰, they are strongly positive, with values
commonly 20– 60‰ in post-glacial rocks, such as the Tapley
Hill, Aralka, Datangpo, Rasthof and lower Court forma-
tions (Fig. 4.1). This extreme, immediately post-glacial trend in
pyrite values is matched by highly 34S-enriched d34Ssulphate
values from CAS, such that D34S  0‰. d34Spyrite values then
decline precipitously in the middle interglacial period, paralleling
a smaller decline in d34Ssulphate, resulting in D34S values approach-
ing 30‰. D34S reduces again just before the onset of the end-
Cryogenian glaciation, coinciding with the Trezona negative
d13C anomaly (Fig. 4.1; Table 4.1).
d34S values are again erratic immediately following the
end-Cryogenian glaciation: a trend of rapidly increasing
d34Ssulphate that coincides with a wide range in d34Spyrite (– 30 to
þ20‰) may be sufficient to distinguish this post-glacial (i.e.
 CHEMOSTRATIGRAPHY AND THE NEOPROTEROZOIC GLACIATIONS
basal Ediacaran) sulphur-isotope record from the middle Cryogen-
ian post-glacial record. However, coeval d34S profiles that vary
from more proximal to more open ocean post-glacial sections in
Namibia (Hurtgen et al. 2006) and South China (Li et al. 2010)
suggest significant heterogeneity in the concentration and
sulphur-isotope composition of seawater sulphate and generally
low sulphate concentrations in the earliest Ediacaran Period.
Therefore, sulphur-isotope chemostratigraphy may be useful in
establishing an early Ediacaran age, but would not be reliable for
detailed correlations for sedimentary rocks of this age.
59
Fig. 4.2. Summary stratigraphic plots and
accompanying d13Ccarb profiles and
87
Sr/86Sr data for key sections spanning the
Trezona and Islay pre-glacial (Cryogenian)
negative d13C anomalies. Note the variable
scales across the non-scaled glaciogenic
intervals in each column. Nucc, Nuccaleena
Formation; Brach, Brachina Formation;
Macdon., Macdonaldryggen Member.
Australia data from McKirdy et al. (2001);
Namibia data from Halverson et al. (2002,
2007a), which includes data from both the
inner and outer continental shelf of the
Ombaatjie Formation; Scotland data from
Brasier & Shields (2000), McCay et al.
(2006) and Prave et al. (2009); Svalbard
data from Halverson et al. (2004, 2007a).
The Ediacaran Period
The beginning of the Ediacaran Period coincides with a major
negative d13Ccarb anomaly (the Maieberg anomaly; Fig. 4.1) that
punctuates the final global glaciation of the Neoproterozoic
(Knoll et al. 2006), as discussed above. Although d13Ccarb values
return to near 0‰ within about 3 million years of the end of glacia-
tion (Condon et al. 2005), they then remain relatively low for much
of the early Ediacaran Period, with the exception of a positive spike
to þ6 –10‰ seen in some sections, such as NW Namibia and NE
60 G. P. HALVERSON & G. SHIELDS-ZHOU
Svalbard (Halverson et al. 2005). This pattern contrasts with the
sustained high d13Ccarb values typical of much of the earlier
Neoproterozoic.
Superimposed on the lower average d13Ccarb values in the Edia-
caran Period, the dominant feature in the record is the so-called
Shuram (or Shuram-Wonoka) anomaly, named after the Shuram
Formation in the Huqf Supergroup of Oman (Burns & Matter
1993). The Shuram anomaly features a precipitous drop to
d13Ccarb values , – 10‰, followed by a protracted recovery back
to values near 0‰ (Fig. 4.1). This extraordinary anomaly has
been the subject of considerable recent research and controversy.
However, despite arguments against a primary origin for the
anomaly (Bristow & Kennedy 2008; Knauth & Kennedy 2009;
Derry 2010), anomalies of strikingly similar magnitude, structure
and approximate age occur in multiple sedimentary basins from
across the globe (Table 4.1), including Sr-rich limestones in south-
central Siberia (Pokrovskii et al. 2006; Melezhik et al. 2009).
Thus, the most parsimonious interpretation is that the Shuram
anomaly reflects a primary oceanographic phenomenon.
Despite the large number of chemostratigraphic data that have
been obtained globally from rocks preserving the Shuram
anomaly, the timing and duration of the anomaly and, indeed,
the number of large Ediacaran negative d13Ccarb anomalies
remain controversial. For example, whereas d13Ccarb records
from thick, seemingly continuous carbonate-rich sections in
Oman (Fike et al. 2006; Le Guerroué et al. 2006) and south-central
Siberia (Pokrovskii et al. 2006) clearly show a single, long-lived
negative d13Ccarb anomaly, certain sections from the Doushantuo
Formation, South China, show a pair of large-magnitude
anomalies separated by d13Ccarb values as high as þ6‰ (Jiang
et al. 2007; McFadden et al. 2008). Because the Doushantuo
pattern is preserved in highly condensed sections (relative to sec-
tions from other successions) from a basin that was not demonstra-
bly fully open to the global ocean and has not been reproduced
globally, the more consistent and higher-resolution patterns
derived from Oman and Siberia are favoured in the compilation
in Figure 4.1. However, we acknowledge that the inconsistent pat-
terns between the Doushantuo Formation and correlative units
worldwide pose a problem for carbon-isotope chemostratigraphy.
Separate data sets from the Wonoka Formation, South Australia
(Calver 2000), Oman (Fike et al. 2006) and South China (McFad-
den et al. 2008) also show a decoupling between the d13Ccarb
record spanning the Shuram anomaly and coeval organic carbon
(d13Corg) data (Calver 2000; Fike et al. 2006; McFadden et al.
2008), a pattern that lends support to the popular hypothesis that
the Ediacaran ocean contained a large, reactive organic carbon
pool that strongly modulated d13Ccarb and the oxidation of the
deep oceans (Rothman et al. 2003; Condon et al. 2005).
However, the relationship between d13Ccarb and d13Corg is not
the same in these three successions. For example, d13Corg values
are much more negative in South China than in Oman or South
Australia, and d13Corg values are more variable through the main
part of the anomaly in Oman than elsewhere. Therefore, the signifi-
cance and reliability of the middle Ediacaran d13Corg record
is unclear.
Of critical importance in reconstructing the connections
between perturbations to the exogenic carbon cycle, global
climate and biospheric evolution in the Ediacaran Period is the
connection, if any, between the Shuram anomaly and the short-
lived c. 580 Ma Gaskiers glaciation (Bowring et al. 2003). Is the
Gaskiers glaciation, like the global Cryogenian glaciations, pre-
ceded by a large negative d13C anomaly? Unfortunately, no unam-
biguous evidence for both the glaciation and the full anomaly
occurs in any single stratigraphic succession. Although Halverson
et al. (2005) interpreted major erosional unconformities in succes-
sions hosting the Shuram anomaly, such as the Wonoka Formation
(South Australia) and the Johnnie Formation (Death Valley), to be
related to Gaskiers glacioeustasy, this argument remains speculat-
ive. The problem of resolving the relative timing between d13C
anomalies and glaciation is exacerbated by the possibility that
whereas the anomaly should be globally synchronous, the glacia-
tion may be diachronous, or there may be multiple discrete glacia-
tions (e.g. Chumakov 2010). Indeed, d13Ccarb values are highly
variable in relation to other reported Ediacaran glacial deposits,
such as in the Kimberley of northern Australia, where positive
d13C values characterize carbonates overlying the Egan Formation
(Corkeron 2007, 2011). Other contrasting d13Ccarb values are
found in the Hankalchough Formation in the Tarim Basin, NW
China (Xiao et al. 2004), and the Hongtiegou Formation in the
Chaidam Basin, NW China (Shen et al. 2010). Some authors
have even argued that some of the purported Ediacaran glacial
deposits, such as the Hankalchough (Chumakov 2010) and the
Hongtiegou (Shen et al. 2010) formations significantly post-date
the Gaskiers glaciation (Chumakov 2009). Germs et al. (2009)
argue for a latest Ediacaran (c. 547 Ma) ‘Schwarzrand’ glaciation
based on evidence from the upper Nama Group, Namibia.
Notwithstanding the ambiguity in the age, timing, number and
distribution of Ediacaran-aged glacial deposits, carbon-isotope
data from the basal Ediacaran Nyborg Formation and the overlying
glaciogenic Mortensnes Formation of the Varanger Peninsula,
northern Norway, do appear to resolve the relative timing
between at least one middle Ediacaran glaciation and the Shuram
d13Ccarb anomaly. Dolomite beds within Member E in the upper-
most Nyborg Formation, in the least-truncated section beneath the
major erosional unconformity at the base of the glaciogenic Mor-
tensnes Formation (Edwards 1984), have d13Ccarb compositions
of –7.6 to –9.9‰, while the matrix of a carbonate-dominated dia-
mictite within the Mortensnes Formation has a d13Ccarb composition
of –10.4‰ (see Rice et al. 2011). Because the Shuram anomaly is
the only interval in the Ediacaran Period where d13Ccarb values are
known to drop below –7‰, these data are interpreted to indicate
that the Mortensnes glaciation post-dated the nadir of the negative
anomaly (Halverson et al. 2005). In this regard, an Ediacaran glacia-
tion, although seemingly not global, may have been triggered by a
similar mechanism as the Cryogenian glaciations.
The return to positive d13Ccarb values following the Shuram
anomaly occurs late in the Ediacaran, by c. 551 Ma, implying
that the Ediacaran fauna first appeared during the anomaly
(Condon et al. 2005). A subsequent, short-lived negative d13C
anomaly coincides with the Precambrian –Cambrian boundary
(Magaritz et al. 1986; Knoll et al. 1995; Saylor et al. 1998;
Amthor et al. 2003), but there is no evidence to link this
anomaly with a glaciation.
Strontium-isotope ratios from the Mackenzie Mountains
indicate little net change in seawater 87Sr/86Sr spanning the end-
Cryogenian glaciation (Fig. 4.2). However, data from strontium-
rich limestones in the lower Maieberg Formation, NW Namibia,
show a rapid rise in 87Sr/86Sr to values .0.7078 following
post-glacial maximum flooding and prior to recovery in d13Ccarb
values to near 0‰ (Halverson et al. 2007; cf. Shields 2007).
This positive spike in 87Sr/86Sr is a predictable result of extremely
high CO2 levels and corresponding elevated silicate weathering
rates following deglaciation (Higgins & Schrag 2003). Simple
geochemical modelling also implies that the spike to more
radiogenic values should be short-lived, with 87Sr/86Sr approach-
ing background values after CO2 declined to more typical
levels within a few million years of deglaciation (Le Hir et al.
2009). This decline may be recorded in data from the Una
Group (Irecê basin) on the São Francisco Craton, Brazil (Misi &
Veizer 1998), where the strontium- and carbonate-isotope
compositions suggest an early Ediacaran age, consistent with
recent detrital zircon data from the equivalent Bambui Group
(Sial et al. 2010).
A much better established pattern in the Ediacaran strontium-
isotope record is the subsequent rise to 87Sr/86Sr  0.7080 that
characterizes the latter part of the Ediacaran Period (e.g.
Kaufman et al. 1997; Jacobsen & Kaufman 1999; Shields 2007;
Melezhik et al. 2009). Sawaki et al. (2010b) suggested additional
 CHEMOSTRATIGRAPHY AND THE NEOPROTEROZOIC GLACIATIONS
structure in the late Ediacaran record based on data they obtained
from a drill core through the Doushantuo Formation, including a
drop to ,0.7080 prior to the Shuram anomaly and a peak in
87
Sr/86Sr of 0.7090 coinciding with the protracted low d13Ccarb
of the anomaly. Regarding the high values, as for Oman, where
similarly high values have been reported for the late Ediacaran
(Burns et al. 1994), none of these highly enriched values is
preserved in samples retaining high strontium concentrations.
Conversely, strontium-rich limestones in the upper Doushantuo
Formation in the core have 87Sr/86Sr values of c. 0.7085, similar
to strontium-isotope data straddling the Precambrian –Cambrian
boundary in other well-studied successions (Brasier et al. 1996;
Kaufman et al. 1997; Saylor et al. 1998). Gaucher et al. (2009)
report 87Sr/86Sr values of ,0.7080 in the late Ediacaran Arroyo
del Soldado Group on the Rio de la Plata craton, lending support
to the possibility that 87Sr/86Sr may have indeed fluctuated
strongly in the late Ediacaran Period, much as it did in the sub-
sequent Early Cambrian (Halverson et al. 2010). Evidently,
much work remains in elucidating the fine structure in the late
Ediacaran 87Sr/86Sr record.
The relationship between middle Ediacaran 87Sr/86Sr and the
Gaskiers (or other Ediacaran) glaciation is virtually unconstrained.
However, strontium-isotope data from highly strontium-enriched
limestones in southern Siberia show 87Sr/86Sr ¼ 0.7081 at about
the level of the nadir in the Shuram anomaly, followed by an
increase towards 0.7086 up-section, but still within the range of
very low d13Ccarb (Pokrovskii et al. 2006; Melezhik et al. 2009).
Anomalously 34S-enriched seawater sulphate compositions
bookend the Ediacaran Period (Fig. 4.1) in many successions. As
discussed above, variable, but predominantly high d34Ssuphate and
low D34S values near 0‰ prevail in the basal Ediacaran Period
(Fig. 4.1). By the early Cambrian, d34Ssulphate, as measured in a
variety of minerals, including evaporites, francolite and carbonate,
was again high (35 –40‰) in most basins (e.g. Holser & Kaplan
1966; Strauss 1993; Shields et al. 1999; Kampschulte & Strauss
2004). However, d34Ssulphate from the late Ediacaran in South
China (McFadden et al. 2008) and southern Namibia (Ries et al.
2009) are generally lower, with concomitant decreases in
D34S. The source of this substantial spatial variability in the late
Ediacaran sulphur-isotope record, whether oceanographic,
related to post-depositional influences or analytical artefacts, is
not resolved.
The Ediacaran sulphur-isotope record as compiled in Figure 4.1
comprises a detailed data set of d34Ssulphate (CAS) and d34Spyrite
data from the Nafun Group, Oman (Fike et al. 2006) and pyrite
data from the Pertatataka Formation, central Australia (Gorjan
et al. 2000). The Oman record, together with other fragmentary
data (Fig. 4.1), indicates that more typical d34Ssulphate values
close to 20‰ are the norm for the Ediacaran Period, although eva-
porite and CAS data appear to be contradictory at times (cf. Fike
et al. 2006 with Schröder et al. 2004). At the same time, data
from both Oman and central Australia display a spike to positive
d34Spyrite values in the early Ediacaran Period, followed by a
gradual decrease through the middle Ediacaran Period, resulting
in a steady increase in d34S, a predictable result of increasing sea-
water sulphate concentrations (Halverson & Hurtgen 2007).
Average d34Spyrite values in the deepwater Conception Group
(Avalon Peninsula, SE Newfoundland; Canfield et al. 2007)
show a decrease from about þ20 to 0‰ across the Gaskiers
glaciation, but the d34Ssulphate record spanning the glaciation is
unknown. Iron-speciation data indicate ferruginous (and presum-
ably low sulphate) conditions during the Gaskiers glaciation
(Canfield et al. 2008). Detailed d34Ssulphate (CAS) records have
been produced across the onset of the Shuram –Wonoka anomaly.
In Oman, a spike in d34Ssulphate to þ29‰ coincides with the
anomaly (Fike et al. 2006). Across the presumably equivalent
interval in Death Valley (Kaufman et al. 2007) and South China
(McFadden et al. 2008), d34Ssulphate decreases to values ,20‰.
Thus, although d34S records show tantalizing connections with
61
other major Ediacaran events, either d34S compositions of late
Neoproterozoic seawater sulphate were highly heterogeneous
(Ries et al. 2009), inferred correlations are inaccurate, or the
measured sulphur-isotope compositions in some or all of the sedi-
mentary successions do not preserve reliable signatures of primary
seawater composition.
Discussion and conclusions
Precise radiometric ages are the gold standard for drawing precise
interbasinal correlations, particularly in the absence of a robust
biostratigraphy. Direct ages are also required for calibrating the
sedimentary record, but they are frustratingly rare in most of the
successions that archive important information about the evolution
of Earth’s surface environment in the Neoproterozoic. Thus, other
tools and many assumptions are required to merge the fragmentary
records from across the globe into one coherent Neoproterozoic
chronology. Isotope chemostratigraphy has long been heralded
as a possible solution (Knoll & Walter 1992). Insofar as the
popular proxies for ocean chemistry (d13C, d34S, 87Sr/86Sr) are
reliable, geologists can characterize the Neoproterozoic record
one fragment at a time. Accordingly, isotope geochemistry has
been widely applied to stratigraphic studies of the Neoproterozoic,
with a strong focus on the glaciogenic record.
An intimate relationship between extreme perturbations to the
global carbon cycle, as recorded in the marine d13Ccarb record,
and episodes of widespread glaciation (Knoll et al. 1986) has
long been the starting point for these studies. However, even as
this fundamental observation about the behaviour of the Neopro-
terozoic Earth has withstood the test of time, diverse studies
from across the world have added important new details to this
broad pattern. Long-lived positive carbon-isotope trends had
been ascribed to pre-glacial oceanographic conditions, while nega-
tive carbon-isotope anomalies were attributed to post-glacial ‘cap
carbonates’, providing for elegantly simple models for the inter-
relation between d13Ccarb, carbon burial, atmospheric CO2 and
glaciation (e.g. Kaufman et al. 1997). The recognition that a
large negative d13Ccarb anomaly actually preceded the end-
Cryogenian glaciation shattered this model of a simple cause-
and-effect relationship between the two and opened the door for
the compelling but controversial Snowball Earth hypothesis
(Hoffman et al. 1998), which continues to dominate the debate
about the Neoproterozoic glaciations.
It is now recognized that negative d13C anomalies also precede
middle Cryogenian (McCay et al. 2006; Prave et al. 2009; Mac-
donald et al. 2010b) and middle Ediacaran glaciations (Halverson
et al. 2005). Furthermore, in all cases, it appears that the initiation
of glaciation followed a return towards positive d13Ccarb values
after the d13Ccarb minima, thus further complicating an already
complex relationship between the anomalies and global cooling.
This motif in the coupled d13C-climate record renders using the
carbon-isotope record alone for global correlations insufficient,
but fortunately it projects upon a marine strontium-isotope
record with a strong unidirectional trend (Fig. 4.1). 87Sr/86Sr
increases steadily during the Neoproterozoic from 0.7055 around
the Mesoproterozoic– Neoproterozoic boundary to 0.7085 at the
Neoproterozoic –Palaeozoic boundary, interrupted by only a few
significant but short-lived declines (Fig. 4.1). Consequently, sea-
water 87Sr/86Sr ratios are different for each of the large negative
anomalies associated with the Neoproterozoic glaciations
(Table 4.1). The potential of strontium isotopes as a tracer of
global environmental change remains to be fully realized.
However, abrupt rises, most notably following the Cryogenian
glaciations, are consistent with enhanced chemical weathering
after glaciation. Conversely, changes in seawater 87Sr/86Sr appear
to have been relatively muted during glaciations.
At this stage, the Neoproterozoic record is now sufficiently well
understood that the combination of carbon- and strontium-isotope
62 G. P. HALVERSON & G. SHIELDS-ZHOU
stratigraphy, particularly if integrated with stratigraphic and sedi-
mentological data, enables confident correlations and assignment
of robust, if only relative, ages on carbonate-rich successions
that otherwise lack firm age control. High-quality data sets span-
ning variable portions of the Neoproterozoic record and tied to
an increasing number of precise U –Pb zircon ages have now
been produced from across the world (e.g. Macdonald et al.
2010b). The result is a convergence in carbon- and strontium-
isotope compilations towards a consistent pattern, even if many
of the finer details remain obscure.
The integrated stratigraphic, d13C and 87Sr/86Sr records serve
as a chronological template to the Neoproterozoic onto which
other secular data sets can be added, such as sulphur-isotope
compositions, redox specific data, biomarkers and biostratigraphy.
The sulphur-isotope record is patchy, particularly for the first
half of the Neoproterozoic, and highly variable, with clear
instances where data sets spanning equivalent-aged rocks from
different parts of the world, and even single basins, are not
consistent with one another. Thus, much work remains to be
done to establish the reliability of the sulphur-isotope proxy and
the causes, either related to oceanographic conditions or diagenetic
process, for the high degree of scatter in the record. Nevertheless,
several of the salient patterns, mostly notably extremely high
d34Ssulphate and d34Spyrite values, are temporally linked to glaciation,
indicating that the record can be used to help reconstruct the
palaeoenvironmental evolution of Neoproterozoic ocean chem-
istry. Specifically, large variations in d34S of sulphate and pyrite
and the difference between these two values in coeval sediments
(D34S) is an indirect measure of fluctuations in the redox state of
global seawater (Hurtgen et al. 2005). The sustained and very
high d34Spyrite (20 –60‰) values recorded in the post-Sturtian
Tapley Hill – Aralka formations in Australia (Gorjan et al. 2000)
and the peak in d34Ssulphate straddling the Precambrian – Cambrian
boundary stand out as unique chronostratigraphic markers in the
otherwise noisy sulphur-isotope record (Fig. 4.1).
Iron-speciation data has become a popular new tool for recon-
structing water column redox conditions (Lyons & Severmann
2006) that has been successfully applied to the Neoproterozoic
(e.g. Canfield et al. 2007, 2008). Although it remains to be
shown that distinct trends can be correlated beyond a single
basin (Johnston et al. 2010), or are even consistent within a
single basin, sharp fluctuations in normalizsed highly reactive
iron content in shales (FeHR/FeT), for example, are closely
linked to glaciation and other proxy evidence for oscillations in
the redox chemistry of the oceans (Canfield et al. 2007, 2008).
Although not discussed here, many other chemostratigraphic
proxies show great promise for tracing evolving redox conditions
through the Neoproterozoic and evaluating the highly unusual
environment in the aftermath of global glaciations (Halverson
et al. 2010, and references therein). Furthermore, biostratigraphy
is increasingly being tightly integrated with chemostratigraphic
data in order to evaluate more closely the connections between
biospheric and biogeochemical change (e.g. McFadden et al.
2008; Nagy et al. 2009; Macdonald et al. 2010a). Thus, there is
little doubt that chemostratigraphy will continue to elucidate the
finer details in the number, timing and correlation of Neoprotero-
zoic glaciations and the environmental conditions leading up to
and following them.
This represents a contribution of the IUGS- and UNESCO-funded IGCP (Inter-
national Geoscience Programme) Project #512.
References
Alene, M., Jenkin, G. R. T., Leng, M. J. & Darbyshire, D. P. F. 2006.
The Tambien Group, Ethiopia: An early Cryogenian (ca. 800– 735
Ma) Neoproterozoic sequence in the Arabian– Nubian Shield. Pre-
cambrian Research, 147, 79 –99.
Allen, P. A. & Etienne, J. L. 2008. Sedimentary challenge to Snowball
Earth. Nature Geoscience, 1, 817– 825.
Amthor, J. E., Grotzinger, J. P., Schröder, S., Bowring, S. A., Rame-
zani, J., Martin, M. W. & Matter, A. 2003. Extinction of Cloudina
and Namacalathus at the Precambrian – Cambrian boundary in Oman.
Geology, 31, 431–434.
Bahlburg, H. & Dobrzinski, N. 2011. A review of the Chemical Index
of Alteration (CIA) and its application to the study of Neoproterozoic
glacial deposits and climate transitions. In: Arnaud, E., Halverson,
G. P. & Shields-Zhou, G. (eds) The Geological Record of Neo-
proterozoic Glaciations. Geological Society, London, Memoirs, 36,
81 – 92.
Bailey, T. R., McArthur, J. M., Prince, H. & Thirlwall, M. F. 2000.
Dissolution methods for strontium isotope stratigraphy: whole rock
analysis. Chemical Geology, 167, 313–319.
Banner, J. L. & Hanson, G. N. 1990. Calculation of simultaneous isoto-
pic and trace element variations during water– rock interaction with
applications to carbonate diagenesis. Geochimica et Cosmochimica
Acta, 54, 3123– 3137.
Bao, H., Lyons, J. R. & Zhou, C. 2008. Triple oxygen isotope evidence
for elevated CO2 levels after a Neoproterozoic glaciation. Nature,
453, 504– 506.
Bartley, J. K., Semikhatov, M. A., Kaufman, A. J., Knoll, A. H.,
Pope, M. C. & Jacobsen, S. B. 2001. Global events across the Meso-
proterozoic – Neoproterozoic boundary: C and Sr isotopic evidence
from Siberia. Precambrian Research, 111, 165– 202.
Borg, G., Kärner, K., Buxton, M., Armstrong, R. & van der Merwe,
S. W. 2003. Geology of the Skorpion supergene zinc deposit,
southern Namibia. Economic Geology, 98, 749–771.
Bowring, S., Myrow, P., Landing, E., Ramezani, J. & Grotzinger, J.
2003. Geochronological constraints on terminal Proterozoic events
and the rise of the Metazoans. Geophysical Research Abstracts
(EGS, Nice), 50, 13219.
Brand, U. & Veizer, J. 1981. Chemical diagenesis of a multicomponent
carbonate system – 2: stable isotopes. Journal of Sedimentary Petrol-
ogy, 51, 987– 997.
Brasier, M. D. & Lindsay, J. F. 1995. A billion years of environmental
stability and the emergence of eukaryotes: New data from northern
Australia. Geology, 26, 555–558.
Brasier, M. D. & Shields, G. 2000. Neoproterozoic chemostratigraphy
and correlation of the Port Askaig glaciation, Dalradian Supergroup
of Scotland. Journal of the Geological Society, London, 157, 909–914.
Brasier, M. D., Shields, G. A., Kuleshov, V. N. & Zhegallo, E. A.
1996. Integrated chemo- and biostratigraphic calibration of early
animal evolution: Neoproterozoic– early Cambrian of southwest
Mongolia. Geological Magazine, 133, 445–485.
Bristow, T. F. & Kennedy, M. J. 2008. Carbon isotope excursions and
the oxidant budget of the Ediacaran atmosphere and ocean.
Geology, 36, 863–866.
Burns, S. J. & Matter, A. 1993. Carbon isotopic record of the latest Pro-
terozoic from Oman. Eclogae Geologicae Helvetiae, 86, 595– 607.
Burns, S. J., Haudenschild, U. & Matter, A. 1994. The strontium
isotopic composition of carbonates from the late Precambrian
(,560–540 Ma) Huqf Group of Oman. Chemical Geology, 111,
269– 282.
Calver, C. 2000. Isotope stratigraphy of the Ediacaran (Neoproterozoic
III) of the Adelaide rift complex, Australia, and the overprint of
water column stratification. Precambrian Research, 100, 121– 150.
Canfield, D. E. 1989. Reactive iron in marine sediments. Geochimica et
Cosmochimica Acta, 53, 619– 632.
Canfield, D. E. & Teske, A. 1996. Late Proterozoic rise in atmospheric
oxygen concentration inferred from phylogenetic and sulfur-isotope
studies. Nature, 382, 127–132.
Canfield, D. E., Poulton, S. W., Knoll, A. H., Narbonne, G. M.,
Ross, G., Goldberg, T. & Strauss, H. 2008. Ferruginous conditions
dominated later Neoproterozoic deep-water chemistry. Science, 321,
949– 952.
Canfield, D. E., Poulton, S. W. & Narbonne, G. M. 2007. Late-
Neoproterozoic deep-ocean oxygenation and the rise of animal life.
Science, 315, 92 – 95.
Chumakov, N. M. 2009. The Baykonurian Glaciohorizon of the Late
Vendian. Stratigraphy and Geological Correlation, 17, 373–381.
 CHEMOSTRATIGRAPHY AND THE NEOPROTEROZOIC GLACIATIONS
Chumakov, N. M. 2010. Neoproterozoic glacial events in Eurasia. In:
Gaucher, C., Sial, A. N., Halverson, G. P. & Frimmel, H. E.
(eds) Neoproterozoic-Cambrian Tectonics, Global Change and
Evolution: A Focus on Southwestern Gondwana. Developments in
Precambrian Geology, 16. Elsevier, Dordrecht, 389– 403.
Condon, D., Zhu, M., Bowring, S., Jin, Y., Wang, W. & Yang, A.
2005. From the Marinoan glaciation to the oldest bilaterians:
U –Pb ages from the Doushantou Formation, China. Science, 308,
95 – 98.
Corkeron, M. 2007. ‘Cap carbonates’ and Neoproterozoic glacigenic
successions from the Kimberly region, north-west Australia. Sedi-
mentology, 54, 871– 903.
Corkeron, M. 2011. Neoproterozoic glacial deposits of the Kimberly
Region and northwestern Northern Territory, Australia. In:
Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The
Geological Record of Neoproterozoic Glaciations. Geological
Society, London, Memoirs, 36, 659–672.
Corsetti, F. A. & Kaufman, A. J. 2003. Statigraphic investigations of
carbon-isotope anomalies and Neoproterozoic ice ages in Death
Valley, California. Geological Society of America Bulletin, 115,
916– 932.
Cozzi, A., Allen, P. A. & Grotzinger, J. P. 2004. Understanding
carbonate ramp dynamics from C profiles: Examples from
the Neoproterozoic Buah Formation of Oman. Terra Nova, 16,
62 – 67.
Derry, L. A. 2010. A burial diagenesis origin for the Ediacaran Shuram-
Wonoka anomaly. Earth and Planetary Science Letters, 295,
152– 162.
Derry, L. A., Kaufman, A. J. & Jacobsen, S. B. 1992. Sedimentary
cycling and environmental change in the Late Proterozoic: evidence
from stable and radiogenic isotopes. Geochimica et Cosmochimica
Acta, 56, 1317–1329.
Detmers, J., Brüchert, V., Habicht, K. S. & Kuever, J. 2001. Diversity
of sulfur isotope fractionations by sulfate-reducing prokaryotes.
Applied and Environmental Microbiology, 67, 888– 894.
Edmond, J. M. 1992. Himalayan tectonic, weathering processes, and
strontium isotope record in marine limestones. Science, 258,
1594– 1597.
Edwards, M. B. 1984. Sedimentology of the Upper Proterozoic glacial
record, Vestertana Group, Finnmark, North Norway. Norges geolo-
giske Undersøkelse Bulletin, 394, 1– 76.
Evans, D. A. D. 2006. Proterozoic low orbital obliquity and axial-
dipolar geomagnetic field from evaporite paleolatitudes. Nature,
444, 51– 55.
Fairchild, I. J., Marshall, J. D. & Bertrand-Sarfati, J. 1990. Strati-
graphic shifts in carbon isotopes from Proterozoic stromatolitic
carbonates (Mauritania): influences of primary mineralogy and dia-
genesis. American Journal of Science, 290A, 46 –79.
Fairchild, I. J., Spiro, B., Herrington, P. M. & Song, T. 2000. Controls
on Sr and C isotope compositions of Neoproterozoic Sr-rich
limestones of East Greenland and North China. In: Grotzinger, J.
& James, N. (eds) Carbonate Sedimentation, Diagenesis in an
Evolving Precambrian World. SEPM Special Publication, 67,
297– 313.
Fanning, C. M. & Link, P. K. 2008. Age constraints for the Sturtian gla-
ciation: data from the Adelaide Geosyncline, South Australia and
Pocatello Formation, Idaho, USA. Selwyn Symposium 2008, Geo-
logical Society of Australia, Extended Abstracts, 91, 57– 62.
Farquhar, J., Bao, H. & Thiemens, M. 2000. Atmospheric influence of
Earth’s earliest sulfur cycle. Science, 289, 756– 758.
Fike, D. A., Grotzinger, J. P., Pratt, L. M. & Summons, R. E. 2006.
Oxidation of the Ediacaran ocean. Nature, 444, 744– 747.
Fölling, P. G. & Frimmel, H. W. 2002. Chemostratigraphic correlation
of carbonate successions in the Gariep and Saldania belts, Namibia
and South Africa. Basin Research, 14, 69 –88.
Frank, T. D., Kah, L. C. & Lyons, T. W. 2003. Changes in organic
matter production and accumulation as a mechanism for isotopic
variation in the Mesoproterozoic ocean. Geological Magazine, 140,
397– 420.
Frimmel, H. 2009. Trace element distribution in Neoproterozoic carbon-
ates as palaeoenvironmental indicator. Chemical Geology, 258,
338– 353.
63
Frimmel, H. W., Klötzi, U. S. & Siegfried, P. R. 1996. New Pb/Pb
single zircon age constraints on the timing of Neoproterozoic glacia-
tion and continental break-up in Namibia. The Journal of Geology,
104, 459–469.
Gaucher, C., Sial, A. N., Poiré, D., Gómez-Peral, L., Ferreira, V. P.
& Pimentel, M. M. 2009. Chemostratigraphy. Neoproterozoic –
Cambrian evolution of the Rı́o de la Plata Palaeocontinent. In:
Gaucher, C., Sial, A. N., Halverson, G. P. & Frimmel, H. E.
(eds) Neoproterozoic –Cambrian Tectonics, Global Change and
Evolution: A Focus on Southwestern Gondwana. Developments in
Precambrian Geology, 16, 115–122.
Germs, G. J. B., Miller, R. McG., Frimmel, H. E. & Gaucher, C. 2009.
Syn- to late-orogenic sedimentary basins of southwestern Africa.
Neoproterozoic to Early Palaeozoic evolution of Southwestern
Africa. In: Gaucher, C., Sial, A. N., Halverson, G. P. &
Frimmel, H. E. (eds) Neoproterozoic– Cambrian Tectonics, Global
Change and Evolution: A Focus on Southwestern Gondwana. Devel-
opments in Precambrian Geology, 16, 183– 203.
Gorjan, P., Veevers, J. J. & Walter, M. R. 2000. Neoproterozoic sulfur-
isotope variation in Australia and global implications. Precambrian
Research, 100, 151–179.
Gorjan, P., Walter, M. R. & Swart, R. 2003. Global Neoproterozoic
(Sturtian) post-glacial sulfide-sulfur isotope anomaly recognised in
Namibia. Journal of African Earth Sciences, 36, 89 –98.
Gorokhov, I. M., Semikhatov, M. A., Baskakov, A. V., Kutyavin, E.
P., Melnikov, N. N., Sochava, A. V. & Turchenko, T. L. 1995. Sr
isotopic composition in Riphean, Vendian, and Lower Cambrian
carbonates from Siberia. Stratigraphy and Geological Correlation,
3, 1 – 28.
Grotzinger, J. P., Bowring, S. A., Saylor, B. Z. & Kaufman, A. J.
1995. Biostratigraphic and geochronologic constraints on early
animal evolution. Science, 270, 598–604.
Halverson, G. P. 2006. A Neoproterozoic Chronology. In: Xiao, S. &
Kaufman, A. J. (eds) Neoproterozoic Geobiology and Paleobiology.
Springer, 231– 271.
Halverson, G. P. & Hurtgen, M. T. 2007. Ediacaran growth of the
marine sulfate reservoir. Earth and Planetary Science Letters, 263,
32– 44.
Halverson, G. P., Hoffman, P. F., Schrag, D. P. & Kaufman, A. J.
2002. A major perturbation of the carbon cycle before the Ghaub
glaciation (Neoproterozoic) in Namibia: prelude to snowball Earth?
Geochemistry, Geophysics, Geosystems, 3, 10.1029/2001GC000244.
Halverson, G. P., Maloof, A. C. & Hoffman, P. F. 2004. The Marinoan
glaciation (Neoproterozoic) in Svalbard. Basin Research, 16,
297– 324.
Halverson, G. P., Hoffman, P. F., Schrag, D. P., Maloof, A. C. & Rice,
A. H. 2005. Towards a Neoproterozoic composite carbon-isotope
record. Geological Society of America Bulletin, 117, 1181–1207.
Halverson, G. P., Dudas, F. O., Maloof, A. C. & Bowring, S. A.
2007a. Evolution of the 87Sr/86Sr composition of Neoproterozoic
seawater. Palaeogeography, Palaeoclimatology, Palaeoecology,
256, 103–129.
Halverson, G. P., Maloof, A. C., Schrag, D. P., Dudas, F. O. &
Hurtgen, M. T. 2007b. Stratigraphy and geochemistry of a ca 800
Ma negative carbon isotope interval in northeastern Svalbard. Chemi-
cal Geology, 237, 5 –27.
Halverson, G. P., Wade, B. P., Hurtgen, M. T. & Barovich, K. M.
2010. Neoproterozoic chemostratigraphy. Precambrian Research,
182, 337–350.
Hayes, J. M., Kaplan, I. R. & Wedeking, K. W. 1983. Precambrian
organic geochemistry, preservation of the record. In: Schopf, J. W.
(ed.) Earth’s Earliest Biosphere. Its Origin and Evolution. Princeton
University Press, 93 – 134.
Hayes, J. M., Strauss, H. & Kaufman, A. J. 1999. The abundance of 13C
in marine organic matter and isotopic fractionation in the global bio-
geochemical cycle of carbon during the past 800 Ma. Chemical
Geology, 161, 103– 125.
Higgins, J. A. & Schrag, A. P. 2003. Aftermath of a snowball Earth.
Geochemistry, Geophysics, Geosystems, 4, 1028.
Hill, A. C. 2005. Stable isotope stratigraphy, GSWA Lancer 1, Officer
Basin, Western Australia. In: Mory, A. J. & Haines, P. W. (eds)
GSWA Lancer 1 Well Completion Report (Interpretive Papers)
64 G. P. HALVERSON & G. SHIELDS-ZHOU
Officer and Gunbarrel Basins, Western Australia. Western Australia
Geological Survey Record 2005/4, 1– 11.
Hill, A. C., Arouri, K., Gorjan, P. & Walter, M. R. 2000. Geochem-
istry of marine and non-marine environments of a Neoproterozoic
cratonic carbonate/evaporite: the Bitter Springs Formation, Central
Australia. In: Grotzinger, J. P. & James, N. P. (eds) Carbonate
Sedimentation and Diagenesis in an Evolving Precambrian World.
SEPM Tulsa Special Publications, 67, 327– 344.
Hoffman, P. F. & Halverson, G. P. 2011. Neoproterozoic glacial record
in the Mackenzie Mountains, northern Canadian Cordillera. In:
Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The
Geological Record of Neoproterozoic Glaciations. Geological
Society, London, Memoirs, 36, 397–411.
Hoffman, P. F. & Li, Z.-X. 2009. A palaeogeographic context for Neopro-
terozoic glaciation. Palaeogeography, Palaeoclimatology, Palaeo-
ecology, 277, 158– 172.
Hoffman, P. F. & Schrag, D. P. 2002. The snowball Earth hypothesis:
testing the limits of global change. Terra Nova, 14, 129–155.
Hoffman, P. F., Kaufman, A. J., Halverson, G. P. & Schrag, D. P.
1998. A Neoproterozoic snowball Earth. Science, 281, 1342– 1346.
Hoffman, P. F., Halverson, G. P., Domack, E. W., Husson, J. M.,
Higgins, J. A. & Schrag, D. P. 2007. Are basal Ediacaran (635
Ma) post-glacial ‘cap dolostones’ diachronous? Earth and Planetary
Science Letters, 258, 114– 131.
Hoffman, P. F., Macdonald, F. A. & Halverson, G. P. 2011. Chemical
sediments associated with Neoproterozoic glaciations: iron for-
mation, cap carbonate, barite and phosphorite. In: Arnaud, E.,
Halverson, G. P. & Shields-Zhou, G. (eds) The Geological
Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 67– 80.
Hoffmann, K. H., Condon, D. J., Bowring, S. A. & Crowley, J. L.
2004. A U– Pb zircon date from the Neoproterozoic Ghaub For-
mation, Namibia: constraints on Marinoan glaciation. Geology, 32,
817– 820.
Holser, W. T. 1997. Geochemical events documented in inorganic carbon
isotopes. Palaeogeography, Palaeoclimatology, Palaeoecology, 132,
173– 182.
Holser, W. T. & Kaplan, I. R. 1966. Isotope geochemistry of sedimen-
tary sulfates. Chemical Geology, 1, 93 –135.
Hough, M. L., Shields, G. A., Evins, L. Z., Strauss, H., Henderson, R.
A. & Mackenzie, S. 2006. A major sulphur-isotope event at c. 510
Ma: a possible anoxia– extinction –volcanism connection during the
Early– Middle Cambrian transition? Terra Nova, 18, 257– 263.
Hsu, K. J., Oberhänsli, H., Gao, J. Y., Shu, S., Haihong, C. & Krähen-
bühl, U. 1985. ‘Strangelove ocean’ before the Cambrian explosion.
Nature, 316, 809–811.
Hurtgen, M. T., Arthur, M. A., Suits, N. & Kaufman, A. J. 2002. The
sulfur isotopic composition of Neoproterozoic seawater sulfate:
implications for snowball Earth? Earth and Planetary Science
Letters, 203, 413–429.
Hurtgen, M. T., Halverson, G. P., Arthur, M. A. & Hoffman, P. F.
2006. Sulfur cycling in the aftermath of a Neoproterozoic (Marinoan)
snowball glaciation: Evidence for a syn-glacial sulfidic deep ocean.
Earth and Planetary Science Letters, 245, 551–570.
Jacobsen, S. B. & Kaufman, A. J. 1999. The Sr, C, and O
isotopic evolution of Neoproterozoic seawater. Chemical Geology,
161, 37 – 57.
Jiang, G., Kaufman, A. J., Christie-Blick, N., Zhang, S. & Wu, H.
2007. Carbon isotope variability across the Ediacaran Yangtzee plat-
form in South China: Implications for a large surface-to-deep ocean
gradient. Earth and Planetary Science Letters, 261, 303– 320.
Johnston, D. T., Schmitz, M. D. et al. 2005. Active microbial
sulfur disproportionation in the Mesoproterozoic. Science, 310,
1477– 1479.
Johnston, D. T., Poulton, S. W., Dehler, C., Porter, S., Husson, J.,
Canfield, D. E. & Knoll, A. H. 2010. An emerging picture of Neo-
proterozoic ocean chemistry: Insights from the Chuar Group, Grand
Canyon, USA. Earth and Planetary Science Letters, 290, 64– 73.
Jones, D. S., Maloof, A. C., Hurtgen, M. T., Rainbird, R. H. &
Schrag, D. P. 2010. Regional and global chemostratigraphic corre-
lation of the early Neoproterozoic Shaler Sueprgroup, Victoria
Island, Northwestern Canada. Precambrian Research, 181, 43 –63.
Kampschulte, A. & Strauss, H. 2004. The sulfur isotopic evolution of
Phanerozoic seawater based on the analysis of structurally substituted
sulfate in carbonates. Chemical Geology, 204, 255–286.
Kasemann, S. A., Hawkesworth, C. J., Prave, A. R., Fallick, A. E. &
Pearson, P. 2005. Boron and calcium isotope composition in Neo-
proterozoic carbonate rocks from Namibia: evidence for extreme
environmental change. Earth and Planetary Science Letters, 231,
73 – 86.
Kaufman, A. J. & Knoll, A. H. 1995. Neoproterozoic variations in the C-
isotopic composition of seawater. Precambrian Research, 73, 27–49.
Kaufman, A. J., Hayes, J. M., Knoll, A. H. & Germs, G. J. B. 1991. Iso-
topic composition of carbonates and organic carbon from upper
Proterozoic successions in Nambia. Precambrian Research, 49,
301– 327.
Kaufman, A. J., Jacobsen, S. B. & Knoll, A. H. 1993. The Vendian
record of Sr and C isotopic variations in seawater: Implications for
tectonics and paleoclimate. Earth and Planetary Science Letters,
120, 409– 430.
Kaufman, A. J., Knoll, A. H. & Narbonne, G. M. 1997. Isotopes, ice
ages, and terminal Proterozoic earth history. Proceedings of the
National Academy of Science (USA), 94, 6600–6605.
Kaufman, A. J., Corsetti, F. A. & Varni, M. A. 2007. The effect of
rising atmospheric oxygen on carbon and sulfur isotope anomalies
in the Neoproterozoic Johnnie Formation, Death Valley, USA.
Chemical Geology, 237, 47 –63.
Kennedy, M. J., Runnegar, B., Prave, A. R., Hoffmann, K.-H. &
Arthur, M. 1998. Two or four Neoproterozoic glaciations?
Geology, 26, 1059–1063.
Kennedy, M. J., Christie-Blick, N. & Sohl, L. E. 2001. Are Proterozoic
cap carbonates and isotopic excursions a record of gas hydrate
destabilization following Earth’s coldest intervals? Geology, 29,
443– 446.
Key, R. M., Liyungu, A. K., Njamu, F. M., Somwe, V., Banda, J.,
Mosley, P. N. & Armstrong, R. A. 2001. The western arm of the
Lufilian Arc in NW Zambia and its potential for copper mineraliz-
ation. Journal of African Earth Sciences, 33, 503– 528.
Knauth, L. P. & Kennedy, M. J. 2009. The late Precambrian greening of
the Earth. Nature, 460, 728– 732.
Knoll, A. H. 2000. Learning to tell Neoproterozoic time. Precambrian
Research, 100, 3– 20.
Knoll, A. H. & Walter, M. R. 1992. Latest Proterozoic stratigraphy and
Earth history. Nature, 356, 673–677.
Knoll, A. H., Hayes, J. M., Kaufman, A. J., Swett, K. & Lambert, I. B.
1986. Secular variation in carbon isotope ratios from Upper Protero-
zoic successions of Svalbard and east Greenland. Nature, 321,
832– 837.
Knoll, A. H., Grotzinger, J. P., Kaufman, A. J. & Kolosov, P. 1995.
Integrated approaches to terminal Proterozoic stratigraphy: An
example from the Olenek Uplift, northeastern Siberia. Precambrian
Research, 73, 251– 270.
Knoll, A. H., Walter, M. R., Narbonne, G. M. & Christie-Blick, N.
2006. The Ediacaran Period: a new addition to the geologic time
scale. Lethaia, 39, 13 – 30.
Kulp, J. L., Turekian, K. & Boyd, D. W. 1952. Strontium content of
limestones and fossils. Geological Society of America Bulletin, 63,
701– 716.
Le Guerroué, E., Allen, P. A., Cozzi, A., Etienne, J. L. & Fanning,
M. 2006. 50 Myr recovery from the largest negative d13C excursion
in the Ediacaran ocean. Terra Nova, 18, 147–153.
Le Hir, G., Donnadieu, Y. et al. 2009. The snowball Earth aftermath:
Exploring the limits of continental weathering processes. Earth and
Planetary Science Letters, 277, 453– 463.
Li, C., Love, G. D., Lyons, T. W., Fike, D. A., Sessions, A. L. & Chu, X.
2010. A stratified redox model for the Ediacaran Ocean. Science,
328, 80– 83.
Lyons, T. W. & Severmann, S. 2006. A critical look at iron paleoredox
proxies: New insights from modern euxinic marine environments.
Geochimica et Cosmochimica Acta, 70, 5698– 5722.
Macdonald, F. A., Jones, D. S. & Schrag, D. P. 2009a. Stratigraphic
and tectonic implications of a newly discovered glacial diamictite-
cap carbonate couplet in southwestern Mongolia. Geology, 37,
123– 126.
 CHEMOSTRATIGRAPHY AND THE NEOPROTEROZOIC GLACIATIONS
Macdonald, F. A., McClelland, W. C., Schrag, D. P. & Macdonald,
W. P. 2009b. Neoproterozoic glaciation on a carbonate platform
margin in Arctic Alaska and the origin of the North Slope subterrane.
Geological Society of America Bulletin, 121, 448–473.
Macdonald, F. A., Cohen, P. A., Dudás, F. O. & Schrag, D. P. 2010a.
Early Neoproterozoic scale microfossils in the Lower Tindir Group of
Alaska and the Yukon Territory. Geology, 38, 143–146.
Macdonald, F. A., Schmitz, M. D. et al. 2010b. Calibrating the Cryo-
genian. Science, 327, 1241– 1243.
Macdonald, F. A., Strauss, J. V., Rose, C., Dudás, F. Ö. & Schrag,
D. P. 2011. Stratigraphy of the Port Nolloth Group of Namibia and
South Africa, and implications for the age of Neoproterozoic iron for-
mations. American Journal of Science, 310, 862– 888.
Magaritz, M., Holser, W. T. & Kirschvink, J. L. 1986. Carbon-isotope
events across the Precambrian/Cambrian boundary on the Siberian
Platform. Nature, 320, 258– 259.
Maloof, A. C., Halverson, G. P., Kirschvink, J. L., Schrag, D. P.,
Weiss, B. P. & Hoffman, P. F. 2006. Combined paleomagnetic,
isotopic, and stratigraphic evidence for true polar wander from the
Neoproterozoic Akademikerbreen Group, Svalbard, Norway. Geo-
logical Society of America Bulletin, 118, 1099– 2014.
Marenco, P. J., Corsetti, F. A., Kaufman, A. J. & Bottjer, D. J.
2008a. Environmental and diagenetic variations in carbonate
associated sulfate: An investigation of CAS in the Lower Triassic
of the western USA. Geochimica et Cosmochimica Acta, 72,
1570– 1582.
Marenco, P. J., Corsetti, F. A., Hammond, D. E., Kaufman, A. J. &
Bottjer, D. J. 2008b. Oxidation of pyrite during extraction of car-
bonate associated sulfate. Chemical Geology, 247, 124– 132.
McCay, G. A., Prave, A. R., Alsop, G. I. & Fallick, A. E. 2006. Glacial
trinity: Neoproterozoic Earth history within the British –Irish Caledo-
nides. Geology, 34, 909–912.
McFadden, K. A., Huang, J. et al. 2008. Pulsed oxidation and biological
evolution in Ediacaran Doushantuo Formation. Proceedings of the
National Academy of Sciences (USA), 105, 3197– 3202.
McKirdy, D. M., Burgess, J. M. et al. 2001. A chemostratigraphic over-
view of the late Cryogenian interglacial sequence in the Adelaide
Fold-Thrust Belt, South Australia. Precambrian Research, 106,
149– 186.
Melezhik, V. A., Gorokhov, I. M., Kuznetsov, A. B. & Fallick, A. E.
2001. Chemostratigraphy of Neoproterozoic carbonates: implications
for ‘blind dating’. Terra Nova, 13, 1 –11.
Melezhik, V. A., Pokrovsky, B. G., Fallick, A. E., Kuznetsov, A. B.
& Bujakaite, M. I. 2009. Constraints on the 87Sr/86Sr of Late
Ediacaran seawater: insights from high-Sr limestones. Journal of
the Geological Society, London, 166, 183–191.
Miller, N. R., Stern, R. J., Avigad, D., Beyth, M. & Schilman, B.
2009. Cryogenian slate-carbonate sequences of the Tambien Group,
Northern Ethiopia (I) ‘Pre-Sturtian’ chemostratigraphy and regional
correlations. Precambrian Research, 170, 129– 156.
Misi, A. & Veizer, J. 1998. Neoproterozoic carbonate sequences of the
Una Group, Irece Basin, Brazil: chemostratigraphy, age and corre-
lations. Precambrian Research, 89, 87 –100.
Misi, A., Kaufman, A. J., Azmy, K., Dardenne, M. A., Sial, A. N. & de
Oliveira, T. F. 2011. Neoproterozoic successions of the São Fran-
cisco Craton, Brazil: The Bambuı́, Una, Vazante and Vaza Barris/
Miaba groups and their glacigenic deposits. In: Arnaud, E., Halver-
son, G. P. & Shields-Zhou, G. (eds) The Geological Record of
Neoproterozoic Glaciations. Geological Society, London, Memoirs,
36, 509– 522.
Nagy, R. M., Porter, S. M., Dehler, C. M. & Shen, Y. 2009. Biotic
turnover driven by eutrophication before the Sturtian low-latitude
glaciation. Nature Geoscience, 2, 415– 418.
Nogueira, A. C. R., Riccomini, C., Sial, A. N., Moura, C. A. V., Trin-
dade, R. I. F. & Fairchild, T. R. 2007. Carbon and strontium isotope
fluctuations and paleoceanographic changes in the late Neoprotero-
zoic Araras carbonate platform, southern Amazon craton, Brazil.
Chemical Geology, 237, 191– 210.
Ono, S., Wing, B., Johnston, D., Farquhar, J. & Rumble, D. 2006.
Mass-dependent fractionation of quadruple stable sulfur isotopes as
a new tracer of sulfur biogeochemical cycles. Geochimica et Cosmo-
chimica Acta, 70, 2238–2252.
65
Pavlov, A. A. & Kasting, J. F. 2002. Mass-independent fractionation of
sulfur isotopes in Archean sediments: strong evidence for an anoxic
Archean atmosphere. Astrobiology, 2, 27 –41.
Paytan, A., Kastner, M., Campbell, D. & Thiemens, M. H. 1998.
Sulfur isotopic composition of Cenozoic seawater sulfate. Science,
282, 1459–1462.
Paytan, A., Kastner, M., Campbell, D. & Thiemens, M. H. 2004.
Seawater sulfur isotopic variations in the Cretaceous. Science, 304,
1663–1665.
Pell, S. D., McKirdy, D. M., Jansyn, J. & Jenkins, R. J. F. 1993, Edia-
caran carbon isotope stratigraphy of South Australia — an initial
study. Transactions of the Royal Society of South Australia, 117,
153– 161.
Pokrovskii, B. G., Melezhik, V. A. & Bujakaite, M. I. 2006. Carbon,
oxygen, strontium, and sulfur isotopic compositions in late Precam-
brian rocks of the Patom Complex, central Siberia: Communication
1. Results, isotope stratigraphy, and dating problems. Lithology and
Mineral Resources, 41, 450– 474.
Poulton, S. W., Fralick, P. W. & Canfield, D. E. 2004. The transition
to a sulphidic ocean 1.84 billion years ago. Nature, 431, 173– 177.
Prave, A. R. 1999. Two diamictites, two cap carbonates, two d13C
excursions, two rifts: The Neoproterozoic Kingston Peak Formation,
Death Valley, California. Geology, 27, 339– 342.
Prave, A. R., Fallick, A. E., Thomas, C. W. & Graham, C. M. 2009.
A composite C-isotope profile for the Neoproterozoic of Scotland
and Ireland. Journal of the Geological Society, London, 166,
845– 857.
Rainbird, R. H., Jefferson, C. W. & Young, G. M. 1996. The early Neo-
proterozoic sedimentary Succession B of northwestern Laurentia:
correlations and paleogeographic significance. Geological Society
of America Bulletin, 108, 454–470.
Raiswell, R., Buckley, F., Berner, R. A. & Anderson, T. F. 1988.
Degree of pyritisation of iron as a paleoenvironmental indicator of
bottom-water oxygenation. Journal of Sedimentary Petrology, 58,
812– 819.
Rice, A. H. N., Edwards, M. B., Hansen, T. A., Arnaud, E. & Halver-
son, G. P. 2011. Glaciogenic rocks of the Neoproterozoic Smalfjord
and Mortensnes Formations, Vestertana Group, E. Finnmark,
Norway. In: Arnaud, E., Halverson, G. P. & Shields-Zhou, G.
(eds) The Geological Record of Neoproterozoic Glaciations. Geo-
logical Society, London, Memoirs, 36, 593–602.
Ries, J. B., Fike, D. A., Pratt, L. M., Lyons, T. W. & Grotzinger, J. P.
2009. Superheavy pyrite (d34Spyr . d34SCAS) in the terminal
Proterozoic Nama Group, southern Namibia: a consequence of
low seawater sulfate at the dawn of animal life. Geology, 37,
743– 746.
Rooney, A. D., Selby, D., Houzait, J.-P. & Renne, P. R. 2010. Re– Os
geochronology of a Mesoproterozoic sedimentary succession, Taou-
deni basin, Mauritania: Implications for basin-wide correlations and
Re– Os organic-rich sediments systematic. Earth and Planetary
Science Letters, 289, 486–496.
Rose, C. V. & Maloof, A. C. 2010. Testing models for post-glacial “cap
dolostone” deposition: Nuccaleena Formation, South Australia. Earth
and Planetary Science Letters, 296, 165–180.
Rothman, D. H., Hayes, J. M. & Summons, R. E. 2003. Dynamics of the
Neoproterozoic carbon cycle. Proceedings of the National Academy
of Sciences (USA), 100, 124– 129.
Sawaki, Y., Kawai, T. et al. 2010a. 87Sr/86Sr chemostratigraphy of Neo-
proterozoic Dalradian carbonates below the Port Askaig glaciogenic
Formation, Scotland. Precambrian Research, 179, 150–164.
Sawaki, Y., Ohno, T. et al. 2010b. The Ediacaran radiogenic Sr isotope
excursion in the Doushantuo Formation in the Three Gorges area,
South China. Precambrian Research, 176, 46 –64.
Saylor, B. Z., Kaufman, A. J., Grotzinger, J. P. & Urban, F. 1998. A
composite reference section for Terminal Proterozoic strata
of southern Namibia. Journal of Sedimentary Research, 68,
1223–1235.
Schrag, D. P., Berner, R. A., Hoffman, P. F. & Halverson, G. P. 2002.
On the initiation of a snowball Earth. Geochemistry, Geophysics,
Geosystems, 31, doi: 10.1029/2001GC000219.
Schidlowski, M., Eichmann, R. & Junge, C. E. 1975. Precambrian
sedimentary carbonates: carbon and oxygen isotope geochemistry
66 G. P. HALVERSON & G. SHIELDS-ZHOU
and implications for the terrestrial oxygen budget. Precambrian
Research, 2, 1 –69.
Schröder, S., Schreiber, C., Amthor, J. E. & Matter, A. 2004.
Stratigraphy and environmental conditions of the terminal
Neoproterozoic – Cambrian Period in Oman: evidence from
sulphur isotopes. Journal of the Geological Society, London, 161,
489– 499.
Shen, B., Xiao, S., Zhou, C., Kaufman, A. J. & Yuan, X. 2010. Carbon
and sulfur isotope chemostratigraphy of the Neoproterozoic Quanji
Group of the Chaidam Basin, NW China: Basin stratification in the
aftermath of an Ediacaran glaciation post-dating the Shuram event?
Precambrian Research, 177, 241– 252.
Shields, G. A. 1999. Working towards a new stratigraphic calibration
scheme for the Neoproterozoic –Cambrian. Eclogae Geologicae
Helvetiae, 92, 221– 233.
Shields, G. A. 2005. Neoproterozoic cap carbonates: a critical appraisal
of existing models and the plumeworld hypothesis. Terra Nova, 17,
299– 310.
Shields, G. A. 2007. A normalised seawater strontium isotope curve:
possible implications for Neoproterozoic –Cambrian weathering
rates and further oxygenation of the Earth. Earth, 2, 35 – 42.
Shields, G. & Veizer, J. 2002. Precambrian marine carbonate isotope
database: Version 1.1. Geochemistry, Geophysics, Geosystems, 3,
doi: 10.1029/2001GC000266.
Shields, G., Stille, P., Brasier, M. D. & Atudorei, N.-V. 1997. Stra-
tified oceans and oxygenation of the late Precambrian environments:
a post glacial geochemical record from the Neoproterozoic. Terra
Nova, 9, 218–222.
Shields, G. A., Strauss, H., Howe, S. S. & Siegmund, H. 1999. Sulphur
isotope compositions of sedimentary phosphorites from the basal
Cambrian of China — implications for Neoproterozoic – Cambrian
biogeochemical cycling. Journal of the Geological Society,
London, 156, 943 –957.
Shields, G., Kimura, H., Yang, J. & Gammon, P. 2004. Sulphur isotopic
evolution of Neoproterozoic – Cambrian seawater: new francolite-
bound sulphate S data and critical appraisal of the existing record.
Chemical Geology, 204, 163–182.
Sial, A. N., Dardenne, M. A. et al. 2010. The São Francisco Paleocon-
tinent. In: Gaucher, C., Sial, A. N., Halverson, G. P. & Frimmel,
H. (eds) Neoproterozoic– Cambrian Tectonics, Global Change and
Evolution: A focus on Southwestern Gondwana. Developments in
Precambrian Geology, Elsevier, Dordrecht, 16, 31– 69.
Strauss, H. 1993. The sulfur isotopic record of Precambrian sulfates: new
data and a critical evaluation of the existing record. Precambrian
Research, 63, 225– 246.
Swanson-Hysell, N. L., Rose, C. V., Calmet, C., Halverson, G. P.,
Hurtgen, M. T. & Maloof, A. C. 2010. Cryogenian glaciation
and onset of carbon-isotope decoupling. Science, 328, 608–611.
Tucker, M. E. 1986. Carbon isotope excursions in Precambrian/
Cambrian boundary beds. Nature, 319, 48 –50.
Turner, E. C. 2009. Lithostratigraphy and stable isotope values of the
early Neoproterozoic Gypsum Formation (Little Dal Group),
Mackenzie Mountains Supergroup, NWT. NWT Open Report,
2009-002.
Veizer, J., Compston, W., Clauer, N. & Schidlowski, M. 1983.
87
Sr/86Sr in late Proterozoic carbonates: evidence for a
‘mantle’ event at 900 Ma. Geochimica et Cosmochimica Acta,
47, 295– 302.
Walter, M. R., Veevers, J. J., Calver, C. R., Gorjan, P. & Hill, A. C.
2000. Dating the 840– 544 Ma Neoproterozoic interval by isotopes of
strontium, carbon, and sulfur in seawater and some interpretive
models. Precambrian Research, 100, 371–433.
Williams, G. E. 1979. Sedimentology, stable-isotope geochemistry and
palaeoenvironment of dolostones capping late Precambrian glacial
sequences in Australia. Journal of the Geological Society of Austra-
lia, 26, 377–386.
Workman, R. K., Grotzinger, J. P. & Hart, S. R. 2002. Constraints on
Neoproterozoic ocean chemistry from C and B analyses of carbonates
from the Witvlei and Nama groups, Namibia. In: Goldschmidt Con-
ference Proceedings (Davos, Switzerland).
Xiao, S., Bao, H. et al. 2004. The Neoproterozoic Quruqtagh Group in
eastern Chinese Tianshan: evidence for a post-Marinoan glaciation.
Precambrian Research, 130, 1 –26.
Xu, B., Xiao, S. et al. 2009. SHRIMP zircon U– Pb age constraints on
Neoproterozoic Quruqtagh diamictites in NW China. Precambrian
Research, 168, 247– 258.
Yoshioka, H., Asahara, Y., Tojo, B. & Kawakami, S. 2003. Systematic
variations in C, O, and Sr isotopes and elemental concentrations in
Neoproterozoic carbonates in Namibia: implications for glacial to
interglacial transition. Precambrian Research, 124, 69 –85.
 Chapter 5

Chemical sediments associated with Neoproterozoic glaciation: iron formation,

cap carbonate, barite and phosphorite

PAUL F. HOFFMAN1,2 *, FRANCIS A. MACDONALD1 & GALEN P. HALVERSON3,4

1
Department of Earth and Planetary Sciences, Harvard University, 20 Oxford Street, Cambridge, MA, 02138, USA
2
School of Earth and Ocean Sciences, University of Victoria, Box 1700, Victoria, BC V6W 2Y2, Canada
3
School of Earth and Environmental Sciences, The University of Adelaide, North Terrace, Adelaide, SA 5005, Australia
4
Present address: Department of Earth and Planetary Sciences, McGill University, 3450 University Street, Montréal,
PQ H3A 2K6, Canada
*Corresponding author (e-mail: paulfhoffman@yahoo.com)

Abstract: Orthochemical sediments associated with Neoproterozoic glaciation have prominence beyond their volumetric proportions
because of the insights they provide on the nature of glaciation and the records they hold of the environment in which they were preci-
pitated. Synglacial Fe formations are mineralogically simple (haematite jaspilite), and their trace element spectra resemble modern sea-
water, with a weaker hydrothermal signature than Archaean– Palaeoproterozoic Fe formations. Lithofacies associations implicate
subglacial meltwater plumes as the agents of Fe(II) oxidation, and temporal oscillations in the plume flux as the cause of alternating
Fe- and Mn-oxide deposits. Most if not all Neoproterozoic examples belong to the older Cryogenian (Sturtian) glaciation. Older and
younger Cryogenian (Marinoan) cap carbonates are distinct. Only the younger have well-developed transgressive cap dolostones,
which were laid down during the rise in global mean sea level resulting from ice-sheet meltdown. Marinoan cap dolostones have a
suite of unusual sedimentary structures, indicating abnormal palaeoenvironmental conditions during their deposition. Assuming the melt-
down of ice-sheets was rapid, cap dolostones were deposited from surface waters dominated by buoyant glacial meltwater, within and
beneath which microbial activity probably catalysed dolomite nucleation. Former aragonite seafloor cement (crystal fans) found in
deeper water limestone above Marinoan cap dolostones indicates carbonate oversaturation at depth, implying extreme concentrations
of dissolved inorganic carbon. Barite is associated with a number of Marinoan cap dolostones, either as digitate seafloor cement associ-
ated with Fe-dolomite at the top of the cap dolostone, or as early diagenetic void-filling cement associated with tepee or tepee-like brec-
cias. Seafloor barite marks a redoxcline in the water column across which euxinic Ba-rich waters upwelled, causing simultaneous barite
titration and Fe(III) reduction. Phosphatic stromatolites, shrub-like structures and coated grains are associated with a glacioisostatically
induced exposure surface on a cap dolostone in the NE of the West African craton, but this appears to be a singular occurrence of phos-
phorite formed during a Neoproterozoic deglaciation.

The association of chemical sediments with Neoproterozoic Fe and Fe – Mn deposits

glacial deposits has long been known. In fact, Neoproterozoic
glaciation was discovered in a number of regions as a result of
the search for economic Fe and Fe–Mn deposits (e.g. SW
Brazil, NW Canada, Namibia). Glacial associated chemical sedi-
ments provide critical evidence concerning the nature of Neopro-
terozoic glaciations and their aftermaths, in the form of
geochemical records of the waters from which they were precipi-
tated, and indirectly the atmosphere with which those waters
interacted.
Here, we briefly review the distribution, lithological associa-
tion, sedimentology and palaeoenvironmental significance of
orthochemical sediments deposited during Neoproterozoic glacia-
tions and deglaciations. They include sedimentary Fe and Fe –Mn
deposits and cap-carbonate sequences, of which ‘cap dolostones’
sensu stricto form the basal transgressive systems tracts. Barite
(BaSO4) and phosphorite mineralization occurs locally within
Marinoan cap dolostones. More information on geologic setting,
stratigraphic relations, geochemistry and isotopic characteristics
is given in the appropriate regional chapters. Palaeogeographies
of the deposits (Fig. 5.1) are based on global model maps for
715 Ma (Sturtian) and 635 Ma (Marinoan), created by the Tec-
tonics Special Research Centre in Perth, Western Australia (Li
et al. 2008; Hoffman & Li 2009). The maps were constructed on
the basis of palaeomagnetic constraints, the mantle plume record
and palaeocontinental tectonic genealogy, factors that are largely
independent of palaeoclimate. The Sturtian glaciation persisted
until 659 + 6 Ma in its type area (Fanning & Link 2008), by
which time the model palaeogeography (Li et al. 2008) was inter-
mediate between 715 and 635 Ma (Fig. 5.1).
Distribution in time and space
The palaeogeographic distribution of synglacial Fe and Fe –Mn
sedimentary deposits is shown in Figure 5.1 (Table 5.1). The econ-
omic Fe – Mn ores of the Jacadigo Group in the Urucum District of
southwestern Brazil and eastern Bolivia (see below) were tenta-
tively assigned to the terminal Cryogenian (Marinoan) glaciation
(Hoffman & Li 2009) through correlation with the Puga diamictite
in the adjacent Paraguay fold belt (Alvarenga & Trompette 1992;
Trompette 1994; Trompette et al. 1998). The Puga diamictite is
capped by a diagnostic Marinoan-type cap-carbonate sequence
(Nogueira et al. 2003, 2007; Trindade et al. 2003; Font et al.
2005, 2006; Alvarenga et al. 2008). However, no cap carbonate
is preserved in the Jacadigo Group, which occupies a transverse
rift-basin on the cratonic foreland of the Paraguay fold belt
(Trompette et al. 1998). The true age of the Jacadigo Group, Stur-
tian or Marinoan, is unknown. Similarly, Fe formation in the gla-
ciogenic Rizu Formation of central Iran was tentatively assigned to
the Marinoan glaciation (Hoffman & Li 2009) on the basis of a
reported cap dolostone (Kianian & Khakzad 2008), but details
are lacking. The reassignment of the Jacadigo and Rizu Fe for-
mations to the Sturtian glaciation (Fig. 5.1) is therefore permiss-
ible, but arbitrary. In either palaeogeographic reconstruction
(Fig. 5.1), the Fe- and Fe –Mn deposits formed disproportionately
within 308 of the palaeoequator, and at the margins of ocean basins
that were both internal (Jiangkou, Sturt, Rapitan, Surprise, Numees
and Chuos) and external (Rizu, Tany) to the fragmented Rodinia
supercontinent.

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 67– 80. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.5
68 P. F. HOFFMAN ET AL.

(a)

Nantuo
“
Blasskranz 635 Ma Namuskluft
Wildrose
Ghaub Vreeland
Petite Stelfox
Tereeken Khongoryn
Landrigan Ta Pod’em
Olympic Marnya
Blaini Ir Ka Dzemkukan
Au SC TM Si
In
Ma Laur CA
Elatina RP NC
Co
Cottons SF
Fiq Ar Fig. 5.1. Palaeogeographic maps for
Shareef Wilsonbreen 715 Ma and 635 Ma (Hoffman & Li 2009)
Bondo Am Storeelv showing distribution of (a) Sturtian and (b)
Superieure WA
Stralinchy-Reelan Marinoan glaciogenic deposits (open stars),
Kodjari Ba Smalfjord
Bakoye and synglacial Fe or Fe– Mn deposits (black
Jbeliat Palestina Puga stars). Palaeocontinents: Ar, Arabia; Am,
Amazonia; Au, Australia; Ba, Baltica; CA,
Chukotka–Arctic Alaska; Co, Congo; In,
(b) India; Ir; central Iran; Ka, Kalahari; Laur,
Jiangkou 715 Ma Numees
Surprise
Laurentia (including Sc, Scotland and Sv,
Blaubekker Svalbard); Ma, Mawson; NC, North China;
Pocatello SF; São Francisco; RP, Rio de la Plata; SC,
Areyonga Ta Toby South China; Si, Siberia; Ta, Tarim; TM,
Rapitan Tuva-Mongolia; WA, West Africa.
Si Tindir
Palaeocontinents constrained by
In SC
TM Maikhan Ul
Rizu Au Chivida penecontemporaneous palaeomagnetic data
Sturt CA NC Kharlukhtakh have heavy lines. The Sturtian glaciation
Ma persisted to c. 660 Ma in some areas (South
Julius River
Konnarock Ka Laur Australia and western USA), by which time
Sv the model palaeogeography was
Gubrah RP Hula Hula
Tambien Petrovbreen intermediate between 715 and 635 Ma.
Ayn
Ar Co Sc Assignment of Fe formations in the Jacadigo
Ulveso
Grand Port Askaig Group (Am) and Rizu Formation (Ir) to the
Chuos SF Am
Ba Tany Sturtian glaciation is uncertain. Correlation
Akwokwo Baykonur of Fe formation in the Gariep Belt (Ka) with
Inferieure WA
Jequitai Chiquerio Jacadigo the glaciogenic Numees Formation follows
Macdonald et al. (2011).

Lithological associations peri- or subglacial in origin based on the presence of faceted,

Impure to pure haematite –jaspilite occurs within intervals of
parallel-laminated argillite, ferruginous argillite and Mn-oxides.
Those identified in Figure 5.1 host outsize clasts (lonestones) of
intrabasinal and extrabasinal (crystalline basement) derivation,
generally interpreted as ice-rafted debris. Many occur in contact
with massive to poorly stratified diamictite, inferred to be
striated and/or preferentially oriented clasts. Sizeable Fe and
Fe –Mn deposits occur in intervals that are sandwiched between
composite diamictite horizons, locally of great thickness (e.g.
Braemar, Holowilena, Jakkalsberg, Sayunei, Tany). Smaller
deposits are intimately interfingered with diamictite (e.g.
Braemar, Chuos). Still other deposits lack diamictite but the
presence of dropstones of intra- and extrabasinal origin suggest a

Table 5.1. Cryogenian glacigenic Fe and Fe– Mn deposits

Palaeocontinent Location Host strata References

Amazonia Urucum Jacadigo Dorr (1945), Urban et al. (1992), Klein & Ladeira (2004)
Australia Braemar Yudnamutana Whitten (1970), Lottermoser & Ashley (2000)
Baltica Middle Urals Vil’va Chumakov (1992, 2007)
Congo Damara Belt Chuos Martin (1965a), Badenhorst (1988), Clifford (2008)
Kalahari Gariep Belt Numees Macdonald et al. (2011)
Laurentia NWT-Yukon Rapitan Young (1976), Yeo (1981, 1986), Klein & Beukes (1993)
Laurentia* Death Valley Surprise Corsetti & Kaufman (2003)
Lut (central Iran) Kerman Rizu Kianian & Khakzad (2008)
South China Yangtze platform Jiangkou Jiafu et al. (1987)
Tuva-Mongolia Erzin Ongoluk Ilyin (2009)

*In the collisional model for Laramide orogeny (Hildebrand 2009), the Death Valley area as well as most other Cryogenian glacial formations in the western USA and
British Columbia are considered to be exotic with respect to Laurentia before the Late Cretaceous.
 CHEMICAL SEDIMENTS ASSOCIATED WITH NEOPROTEROZOIC GLACIATION
subaqueous proglacial setting (e.g. Urucum). Parallel-laminated
facies hosting haematite –jaspilite are commonly interpreted as
interglacial or interstadial, but they might alternatively represent
maximum glacial stages, when outlet glaciers were blocked by
thick multi-annual sea ice (Dowdeswell et al. 2000).
Geochemical characteristics
Compared with Archaean –Palaeoproterozoic banded Fe for-
mations, Cryogenian Fe deposits are mineralogically simple, con-
sisting of haematite, chert (jasper) and minor carbonate. Fe
contents range up to 50% Fe2O3. Organic contents are negligible
and carbonates are moderately depleted in 13C, suggesting that
organic matter originally present was respired with Fe(III)
serving as electron acceptor. Normalized rare earth element
profiles are distinct from Archaean–Palaeoproterozoic Fe for-
mations, being more strongly depleted in light rare earth elements,
and having much weaker positive Eu anomalies or none at all
(Klein & Beukes 1993; Graf et al. 1994; Klein & Ladeira 2004;
Klein 2005). The Cryogenian profiles are more similar to those
of modern seawater, and hydrothermal contributions are more
diluted than in more ancient Fe formations.
Notable examples
Jacadigo Group, Urucum District, Mato Grosso do Sul, Brazil. Eco-
nomic Fe–Mn oxide ores of the Jacadigo Group occur in strati-
graphically isolated fault blocks in the Urucum District
of southwestern Brazil and adjacent Bolivia (Dorr 1945; Walde
et al. 1981; Urban et al. 1992; Trompette et al. 1998). A
lower Urucum Formation, comprising conglomerate, sandstone
and black shale, is overlain by the Santa Cruz Formation, a
sequence of Mn ore horizons, chiefly composed of cryptomelane
(K2Mn8O16), interspersed with ferruginous sandstone and up to
270 m of haematite –jaspilite. All units of the Santa Cruz For-
mation carry dropstones of basement and subordinate carbonate
lithologies. They are interpreted to be ice-rafted and the Jacadigo
Group to have been deposited in a partly ice-covered fjord-like
basin (Urban et al. 1992). The alternation of Mn and Fe ores is
attributed to cycles of glacial advance and retreat. Enhanced
fluxes of oxygenated subglacial meltwater during glacial retreats
favoured Mn precipitation in ice-covered anoxic parts of the
basin. Reduced fluxes of subglacial meltwater during glacial
advances favoured Fe deposition (Urban et al. 1992). The Jacadigo
Group lies in the cratonic foreland of southern Amazonia, close to
the Paraguay fold belt of Early Cambrian (Brasiliano) age within
which glaciogenic diamictite of the Puga Formation (Alvarenga
& Trompette 1992) is sharply overlain by a typical Marinoan-type
cap dolostone (Nogueira et al. 2003, 2007; Trindade et al. 2003;
Font et al. 2005, 2006; Alvarenga et al. 2008). No cap carbonate
is preserved on the Jacadigo Group, however, so the correlation
of the Jacadigo and Puga groups is uncertain, as is, therefore, the
age of the Urucum Fe– Mn deposits.
Rapitan Group, northern Canadian Cordillera. Haematite– jaspilite
and ferruginous argillite occur within glaciogenic diamictite of
the Rapitan Group and correlatives discontinuously for c. 800 km
along the strike of the Ogilvie and Mackenzie mountains (Young
1976; Yeo 1981, 1986; Eisbacher 1985; Klein & Beukes 1993;
Macdonald et al. 2011; Hoffman & Halverson 2011; Macdonald
& Cohen 2011). The glacial onset coincided with a major flood
basalt episode across Arctic Laurentia, bimodal representatives
of which underlie and intercalate the basal Rapitan diamictite in
the Ogilvie Mountains, constraining its age at 717 Ma (Macdonald
et al. 2010). Haematite –jaspilite is up to 120 m thick in an ice-
proximal section (Iron Creek, NW Mackenzie Mtns), where it is
sandwiched between composite diamictite complexes hundreds
69
of metres thick. Jaspilite first appears as 2-m-thick septa between
individual diamictite bodies in the upper part of the lower
complex (Mount Berg Formation). In a more distal section
(Hayhook Lake, SE Mackenzie Mountains), 14 m of ferruginous
argillite and haematite –jaspilite conformably overlie 650 m of
maroon-coloured siltstones with graded sandy beds and a sprink-
ling of small dropstones, some of which were redeposited as
debrites. The jaspilite itself contains rounded boulders of quartz-
monzonite and is overlain disconformably by the upper diamictite
complex (Shezal Formation). The basal part of the otherwise
olive-coloured Shezal diamictite acquired its maroon colour and
numerous jaspilite clasts from the underlying Sayunei Formation.
Fe-isotope and Ce anomaly profiles suggest that the haematite –
jaspilite records subsidence of the basin floor across a redox-
cline in the water column (Klein & Beukes 1993; Halverson
et al. 2011).
Jakkalsberg Member (Numees Formation, Port Nolloth Group), Gariep
Belt, Namibia and South Africa. Haematite- and magnetite –jaspilite
with basement-derived lonestones make up the Jakkalsberg
Member of the Numees Formation in thrust sheets intersected by
the Orange River (Frimmel & von Veh 2003). The Numees For-
mation has long been assumed to be the younger of two glacio-
genic horizons in the Gariep Belt, the older being the Kaigas
Formation (Frimmel 2008, 2011). Recently, a mid-Ediacaran age
for the Numees glaciation has been advocated from carbonate
Pb/Pb dating (Fölling et al. 2000), 87Sr/86Sr ratios of 0.7082–
0.7085 (Fölling & Frimmel 2002), correlations with dated South
American strata (Frimmel 2004, 2008) and micropalaeontological
findings (Gaucher et al. 2005). The microfossils do not appear to
be diagnostic, however, and the Sr isotope ratios (0.001
higher than typical Marinoan values; Halverson et al. 2007) are
potentially attributable to diagenesis, as are the Pb/Pb dates. Sub-
sequent remapping of the least-deformed and least-allochthonous
sections prompted a reassessment of stratigraphic correlations
between the autochthon and thrust sheets within the belt
(Macdonald et al. 2011). The younger autochthonous diamictite,
previously correlated with the Numees Formation, is now pro-
posed as an independent glaciogenic horizon, the Namuskluft
diamictite (Macdonald et al. 2011). It is capped by a pale dolostone
(Dreigratberg Member, previously conflated with the Bloeddrif
Member, of the Holgat Formation) containing hallmark features
of Marinoan cap dolostones –sheet-crack cements, tubestone
stromatolites and giant wave ripples (Hoffman & Macdonald
2010). In contrast, the jaspilite-bearing Numees diamictite is
overlain by a dark microbial limestone with roll-up structures,
characteristic of Sturtian cap carbonates (see below). Accordingly,
the Jakkalsberg jaspilite is probably older Cryogenian (Sturtian)
in age, and the Kaigas Formation may represent a pre-
Sturtian (c. 0.74 Ga) diamictite of uncertain origin (Macdonald
et al. 2011).
Central Flinders Ranges (Holowilena) and Nackara arc (Braemar),
South Australia. Like the more ice-proximal (Iron Creek) section
in NW Canada, haematite – jaspilite and associated ferruginous
argillite in South Australia are sandwiched between a conformably
underlying diamictite complex (Pualco ‘Tillite’) and a discon-
formably overlying complex (Wilyerpa Formation) dominated
by diamictite (Whitten 1970; Preiss 1987; Lottermoser & Ashley
2000). In the Flinders Ranges, ferruginous argillite hosts sporadic
lenses of boulders, some of which are grooved and striated,
ferruginous diamictite, and rare lenses of haematite – jaspilite
with rafted dropstones. In the more basinward Nackara arc,
haematite –jaspilite is interrupted by thin units of ferruginous
diamictite. Sorted sandstones occur within the diamictite com-
plexes in both areas. Correlation of the Holowilena and Braemar
jaspilites is uncertain, but they and the bounding Pualco and
Wilyerpa diamictites are referred to the Sturtian glaciation
(Preiss 1987).
70 P. F. HOFFMAN ET AL.

Cap-carbonate sequences
The continuous layers of carbonate that blanket Cryogenian gla-
ciogenic sequences or their equivalent disconformities are called
‘cap’ carbonates. Clearly associated with syndeglacial flooding
(Kennedy 1996; Bertrand-Sarfati et al. 1997; Hoffman et al.
2007), they occur on virtually every palaeocontinent and even in
siliciclastic-dominated successions (Table 5.2). Lithologically
and isotopically, Sturtian and Marinoan cap carbonates are dis-
tinct, both from each other (Kennedy et al. 1998; Halverson &
Shields 2011) and from most other Neoproterozoic carbonates
(Hoffman 2011). Hoffman & Schrag (2002) proposed that deposi-
tional sequences related to Neoproterozoic syndeglacial flooding
be called ‘cap-carbonate sequences’. ‘Cap dolostones’, sensu
stricto, are the transgressive tracts of cap-carbonate sequences
and feature a suite of idiosyncratic sedimentary features (e.g. size-
graded peloids, sheet-crack cements, tubestone stromatolites,
giant wave ripples, primary and early diagenetic barite). Trans-
gressive cap dolostones have rarely been documented in Sturtian
cap-carbonate sequences and are thin (,1.0 m) where present
(Smith et al. 1994). Cap-carbonate sequences have aroused
intense interest because they are unique to Proterozoic glaciations
and record physical, chemical and biological conditions during and
immediately after global deglaciation (Aitken 1991; Grotzinger &
Knoll 1995; Kennedy, 1996; Hoffman et al. 1998, 2007; James
et al. 2001; Kennedy et al. 2001; Higgins & Schrag 2003; Ridgwell
et al. 2003; Trindade et al. 2003; Shields 2005; Allen & Hoffman
2005; Font et al. 2005, 2006; Hurtgen et al. 2006; Jiang et al. 2006;
Bao et al. 2008; Le Hir et al. 2009; Hoffman 2011).
Marinoan-type (basal Ediacaran) cap-carbonate sequences
The extent and uniqueness of Marinoan cap dolostones was first
appreciated in Australia (Dunn et al. 1971; Rankama 1973) and
the presence of barite in cap dolostones was first recognized in
West Africa (Deynoux & Trompette 1976). The smooth and
abrupt, yet conformable, contact between glaciogenic and related
detritus and cap dolostones is so distinctive and widespread that
it was selected as the basis of the Global Stratotype Section and
Point (GSSP) for the Ediacaran Period (Knoll et al. 2006), the
first Period boundary to be defined strictly on lithologic grounds.
A tuff at the top of the presumed correlative cap dolostone in
South China yields a 238U – 206Pb (IDTIMS) zircon date of
635.2 + 0.4 Ma (Condon et al. 2005). A statistically indistin-
guishable 238U – 206Pb (SHRIMP) zircon date of 636.3 + 4.9 Ma
was obtained from a tuff near the base of the underlying Nantuo
glaciogenic diamictite (Zhang et al. 2008). The presumed

Table 5.2. Average thickness and idiosyncratic features of Marinoan cap dolostones in numerical order of abundance

Palaeocontinent Glaciation Cap dolostone (ref.) Metres LAC PEL GWR SCC TBS TPB DGB SFB
p p p p
Amazonia Puga Mirassol d’Oeste (1)† 24 – – – –
Arabia Gadir Manqil (Fiq) Hadash (2) 4.5 – – – – – – – –
p p p p
Arctic Alaska No deposits Nularvik (3)† 35 – – – –
p p p p p
Australia Olympic Mount Doreen (4) 4 – – –
p p
Australia Elatina Nuccaleena (5) 5 – – – – – –
p p p
Australia Cottons Cumberland Creek (6) 6 – – – – –
p p
Australia Landrigan Lower Stein (7) 8.5 – – – – – –
p p
Baltica Smalfjord Lower Nyborg (8) 5 – – – – – –
p p p p p
Congo Ghaub Keilberg (9)† 38 – – –
Congo Petit Conglomérat Calcaire rosé (10) 10
Congo Upper Tilloid C1 Dolomie rosé (11) 10
India Blaini Upper Blaini (12) 10
p p p p p
Kalahari Namuskluft Dreigratberg (13) 25 – – –
p
Kalahari Blässkranz Tsabesis (14) 21 – – – – – – –
p p p
Kalahari No deposits Bildah (15)† 80 – – – – –
p p p p p p
Laurentia Stelfox (Ice Brook) Ravensthroat (16)† 12 – –
p
Laurentia Storeelv Lower Canyon (17) 10 – – – – – – –
p p p
Laurentia Wilsonbreen Lower Dracoisen (18) 10 – – – – –
p p p
Laurentia Stralinchy-Reelan Cranford (19) 4 – – – – –
p
Laurentia Stelfox (Ice Brook) Hard Luck (20) 4 – – – – – – –
p p p p
Laurentia* Wildrose Noonday (21) 175 – – – –
p
Siberia Ulyakha Lower Ozerki (22) 35 – – – – – – –
p p p p
South China Nantuo Lower Doushantuo (23) 4 – – – –
p
Tarim Tereeken Lower Zhamoketi (24) 6 – – – – – – –
p p p p p p p
Tuva-Mongolia Khongoryn Ol (25)† 15 –
p p
Tuva-Mongolia Khesen Baxha (26) 4 – – – – – –
p p
West Africa Fersiga Oued Djouf (27) 6 – – – – – –
p p p
West Africa Jbéliat Amogjar (28) 5 – – – – –
p p p p p
West Africa Banboli Mid Sud-Banboli (29) 1.5 – – –

*In the collisional model for Laramide orogeny (Hildebrand 2009), Wildrose-Noonday strata and most other Cryogenian glacial formations in the western USA and British
Columbia are considered to be exotic with respect to Laurentia before the Late Cretaceous.
†
Former-aragonite seafloor cement above cap dolostone. LAC, low-angle cross-laminae; PEL, peloids; GWR, giant wave ripples; SCC, sheet-crack cements; TBS,
tubestone stromatolite; TPB, tepee breccia; DGB, diagenetic barite; SFB, seafloor barite.
References: (1) Nogueira et al. (2003); (2) Allen et al. (2004); (3) Macdonald et al. (2009); (4) Kennedy (1996); (5) Plummer (1978); (6) Calver & Walter (2000);
(7) Corkoron (2007); (8) Edwards (1984); (9) Hoffman et al. (2007); (10) Cahen & Lepersonne (1981); (11) Cahen (1950); (12) Kaufman et al. (2006); (13) Hoffman &
Macdonald (2010); (14) PFH observations; (15) Hegenberger (1993); Prave et al. (2011); (16) James et al. (2001); (17) Hambrey & Spencer (1987);
(18) Halverson et al. (2004); (19) McCay et al. (2006); (20) Macdonald & Cohen (2011); (21) Corsetti & Grotzinger (2005); (22) Sovetov & Komlev (2005);
(23) Jiang et al. (2006); (24) Xiao et al. (2004); (25) Macdonald (2011); (26) Macdonald & Jones (2011); (27) Bertrand-Sarfati et al. (1997); (28) Shields et al. (2007);
(29) Nédélec et al. (2007).
 CHEMICAL SEDIMENTS ASSOCIATED WITH NEOPROTEROZOIC GLACIATION 71

correlative Ghaub diamictite in Namibia has an indistinguishable
238
U – 206Pb (IDTIMS) zircon date of 635.6 + 0.5 Ma (Hoffmann
et al. 2004). Marinoan cap dolostones are therefore assumed to
have been deposited at 635 Ma.
Distribution and thickness. The median and average thicknesses of
29 Marinoan cap dolostones on 15 palaeocontinents (Table 5.2)
are 9 m and 18 m, respectively (Hoffman et al. 2007). As there
are large facies-controlled variations in thickness in some areas,
these estimates are based on area-weighted average thicknesses,
where data permit. On the Otavi platform (Congo palaeocontinent)
of Namibia, for example, the Keilberg cap dolostone is 6– 10 m
thick on the distal foreslope, up to 100 m on the upper foreslope,
up to 75 m on the raised outer platform, and 20– 25 m on the dee-
pened inner platform (Fig. 5.2). In general, cap dolostones are
thickest on palaeotopographic highs and thinnest in lows, converse
to underlying glacial deposits. Observed inner-shelf deepening is
characteristic of glaciated margins (Anderson 1999). If cap dolo-
stones averaged 18 m in thickness over 20% of the present conti-
nental surface area (including shelves), they would contain
c. 2.6  105 Pg (Gt) of C. This dwarfs the 3.8  104 Pg of dis-
solved inorganic C in the present ocean. According to the palaeo-
geography in Figure 5.1a, the maximum thickness of cap
dolostones increased from 5 m above 408 palaeolatitude, to 10 m
above 308, 25 m above 208, 38 m above 108 and 175 m at the
palaeoequator (Hoffman & Li 2009).
Expanded and condensed sequences. The decompacted thickness of
a cap-carbonate sequence (as distinct from a cap dolostone), minus
the effect of sediment loading, is a measure of the accommodation
created before and during its deposition. The lowering and raising
of global mean sea level attending the growth and decay of ice
sheets creates no permanent accommodation if global ice-sheet
volume was the same before and after a glaciation. Permanent
accommodation (i.e. outlasting isostatic adjustments) was
created during Cryogenian glaciations by net erosion, compaction
and tectonic subsidence. Glacial palaeofjords, for example,
contain expanded cap-carbonate sequences compared with adja-
cent uplands (Fig. 5.3). Similarly, areas undergoing rapid tectonic
subsidence like the Otavi platform, on which the Sturtian and
Marinoan cap-carbonate sequences are each 250–400 m thick,
contrast with areas like the Taoudeni Basin of the West African
craton (Deynoux 1985; Shields et al. 2007), where subsidence
rates were low and cap-carbonate sequences are highly condensed,
outside of incised palaeovalleys (Fig. 5.3). Condensation may also
occur in fully marine settings where sedimentation rates were
inadequate to fill the existing accommodation space. On the
distal foreslope of the Otavi platform, for example, drowning at
the terminal deglaciations was effectively permanent (Halverson
et al. 2005). The cause of condensation of the lower Doushantuo
Formation cap-carbonate sequence of South China (low sedimen-
tation rate or lack of accommodation?) depends on whether tepee-
like breccias and associated early diagenetic barite mineralization
in the cap dolostone are submarine (Jiang et al. 2006) or subaerial
(Zhou et al. 2010) in origin, respectively.
Transgressive cap dolostones in expanded sequences. Cap dolostones
are typically pale pinkish (yellowish-grey weathering) dolomi-
crites, with palimpsest micro- to macro-peloidal textures indicat-
ing deposition as silt and sand- and granule-sized aggregates.
Those in expanded sequences lack the diagenetic overprint experi-
enced by those in condensed sequences, which we treat separately.
The former (e.g. Cumberland Creek, lower Canyon, lower Dracoi-
sen, Dreigratberg, Keilberg, Mirassol d’Oeste, Mount Doreen,
Nuccaleena, Ol, lower Pertatataka, Ravensthroat, Sentinel Peak,
lower Stein, Tsabesis, and lower Zhamoketi; see Table 5.2) are
invariably laminated, with each lamina representing a sedimen-
tation unit composed of well-sorted, graded or reverse-graded
peloids, up to 3 mm in diameter, terminating with a micropeloidal
drape (Aitken 1991; Kennedy 1996; Calver & Walter 2000; James

S Distal Slope Upper Slope Outer Platform Inner Platform N

5 km 5 km 100 km
80
LEGEND (e) (f) (g)
Maieberg Formation 20
20
δ C13
-5.0 0.0 60
pink marly limestone rhythmite
100 0
(d)
0 -5.0 δ13C 0.0
with sea-floor crystal fans P4017
40 -5.0 δ13C 0.0
G2008
Keilberg Member cap dolostone
80
marly micropeloidal dololutite turbidites
peloidal dolarenite, giant wave ripples 20
as above, swaley low-angle crossbedding
60 LEGEND
tubestone stromatolitic dolostone
peloidal dolarenite, sheet-crack cements 0 δ 13 C carb sediment
marly micropeloidal dolostone turbidites -5.0 δ13C 0.0
40 P6005 δ 13 C carb sheet-crack cement
10
-5.0 0.0 (c)
-5.0 0.0 10
10 Bethanis Member of the Ghaub Formation
(b) 20
(a) stratified proglacial detrital carbonate with ice-rafted debris
5
5 Ombaatjie Formation cycle b7-b8
5
0 peritial cycles of dolostone ribbonite, stromatolite, grainstone
0 P7016 and microbialaminite)
0 -5.0 δ13C 0.0
0 P7009
δ13C
δ13C P7002
P1607

Fig. 5.2. Representative lithologic columns and d13C profiles of the Keilberg cap dolostone (Marinoan) in contiguous palaeobathymetric zones of the Otavi
platform, northern Namibia (Hoffman et al. 2007). The datum for platform columns (e– g) is the inflection point (arrows) in isotopic profiles. Each column records part of
an overall sigmoidal profile, implying diachronous deposition during progressive marine inundation. The systematic nature of profiles rules out a detrital origin for
dolomite and a diagenetic origin for d13C values. Note that values for isopachous dolomite sheet-crack cement (crosses) in (b) and (c) are indistinguishable from host
dolopelmicrite (dots).
72
(m) SW measured sections ( )
0
–50
–100
0 5 10
Teniagouri Group
Atar green and red shale, siltstone
West
African
aquamarine chert (silexite), shale
craton 6 violet and green limestone
Jbeliat Group
5 dolomitic quartz sandstone, granulestone
4 reddish siltstone grading to fine sandstone
3 cap dolostone, tepee breccia, barite crusts
2 pebbly regolith, sand-wedge polygons
1 terrestrial lodgement tillite and outwash
Mesoproterozoic Atar, Tifounke, etc. groups
microbial limestone, dolostone; fine clastics
Fig. 5.3. Stratigraphic cross-section of the glaciogenic Jbéliat Group
(Marinoan?) based on sections measured along Atar Cliff, Mauritania, by
present authors PFH and GPH and Adam C. Maloof (Hoffman & Schrag 2002).
The cap-carbonate sequence (CCS) includes cap dolostone (unit 3) and
regressive highstand deposits (units 4 and 5) preserved only in palaeovalleys.
The datum is the top of the cap-carbonate sequence. Silicified shale (silexite)
belongs to the succeeding Téniagouri depositional sequence and need not be
genetically related to glaciation.
et al. 2001; Halverson et al. 2004; Xiao et al. 2004; Font et al.
2006; Nédélec et al. 2007). Locally, cap dolostones begin with a
coarsening-upward interval of discrete, parallel-sided, normally
graded, sedimentation units interpreted as turbidites, separated
by argillaceous partings (Lithofacies I of Kennedy 1996;
Hoffman & Macdonald 2010). The turbidites, where present,
grade upward into laminated but less discretely bedded dolostone
with ubiquitous low-angle crossbedding, including toplaps, onlaps
and downlaps (Lithofacies II of Kennedy 1996). Kennedy (1996)
attributes the low-angle crossbedding to asymmetric synsedimen-
tary dissolution and/or semi-plastic slumping of originally
parallel-sided strata deposited below the storm wave base. A
simpler and more conventional interpretation is that low-angle
crossbedding records wave action above the storm wave base.
This interpretation is consistent with the sorted macropeloids and
implies initial shoaling subsequent to the underlying turbidites
(Hoffman & Macdonald 2010). Intertidal– supratidal indicators
(e.g. fenestral texture or other desiccation features, abraded
intraclasts, beach ‘rosettes’, channels and levees, polygonal
tepees and related breccias) are absent (Kennedy 1996). The low-
angle cross-bedded lithofacies is dominant and within it occurs
a triad of highly idiosyncratic structures in a broadly consistent
vertical sequence, although all three are rarely present in a single
section (Fig. 5.2). The structures are described in more detail
in Hoffman (2011, and references therein) and are, in order of
appearance:
† sheet-crack cements (Kennedy 1996; Corkeron 2007; Hoffman
& Macdonald 2010) – bedding-parallel, variably buckled,
fibrous isopachous, void-filling dolospar, confined to a continu-
ous metre-thick zone of variable intensity near the base of the
cap dolostone in distal slope sections, typically just above
basal turbidites;
P. F. HOFFMAN ET AL.
NE † tubestone (geoplumb) stromatolites (Cloud et al. 1974; Hegen-
berger 1993; Corsetti & Grotzinger 2005) – metre- to
decametre-scale mounds within which arched microbial
growth laminae are interrupted by tube-like structures, invari-
ably oriented palaeovertically, defined by parallel-laminated
dolomicrite with meniscus-like curvature, variably replaced
by late-stage void-filling cement;
† giant wave ripples (Allen & Hoffman 2005; Hoffman & Li
2009) – trochoidal megaripples with high aspect ratios, aggra-
dational development, and crestward-coarsening bidirectional
laminae that interdigitate in the crestal region.
Interpretations of these structures differ. Kennedy et al. (2001)
relate sheet-crack cements to permafrost clathrate destabili-
15 km zation by marine flooding, but d13C values of the isopachous
dolospar cement are not extraordinarily depleted (Fig. 5.2), nor
is d34S of cap dolostone enriched as predicted (Shields 2005).
Corkeron (2007) attributes sheet-crack cements to pore-fluid over-
pressure associated with differential shale-carbonate compaction.
Hoffman & Macdonald (2010) suggest that pore-fluid overpres-
CCS
sures signal rapid falls in regional sea level associated with the
disappearance of ice sheets and the loss of their gravitational
‘pull’ on adjacent ocean waters. Cloud et al. (1974) interpret the
tubular structure in stromatolites as fluid- or gas-escape channels
on account of their geoplumb orientation. Corsetti & Grotzinger
(2005) interpret them as microbial growth structures. Hoffman
(2011) attempts to estimate the inclination of the upper foreslope
of the Otavi platform from the mean dip of stromatolitic layering
after restoring the contained tubes to vertical. Giant wave ripples
were first interpreted as supratidal tepee structures (Eisbacher
1985), but their crestlines are straight and parallel, not polygonal
like true tepees, and they lack the breccias and void-filling
cements diagnostic of supratidal tepees (Kendall & Warren
1987). Gammon et al. (2005) describe a tepee-like structure in
the Nuccaleena cap dolostone and relate it to a growth fault, but
giant wave ripples in the same formation lack faults and the one
described by Gammon et al. (2005), which is oriented perpendicu-
lar to bedding, could not have had a slip vector that was parallel or
subparallel to the outcrop surface (i.e. the plane of their two-
dimensional kinematic analysis), compromising the growth fault
interpretation. Allen & Hoffman (2005) interpret them as unu-
sually large, steep and aggradational wave ripples, related to the
action of long-period waves in the lower part of the ocean mixed
layer, driven by strong sustained winds (not hurricanes).
Hoffman & Li (2009) find that crestal orientations globally had
meridional (north –south) mean orientations and that zonal
(east – west) orientations were absent, consistent with zonal and
not with cyclonic winds.
Sorted peloids, low-angle crossbedding, stromatolites and giant
wave ripples signify that cap dolostones were deposited above the
storm wave base and at least partly in the euphotic zone. The
occurrence of shallow-water structures over a large range of
palaeodepths implies diachronous deposition attending a large
rise in sea level, consistent with the fragmented d13C records of
cap dolostones from different palaeodepths (Fig. 5.3). The esti-
mated magnitude of the rise of more than a kilometre implicates
global ice-sheet melting (so-called ‘glacioeustasy’) and links the
timescales of deglaciation and cap-dolostone sedimentation
(Kennedy 1996; Bertrand-Sarfati et al. 1997; James et al. 2001;
Hoffman et al. 2007). Cap dolostones are locally postglacial, but
globally syndeglacial.
Seafloor carbonate cement. In expanded sequences, cap dolostones
are overlain conformably by deeper water limestone, limestone
with dolostone turbidites, marlstone or shale. Seafloor cement, in
the form of crystal fans of pseudomorphosed prismatic aragonite,
built masses up to 100 m thick, localized by sea-bottom topogra-
phy (Hoffman 2011). In magnitude, they dwarf other Neoprotero-
zoic seafloor cements and differ from volumetrically comparable
 CHEMICAL SEDIMENTS ASSOCIATED WITH NEOPROTEROZOIC GLACIATION
Archaean –Palaeoproterozoic examples in having formed in the
presence of micritic sedimentation (Sumner 2002).
Highstand deposits of cap-carbonate sequences. Regressive high-
stand deposits in expanded cap-carbonate sequences may be car-
bonate or clastic-dominated. The former begin with limestone
rhythmite and shoal to dolostone grainstone beneath a well-
developed subaerial sequence boundary. Grain-sized equivalent
terrigenous deposits (shales, siltstones and sandstones) conform-
ably overlie cap dolostones in many areas (e.g. in the Brachina
Formation and the ABC Range Quartzite in South Australia). In
the cratonic Taoudeni Basin of West Africa, highstand deposits
are found in palaeovalleys but are missing on palaeotopographic
highs (Fig. 5.3). This has led to misinterpretation. Silicified
shale (silexite) of the Téniagouri Group (Fig. 5.3) and correlatives
rest directly on the altered surface of the cap dolostone over
vast areas of the craton, leading to the concept of a ‘triad’:
tillite –carbonate–silexite. We concur with Bertrand-Sarfati
et al. (1997), in that this concept should be abandoned.
Deposition of the Téniagouri Group (and correlative Azlaf
Group, Deynoux et al. 2006) cannot be related to the glacio-
eustatic transgression and could be much younger (Bertrand-
Sarfati et al. 1997).
Transgressive cap dolostones in condensed sequences. Where accom-
modation was lacking, cap dolostones aggraded or prograded to
the supratidal zone, where they were subjected to intense meteoric
and vadose diagenesis (Fig. 5.3). In such cases, primary marine
dolopelmicrite was repeatedly fractured and re-cemented in
tepee structures and related breccias (Kendall & Warren 1987).
Barite is a major void-filling cement in tepee brecciated cap dolos-
tone on the Yangtze platform of South China (Jiang et al. 2006)
and the West African craton (Shields et al. 2007). Where cap
dolostones were subaerially exposed (Fig. 5.3), highstand deposits
are typically absent.
Sturtian-type (Cryogenian) cap-carbonate sequences
The contacts between Sturtian glaciogenic deposits and cap car-
bonates, like Marinoan ones, are characteristically singular,
smooth, abrupt and conformable. Sturtian cap carbonates are litho-
logically distinct: where Marinoan cap dolostones are pale in
colour (,0.1 wt% total organic carbon, TOC), arenaceous in
texture and display wave-generated structures. Sturtian cap-
carbonate sequences typically begin with fetid, dark grey, parallel
laminated, micritic limestone (rarely rhodochrosite, MnCO3),
with or without parallel-sided graded beds (turbidites) and debrites
(Tojo et al. 2007). In Namibia, the Sturtian cap-carbonate sequence
continues with up to 200 m of continuous sublittoral dolostone
microbialaminite, featuring microbial rollup structures, neptunian
dykes and inclined zones of high primary porosity overlain by
humped microbialaminite (Pruss et al. 2010). The sublittoral
microbialaminite extravaganza eventually grades up into cross-
bedded grainstone, ending at a tepee brecciated sequence boundary.
Compared with Marinoan cap-carbonate sequences, where
transgressive tracts (i.e. cap dolostones) are well developed,
shallow-water transgressive tracts are thin (Smith et al. 1994) or
absent in Sturtian cap-carbonate sequences. Typically, deepwater
deposits directly overlie beveled supraglacial surfaces. The sim-
plest explanation for the difference is a below-critical saturation
state with respect to carbonate in the surface ocean during Sturtian
deglaciation (Hoffman & Schrag 2002).
Mid-Ediacaran cap carbonate sequences
Most mid-Ediacaran glaciogenic horizons lack cap carbonates
(e.g. Serra Azul, Squantum, Mortensnes, Moelv, Vil’chitsi, Loch
73
ne Cille, Luoquan, Hankalchough and Croles Hill diamictites).
Up to 40 cm of argillaceous limestone overlies glacial marine dia-
mictite in two sections of the short-lived, 582 Ma Gaskiers For-
mation on the Avalon Peninsula of eastern Newfoundland,
Canada (Myrow & Kaufman 1999). This cap carbonate is signifi-
cant as the sole carbonate bed in the nearly 15-km-thick host suc-
cession. It lithologically and isotopically resembles the base of
many Sturtian cap-carbonate sequences, although the amount of
carbonate deposited overall is far smaller. Finally, the Egan
diamictite represents a short-lived glacial incursion onto a
mid-Ediacaran carbonate platform in the Kimberley region of
Western Australia (Corkeron & George 2001). Diamictite and
overlying conglomerate are overlain by c.15 m of dolostone with
herringbone and trough crossbedding, and an additional c.15 m
of interbedded quartz-arenite and silty dolostone with a distinctive
stromatolite (Tungussia julia) horizon near the top. The carbonate
sequence following the Egan glaciation resembles no other
cap-carbonate sequence, including the one on the Landrigan dia-
mictite in the same succession, stratigraphically well below the
Egan, which has a recognizable Marinoan-type cap dolostone
(Corkeron 2007).
Barite in Marinoan-type (basal Ediacaran) cap dolostones
Barite (BaSO4) is a major constituent of certain Marinoan cap
dolostones (Table 5.2). The mineral is easily recognizable in the
field from its vitreous pearly lustre, bladed crystal habit and high
specific gravity (4.3 – 5.0). Even where pseudomorphosed by
calcite, its bladed crystals are distinct from aragonite, which is aci-
cular (needlelike) in habit, forming pseudohexagonal prisms due to
polysynthetic twinning. The palaeoenvironmental significance of
barite rests with its highly redox-sensitive solubility in S-bearing
aqueous solutions: the concentration of Ba in modern (oxic) sea-
water is only 1.4 ppb by atom, compared with 51 ppm in crustal
rocks from which Ba in seawater is derived. Consequently,
barite leached by anoxic pore waters from deep-sea carbonate
ooze reprecipitates cumulatively at the sub-seafloor redox front
as it migrates through the sediment column in response to sediment
accumulation. However, it is doubtful that this process alone could
provide a satisfactory explanation for either barite type in
Marinoan cap dolostones, given their limited thickness.
Two types of barite in cap dolostones: primary and early diagene-
tic. Two types of barite have been described from Marinoan cap
dolostones (Table 5.2): primary barite in the form of seafloor
cement (Kennedy 1996; Hoffman & Halverson 2011) and early
diagenetic barite associated with tepee and tepee-like breccias
(Jiang et al. 2006; Shields et al. 2007). Seafloor barite forms at a
laterally continuous horizon in the top few centimetres of cap
dolostones. Tiny barite crystals self-assemble into macroscopic
crystal fans or digitate aggregates. Interaction between barite
growth and particulate sedimentation forms the basis for determin-
ing a seafloor origin, denoting precipitation of barite from the
ambient water column. Early diagenetic barite forms void-filling
isopachous crustose cements within breccias associated with
tepee-like structures. The cause of tepee formation and brecciation
is attributed to submarine methane venting (Jiang et al. 2006), or
alternatively to evaporative pumping due to subaerial (supratidal)
exposure in zones of marine – meteoric groundwater mixing
(Shields et al. 2007). The second alternative is the conventional
interpretation of tepee structures and associated breccias (Assereto
& Kendall 1977; Kendall & Warren 1987).
Seafloor barite in central Australia. In the Amadeus and Ngalia
basins of central Australia, seafloor barite is associated with ferru-
ginous domal stromatolite at the top of cap dolostones overlying
Marinoan glaciogenic deposits of the Olympic and Mount
Doreen formations, respectively (Kennedy 1996). Barite occurs
74 P. F. HOFFMAN ET AL.
either as isolated bladed crystals (1 –2 mm) or as large upward-
oriented rosettes of bladed crystal (up to 4 cm high). The crystals
‘grew at the sediment –water interface, as indicated by the
upward-oriented growth habit, sediment drape defining crystal ter-
minations, and presence of crystal fragments as detrital material
within overlying sediment’ (Kennedy 1996). In the Amadeus
Basin, the barite-rich horizon is conformably overlain by reddish-
grey laminated siltstone with limestone turbidites (Pertatataka
Formation); in the Ngalia Basin it is overlain by red laminated
siltstone (Red Shale Member of the Mount Doreen Formation).
In both cases, in our view, the barite horizon marks a conformable
transition from deposition under the influence of storm waves (i.e.
size-sorted peloids, low-angle crossbedding) to sedimentation
below the storm wave base. Walter & Bauld (1983) suggested
that the barite was secondary after primary anhydrite (CaSO4), con-
tingent on their interpretation that the carbonate and sulphate were
products of intense evaporation in glacial lakes such as those of the
Antarctic Dry Valleys. This is not a credible scenario in an open
marine setting in the absence of a local hydrothermal source of Ca.
Seafloor barite in northwestern Canada. In the Mackenzie Moun-
tains of northwestern Canada, the top 4– 10 cm of the Ravensthroat
cap dolostone contains digitate barite cement (variably calcitized)
over a strike length of nearly 200 km (Hoffman & Halverson
2011). The barite coincides with a change in colour of the peloidal
dolostone from pale cream to chocolate brown. Locally, the
barite-rich layer is developed on a train of giant wave ripples.
The barite first appears as randomly oriented, millimetre-scale,
bladed crystals, which appear to have formed just below the
sediment –water interface. Upwards, the crystals self-organize to
form centimetre-scale digitate structures, commonly overstee-
pened toward the SW (seaward). The digitate structures display
coral-like growth laminae, defined by films of opaque minerals.
Where calcitized, they resemble microdigitate stromatolites
except for their diagnostic bladed crystal habit (so-called barite
‘rosettes’). The flanks of the digitate structures are ragged due to
outward growth of barite after each lamina of peloidal Fe-dolomite
was deposited between the digits. The interplay between crystal
growth and burial by peloids demonstrates that the barite precipi-
tated from the water column, into which the digitate structures
projected up to 1.0 cm above the seafloor. The Fe-dolostone
layer with barite cement makes a conformable contact with over-
lying micritic limestone (Hayhook Formation) – epigenetically
dolomitized locally – containing acicular crystal fans of pseudo-
morphosed aragonite cement and reworked detrital dolomite
basally (James et al. 2001).
Early diagenetic barite in the Taoudeni Basin, West Africa. The
common occurrence of barite in cap dolostones terminating the
Jbéliat glaciation of the West African craton has been known for
half a century (Deynoux & Trompette 1981). The type area is
the Atar Cliff, Mauritania, on the northern edge of the Taoudeni
Basin (Fig. 5.3). There, the cap dolostone directly overlies a peri-
glacial permafrost regolith with deep polygonal sand wedges,
developed above a terrestrial sub- and proglacial succession of
northerly derivation and vast extent (Deynoux 1982, 1985;
Deynoux et al. 2006). The cap dolostone is variably brecciated
and spectacular metre-scale tepee structures (Kendall & Warren
1987) are developed toward the top near Amogjar, east of Atar.
Isopachous barite forms thick crusts around blocks of brecciated
dolostone, as well as secondary (remobilized?) vein-fillings
(Shields et al. 2007). An early diagenetic origin for the cavity-
filling, crustose barite is indicated by detrital barite clasts in
coarse-grained quartz sand- and granulestone of the cap-carbonate
sequence highstand tract (Fig. 5.3).
Early diagenetic barite on the Yangtze Platform, South China. A
relatively thin (2.5 –4.0 m) cap dolostone (lower Doushantuo
Formation units C1 –3 of Jiang et al. 2006) blankets terminal
Cryogenian (Marinoan) glaciogenic deposits of the Nantuo For-
mation on the Yangtze platform (Jiang et al. 2006). Void-filling
barite fans occur in the lower part (unit C1) of the cap dolostone
on the inner platform and on the slope and basin to the SE. Unit 1
is brecciated everywhere and hosts a variety of sheet-crack
cements, cement-filled stromatactis-like cavities, and tepee-like
structures (Jiang et al. 2006). Importantly, the base of unit C1 is a
sharp, smooth, undulating surface; neither it nor the upper
Nantuo Formation experienced the intense brecciation typically
observed within unit C1 (Jiang et al. 2006). In addition, possible
seafloor barite is described as ‘layer-parallel barite fans growing
out of the dolomicrite substrate’ in the lower part of unit C2 in
the basin (Jiang et al. 2006). Tepee-like structures occur locally
in the lower part of unit C2 on the platform. Unit 3 and the upper
part of unit C2 are not brecciated, consisting of parallel- and
small-scale cross-laminated peloidal dolopackstone with graded
layers. Unit 3 has increased silt and limestone (Jiang et al. 2006).
Early diagenetic barite on the Dzabkhan Platform, western Mongo-
lia. A characteristic Marinoan cap dolostone (Table 5.2) overlies
the younger of two Cryogenian glaciogenic horizons (Khongoryn
diamictite) on the Dzabkhan platform of western Mongolia (Mac-
donald et al. 2009, 2011). Barite occurs in the mountain pass (Hoh
Davaa) between Bayan-Uul and Jargalan in northern Govi-Altay.
There the cap dolostone includes two units, separated by a sharp
smooth disconformity veneered by dark brown Fe-dolostone.
The first unit (11.1 m) is a pale tan peloidal dolostone with low-
angle crossbedding. Isopachous sheet-crack cements occur near
its base, above the basal 0.7 m of marly brown ribbon beds. The
second unit (8.4 m) consists of variably brecciated medium-grey
dolostone with thick crusts of isopachous void-filling barite
cement. Toward the top, sheet-crack and stratiform barites form
domal structures with up to 1.0 m of relief. The barite domes
were ultimately onlapped and buried by a third unit consisting of
unbrecciated medium-dark grey dolostone ribbons with subaqu-
eous microbial textured intervals, including rollup structures,
that lead into the scree-covered maximum flooding horizon of
the Ol cap-carbonate sequence. Although onlap relations clearly
show that the domal barite was exposed to seawater at the top of
the second unit, proof that it formed on the seafloor in the form
of detailed interaction between sedimentation and precipitation
has yet to be observed.
Phosphorite in cap-carbonate sequences
Taoudeni Basin, West Africa
On the northeastern margin of the Taoudeni Basin, diamictite of
the Jbéliat glaciation (Fersiga Formation) is overlain by a cap-
carbonate sequence correlative with that found in palaeovalleys
on the Atar Cliffs (Fig. 5.3), 1250 km to the SW (Bertrand-Sarfati
et al. 1997; Deynoux et al. 2006). A thin but continuous cap dolos-
tone (Oued Djouf Formation) is overlain by a regressive highstand
sequence (Grizim Formation) up to 80 m thick, composed of glau-
conitic green shale, siltstone and sandstone, ultimately aeolian
(Bertrand-Sarfati et al. 1997). Apatitic phosphorite is concentrated
at the base of the Grizim Formation, forming stromatolitic domes,
microdigitate clusters and shrub-like colonies over moraines of
Fersiga diamictite, where the cap dolostone is brecciated due to
subaerial exposure, and crossbedded phospharenite composed of
phosphate-coated grains and oncolites in the inter-moraine
depressions (Bertrand-Sarfati et al. 1997). Authigenic glauconite
occurs in both phosphorite facies and void-filling barite is a
minor constituent of the shrub-like phosphate colonies. The phos-
phorites formed rapidly, during the initial stages of the glacio-
eustatic transgression (Bertrand-Sarfati et al. 1997).
Phosphorites of early Ediacaran age occur in other areas,
but their connection to glaciation is tenuous. Khodjari-type
 CHEMICAL SEDIMENTS ASSOCIATED WITH NEOPROTEROZOIC GLACIATION
phosphorite in the northern Volta basin was tentatively linked to
the terminal Cryogenian deglaciation of West Africa (Trompette
et al. 1980), but the phosphorite is younger than regionally exten-
sive silicified green argillite (‘silexite’), which itself disconform-
ably overlies the basal Ediacaran (Marinoan) cap-carbonate
sequence in the Taoudeni basin (Fig. 5.3; Bertrand-Sarfati et al.
1997; Shields et al. 2007). Similarly, economic phosphorite in
the Khubsugul area of northern Mongolia (Ilyin et al. 1986) was
genetically related to the underlying terminal Cryogenian
Khesen diamictite (Sheldon 1984). However, Ilyin (2009) and
recent mapping has shown that the phosphorite is separated from
post-glacial Baxha cap-carbonate sequence by a major hiatus
(Macdonald & Jones 2011).
Genesis and significance of glacial-associated chemical
sediments
Fe and Fe – Mn oxide deposits
Ferrous v. euxinic anoxia. Anoxia due to ice cover has long been
invoked as a means of mobilizing reduced Fe and Mn in solution
as a source for oxide ores within Cryogenian glaciogenic
sequences (Martin 1965b; Urban et al. 1992; Klein & Beukes
1993; Lottermoser & Ashley 2000; Klein & Ladeira 2004).
Fe(II) concentration in the presence of H2S is limited by pyrite
(FeS2) saturation. High degrees of continental ice cover and conse-
quent large sea-level falls favour Fe over S in ocean waters because
of reduced riverine sulphate supply and higher Fe/S in deep-sea
hydrothermal vent fluids, respectively (Canfield & Raiswell
1999; Kump & Seyfried 2005).
Subglacial sulphate-rich ferrous waters. The discovery of modern
Fe- and sulphate-rich (3.45 and 50 mM, respectively) waters
trapped beneath the Taylor Glacier in the McMurdo Dry Valleys
of East Antarctica gives a new perspective on Cryogenian jaspilites
(Mikucki et al. 2009). Fe is not titrated as pyrite because there is
insufficient organic primary production, due to the glacier cover,
to fuel bacterial sulphate reduction. The source of Fe in the
trapped former seawater is subaqueous bedrock ‘weathering’,
enhanced by the low pH (6.2) of the waters. Fe-oxy-hydroxide
forms instantaneously at Blood Falls, where the anoxic waters dis-
charge at the glacier terminus (Mikucki et al. 2009). This demon-
strates that waters rich in both dissolved Fe and sulphate can exist
if ice cover is perennial and sufficiently thick to preclude photo-
trophic primary production.
Localization of oxidative titration. In anoxic Fe-rich waters, the
localization of Fe (and Mn) deposition could relate to the avail-
ability of oxygenated subglacial meltwater at the grounding-lines
of wet-base ice streams within cold-base ice sheets. This would
be consistent with the observed occurrence of Cryogenian jaspi-
lites as lenticular bodies in contact with diamictite complexes
whose lithofacies and internal organization is most consistent
with deposition in ice-sheet grounding zones. Alternation of Fe
and Mn ores could relate to variations of the meltwater flux associ-
ated with glacial retreat and advance, respectively, as proposed by
Urban et al. (1992).
Cap-carbonate sequences
Accumulation rates for syndeglacial cap dolostones. Marinoan cap
dolostones were deposited during marine transgressions resulting
from the meltdown of grounded ice sheets (Preiss et al., 1978;
Kennedy 1996; Bertrand-Sarfati 1997; James et al. 2001; Shields
2005; Fairchild & Kennedy 2007; Hoffman et al. 2007). The time-
scales for Cryogenian deglaciations have not been specifically tar-
geted in climate modelling. Nevertheless, deglaciation histories of
75
the temperate Quaternary ice sheets, simplified Neoproterozoic
deglacial simulations, and the apparent absence of polar continents
in Cryogenian palaeogeographic reconstructions (Fig. 5.1) all
suggest that Sturtian and Marinoan ice sheets disappeared
rapidly. How rapidly? If those middle and low-latitude ice sheets
disappeared at the rate most temperate mountain glaciers have
been lowered in recent decades (1.1 m/a), their demise would
have taken 2 ka assuming their average thickness was compar-
able with the present East Antarctic Ice Sheet (Lythe et al.
2001). This is close to the pan-deglacial timescale of ,2 ka in a
climate model (Hyde et al. 2000). Positive feedbacks – ice-albedo,
ice-elevation and various greenhouse-gas feedbacks notably water
vapour feedback – contribute to rapid deglaciation. The energy
demands for such a deglaciation, amounting to a globally averaged
energy flux of c. 11 W m2 (Wallace & Hobbs 1977, p. 320), are
small compared to the Neoproterozoic solar irradiance
of c. 362 W m2 (c. 94% of present), a high surface albedo notwith-
standing. Pending detailed modelling of a global deglaciation, we
conservatively take 8 ka, the timescale for Quaternary deglacia-
tions, as an upper limit for the Marinoan deglaciation. Given
median (9 m) and maximum (175 m) thicknesses of Marinoan
cap dolostones globally (Hoffman et al. 2007), average accumu-
lation rates were then c. 1 and 22 mm/a for cap dolostones of
typical and extreme thickness, respectively. If deglaciation
occurred in 2 ka, the respective accumulation rates were four
times larger. These must be considered minimum rates, as cap
dolostones at any given location represent only a fraction of the
total deglaciation (Fig. 5.2). We consider it likely that unusual
structures in cap dolostones (e.g. sheet-crack cements, tubestone
stromatolites and highly aggradational wave ripples) may be fun-
damentally related to extreme rates of accumulation.
The record of geomagnetic reversals and excursions in the
Mirassol d’Oeste and Nuccaleena cap dolostones creates a
dilemma (Trindade et al. 2003; Raub 2008). Conventional esti-
mates of their duration and frequency (Johnson et al. 1995;
Gubbins 1999; Roberts 2008) imply a timescale on the order of
0.1– 1.0 Ma for those cap dolostones, implying that accumulation
rates were at least 10 –100 times slower than the ‘conservative’
estimates given in the preceding paragraph. This makes for an
irreconcilable conflict between conventional interpretations of
geomagnetic polarity reversals and excursions on the one hand,
and estimated deglaciation rates on the other. If one accepts a
conventional magnetostratigraphic interpretation, cap dolostones
are highly condensed (Fairchild & Kennedy 2007). However,
compared with other depositional sequences in the same
(or other) successions, Marinoan cap dolostones are relatively
thick and expanded among transgressive tracts (e.g. Hoffman &
Halverson 2008). The resolution of geomagnetic field reversals
and excursions is limited by the accumulation rates of the sedi-
ments in which the records are kept (Channell & Lehman 1997;
Roberts & Lewin-Harris 2000; Roberts 2008). If Marinoan cap
dolostones were actually deposited at rates far exceeding the
best Quaternary deep-sea records (Channell & Lehman 1997;
Laj et al. 2006), might they not display hitherto unsuspected
phenomena? This would most likely be the case if the absolute
strength of the geomagnetic field was collapsed in 635 Ma, just
as it does during field reversals and excursions. The slowness
with which the solid inner core reverses, by diffusion, does not
limit the speed with which the outer core alone reverses in a
weak field (Gubbins 1999). In our view, the issue is unresolved,
but neither the sedimentological evidence for high sedimentation
rates nor the physical arguments for rapid deglaciation should be
dismissed lightly.
Are syndeglacial cap dolostones non-marine?
If Marinoan ice sheets had an average thickness of c. 2.2 km over
all continents and shelves (Fig. 5.1), their disappearance in 10 ka
76 P. F. HOFFMAN ET AL.
would produce meltwater at a global average rate of
c. 0.6  106 m3 s21 (2.01  1017 m3/3.15  1010 s), or c. 6.4 Sv
(Sverdrups, where 1.0 Sv ¼ 106 m3 s21). Deglaciation in 2 ka
would result in a global meltwater flux of c. 32 Sv. These values
correspond to approximately 6 and 32 times present total runoff,
respectively. Such an increase in freshwater input, followed by
strong surface warming after the ice disappeared, must have
impeded ocean mixing by creating a more stable density stratifica-
tion (Shields 2005). Cold saline deepwater that evolved during the
Marinoan glaciation would have been capped by a thickening
meltwater-dominated lid. The melting of sea ice, or a sea glacier
in the case of a Snowball Earth (Warren et al. 2002; Goodman
& Pierrehumbert 2003), followed by the melting of ice sheets,
would have contributed up to 0.7 and c. 1.0 km of freshwater
respectively. Ocean mixing time would have been many times
the present 103 years, ensuring that the surface ocean was brackish
when cap dolostones were deposited (Shields 2005). This
meltwater-dominated lid has been dubbed ‘plumeworld’ (Shields
2005), and we refer to the Marinoan example as ‘Glacial Lake
Harland’ (GLH). (W. Brian Harland of Cambridge University
was a pioneer advocate of Neoproterozoic glaciations as chronos-
tratigraphic markers.) Shields (2005, p. 305) gives many reasons
why GLH was more favourable than seawater for rapid carbonate
production.
Abiotic or biogenic dolomite?
The dolomicrospar of cap dolostones preserves primary structures,
makes conformable contact with overlying and (rare) underlying
limestones, and near the base of the former is reworked as intra-
clastic dolomite grains. These are good reasons to conclude that
the dolomite was either primary or a product of very early diagen-
esis near the sediment – water interface. It has recently been
demonstrated that dolomite-forming reactions are catalysed at
low temperatures by microbial activity utilizing different meta-
bolic pathways: sulphate reduction (van Lith et al. 2003), anaero-
bic methane oxidation (Moore et al. 2004) and methanogenesis by
Archaea (Kenward et al. 2009). The first of these has been applied
to Marinoan cap dolostones (Shields 2005; Font et al. 2006;
Nédélec et al. 2007). However, methanogenesis by Archaea is
particularly appealing because the formation of ordered dolomite
at low temperature (48C) has been demonstrated experimentally
in a low sulphate, low Mg:Ca (,1), acidic (pH ¼ 6.7), freshwater
environment, not unlike GLH (see above). As the methanogenic
consortium (Kenward et al. 2009) is anaerobic, formation of
cap dolostone by this means must have occurred below the
sediment –water interface, assuming deglacial surface waters
were well oxygenated.
Sources of alkalinity for cap dolostones. Syndeglacial carbonate pro-
duction in surface waters was driven by warming (DT  508C;
Pierrehumbert 2002), inundation of shallow shelves and platforms
where carbonate burial is favoured (Ridgwell et al. 2003), alka-
linity provided by anaerobic respiration (Kennedy et al. 2001;
but see Shields 2005 for critical discussion), and carbonate weath-
ering (Higgins & Schrag 2003; Anderson 2007; see Le Hir et al.
2009 for critical discussion, but note that they consider only sili-
cate weathering as a source of alkalinity, not carbonate weathering
which is orders of magnitude more rapid).
Significance of seafloor aragonite cements. Seafloor cements com-
posed of former aragonite occurring in limestones above Marinoan
cap dolostones formed below the storm wave base, at or near the
maximum flooding horizon (Grotzinger & Knoll 1995; James
et al. 2001; Nogueira et al. 2003; Hoffman 2011). They represent
a brief revival of more ancient conditions following a long decline
in the importance of seafloor cement in the latter half of the Proter-
ozoic (Grotzinger & James 2000; Sumner 2002). What was the
source of alkalinity driving carbonate production in deep water?
One possibility is anaerobic respiration, which operates as an
alkalinity pump in modern super-anoxic fjords (Anderson et al.
1987). Another is that extreme concentrations of dissolved inor-
ganic carbon (DIC), expected for an ocean equilibrated with
carbonate sediments and a c. 0.1 bar CO2 atmosphere, flattened
the negative gradient in carbonate saturation (V) with increasing
depth, favouring the precipitation of cement on the seafloor,
irrespective of the oxidant or intensity of organic carbon cycling
(Higgins et al. 2009).
Barite in cap dolostones
Seafloor barite. Seafloor barites in basal Ediacaran (Marinoan) cap
dolostones of central Australia and northwestern Canada are
strikingly similar in form, dimensions and precise stratigraphic
position within their respective cap-carbonate sequences. We
infer that they formed where the seabed intersected the interface
between Ba-rich euxinic deeper water and an oxic Fe(III)-
laden mixed layer. Localized upwelling would then have caused
sulphate production and barite titration. Simultaneously, anaerobic
respiration, where Fe was the electron acceptor, caused Fe(III)
reduction, leading to Fe-carbonate (Fe-dolostone) production
coincident with the barite cement horizon. It is well established
that mixed-layer waters from which cap dolostones were deposited
were Fe(III)-laden (Embleton & Williams 1986; Li 2000; Font
et al. 2005), accounting for the pinkish tint of many cap dolostones.
Ba in euxinic glacial deep waters would derive from seawater –
basalt exchange reactions and seafloor weathering of detrital feld-
spar. We see no need for the involvement of methane venting, for
which there is neither physical nor isotopic evidence at seafloor
barite horizons in Marinoan cap dolostones (Kennedy 1996;
Hoffman & Halverson 2011). Euxinic deep water that evolved
during the Marinoan glaciation (Hurtgen et al. 2006) may have
contrasted with Fe(II) sulphate deep water (Mikucki et al. 2009),
which may have developed during the Sturtian glaciation, account-
ing for the deposition of haematite –jaspilite at that time.
Early diagenetic (void-filling) barite. Tepee breccias in cap dolo-
stones and associated void-filling isopachous barite cement have
been ascribed to vadose diagenesis in West Africa (Bertrand-
Sarfati et al. 1997; Shields et al. 2007). In South China, similar fea-
tures have been attributed to submarine cold seeps (Jiang et al.
2006) or, alternatively, to vadose diagenesis (Zhou et al. 2010).
In both interpretations, barite precipitation is thought to have
resulted from mixing of an oxic sulphate-rich fluid (seawater)
and a highly reducing, Ba- and possibly methane-rich fluid,
derived from the destabilization of permafrost gas hydrate in under-
lying glaciogenic sediments from which the Ba was scavenged
(Jiang et al. 2006; Shields et al. 2007). It remains to be seen if
this attractive scenario is compatible with a basic field observation
in both areas, that the contact surface between the glaciogenic dia-
mictite and the cap dolostone is smooth and unbroken (authors’
observations). Is it conceivable that methane and other gases, gen-
erated within the till (permafrost melting), could thoroughly brecci-
ate the cap dolostone on the seafloor, while leaving their mutual
contact surface undisturbed? Methane involvement is supported
by extreme d13C depletion (below –10‰ Vienna Pee Dee Belem-
nite standard) in certain void-filling cement phases in brecciated
cap dolostone on the Yangtze platform (Jiang et al. 2003; Wang
et al. 2008). However, such strong depletions are not observed else-
where, despite thousands of published analyses, and those in South
China come mostly from late-stage calcite void-fillings that
demonstrably post-date coexisting barite and void-filling dolomite
cements (Bristow et al. 2011; Zhou et al. 2011). Moreover, d34S
values in cap dolostones are not strongly enriched, as predicted if
anaerobic methane oxidation (an alkalinity pump) was fueling bac-
terial sulphate reduction (Shields 2005).
 CHEMICAL SEDIMENTS ASSOCIATED WITH NEOPROTEROZOIC GLACIATION
Phosphorite in cap dolostones
It is not possible to generalize about Neoproterozoic glacial-related
phosphorite from a single example, no matter how impressive and
well documented it is (Bertrand-Sarfati et al. 1997). Its existence
supports assumed high productivity during deglaciation (Shields
2005; Font et al. 2006; Nédélec et al. 2007).
Conclusions
Synglacial Fe and Fe–Mn deposits are mostly, if not all, associated
with the older Cryogenian (Sturtian) glaciation. They indicate
anoxic (not euxinic) deep water. Lithofacies associations suggest
that subglacial meltwater plumes provided oxidant. Alternating
Fe and Mn deposits may reflect oscillating plume fluxes associated
with glacial cycles.
Sturtian and Marinoan cap carbonates are distinct. Transgres-
sive cap dolostones are almost entirely limited to the Marinoan
deglaciation, and were deposited diachronously, mainly above
the storm wave base, during coastal inundation by meltwater-
dominated surface waters. Biogenic dolomite nucleation in
waters of low ionic strength and low Mg:Ca ratio was likely
microbially mediated.
Barite occurs in Marinoan cap dolostones both as seafloor
cements in terminal Fe-dolomite beds and as early diagenetic void-
filling crusts within tepee and tepee-like breccias. Seafloor barite
cement formed where euxinic Ba-rich deep waters upwelled into
oxic Fe(III)-rich surface waters. Early diagenetic void-filling
barite cement in cap dolostones has been attributed to subaerial
and submarine methane seepage, driven by destabilization of
permafrost hydrates in underlying sediments following marine
inundation. Although attractive in principle, undisturbed contact
surfaces between glaciogenic diamictites and internally brecciated
cap dolostones may be difficult to reconcile with seepage from
below, unless brecciation was strictly limited to the horizon of
methane oxidation.
Fieldwork was supported by research grants from the Earth System History
(ESH), Arctic Natural Science (ANS), and Geobiology & Environmental Geo-
chemistry (GEG) programmes of the US National Science Foundation (NSF).
PFH receives additional support from the Canadian Institute for Advanced
Research (CIFAR), Harvard University Center for the Environment (HUCE),
US Social Security, and the Canada Pension Plan. FAM and GPH thank the
Yukon Geological Survey for additional support. We gratefully acknowledge
constructive reviews by G. Jiang and P. K. Link, which prompted several
improvements. This represents a contribution of the IUGS- and UNESCO-
funded IGCP (International Geoscience Programme) Project #512.
References
Aitken, J. D. 1991. The Ice Brook Formation and post-Rapitan, late
Proterozoic glaciation, Mackenzie Mountains, Northwest Territories.
Geological Survey of Canada Bulletin, 404, 1 – 43.
Allen, P. A. & Hoffman, P. F. 2005. Extreme winds and waves in the
aftermath of a Neoproterozoic glaciation. Nature, 433, 123–127.
Allen, P., Leather, J. & Brasier, M. D. 2004. The Neoproterozoic Fiq
glaciation and its aftermath, Huqf Supergroup of Oman. Basin
Research, 16, 507– 534.
Alvarenga, C. J. S. de & Trompette, R. 1992. Glacially influenced sedi-
mentation in the Later Proterozoic of the Paraguay belt (Mato Grosso,
Brazil). Palaeogeography, Palaeoclimatology, Palaeoecology, 92,
85 – 105.
Alvarenga, C. J. S. de, Dardenne, M. A. et al. 2008. Isotope stratigra-
phy of Neoproterozoic cap carbonates in the Araras Group, Brazil.
Gondwana Research, 13, 469–479.
Anderson, J. B. 1999. Antarctic Marine Geology. Cambridge University
Press, Cambridge.
Anderson, S. P. 2007. Biogeochemistry of glacial landscape systems.
Annual Review of Earth and Planetary Sciences, 35, 375– 399.
77
Anderson, L. G., Dyrssen, D. & Skei, J. 1987. Formation of chemogenic
calcite in super-anoxic seawater –Framvaren, Southern-Norway.
Marine Chemistry, 20, 361–376.
Assereto, R. L. A. M. & Kendall, C. G. St. C. 1977. Nature, origin and
classification of peritidal tepee structures and related breccas. Sedi-
mentology, 24, 153–210.
Badenhorst, F. P. 1988. The lithostratigraphy of the Chuos mixtite in part
of the southern central zone of the Damara Orogen, South West
Africa. Communications of the Geological Survey of South West
Africa/Namibia, 4, 103– 110.
Bao, H., Lyons, J. R. & Zhou, C. 2008. Triple oxygen isotope evidence
for elevated CO2 levels after a Neoproterozoic glaciation. Nature,
452, 504–506.
Bertrand-Sarfati, J., Flicoteaux, R., Moussine-Pouchkine, A. &
Aı̈t Kaci Ahmed, A. 1997. Lower Cambrian apatitic stromatolites
and phospharenites related to the glacio-eustatic cratonic rebound
(Sahara, Algeria). Journal of Sedimentary Research, 67, 957–974.
Bristow, T. F., Boniface, M., Derkowski, A., Eller, J. M. & Grotzin-
ger, J. P. 2011. A hydrothermal origin for isotopically anomalous cap
dolostone cements from south China. Nature, 474, 68 –72.
Cahen, L. & Lepersonne, J. 1981. Proterozoic diamictites of Lower
Zaire. In: Hambrey, M. J. & Harland, W. B. (eds) Earth’s Pre-
Pleistocene Glacial Record. Cambridge University Press, Cam-
bridge, 153– 157.
Cahen, L. 1950. Le Calcaire de Sekelolo, le Complexe tillitique et la
Dolomie rose C1 dans l’Anticlinal de Congo dia Kati (Bas-Congo).
Annales du Musée du Congo Belge, Sciences Géologiques, 7, 13– 54.
Calver, C. R. & Walter, M. R. 2000. The late Neoproterozoic Grassy
Group of King Island, Tasmania: correlation and palaeogeographic
significance. Precambrian Research, 100, 299– 312.
Canfield, D. E. & Raiswell, R. 1999. The evolution of the sulfur cycle.
American Journal of Science, 299, 697–723.
Channell, J. E. T. & Lehman, B. 1997. The last two geomagnetic
polarity reversals recorded in high-deposition-rate sediment drifts.
Nature, 389, 712– 715.
Chumakov, N. M. 1992. The problems of old glaciations: Pre-Pleistocene
glaciogeology in the USSR. Soviet Science Reviews, Section G
Geology, 1, 1 –208.
Chumakov, N. M. 2007. Climates and climate zonality of the Vendian:
geological evidence. In: Vickers-Rich, P. & Komarower, P. (eds)
The Rise and Fall of the Ediacaran Biota. Geological Society,
London, Special Publication, 286, 15– 26.
Clifford, T. N. 2008. The geology of the Neoproterozoic Swakop– Otavi
transition zone in the Outjo Distroct, northern Damara Orogen,
Namibia. South African Journal of Geology, 111, 117– 140.
Cloud, P., Wright, L. A., Williams, E. G., Diehl, P. & Walter, M. R.
1974. Giant stromatolites and associated vertical tubes from the upper
Proterozoic Noonday Dolomite, Death Valley region, eastern Califor-
nia. Geological Society of America Bulletin, 85, 1869– 1882.
Condon, D., Zhu, M., Bowring, S. A., Wang, W., Yang, A. & Jin, Y.
2005. U– Pb ages from the Neoproterozoic Doushantuo Formation,
China. Science, 308, 95 –98.
Corkeron, M. 2007. ‘Cap carbonates’ and Neoproterozoic glacigenic
successions from the Kimberley region, north-west Australia. Sedi-
mentology, 54, 871–903.
Corkeron, M. L. & George, A. D. 2001. Glacial incursion on a Neopro-
terozoic carbonate platform in the Kimberley region, Australia. Geo-
logical Society of America Bulletin, 113, 1121– 1132.
Corsetti, F. A. & Grotzinger, J. P. 2005. Origin and significance of tube
structures in Neoproterozoic post-glacial cap carbonates: example
from Noonday Dolomite, Death Valley, United States. Palaios, 20,
348– 363.
Corsetti, F. A. & Kaufman, A. J. 2003. Stratigraphic investigations of
carbon isotope anomalies and Neoproterozoic ice ages in Death
Valley, California. Geological Society of America Bulletin, 115,
916– 932.
Deynoux, M. 1982. Periglacial polygonal structures and sand wedges in
the late Precambrian glacial formations of the Taoudeni Basin in
Adrar of Mauretania (West Africa). Palaeogeography, Palaeoclima-
tology, Palaeoecology, 39, 55 –70.
Deynoux, M. 1985. Terrestrial or waterlain glacial diamictites? Three
case studies from the late Proterozoic and late Ordovician glacial
78 P. F. HOFFMAN ET AL.
drifts in West Africa. Palaeogeography, Palaeoclimatology,
Palaeoecology, 51, 97– 141.
Deynoux, M. & Trompette, R. 1976. Discussion: Late Precambrian
mixtites: glacial and/or nonglacial? Dealing especially with the
mixtites of West Africa. American Journal of Science, 276,
1302– 1315.
Deynoux, M. & Trompette, R. 1981. Late Precambrian tillites of the
Taoudeni Basin, West Africa. In: Hambrey, M. J. & Harland,
W. B. (eds) Earth’s Pre-Pleistocene Glacial Record. Cambridge
University Press, Cambridge, 123– 131.
Deynoux, M., Affaton, P., Trompette, R. & Villeneuve, M.
2006. Pan-African tectonic evolution and glacial events registered
in Neoproterozoic to Cambrian cratonic and foreland basins of
West Africa. Journal of African Earth Sciences, 46, 397–426.
Dorr, J. V. N. II. 1945. Manganese and iron deposits of Morro do Urucum,
Mato Grosso, Brazil. United States Geological Survey Bulletin,
946-A, 1– 47.
Dowdeswell, J. A., Whittington, J. A., Jennings, A. E., Andrews,
J. T., Mackensen, A. & Marienfield, P. 2000. An origin for lami-
nated glacimarine sediments through sea-ice build-up and suppressed
iceberg rafting. Sedimentology, 47, 557– 576.
Dunn, P. R., Thompson, B. P. & Rankama, K. 1971. Late Pre-Cambrian
glaciation in Australia as a stratigraphic boundary. Nature, 231,
498– 502.
Edwards, M. B. 1984. Sedimentology of the Upper Proterozoic glacial
record, Vestertana Group, Finnmark, North Norway. Norges Geolo-
giske Undersøkelse Bulletin, 394, 76.
Eisbacher, G. H. 1985. Late Proterozoic rifting, glacial sedimentation and
sedimentary cycles in the light of Windermere deposition, western
Canada. Palaeogeography, Palaeoclimatology, Palaeoecology, 51,
231– 254.
Embleton, B. J. J. & Williams, G. E. 1986. Low latitude of deposition for
late Precambrian periglacial varvites in South Australia: implications
for palaeoclimatology. Earth and Planetary Science Letters, 79,
419– 430.
Fairchild, I. J. & Kennedy, M. J. 2007. Neoproterozoic glaciation in the
Earth System. Journal of the Geological Society, London, 164,
895– 921.
Fanning, C. M. & Link, P. K. 2008. Age constraints for the Sturtian
glaciation: data from the Adelaide Geosyncline, South Australia
and Pocatello Formation, Idaho, USA. Selwyn Symposium
2008, Geological Society of Australia, Extended Abstracts, 91,
57 –62.
Fölling, P. G. & Frimmel, H. E. 2002. Chemostratigraphy correlation of
carbonate successions in the Gariep and Saldania Belts, Namibia and
South Africa. Basin Research, 14, 69 – 88.
Fölling, P. G., Zartman, R. E. & Frimmel, H. E. 2000. A novel
approach to double-spike Pb/Pb dating of carbonate rocks: examples
from Neoproterozoic sequences in southern Africa. Chemical
Geology, 171, 97 –122.
Font, E., Nédélec, A., Trindade, R. I. F., Macouin, M. & Charrière,
A. 2006. Chemostratigraphy of the Neoproterozoic Mirassol d’Oeste
cap dolostones (Mato Grosso, Brazil): an alternative model for
Marinoan cap dolostone formation. Earth and Planetary Science
Letters, 250, 89 –103.
Font, E., Trindade, R. I. F. & Nédélec, A. 2005. Detrital remanent mag-
netization in haematite-bearing Neoproterozoic Puga cap dolostone,
Amazon craton: a rock magnetic and SEM study. Geophysical
Journal International, 163, 491–500.
Frimmel, H. E. 2004. Neoproterozoic sedimentation rates and timing
of glaciations – a southern African perspective. In: Eriksson,
P. G., Altermann, W., Nelson, D. R., Mueller, W. U. &
Catuneanu, O. (eds) The Precambrian Earth: Tempos and
Events. Elsevier, Amsterdam, 459– 473.
Frimmel, H. E. 2008. Neoproterozoic Gariep Orogen. In: Miller,
R. McG. (ed.) The Geology of Namibia (In Three Volumes),
Volume 2 (Neoproterozoic to Lower Palaeozoic). Geological
Survey of Namibia, Windhoek, 14-1–14-39.
Frimmel, H. E. & Von Veh, M. W. 2003. Numees formation (including
the Jakkalsberg Member). In: Johnson, M. R. (ed.) Catalogue of
South African Lithostratigraphic Units. South African Committee
for Stratigraphy, Pretoria, 7-25– 7-28.
Frimmel, H. E. 2011. The Kaigas and Numees Formations, Port Nolloth
Group, in South Africa and Namibia. In: Arnaud, E., Halverson,
G. P. & Shields-Zhou, G. (eds) The Geological Record of Neopro-
terozoic Glaciations. Geological Society, London, Memoirs, 36,
223– 231.
Gammon, P. R., McKirdy, D. M. & Smith, H. D. 2005. The timing and
environment of tepee formation in a Marinoan cap carbonate. Sedi-
mentary Geology, 177, 195– 208.
Gaucher, C., Frimmel, H. E. & Germs, G. J. B. 2005. Organic-walled
microfossils and biostratigraphy of the upper Port Nolloth Group
(Namibia): implications for latest Neoproterozoic glaciations. Geo-
logical Magazine, 142, 539–559.
Goodman, J. & Pierrehumbert, R. T. 2003. Glacial flow of floating
marine ice in ‘Snowball Earth’. Journal of Geophysical Research,
108, 3308, doi: 10.1029/2002JC001471.
Graf, J. L., Jr, O’Connor, E. A. & Van Leeuwin, P. 1994. Rare earth
element evidence of origin and depositional environment of Late
Proterozoic ironstone beds and manganese-oxide deposits, SW
Brazil and SE Bolivia. Journal of South American Earth Sciences,
7, 115– 133.
Grotzinger, J. P. & James, N. P. 2000. Precambrian carbonates: evol-
ution of understanding. In: Grotzinger, J. P. & James, N. P. (eds)
Carbonate Sedimentation and Diagenesis in the Evolving Precam-
brian World. SEPM (Society for Sedimentary Geology) Special
Publication, 67, 3– 20.
Grotzinger, J. P. & Knoll, A. H. 1995. Anomalous carbonate precipi-
tates: is the Precambrian the key to the Permian? Palaios, 10,
578– 596.
Gubbins, D. 1999. The distinction between geomagnetic excursions and
reversals. Geophysical Journal International, 137, F1– F3.
Halverson, G. P. & Shields, G. 2011. Chemostratigraphy and
the Neoproterozoic glaciations. In: Arnaud, E., Halverson, G. P.
& Shields, G. (eds) The Geological Record of Neoproterozoic
Glaciations. Geological Society, London, Memoirs, 36, 51 –66.
Halverson, G. P., Maloof, A. C. & Hoffman, P. F. 2004. The Marinoan
glaciation (Neoproterozoic) in northeast Svalbard. Basin Research,
16, 297– 324.
Halverson, G. P., Hoffman, P. F., Schrag, D. P., Maloof, A. C. & Rice,
A. H. N. 2005. Toward a Neoproterozoic composite carbon-isotope
record. Geological Society of America Bulletin, 117, 1181– 1207.
Halverson, G. P., Dudás, F. Ö., Maloof, A. C. & Bowring, S. A.
2007. Evolution of the 87Sr/86Sr composition of Neoproterozoic
seawater. Palaeogeography, Palaeoclimatology, Palaeoecology,
256, 103– 129.
Halverson, G. P., Poitrasson, F., Hoffman, P. F., Nédélec, A.,
Montel, J.-M. & Kirby, J. 2011. Fe isotope and trace element geo-
chemistry of the Neoproterozoic syn-glacial Raptian iron formation.
Earth and Planetary Science Letters, 309, 100–112.
Hegenberger, W. 1993. Stratigraphy and sedimentology of the Late Pre-
cambrian Witvlei and Nama Groups, East of Windhoek. Geological
Survey of Namibia Memoir, 17, 82.
Higgins, J. A. & Schrag, D. P. 2003. Aftermath of a snowball Earth. Geo-
physics, Geochemistry, Geosystems, 4, doi:10.1029/2002GC000403.
Higgins, J. A., Fischer, W. W. & Schrag, D. P. 2009. Oxygenation of the
ocean and sediments: consequences for the seafloor carbonate
factory. Earth and Planetary Science Letters, 284, 25 – 33.
Hildebrand, R. S. 2009. Did westward subduction cause Cretaceous–
Tertiary orogeny in the North American Cordillera? Geological
Society of America, Special Paper, 457, 71.
Hoffman, P. F. 2011. Strange bedfellows: glacial diamictite and cap
carbonate from the Marinoan (635 Ma) glaciation in Namibia. Sedi-
mentology, 58 – 119.
Hoffman, P. F. & Halverson, G. P. 2008. Otavi group of the western
northern platform, the eastern Kaoko Zone and the western northern
Margin Zone. In: Miller, R. McG. (ed.) The Geology of Namibia,
Vol. 2: Neoproterozoic to Lower Palaeozoic. Handbook of the
Geological Survey of Namibia, Windhoek, 13.69– 13.136.
Hoffman, P. F. & Halverson, G. P. 2011. Neoproterozoic glacial record
in the Mackenzie Mountains, northern Canadian Cordillera. In:
Arnaud, E., Halverson, G. P. & Shields, G. (eds) The Geological
Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 397 –411.
 CHEMICAL SEDIMENTS ASSOCIATED WITH NEOPROTEROZOIC GLACIATION
Hoffman, P. F. & Li, Z. X. 2009. A palaeogeographic context for
Neoproterozoic glaciation. Palaeogeography, Palaeoclimatology,
Palaeoecology, 277, 158–172.
Hoffman, P. F. & Macdonald, F. A. 2010. Sheet-crack cements and
early regression in Marinoan (635 Ma) cap dolostones: regional
benchmarks of vanishing ice-sheets? Earth and Planetary Science
Letters, 300, 374– 384.
Hoffman, P. F. & Schrag, D. P. 2002. The snowball Earth hypothesis:
testing the limits of global change. Terra Nova, 14, 129– 155.
Hoffman, P. F., Kaufman, A. J., Halverson, G. P. & Schrag, D. P.
1998. A Neoproterozoic Snowball Earth. Science, 281, 1342– 1346.
Hoffman, P. F., Halverson, G. P., Domack, E. W., Husson, J. M.,
Higgins, J. A. & Schrag, D. P. 2007. Are basal Ediacaran (635
Ma) post-glacial ‘cap dolostones’ diachronous? Earth and Planetary
Science Letters, 258, 114– 131.
Hoffmann, K.-H., Condon, D. J., Bowring, S. A. & Crowley, J. L.
2004. U– Pb zircon date from the Neoproterozoic Ghaub Formation,
Namibia: constraints on Marinoan glaciation. Geology, 32, 817– 820.
Hurtgen, M. T., Halverson, G. P., Arthur, M. A. & Hoffman, P. F.
2006. Sulfur cycling in the aftermath of a 635-Ma snowball
glaciation: evidence for a syn-glacial sulfidic deep ocean. Earth
and Planetary Science Letters, 245, 551–570.
Hyde, W. T., Crowley, T. J., Baum, S. K. & Peltier, W. R. 2000.
Neoproterozoic ‘snowball Earth’ simulations with a coupled
climate/ice-sheet model. Nature, 405, 425–429.
Ilyin, A. V. 2009. Neoproterozoic banded iron formations. Lithology and
Mineral Resources, 44, 78 –86.
Ilyin, A. V., Zaitsev, N. S. & Bjamba, Z. 1986. Khubsugul, Mongolia
people’s republic. In: Cook, P. J. & Shergold, J. H. (eds) Phosphate
Deposits of the World, Vol. 1: Proterozoic and Cambrian Phosphor-
ites. Cambridge University Press, Cambridge.
James, N. P., Narbonne, G. M. & Kyser, T. K. 2001. Late Neoprotero-
zoic cap carbonates: Mackenzie Mountains, northwestern Canada:
precipitation and global glacial meltdown. Canadian Journal of
Earth Sciences, 38, 1229–1262.
Jiafu, T., Heqin, F. & Zhioiu, G. 1987. Stratigraphy, type and formation
conditions of the Late Precambrian banded iron ores in South China.
Chinese Journal of Geochemistry, 6, 332–351.
Jiang, G., Kennedy, M. J. & Christie-Blick, N. 2003. Stable isotopic
evidence for methane seeps in Neoproterozoic postglacial cap car-
bonates. Nature, 426, 822– 826.
Jiang, G., Kennedy, M. J., Christie-Blick, N., Wu, H. & Zhang, S.
2006. Stratigraphy, sedimentary structures and textures of the late
Neoproterozoic Doushantuo cap carbonate in South China. Journal
of Sedimentary Research, 76, 978– 995.
Johnson, H. P., Van Patten, D., Tivey, M. & Sager, W. W. 1995. Geo-
magnetic polarity reversal rate for the Phanerozoic. Geophysical
Research Letters, 22, 231– 234.
Kaufman, A. J., Jiang, G., Christie-Blick, N., Banerjee, D. M. & Rai,
V. 2006. Stable isotope record of the terminal Neoproterozoic Krol
platform in the Lesser Himalayas of northern India. Precambrian
Research, 147, 156– 185.
Kendall, C. G. St. C. & Warren, J. 1987. A review of the origin and
setting of tepees and their associated fabrics. Sedimentology, 34,
1007– 1027.
Kennedy, M. J. 1996. Stratigraphy, sedimentology and isotopic geochem-
istry of Australian Neoproterozoic postglacial cap dolostones: degla-
ciation, d13C excursions and carbonate precipitation. Journal of
Sedimentary Research, 66, 1050– 1064.
Kennedy, M. J., Runnegar, B., Prave, A. R., Hoffmann, K.-H. &
Arthur, M. A. 1998. Two of four Neoproterozoic glaciations?
Geology, 26, 1059– 1063.
Kennedy, M. J., Christie-Blick, N. & Sohl, L. E. 2001. Are Proterozoic
cap carbonates and isotopic excursions a record of gas hydrate desta-
bilization following Earth’s coldest intervals? Geology, 29, 443– 446.
Kenward, P. A., Goldstein, R. H., González, L. A. & Roberts, J. A.
2009. Precipitation of low-temperature dolomite from an aerobic
microbial consortium: the role of methanogenic Archaea. Geobiol-
ogy, 7, 1 – 10.
Kianian, M. & Khakzad, A. 2008. Geochemistry of glaciogenic Neopro-
terozoic banded iron-formations from Kerman District (Iran). 33rd
International Geological Congress, Oslo, Abstracts, Session CGC-04.
79
Klein, C. 2005. Some Precambrian banded iron-formations (BIFs) from
around the world: their age, geologic setting, mineralogy, meta-
morphism, geochemistry and origin. American Mineralogist, 90,
1473–1499.
Klein, C. & Beukes, N. J. 1993. Sedimentology and geochemistry of
the glacigenic Late Proterozoic Rapitan iron-formation in Canada.
Economic Geology, 88, 542–565.
Klein, C. & Ladeira, E. A. 2004. Geochemistry and mineralogy of
Neoproterozoic banded iron-formations and some selected, siliceous
manganese formations from the Urucum District, Mato Grosso do
Sul, Brazil. Economic Geology, 99, 1233– 1244.
Knoll, A. H., Walter, M. R., Narbonne, G. M. & Christie-Blick, N.
2006. The Ediacaran Period: a new addition to the geologic time
scale. Lethaia, 39, 13– 30.
Kump, L. R. & Seyfried, W. E. Jr. 2005. Hydrothermal Fe fluxes during
the Precambrian: effect of low oceanic sulfate concentrations and low
hydrostatic pressure on the composition of black smokers. Earth and
Planetary Science Letters, 235, 654– 662.
Laj, C., Kissel, C. & Roberts, A. P. 2006. Geomagnetic field behavior
during the Iceland Basin and Laschamp geomagnetic excursions:
a simple transitional field geometry? Geochemistry, Geophysics,
Geosystems, 7, Q03004, doi:10.1029/2005GC001122.
Le Hir, G. & Donnadieu, Y. et al. 2009. The snowball Earth aftermath:
exploring the limits of continental weathering processes. Earth and
Planetary Science Letters, 277, 453– 463.
Li, Z. X. 2000. New palaeomagnetic results from the ‘cap dolomite’ of the
Neoproterozoic Walsh Tillite, northwestern Australia. Precambrian
Research, 100, 359–370.
Li, Z. X. & Bogdanova, S. V. et al. 2008. Assembly, configuration and
break-up history of Rodinia: a synthesis. Precambrian Research, 160,
179– 210.
Lottermoser, B. G. & Ashley, P. M. 2000. Geochemistry, petrology and
origin of Neoproterozoic ironstones in the eastern part of the Adelaide
Geosyncline. Precambrian Research, 101, 49 –67.
Lythe, M. B., Vaughan, D. G. & BEDMAP CONSORTIUM
2001. BEDMAP: a new thickness and subglacial topographic
model of Antarctica. Journal of Geophysical Research, 106,
11335–11351.
Macdonald, F. A. 2011. The Tsagaan Oloom Formation, southwestern
Mongolia. In: Arnaud, E., Halverson, G. P. & Shields, G. (eds)
The Geological Record of Neoproterozoic Glaciations. Geological
Society, London, Memoirs, 36, 331– 337.
Macdonald, F. A. & Cohen, P. A. 2011. The Tatonduk inlier, Alaska–
Yukon border. In: Arnaud, E., Halverson, G. P. & Shields, G.
(eds) The Geological Record of Neoproterozoic Glaciations.
Geological Society, London, Memoirs, 36, 389–396.
Macdonald, F. A. & Jones, D. S. 2011. The Khubsugul Group, northern
Mongolia. In: Arnaud, E., Halverson, G. P. & Shields, G. (eds)
The Geological Record of Neoproterozoic Glaciations. Geological
Society, London, Memoirs, 36, 339– 345.
Macdonald, F. A., Mcclelland, W. C., Schrag, D. P. & Macdonald,
W. P. 2009. Neoproterozoic glaciation on a carbonate platform
margin in Arctic Alaska and the origin of the North Slope subterrane.
Geological Society of America Bulletin, 121, 448– 473.
Macdonald, F. A., Schmidtz, M. D. et al. 2010. Calibrating the Cryo-
genian. Science, 327, 1241– 1243.
Macdonald, F. A., Strauss, J. V., Rose, C., Dudás, F. Ö. & Schrag,
D. P. 2011. Stratigraphy of the Port Nolloth Group of Namibia and
South Africa and implications for the age of Neoproterozoic iron
formations. American Journal of Science, 310, 862–888.
Martin, H. 1965a. The Precambrian Geology of South West Africa and
Namaqualand. University of Cape Town, Precambrian Research
Unit Bulletin, 1, 1– 159.
Martin, H. 1965b. Beobachtungen zum Problem der jung-präkam-
brischen Glazialen Ablagerungen in Südwestafrika (Observations
concerning the problem of the late Precambrian glacial deposits in
South West Africa). Geologische Rundschau, 54, 115– 127.
Mikucki, J. A., Pearson, A. et al. 2009. A contemporary microbially
maintained subglacial ferrous ‘ocean’. Science, 324, 397–400.
Moore, T. S., Murray, R. W., Kurtz, A. C. & Schrag, D. P. 2004.
Anaerobic methane oxidation and the formation of dolomite. Earth
and Planetary Science Letters, 229, 141–154.
80 P. F. HOFFMAN ET AL.
Myrow, P. M. & Kaufman, A. J. 1999. A newly discovered cap carbonate
above Varanger-age glacial deposits in Newfoundland, Canada.
Journal of Sedimentary Research, 69, 784– 793.
Nédélec, A., Affaton, P., France-Lanord, C., Charrière, A. &
Alvaro, J. 2007. Sedimentology and chemostratigraphy of the
Bwipe Neoproterozoic cap dolostones (Ghana, Volta Basin): a
record of microbial activity in a peritidal environment. C. R.
Geoscience, 339, 223– 239.
Nogueira, A. C. R., Riccomini, C., Sial, A. N., Moura, C. A. V. &
Fairchild, T. R. 2003. Soft-sediment deformation at the base of
the Neoproterozoic Puga cap carbonate (southwestern Amazon
craton, Brazil): confirmation of rapid icehouse to greenhouse tran-
sition in snowball Earth. Geology, 31, 613–616.
Nogueira, A. C. R., Riccomini, C., Sial, A. N., Moura, C. A. V., Trin-
dade, R. I. F. & Fairchild, T. R. 2007. Carbon and strontium isotope
fluctuations and paleoceanographic changes in late Neoproterozoic
Araras carbonate platform, southern Amazon craton, Brazil. Chemi-
cal Geology, 237, 168–190.
Pierrehumbert, R. T. 2002. The hydrologic cycle in deep-time climate
problems. Nature, 419, 191–198.
Plummer, P. S. 1978. Note on the palaeoenvironmental significance of the
Nuccaleena Formation (upper Precambrian), central Flinders Ranges,
South Australia. Journal of the Geological Society of Australia, 25,
395– 402.
Prave, A. R., Hoffmann, K.-H., Hegenberger, W. & Fallick, A. E.
2011. The Witvlei Group of east-central Namibia. In: Arnaud, E.,
Halverson, G. P. & Shields, G. (eds) The Geological Record
of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 211– 216.
Preiss, W. V. 1987. The Adelaide Geosyncline: Late Proterozoic stratigra-
phy, sedimentation, palaeontology and tectonics. Geological Survey
of South Australia Bulletin, 53, 438.
Preiss, W. V., Walter, M. R., Coates, R. P. & Wills, A. T. 1978. Litho-
logical correlation of the Adelaidean glaciogenic rocks in parts of the
Amadeus, Ngalia & Georgina basins. Bureau of Mineral Resources,
Journal of Australian Geology and Geophysics, 3, 43 –53.
Pruss, S. B., Bosak, T., Macdonald, F. A., Mclane, M. & Hoffman,
P. F. 2010. Microbial facies in a Sturtian cap carbonate, the Rasthof
Formation, Otavi Group, northern Namibia. Precambrian Research,
181, 187– 108.
Rankama, K. 1973. The Late Precambrian glaciation, with particular
reference to the Southern Hemisphere. Journal and Proceedings,
Royal Society of New South Wales, 106, 89– 97.
Raub, T. D. 2008. Prolonged deglaciation of ‘Snowball Earth’. PhD
thesis, Yale University, CT.
Ridgwell, A. J., Kennedy, M. J. & Caldeira, K. 2003. Carbonate depo-
sition, climate stability and Neoproterozoic ice ages. Science, 302,
859– 862.
Roberts, A. P. 2008. Geomagnetic excursions: knowns and unknowns.
Geophysical Research Letters, 35, L17307, doi: 10.1029/
2008GL034719.
Roberts, A. P. & Lewin-Harris, J. C. 2000. Marine magnetic anomalies:
evidence that ‘tiny wiggles’ represent short-period geomagnetic
polarity intervals. Earth and Planetary Science Letters, 183,
375– 388.
Sheldon, R. P. 1984. Ice-ring origin of the Earth’s atmosphere and hydro-
sphere and Late Proterozoic –Cambrian hypothesis. Geological
Survey of India Special Publication, 17, 17 –21.
Shields, G. A. 2005. Neoproterozoic cap carbonates: a critical appraisal
of existing models and the plumeworld hypothesis. Terra Nova, 17,
299– 310.
Shields, G. A., Deynoux, M., Strauss, H., Paquet, H. & Nahon, D.
2007. Barite-bearing cap dolostone of the Taoudéni Basin, northwest
Africa: sedimentary and isotopic evidence for methane seepage
after a Neoproterozoic glaciation. Precambrian Research, 154,
209– 235.
Smith, L. H., Kaufman, A. J., Knoll, A. H. & Link, P. K. 1994.
Chemostratigraphy of predominantly siliciclastic Neoproterozoic
successions: a case study of the Pocatello Formation and lower
Brigham Group, Idaho. Geological Magazine, 131, 301–314.
Sovetov, Yu. K. & Komlev, D. A. 2005. Tillites at the base of the Oselok
Group, foothills of the Sayan Mountains, and the Vendian lower
boundary in the southwestern Siberian Platform. Stratigraphy and
Geological Correlations, 13, 337–366.
Sumner, D. Y. 2002. Decimetre-thick encrustations of calcite and arago-
nite on the sea-floor and implications for Neoarchaean and Neoproter-
ozoic ocean chemistry. Special Publications of the International
Association of Sedimentologists, 33, 107–120.
Tojo, B., Katsuta, N., Takano, M., Kawakami, S. & Ohno, T. 2007.
Calcite – dolomite cycles in the Neoproterozoic Cap carbonates,
Otavi Group, Namibia. In: Vickers-Rich, P. & Komarower, P.
(eds) The Rise and Fall of the Ediacaran Biota. Geological Society,
London, Special Publication, 286, 103– 113.
Trindade, R. I. F., Font, E., D’agrella-Filho, M. S., Nogueira,
A. C. R. & Riccimini, C. 2003. Low-latitude and multiple geomag-
netic reversals in the Neoproterozoic Puga cap carbonate, Amazon
craton. Terra Nova, 15, 441–446.
Trompette, R. 1994. Geology of Western Gondwana (2000– 500 Ma):
Pan-African –Brasiliano Aggregation of South America and Africa.
Balkema, Amsterdam.
Trompette, R., Affaton, P., Joulia, F. & Marchand, J. 1980. Strati-
graphic and structural controls of late Precambrian phosphate depos-
its of the Northern Volta Basin in Upper Volta, Niger and Benin, West
Africa. Economic Geology, 75, 62– 70.
Trompette, R., de Alvarenga, C. J. A. & Wade, D. 1998. Geological
evolution of the Neoproterozoicv Corumbá graben system (Brazil).
Depositional context of the stratified Fe and Mn ores of the Jacadigo
Group. Journal of South American Earth Sciences, 11, 587–597.
Urban, H., Stribny, B. & Lippolt, H. J. 1992. Iron and manganese
deposits of the Urucum District, Mato Grosso do Sul, Brazil. Econ-
omic Geology, 87, 1375–1892.
van Lith, Y., Warthmann, R., Vasconcelos, C. & McKenzie, J. A.
2003. Sulphate-reducing bacteria induce low-temperature Ca-
dolomite and high Mg-calcite formation. Geobiology, 1, 71 – 79.
Walde, D. H. G., Gierth, E. & Leonardos, O. H. 1981. Stratigraphy and
mineralogy of the manganese ores of Urucum, Mato Grosso, Brazil.
Geologische Rundschau, 70, 1077– 1085.
Wallace, J. M. & Hobbs, P. V. 1977. Atmospheric Science: An Introduc-
tory Survey. Academic Press, San Diego.
Walter, M. R. & Bauld, J. 1983. The association of sulphate evaporites,
stromatolitic carbonates and glacial sediments: examples from the
Proterozoic of Australia and the Cainozoic of Antarctic. Precambrian
Research, 21, 129– 148.
Wang, J., Jiang, G., Xiao, S., Li, Q. & Wei, Q. 2008. Carbon isotope evi-
dence for widespread methane seeps in the ca 635 Ma Doushantuo
cap carbonate in South China. Geology, 36, 347– 350.
Warren, S. G., Brandt, R. E., Grenfell, T. C. & Mckay, C. P. 2002.
Snowball Earth: ice thickness on the tropical ocean. Journal of
Geophysical Research, 107, 3167, doi: 10.1029/2001JC001123.
Whitten, G. F. 1970. The investigation and exploitation of the Razorback
Ridge iron deposit. Geological Survey of South Australia Reports of
Investigations, 33, 165.
Xiao, S., Bao, H. et al. 2004. The Neoproterozoic Quruqtagh Group in
eastern Chinese Tianshan: evidence for a post-Marinoan glaciation.
Precambrian Research, 130, 1 –26.
Yeo, G. M. 1981. The Late Proterozoic Rapitan glaciation in the northern
Cordillera. In: Campbell, F. H. A. (ed.) Proterozoic Basins of
Canada. Geological Survey of Canada Paper, 81– 10, 25 –46.
Yeo, G. M. 1986. Iron-formation in the late Proterozoic Rapitan Group,
Yukon and Northwest Territories. In: Morin, J. A. (ed.) Mineral
Deposits of Northern Cordillera. Canadian Institute of Mining and
Metallurgy Special Volume, 37, 142– 153.
Young, G. M. 1976. Iron-formation and glaciogenic rocks of the Rapitan
Group, Northwest Territories, Canada. Precambrian Research, 3,
137– 158.
Zhang, S., Jiang, G. & Han, Y. 2008. The age of the Nantuo Formation
and Nantuo glaciation in South China. Terra Nova, 20, 289–294.
Zhou, C., Bao, H., Peng, Y. & Yuan, X. 2010. Timing of deposition of
17
O-depleted barite at the Nantuo glacial meltdown in South China.
Geology, 38, 903–906.
 Chapter 6

A review of the Chemical Index of Alteration (CIA) and its application to the study
of Neoproterozoic glacial deposits and climate transitions

HEINRICH BAHLBURG1* & NICOLE DOBRZINSKI2

1
Institut für Geologie und Paläontologie, Westfälische Wilhelms-Universität, 48149 Münster, Germany
2
RWE Dea AG, Wietze Laboratory, 29323 Wietze, Germany
*Corresponding author (e-mail: hbahlburg@uni-muenster.de)

Abstract: The Chemical Index of Alteration (CIA) is the most accepted of available weathering indices. Past conditions of physical and
chemical weathering can be reliably inferred if application of the CIA is combined with a comprehensive facies analysis. When applied to
the reconstruction of climate conditions during Neoproterozoic times, CIA data provide crucial insights into the changes in the relative
contributions of chemical and physical weathering in the production of sedimentary detritus. CIA data are thus instrumental not only in
documenting changes between icehouse and greenhouse climates, but also in recognizing shorter-term climate oscillations between
glacial and warm– humid conditions. Concerning the Neoproterozoic glacial periods, sedimentological and CIA data sets give strong
evidence of a functioning hydrological cycle, operative sediment routing systems, and variable climate conditions oscillating
between dry–cool and glacial, and warm–humid and interglacial. These findings are incompatible with the hypothesis of a totally
ice-covered Snowball Earth.

Climate exerts the major control on weathering processes affecting
the upper continental crust. Direct evidence of past weathering
conditions and thus climate can be obtained from palaeosols
through a combination of, among other things, field observation,
petrography (particularly diagenetic phases), X-ray diffractometry
and whole rock and isotope geochemistry. However, palaeosols
are restricted to continental environments, which commonly
have a relatively low preservation potential. Even though palaeo-
soils as old as Palaeoproterozoic are reported in the literature
(e.g. Gutzmer & Beukes 1998; Nedachi et al. 2005), most infor-
mation on weathering conditions and thus climate has to be
gleaned from reworked siliciclastic material deposited in marine
or lacustrine environments. In the Precambrian eon, which
represents nearly 90% of Earth history, marine siliciclastic sedi-
mentary rocks are the most widely available source of information
on past climates.
In Phanerozoic rocks, information derived from siliciclastic
sedimentary rocks may be supplemented by ecological indicators
derived from fossils. However, the bearing of palaeofaunal data
on the interpretations of palaeoclimate conditions needs to be con-
sidered very carefully, because differences between continental
regimes of weathering and marine environments of deposition
may be very large. This is illustrated for example by the
pronounced difference at the coast of northern Chile between con-
ditions of extreme subtropical aridity in the Atacama desert and the
cold waters of the Humboldt current originating in the peri-
Antarctic westwind drift.
Extreme climate states are of particular interest to palaeoclimate
research. Among these, those that figure most prominently are the
icehouse climates that happened in at least five cycles of highly
variable duration and extent during Earth history, that is, the
Palaeoproterozoic and Neoproterozoic, Ordovician, Carbonifer-
ous –Permian and Neogene glaciations (Crowley & North 1991;
Crowell 1999). Of these, the Neoproterozoic glaciations are
some of the most discussed and contentious (e.g. Hoffman et al.
1998; Evans 2000; Hoffman & Schrag 2002; Eyles & Januszczak
2007; Fairchild & Kennedy 2007; Etienne et al. 2008; Eyles 2008).
The timing and causes of the inception and termination of Neopro-
terozoic glacial states as well as variations in the severity of the
icehouse conditions themselves remain unresolved issues (e.g.
Schrag et al. 2002; Baum & Crowley 2003; Fairchild &
Kennedy 2007; Shields 2008).
Unfortunately, in many cases, facies analysis of sedimentary
rocks does not supply unambiguous information on palaeoclimate
conditions. This conundrum is well illustrated by the mutually
exclusive interpretations of the Neoproterozoic, that is, Marinoan,
Ghaub Formation diamictites (Otavi Group) in northern Namibia.
These have been interpreted either as glaciogenic sediments
(Gevers 1931; Martin 1964; Hoffmann & Prave 1996; Hoffman
& Schrag 2002) or as tectonically triggered mass flow deposits
(Schermerhorn 1974; Martin et al. 1985; Eyles & Januszczak
2007). However, these issues may be moved closer to resolution
through a combination of facies analysis with considerations of
weathering as a function of climate (e.g. Young & Nesbitt 1999;
Young 2001; Scheffler et al. 2003).
In this contribution we will review the most important weather-
ing proxies applied to siliciclastic sedimentary rocks. We will
concentrate on the CIA (Nesbitt & Young 1982) and its potential
and limitations for deriving palaeoclimate information from
glacial and pre- and post-glacial siliciclastic successions. We
will preferentially but not exclusively address Neoproterozoic
glacial successions based on a combination of published infor-
mation from the Nanhuan –Sinian glacial succession in South
China (Dobrzinski et al. 2004; Dobrzinski & Bahlburg 2007)
and new data on the Port Askaig Formation (Scotland), the
Ghaub Formation (northern Namibia) and the Mortensnes and
Smalfjord formations (northern Norway).
Weathering and weathering indices
Exposed rocks are affected to variable degrees by a combination of
chemical and physical weathering (Bland & Rolls 1998). Progress-
ive chemical weathering of labile minerals like feldspar leads to
the loss of Ca2þ, Kþ and Naþ and the transformation to minerals
more stable under surface conditions (Fedo et al. 1995). Ulti-
mately, it results in the formation of shales rich in clay minerals
such as illite and kaolinite, and Fe-oxyhydrates such as goethite.
Physical weathering, in turn, leads to the degradation of rocks to
smaller grain sizes, ideally without causing geochemical and

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 81– 92. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.6
82 H. BAHLBERG & N. DOBRZINSKI
mineralogical changes. If physical weathering, and the grinding
action of moving ice-lodged debris in particular, degrades a
source rock into a clay-sized deposit, it should essentially preserve
the mineralogical and geochemical composition of the original
rock (Nesbitt & Young 1982, 1996). Consequently, the character
of the climate framework and the way it governs weathering con-
ditions and reactions is reflected by the mineralogical and mobile
element geochemical composition of the resulting deposits, and
the shales and clay-sized materials in particular.
Ultimately, chemical weathering of silicate rocks leads
through hydrolysis to an exchange of the cations Naþ, Kþ, Ca2þ
and Mg2þ for Hþ, and maybe a loss of Si4þ (Kramer 1968).
Naþ, Kþ and Ca2þ are commonly supplied by the weathering
of feldspar and volcanic glass, together accounting for c. 58% of
the exposed crust. Mg2þ is derived from glasses, sheet silicates
and mafic minerals and resides in chloritic and smectitic clays
(Nesbitt & Young 1984; Pettijohn et al. 1987). In general, hydro-
lytic weathering causes a progressive transformation of affected
components into clay minerals, ultimately kaolinite. In a qualitat-
ive way, mineralogical changes can be detected in shales and
sandstones using X-ray diffractometry. Optical analysis is a
quantitative option in sandstones only. Here, the Mineralogical
Index of Alteration (MIA), given by the ratio of quartz to the
sum of quartz þ K-feldspar þ plagioclase, permits an assessment
of weathering effects (Johnsson 1993). Quantitative estimations
of weathering in fine-grained as well as coarse-grained rocks are
relatively easily achieved by using whole-rock geochemical data
to calculate geochemical weathering proxies. A comprehensive
review of a great number of geochemical weathering indices has
been presented by Duzgoren-Aydin et al. (2002).
Recently, von Eynatten et al. (2003) and von Eynatten (2004)
presented a quantitative approach, the t-index, to statistically
model linear compositional and weathering trends. If corroborated
by further studies the t-index will likely lead to more quantitative
definitions of weathering trends. A related statistical approach has
been taken by Ohta & Arai (2007), whose W index is based on prin-
cipal component analysis of eight major oxides. As yet, it has been
applied only to igneous rocks.
Geochemical estimations of weathering effects need to be con-
sidered carefully, because the major cations Naþ, Kþ and Ca2þ
may also be mobile under diagenetic conditions (Wintsch &
Kvale 1994). Geochemical weathering proxies make use of the
changes of bulk-rock geochemical composition caused by chemi-
cal alteration. A very simple proxy is the Ruxton Ratio R (Ruxton
1968) given by the SiO2/Al2O3 ratio. This assumes that Al2O3
remains immobile during weathering, and changes in R therefore
reflect silica loss as a proxy for total element loss. The Ruxton
Ratio may be useful when weathering profiles on rocks of felsic
and intermediate composition are considered, but was found to
be poorly correlated to the actual weathering grade of a silicate
rock (Duzgoren-Aydin et al. 2002).
As the transformation of feldspar to clay minerals and the
coincident mobility of the main cations is a major process of
chemical (i.e. hydrolytical) weathering, Parker (1970) considered
it more useful to mirror changes in Naþ, Kþ, Ca2þ and Mg2þ
and created a weathering index (WIP, Weathering Index of
Parker) given by
  
Na Mg K Ca
WIP ¼ :35 þ :9 þ :25 þ :7  100 (1)
0 0 0 0
where the cations* represent the atomic percentage of an element
divided by the atomic weight. Parker (1970) also considered the
susceptibility of these elements to weathering by including in the
denominator Nicholls’ values of bond strength as a measure of
the energy necessary to break the cation-to-oxygen bonds of
the respective oxides. These different values are considered to
reflect the probability of an element being mobilized during the
weathering process. Values of WIP are commonly between
100 and 0, with the least weathered rocks having the highest
values (Fig. 6.1). The WIP implicitly assumes that all Ca2þ in a
silicate rock is contained in silicate minerals. This simplification
is a source of imprecision of the index, particularly if larger
amounts of carbonate detritus or cements are present in the rock.
More problematic still is the lack of consideration of a relatively
immobile reference phase like Al2O3 in the formula, which
would help to monitor relative changes of composition of the
relevant mineral components.
The disadvantages of the WIP are overcome in the Chemical
Index of Alteration (CIA) using whole-rock geochemical data of
major element oxides (Nesbitt & Young 1982). The index is essen-
tially based on the same considerations that led Kramer (1968) to
Fig. 6.1. Relationship between two
weathering proxies, WIP (Parker 1970) and
CIA (Nesbitt & Young 1982). Shown are
data obtained from the matrix of
Neoproterozoic glacial diamictites from
South China (Dobrzinski et al. 2004), the
Port Askaig Formation, Scotland (Panahi &
Young 1997, and this study, Table 6.3), the
Ghaub Formation of northern Namibia
(Table 6.4), and the Smalfjord and
Mortensnes formations, northern Norway
(Table 6.4). Also shown are data from
non-glacial shales and siltstones including
those underlying, intercalated with and
overlying the South China glacial
diamictites, and well-weathered Palaeozoic
shales from the Cantabrian mountain belt of
Hercynian age in northern Spain
(Table 6.1).
 A REVIEW OF THE CHEMICAL INDEX OF ALTERATION 83

Table 6.1. Sampled formations and compositional groupings discussed in this contribution

Lithostratigraphic units, Formation Region Rock types Sampled Depositional age Ref.
terminology, this paper lithology

Devonian and Vergaño, Vañes, Northern Spain Quartz Shale, siltstone Late Devonian to Wagner & Wagner-Gentis
Carboniferous Carmen, Potes sandstone, Carboniferous (1963), IGME (1984), ITGE
formations, Group, Murcia sandstone, (1994), Keller et al. (2008)
Cantabria siltstone,
shale
Mortensnes and Mortensnes, Northern Diamictite, Siltstone, shale Ediacaran, Rice & Hoffmann (2001),
Smalfjord Smalfjord Norway shale Neoproterozoic Arnaud & Eyles (2002), Rice
formations et al. (2011)
Port Askaig Port Askaig Scotland Diamictite Diamictite Cryogenian, Panahi & Young (1997),
Formation matrix: shale, Neoproterozoic Arnaud & Eyles (2006),
siltstone Benn & Prave (2006),
Arnaud & Fairchild (2011)
Ghaub diamictites, Ghaub Otavi Diamictite, Silicate shale Cryogenian, Hoffman & Schrag (2002),
Namibia Mountains, siltstone matrix, Neoproterozoic Eyles (2007), Hoffman
northern siltstone (2011)
Namibia
Non-glacial, South Doushantuo, Yangtze Carbonate, Shale Ediacaran, Dobrzinski et al. (2004),
China (post-glacial) Jinjiadong platform, shale Neoproterozoic Dobrzinski & Bahlburg
South China (2007), Zhang et al. (2011)
Glacial, South China, Nantuo, Hongjiang, Yangtze Diamictite Diamictite Cryogenian, Dobrzinski et al. (2004),
(upper diamictites) Leigongwu platform, matrix: shale, Neoproterozoic Dobrzinski & Bahlburg
South China siltstone (2007), Zhang et al. (2011)
Non-glacial, South Datanpo, Xianmeng, Yangtze Carbonate, Shale Cryogenian, Dobrzinski et al. (2004),
China (inter-glacial) Lantian platform, shale Neoproterozoic Dobrzinski & Bahlburg
South China (2007), Zhang et al. (2011)
Glacial, South China, Dongshanfeng, Yangtze Diamictite Diamictite Cryogenian, Dobrzinski et al. (2004),
(lower diamictites) Jiangkou, Tie-si-ao platform, matrix: shale, Neoproterozoic Dobrzinski & Bahlburg
South China siltstone (2007), Zhang et al. (2011)
Non-glacial, South Xieshuihe, Yangtze Sandstone, Shale Tonian and Dobrzinski et al. (2004),
China (pre-glacial) Wuqiangxi, Zitang platform, shale Cryogenian, Dobrzinski & Bahlburg
South China Neoproterozoic (2007), Zhang et al. (2011)

conclude that monitoring the hydrolysis of feldspar and volcanic Ca in phosphates may not be considered in calculation of the
glass and the respective changes in the content of the major CIA, because CIA values will increase by not more than c. 1
cations offers the best quantitative measure of chemical weather- unit if all P2O5 present in common siliciclastic rocks is assigned
ing. It represents a ratio of predominantly immobile Al2O3 to the to apatite.
mobile cations Naþ, Kþ and Ca2þ given as oxides. The CIA is The CIA has been successfully applied in a large number of
defined as studies involving glacial deposits (e.g. Young & Nesbitt 1999;
Young 2001; Condie et al. 2001; Scheffler et al. 2003; Dobrzinski
  et al. 2004; Young et al. 2004; Rieu et al. 2007) and other deposi-
Al2 O3 tional environments (e.g. Gallet et al. 1998; Aristizábal et al. 2005;
CIA ¼  100 (2) Kahmann et al. 2008).
Al2 O3 þ Na2 O þ K2 O þ CaO

where the major element oxides are given in molecular pro-
portions. CaO* represents the CaO content of silicate minerals
only (Fedo et al. 1995) and thus eliminates one of the disadvan-
tages of the WIP. Kaolinite has a CIA value of 100 and represents
the highest degree of weathering. Illite is between 75 and c. 90,
muscovite is at 75, and the feldspars at 50. Fresh basalts have
values between 30 and 45, fresh granites and granodiorites
between 45 and 55 (Nesbitt & Young 1982; Fedo et al. 1995;
Fig. 6.1).
Careful attention has to be paid to the potential presence of
clastic carbonate grains or of carbonate cement in the sedimentary
rock. If undetected but abundant, both would lead to very low and
unrealistic CIA values. The presence of carbonate grains or
cements (calcite or dolomite) and the ratio of calcite to dolomite
have to be determined petrographically. Both can be effectively
accounted for geochemically by determining total MgO and CaO
and the total inorganic carbon (TIC) content of a sample, all of
which can then be used to calculate relative contributions to
CO2, and finally CaO* (Fedo et al. 1995; Tables 6.2, 6.3, 6.4).
A brief discussion of WIP and CIA
Results of a comparison of WIP and CIA are displayed in Fig. 6.1.
It combines data on Neoproterozoic glaciomarine deposits from
South China (Dobrzinski et al. 2004) and the Port Askaig For-
mation (Scotland; Panahi & Young 1997; and our own analyses)
with new data on the Ghaub Formation (Namibia) and the
Mortensnes and Smalfjord formations of northern Norway
(Tables 6.1 to 6.4). For comparison we also show hitherto unpub-
lished results on a number of late Devonian and early Carbonifer-
ous siliciclastic shales and siltstones from the Palentinian foreland
basin (Keller et al. 2007, 2008) of the Hercynian Cantabrian
Mountain belt in northern Spain (Table 6.1). In the Late Palaeo-
zoic, the Palentinian Basin was located at near-equatorial latitudes
(Weil et al. 2001). We use the Cantabrian data as an example of
well-weathered detritus developed under non-glacial conditions.
The WIP values of the glacial Ghaub, Port Askaig, Mortensnes
and Smalfjord formations fall between 29 and 107, with a majority
between 40 and 80 (averages: Ghaub, 76; Mortensnes and
84 H. BAHLBERG & N. DOBRZINSKI
Fig. 6.2. The Yangjiaping section in South China and its Neoproterozoic
climate record indicated by CIA values. PG, preglacial units; LD, lower
diamictite; DF, Datangpo Formation; UD, upper diamictite; DSF, Doushantuo
Formation (Table 6.1). Note the initially high CIA values in the lower part of the
upper diamictite unit followed by a decrease interpreted as the initial and then
waning incorporation of chemically weathered detritus into the glacial deposit.
Source: modified from Dobrzinski et al. (2004).
Smalfjord, 56; Port Askaig, 66; Fig. 6.1). In contrast, the Cantab-
rian formations demonstrate their well-weathered character with
values between 0 and 75, with a majority below 40 (average,
19). Considering these formations, the WIP appears to differentiate
well between glacial and non-glacial weathering types.
The CIA values of the glacial units fall within 55 and 77 with a
majority below 70 (averages: Ghaub, 63; Mortensnes and Smalf-
jord, 66; Port Askaig, 70; Fig. 6.1). These are typical values of
unweathered to slightly weathered detritus and conform well to a
glacial weathering regime. The Cantabrian shales and siltstones,
in turn, reflect the dominance of warm –humid weathering
conditions with a CIA between 56 and 100 and an average of 79.
The glacial and non-glacial deposits from South China seem to
complicate the picture because they both have relatively low as
well as high WIP values typical of intense and weak chemical
weathering, respectively (Fig. 6.1). The WIP range of the glacial
deposits is 31–121 with an average of 57. The respective values
of the non-glacial formations are 2 –118 and 52. CIA values are
distributed between 50 and 85 in the glacial deposits (average,
68) and between 55 and 85 with the rare sample having a value
of close to 100 (average, 71; Fig. 6.1; Dobrzinski et al. 2004).
The values of both indices seem to demonstrate that both the
glacial and non-glacial Neoproterozoic sedimentary rocks in
South China consist of a mixture of detritus of glacial and non-
glacial weathering provenance, albeit in different proportions.
The initial incorporation of weathered older detritus into the Neo-
proterozoic glacial deposits was demonstrated by Dobrzinski et al.
(2004) by upward trends to lower CIA values in several sections in
South China (Fig. 6.2). Similar observations were made by Panahi
& Young (1997) in the Port Askaig Formation, Scotland.
K-feldspar and plagioclase are relatively common in the
Chinese glacial and non-glacial deposits. The presence of labile
feldspar indicates that the non-glacial rocks did not form predomi-
nantly under humid and warm –humid conditions. Regarding the
weathering sensitive cations Kþ and Naþ, average values for
Na2O in both the glacial and non-glacial deposits from South
China are 1.3 wt%, whereas the average value of detritus derived
from the upper continental crust is 1.2 wt% (PAAS, Post-Archaean
average Australian shale, Taylor & McLennan 1985). Condie
(1993) gives an estimate of Na2O in the average Proterozoic
shale (APS) derived from the Proterozoic upper crust of 1.1 wt%.
Average K2O values of 3.3 wt% coincide with the Proterozoic
upper crust, upper continental crust and Average Proterozoic
shale values of 3.3, 3.4 and 3.6 wt%, respectively (Condie 1993;
McLennan 2001). Considering the calculations of the respective
indices, an increase in, for example, Na2O of 0.5 wt% causes a
decrease in the CIA of c. 2.5 units, and an increase of WIP of
5 units. A decrease by 0.5 wt% results in similar, but opposing,
changes in magnitude. In the case of K2O the effect is similar;
however, the change in CIA units is c.1.5 per 0.5 wt% change.
The respective changes in WIP are c. + 4 units.
Our data demonstrate that both the WIP and CIA appear to be
equally sensitive to small changes in the concentrations of the
major cations, and thus, by inference, of dominant warm –humid
or dry –cold weathering conditions. Disadvantages of the WIP
include the lack of consideration of mobile relative to immobile
phases, and the fact that it is based on whole-rock CaO and not
on CaO* of the silicate fraction only. Another advantage of the
CIA lies in the way it permits the definition and prediction of
weathering trends of silicate rocks (Nesbitt & Young 1982,
1984). Data and trends can then be displayed well in A – CN– K
(Al2O3 – CaO* þ Na2O – K2O) ternary diagrams (Fig. 6.3;
Nesbitt & Young 1984). The combination of these features
makes the CIA the presently preferred weathering index.
Provenance and reconstruction of original compositions
Provenance pathways and the distribution of lithologies in source
regions exert a first-order control on the composition of siliciclastic
deposits (Johnsson 1993). Differences in provenance are therefore
reflected also in the weathering indices (Fedo et al. 1995). In the
ternary A –CN –K diagram of Figure 6.3, the provenance compo-
sitions and weathering trends can be depicted and predicted.
Hydrolytic weathering leads to a loss of Naþ and Ca2þ and
changes rock compositions towards the A apex and ever higher
CIA values.
Transport sorting exerts another major influence on the compo-
sition of clastic sediments. Along a river and transport path, labile
mineral grains will be comminuted preferentially. In the presence
of chemical weathering, labile grains like feldspar will be prone to
progressive decay along cleavages, cracks and other zones of
lattice weakness. Feldspars will eventually be transformed into
clay minerals and a marked compositional difference, in addition
to the comminution of mineral grains, between source and sedi-
ment will become evident (Johnsson et al. 1988). Sorting will
thus lead to a higher proportion of clay minerals downriver and
consequently higher CIA values (Nesbitt et al. 1997; Fig. 6.3).
Thus, when CIA values of different successions are compared,
similar grain sizes need to be considered.
Only exceptional circumstances permit the transfer of labile
mineral grains from source to sink without any mineralogical
 A REVIEW OF THE CHEMICAL INDEX OF ALTERATION 85

Fig. 6.3. Major oxides in molecular ratios with compositions of typical magmatic source rock types (Fedo et al. 1995) and average compositions of the UCC
through time (UCC, McLennan 2001; PAAS, Taylor & McLennan 1985; all others, Condie 1993). Note that the lower part of the diagram with A , 40 is not shown. 1,
gabbro; 2, tonalite; 3, granodiorite; 4, granite; 5, A-type granite; 6, charnokite; 7, potassic granite. Ideal weathering trends of UCC-type source lithologies would be
parallel to the predicted weathering trend (Nesbitt & Young 1984). The statistically modelled t-index weathering trend (von Eynatten 2004) is based on data obtained from
the world’s major rivers and erosional products of major denudation areas (McLennan 1993). Trends of K-metasomatism and its graphical correction to pre-metasomatic
values on predicted weathering trends is indicated by grey arrows originating or pointing at the K apex of the diagram (Fedo et al. 1995). Alteration trend of
Neoproterozoic glacial successions estimated from the distribution of samples in Fig. 6.4. The dotted double-headed arrow outlines the effect of grain sorting on the
chemical composition of sediments and consequently on CIA values; c, coarse, sand; f, fine, clay (Nesbitt et al. 1997).

and thus compositional changes, that is, without the influence of
any chemical weathering. In such an ideal case transport will
lead only to comminution of mineral grains; the final fine-grained
deposit will have the same composition as the coarser source
(Nesbitt & Young 1996), and CIA values will be unaffected.
Diagenetic addition of potassium to a deposit during illitization
of kaolinite and the replacement of plagioclase or illite by
K-feldspar, the so called K-metasomatic effect of Fedo et al.
(1995), causes weathered compositions to plot closer to the K
apex (Fig. 6.3), thus resulting in diagenetically lowered CIA
values. This can be checked by comparisons with average upper
crustal compositions (see above) and by comparing CIA values
with K/Cs ratios (McLennan et al. 1993); a K-metasomatic
effect would disturb the pronounced negative correlation
between the K/Cs ratio and weathering intensity. The distribution
of data in the A –CN –K diagram (Fig. 6.4) indicates that in
particular some of the Chinese sedimentary rocks discussed in
this paper show a weak K-metasomatic effect (Dobrzinski et al.
2004). If projected back to the predicted weathering trend origi-
nating from average granodiorite or the upper continental crust
(UCC) (Fig. 6.3), the originally higher CIA values of these
samples can be reconstructed (Tables 6.3 and 6.4), thus permitting
the subsequent inference of original provenance compositions
(Figs 6.3 and 6.4; Fedo et al. 1995).
As the CIA is founded essentially in considerations of the mobi-
lity of the major cations, geochemical estimations of provenance

Fig. 6.4. A–CN– K (Al2O3 –CaO*þ

Na2O– K2O) diagram with data of
Neoproterozoic glacial units and reference
units and their composition (PAAS, Taylor
& McLennan 1985; UCC, McLennan 2001;
all others, Condie 1993). Note that the lower
part of the diagram with A , 40 is not
shown. The left side of the figure shows the
range and averages (black dots) of CIA
values of the Neoproterozoic units corrected
for K-metasomatism according to Fedo
et al. (1995). For comparison, the figure also
shows data obtained from shales of several
late Devonian and Carboniferous
formations from the Palentinian foreland
basin of the Hercynian Cantabrian mountain
belt in northern Spain (Keller et al. 2008),
which conform to the predicted weathering
trend (Tables 6.1, 6.2; Nesbitt & Young
1984).
86 H. BAHLBERG & N. DOBRZINSKI
should be combined preferably with ratios of immobile elements
including the high field strength elements La, Th, Sc and Zr
(Bhatia & Crook 1986; McLennan et al. 1993; Bahlburg 1998).
All the Neoproterozoic sedimentary rocks considered in this paper
have Zr/Sc and Th/Sc ratios similar to or higher than the Upper
Continental Crust (UCC) and PAAS (Fig. 6.5; Taylor & McLen-
nan 1985; McLennan 2001). The rock compositions all cluster
around the composition of the APS (Fig. 6.5; Condie 1993) as a
proxy for material derived from the average Proterozoic upper
crust. This provides another indication that processes beyond orig-
inal weathering and recycling did not significantly alter the compo-
sition of the studied Neoproterozoic sedimentary rocks.
Weathering trends
Thermodynamic, kinetic, experimental and observational evi-
dence define weathering trends for silicate rocks that are (sub)par-
allel to the CN– A join of the A – CN– K diagram (Nesbitt & Young
1982, 1984). Weathering of average granodiorite or UCC along the
predicted weathering trend (Fig. 6.3) will result in the transform-
ation of labile components including the feldspars first to illite,
thus causing the sample’s composition to plot ever closer to the
A –K join and the illite composition in A –CN –K space (Nesbitt
& Young 1984; Fig. 6.3). Increasing formation of kaolinite
during progressive weathering will curve the weathering trend
towards the A apex when approaching the A – K join (Nesbitt &
Young 1984) in a way similar to the calculated weathering trend
of von Eynatten et al. (2004; Fig. 6.3). The validity of this
general weathering behaviour and trend is substantiated by the
fact that the compositions of shale composites like PAAS and
APS as averaged samples of the weathered and recycled upper
crust plot very near the predicted weathering trend (Fig. 6.3).
Also, the Late Palaeozoic reference samples from Cantabria,
Spain, which were well weathered under non-glacial conditions,
plot along this trend as predicted by CIA systematics (Figs 6.3 &
6.4; Nesbitt & Young 1982, 1984).
In many cases, sedimentary rocks do not plot on the predicted
weathering trend but below it, thus reflecting an increase in Kþ
in the samples (e.g. Panahi & Young 1997; Rieu et al. 2007). A
correction of this K-metasomatic effect according to Fedo et al.
(1995) of those CIA data points below the predicted weathering
trend line results in an increase of CIA values (Figs 6.3, 6.4;
Tables 6.3, 6.4). This increase depends on the compositional
difference between the sample and the predicted original value
and reflects the degree of K-metasomatism having affected a
sample’s detritus if the original composition was granodioritic
and close to that of the UCC (Fig. 6.3; Fedo et al. 1995). If the orig-
inal composition was richer in Kþ than granodiorite or UCC, the
predicted weathering trend would shift to the right but remain
Fig. 6.5. Zr/Sc v. Th/Sc diagram of
McLennan et al. (1993) demonstrating that
all studied sedimentary rocks have an upper
crustal composition and were affected to
minor degrees by recycling, which is
indicated by high Zr/Sc values. This effect
is most prominently shown in some of the
Late Palaeozoic sedimentary rocks from
Cantabria. It is noteworthy that all the
Neoproterozoic samples cluster around the
composition of the average Proterozoic
shale (APS, Condie 1993). Reference units
and their composition: PAAS, Taylor &
McLennan (1985); UCC, McLennan
(2001); all others, Condie (1993).
(sub)parallel to the predicted weathering trend (Nesbitt & Young
1984). Corrections for K-metasomatism and resulting increases
in CIA values would be correspondingly smaller. A correction to
the predicted weathering trend thus represents a maximum correc-
tion if average granodiorite or UCC is the starting point.
The average compositions of post Archaean shale and Protero-
zoic shale (PAAS, Taylor & McLennan 1985; APS, Condie
1993) plot below the predicted weathering trend towards the K
apex of the A –CN –K diagram. This may be due either to starting
compositions richer in Kþ than UCC and the average Proterozoic
upper crust (Fig. 6.3), or a mild K-metasomatic effect including the
conversion under increased temperature of illite to secondary
K-feldspar (Nesbitt & Young 1989; Fedo et al. 1995).
The composition of almost all considered Neoproterozoic units
and samples appears to systematically deviate from the predicted
weathering trend, in a way here preliminarily called the observed
Neoproterozoic alteration trend (Fig. 6.3). In discussing this
trend, two things have to be borne in mind: the similarity of the
estimates for the UCC and the average Proterozoic crust in A –
CN –K space, and the fact that glaciogenic deposits commonly rep-
resent averaged samples of the available upper crust exposed
to denudation.
It is a reasonable assumption that the original average geochem-
ical composition of the silicate sources feeding the Neoproterozoic
glaciogenic deposits was similar to the average upper crustal com-
positions. Repeated weathering and recycling processes may lead
to a stepwise increase of Kþ in the resulting deposits, as indicated
by the offsets of the average shale compositions from the predicted
weathering trend (Fig. 6.3). The observed Neoproterozoic altera-
tion trend may consequently reflect the degree to which unweath-
ered and weathered crust and recycled sedimentary rocks have
been mixed and incorporated into the glaciogenic deposits. The
correction of CIA values for a K-metasomatic effect according
to Fedo et al. (1995) will thus reconstitute CIA values to the pre-
dicted weathering trend originating in fresh upper crust, which
may, however, be unrepresentative of the actually eroded conti-
nental surface. If CIA values were referred back only to unweath-
ered compositions, irrespective of the regional context and the
recycling history, estimates of weathering effects during the last
sedimentary cycle will be exaggerated and may consequently
be misleading.
The evidence of glaciomarine deposits
Successions of modern and ancient glacial deposits and of glacially
derived detritus are ideally characterized by low CIA values
reflecting the dominance of physical over chemical weathering
processes (Nesbitt & Young 1982; Young 2001). A typical
example is the Palaeoproterozoic Gowganda Formation (Young
 A REVIEW OF THE CHEMICAL INDEX OF ALTERATION 87

Table 6.2. Major element whole-rock compositions and weathering indices of shales

Sample* SiO2 Al2O3 Fe2O3 MgO CaO‡ Na2O K2O TiO2 P2O5 MnO LOI Total TC TIC WIP CIA†

BT Nr.64 96.83 1.85 0.60 0.04 – – – 0.23 0.08 0.01 0.42 100.06 0 100
MB 87-K 95.33 2.40 0.61 0.02 – – 0.04 0.23 0.04 0.01 0.72 99.41 0 98
MB P-05 87.40 3.54 0.83 0.12 2.49 0.07 0.41 0.20 0.07 0.05 3.02 98.21 0.72 0.54 11 86
MB P-25 83.00 3.47 2.19 0.29 0.61 0.03 0.73 0.09 0.03 0.03 2.20 93.28 9 64
MB P-16 60.87 19.13 5.87 0.60 3.14 0.60 2.68 1.02 0.09 0.07 6.79 100.87 0.94 0.45 38 83
BT PO 28 84.47 6.02 2.79 0.66 1.40 0.15 1.08 0.62 0.08 0.03 1.57 100.28 16 60
BT PO 29 92.10 3.78 0.82 0.14 1.02 0.21 0.76 0.11 0.02 0.03 1.57 101.58 11 56
PO 1 75.34 8.04 7.37 1.04 0.55 0.03 1.06 0.52 0.07 0.18 4.30 98.50 0.74 0.58 14 87
PO 2-1 79.79 9.72 3.36 0.62 0.05 0.10 1.45 0.56 0.04 0.01 2.55 98.30 15 84
PO 3 88.09 7.16 1.74 0.15 – – 0.75 0.53 0.02 0.00 1.62 100.06 7 90
PO 4-2 74.77 9.96 6.38 0.42 0.92 0.40 1.23 0.58 0.09 0.15 3.67 98.56 0.30 0.16 18 83
PO 4-4 67.92 6.79 9.36 0.87 4.40 0.34 1.43 0.40 0.07 0.28 6.74 98.59 1.31 1.21 29 76
PO 5 37.88 3.92 4.89 0.45 27.10 0.03 0.49 0.19 0.07 0.09 23.38 98.49 6.04 5.92 75 87
PO 6 45.19 2.17 1.41 0.34 26.81 0.02 0.16 0.11 0.14 0.12 21.91 98.38 5.86 5.76 71 91
PO 9 54.00 3.36 2.81 0.39 19.65 0.12 0.34 0.16 0.16 0.17 16.83 97.99 4.42 4.33 55 86
PO 11 84.53 6.15 3.20 0.18 0.34 0.05 0.93 0.46 0.09 0.03 2.04 98.35 10 78
BT PO 4 71.88 6.79 4.90 1.22 4.29 0.25 1.19 0.37 0.08 0.08 6.91 97.97 1.80 1.59 27 80
BT PO 5 68.46 6.44 5.28 1.15 6.87 0.08 0.94 0.35 0.08 0.08 9.26 98.99 2.40 2.24 29 85
BT PO 16 72.33 7.15 8.37 1.27 2.11 0.21 1.04 0.46 0.20 0.13 7.16 100.44 1.83 1.22 20 83
BT PO 30/2 68.44 12.29 5.23 1.17 2.68 0.31 1.72 0.70 0.09 0.06 5.60 98.30 1.21 0.75 28 84
FC 1 73.68 2.87 7.31 0.50 6.40 0.15 0.55 0.21 0.14 0.24 7.90 99.98 1.92 1.81 24 77
FC 2 77.89 3.83 10.22 0.80 0.40 0.15 0.53 0.24 0.09 0.21 5.60 99.97 9 71
FC 3 87.51 4.35 4.40 0.43 0.07 0.20 0.54 0.33 0.07 0.06 2.20 100.18 8 81
FC 5 74.16 6.86 11.27 0.57 0.13 0.40 1.20 0.50 0.14 0.23 4.60 100.10 16 76
CRM 1A 85.36 6.00 3.72 0.28 0.38 0.04 1.21 0.41 0.10 0.05 2.10 99.67 12 74
CRM 2 84.48 4.90 3.58 0.30 1.78 0.05 1.03 0.34 0.08 0.06 3.10 99.72 15 52
CRM 3 84.32 6.80 3.88 0.17 0.11 0.40 1.05 0.41 0.11 0.06 2.40 99.74 13 77
CM 1A 79.37 3.11 9.94 0.86 0.64 0.11 0.36 0.21 0.06 0.20 5.00 99.87 8 64
CM 3A 84.89 3.25 6.37 0.60 0.22 0.12 0.36 0.25 0.04 0.16 3.60 99.87 6 77
CM 3B 90.78 3.82 2.13 0.32 0.04 0.22 0.47 0.25 0.03 0.01 2.00 100.08 7 80
CM 4A 89.18 4.10 3.27 0.44 0.04 0.23 0.51 0.30 0.05 0.01 1.40 99.55 8 80
CM 4A_2 89.09 4.16 3.43 0.45 0.05 0.23 0.51 0.31 0.06 0.01 1.40 99.71 8 80
CM 5A 76.72 5.54 9.19 1.02 0.34 0.36 0.81 0.61 0.08 0.33 4.80 99.82 14 73
VN 1 79.31 10.69 3.19 0.82 0.06 0.51 1.65 0.65 0.08 0.02 2.50 99.52 21 80
VN 2 79.70 10.79 2.82 0.81 0.06 0.52 1.76 0.66 0.10 0.01 2.50 99.77 22 79
VN 3A 81.63 9.43 2.84 0.79 0.07 0.55 1.50 0.59 0.09 0.02 2.10 99.64 20 78
VN 3B 82.34 8.93 2.78 0.75 0.06 0.52 1.50 0.55 0.10 0.01 2.30 99.87 20 78
VN 4 79.00 11.45 2.83 0.82 0.04 0.48 2.00 0.66 0.07 0.01 2.60 100.00 24 79
VN 5 81.63 9.40 2.62 0.75 0.11 0.42 1.44 0.81 0.14 0.01 2.40 99.77 18 79

*Formations, late Devonian– Carboniferous, of the Cantabrian Mountains, northern Spain: Vergaño, VN; Vañes, CM; Carmen, FC, CRM; Potes Group, PO, BT PO; Murcia,
BT, MB. Major elements were determined by ICP-ES after a LiBO2 fusion at ACME Analytical Labs, Vancouver, Canada.
†
CIA has been calculated according to Nesbitt and Young (1982) and Fedo et al. (1995). Ca in phosphates has not been considered in the calculation of the CIA, because CIA
values increase by only c.1 unit if all P2O5 is assigned to apatite.
‡
Calculation of CaO* is based on values of TC (total carbon) and TIC (total inorganic carbon) obtained by analysis using CS-Mat 5500, in the Department of Geology and
Paleontology, Münster University. Only samples with increased values of CaO were selected. Those with low values did not show a reaction with HCl (10%); dolomite
was not observed in thin section. CaO was consequently considered to be equal to CaO*.
–, below detection limit; blank spaces, not measured.

& Nesbitt 1999). Within this unit, rocks of diamictite facies com-
monly have the lowest CIA values (between 50 and 70). Diamic-
tites in general represent either lodgement tills or moraine
material that was rapidly redeposited by mass wasting in glacio-
marine environments (e.g. Scheffler et al. 2003; Dobrzinski et al.
2004; Young et al. 2004; Rieu et al. 2007; Fig. 6.4). Such deposits
with low CIA values are frequently associated with finer-grained
and laminated glaciomarine sandstones, siltstones and shales,
which may include ice-rafted debris, with higher CIA values
between 70 and 85 (Fig. 6.4; Panahi & Young 1997; Dobrzinski
et al. 2004; Young et al. 2004; Rieu et al. 2007).
Lower CIA values in glaciomarine deposits can be caused by
several factors acting alone or in concert. K-metasomatism in con-
nection with the conversion of illite to K-feldspar, or the illitization
of kaolinite, is the most commonly cited cause (Nesbitt & Young
1989; Fedo et al. 1995). However, the grain-size of the analysed
rock is also an important factor. Sands formed from chemically
pre-weathered rocks and soils generally have lower CIA values
and plot lower in the A –CN –K triangle than associated muds,
because feldspars and other labile minerals tend to be concentrated
in the coarser-grained fraction (Nesbitt et al. 1996; Fig. 6.3).
Furthermore, sorting or mixing of known and cryptic sources
may result in an unexpected change of the amount of feldspar in
a deposit.
A sedimentary enrichment of K-feldspar relative to the assumed
source rock causes the distribution of data in the A –CN –K
diagram to mimic a K-metasomatic effect. In such a case the pre-
dicted weathering trend would shift to the right and point more
closely to a muscovite rather than an illite composition on the
A –K join of the diagram, or a CIA of 80 instead of 90 (Fig. 6.4;
Nesbitt et al. 1997). A correction of the CIA data for an alleged
K-metasomatism according to Fedo et al. (1995; Figs 6.3 and
6.4) would in this case lead to an overadjustment. A cause of
increased or high CIA values in glaciomarine rocks may also be
88 H. BAHLBERG & N. DOBRZINSKI

Table 6.3. CIA and WIP values of the Neoproterozoic Port Askaig Formation, the incorporation of older, strongly weathered material into the
Scotland, based on the data of Panahi & Young (1997) and our own analyses glacial deposits (Nesbitt & Young 1997; Dobrzinski et al. 2004;
Fig. 6.2).
Sample TC TIC WIP CIA* CIAcorr† Many Neoproterozoic glaciomarine successions are character-
ized by a deviation of the observed weathering trend from the pre-
94-102 0.58 0.45 33 68 84 dicted one and towards the K apex (Figs 6.3 and 6.4), including the
94-106 0.68 0.44 56 65 73 lower and upper diamictites of the Yangtze platform in South
94-108 0.90 0.62 61 64 72 China (Table 6.1; Dobrzinski et al. 2004), the Port Askaig For-
94-78 4.14 3.39 75 74 83 mation (Nesbitt & Young 1997; Tables 6.1 and 6.3), the Ghaub
94-79 4.52 4.17 72 74 86 Formation of northern Namibia, the Norwegian Smalfjord and
94-87 1.95 1.23 57 63 72 Mortensnes formations (Tables 6.1 and 6.4), and the Fiq Formation
94-97 0.56 0.45 36 64 71
in Oman (Rieu et al. 2007). The same effect is also evident in the
95-62 3.39 2.85 69 77 87
glacial Ordovician Table Mountain Group in South Africa (Young
95-66 5.80 5.00 84 74 87
et al. 2004).
95-67 5.11 4.75 75 72 87
95-69 3.37 2.91 75 70 86
Regarding the Port Askaig and Fiq formations, the Chinese
95-70 3.92 3.37 70 68 78 diamictites and the Table Mountain Group, K-metasomatism has
95-71 6.19 4.90 85 71 87 in fact been cited as the cause of the deviation from the commonly
95-73 4.93 4.04 79 71 84 used predicted weathering trend (Nesbitt & Young 1997;
95-76 5.21 4.25 78 75 83 Dobrzinski et al. 2004; Young et al. 2004; Rieu et al. 2007).
95-79 3.57 2.93 64 73 87 However, almost all the mentioned cases include analyses of
95-80 5.03 4.66 80 70 81 matrix material of glaciomarine diamictites and associated coarse-
95-83 1.69 1.32 53 66 74 grained strata. Some of the Neoproterozoic diamictites, including
95-84 1.56 0.51 56 61 66 the ones from South China (Dobrzinski & Bahlburg 2007), have an
impure shaley matrix rich in silt or fine sand. The presence of this
*Ca in phosphates has not been considered in the calculation of the CIA, because coarser material in the matrix of the diamictites may have contrib-
CIA values increase by only c.1 unit if all P2O5 is assigned to apatite. CIA has uted consequently to a deviation of the observed weathering trends
been calculated according to Nesbitt & Young (1982) and Fedo et al. (1995). from the predicted one.
†
CIAcorr: CIA value corrected to the predicted weathering trend according to Fedo A combination of grain-size effects and the variable incorpor-
et al. (1995). ation of older weathered material into the Neoproterozoic glacio-
Calculation of CaO* is based on values of TC (total carbon) and TIC (total marine deposits may account for the measured CIA values
inorganic carbon) obtained by analysis using CS-Mat 5500, in the (Fig. 6.4; Tables 6.2 to 6.4). In view of the corrected CIA values
Department of Geology and Paleontology, Münster University.

Table 6.4. Major element whole-rock compositions and weathering indices

Sample*† SiO2 Al2O3 Fe2O3 MgO CaO} Na2O K2O TiO2 P2O5 MnO LOI Total TC TIC WIP CIA§ CIAcorrk

GF 1 70.77 6.66 15.52 0.4 0.05 0.05 3.17 1.24 0.12 0.01 1.50 99.50 0.03 –‡ 29 65 87
GF 2 81.78 8.45 2.31 0.42 0.01 0.1 5.22 0.43 0.06 0.01 1.20 99.98 0.01 – 46 59 87
GF 3 72.25 7.50 10.25 0.95 0.47 0.09 4.79 0.34 0.35 0.04 2.70 99.73 0.06 0.01 45 55 79
GF 4 28.42 6.46 2.82 10.93 19.89 0.86 2.14 0.40 0.12 0.06 27.80 99.91 7.48 7.26 107 63 72
GF 5 38.51 7.33 3.44 9.29 15.52 0.83 2.55 0.43 0.14 0.06 21.70 99.80 5.66 5.34 94 64 74
GF 7 27.84 4.28 3.20 12.98 19.08 0.22 1.79 0.24 0.10 0.08 30.00 99.82 7.91 7.65 102 65 81
GF 8 59.81 16.53 7.32 3.15 0.70 0.74 5.36 0.87 0.18 0.11 5.00 99.80 0.23 0.09 63 69 80
GF 9 33.92 7.96 4.97 10.47 15.03 0.91 2.82 0.47 0.14 0.14 23.00 99.83 6.09 5.62 99 64 74
GF 11 71.33 13.72 3.68 1.11 0.56 2.20 4.93 0.44 0.14 0.03 1.70 99.84 0.07 0.06 67 59 68
GF 12 73.82 12.63 3.39 0.96 0.31 2.06 4.49 0.42 0.12 0.05 1.60 99.86 0.03 0.02 61 59 68
GF 13 7.93 0.84 0.48 19.65 27.73 0.02 0.29 0.06 0.07 0.60 42.30 99.98 11.13 10.92 127 71 85
S1 73.68 10.67 3.59 1.89 1.42 1.45 2.82 0.5 0.12 0.03 3.20 99.38 0.36 0.27 46 65 70
S2 73.22 11.95 4.10 1.72 0.41 1.03 3.66 0.65 0.11 0.02 2.50 99.38 0.08 – 46 65 73
S3 67.36 13.36 5.94 2.94 0.27 1.41 3.91 0.83 0.17 0.10 3.10 99.40 0.09 0.04 55 67 74
S4 59.38 11.26 4.65 4.85 4.74 1.37 3.48 0.79 0.16 0.06 8.80 99.55 1.88 1.56 68 65 73
S5 69.49 12.75 3.91 2.33 1.48 1.91 3.39 0.61 0.11 0.03 3.40 99.42 0.46 0.32 57 65 70
S6 65.02 14.83 5.73 2.6 1.05 1.86 4.00 0.70 0.13 0.04 3.50 99.48 0.24 0.17 61 65 69
M1 64.35 15.52 6.23 3.13 0.33 2.05 3.58 0.78 0.14 0.05 3.30 99.48 0.06 0.06 59 68 72
M2 64.14 15.93 5.49 2.67 0.89 2.36 3.76 0.82 0.15 0.04 3.20 99.46 0.16 0.14 63 65 69
M3 69.19 10.5 3.88 3.05 2.98 1.45 2.45 0.51 0.11 0.05 5.20 99.38 0.95 0.73 50 68 73
M4 67.96 11.91 4.44 2.88 2.35 1.96 2.42 0.58 0.12 0.06 4.70 99.40 0.78 0.56 53 67 69
M5 65.56 11.05 5.19 3.01 3.43 1.85 2.13 0.54 0.10 0.10 6.20 99.17 1.11 0.79 52 67 69
M6 63.64 14.2 5.79 3.39 2.00 2.45 2.61 0.69 0.13 0.10 4.40 99.43 0.46 0.29 59 64 64

*Neoproterozoic Ghaub Formation (GF), northern Namibia, and the Smalfjord (S) and Mortensnes (M) formations, northern Norway. Samples of the Ghaub Formation
diamictites were taken on the Ghaub and Jakkalumuramba farms NE of the town of Otavi. Analyses represent matrix compositions.
†
Major elements were determined by ICP-ES after a LiBO2 fusion at ACME Analytical Labs, Vancouver, Canada.
–below detection limit.
§
CIA has been calculated according to Nesbitt & Young (1982) and Fedo et al. (1995). Ca in phosphates has not been considered in the calculation of the CIA, because CIA
values increase by only c.1 unit if all P2O5 is assigned to apatite.
k
CIAcorr: CIA value corrected to the predicted weathering trend according to Fedo et al. (1995).
}
Calculation of CaO* is based on values of TC (total carbon) and TIC (total inorganic carbon) obtained by analysis using CS-Mat 5500, in the Department of Geology and
Paleontology, Münster University.
 A REVIEW OF THE CHEMICAL INDEX OF ALTERATION
Fig. 6.6. Variations in the CIA with stratigraphic height in the Fiq Formation,
Oman. MB Fm., Masirah Bay Formation. Modified from Rieu et al. (2007).
shown in Figure 6.4 and the discussion in the preceding para-
graphs, we conclude that in particular the high values are overcor-
rected maximum values very likely exaggerating the magnitude of
changes due to chemical weathering.
Implications for the study of Neoproterozoic climate change
The origin of Neoproterozoic diamictite successions was debated
long before the ‘Snowball Earth’ hypothesis was proposed
(Kirschvink 1992; Hoffman et al. 1998; Hoffman & Schrag
2002). Classic examples like the Ghaub diamictites in northern
Namibia were repeatedly interpreted either as tillites and glacio-
marine sediments (e.g. Gevers 1931; Martin 1964; Hoffmann &
Prave 1996; Hoffman & Schrag 2002) or as tectonically triggered
mass flow deposits (e.g. Schermerhorn 1974; Martin et al. 1985;
Eyles & Januszczak 2007). Furthermore, there are also cases
where alleged glacial deposits were proven to be in fact of deposi-
tional origin unrelated to ice action and cold climates. Examples
include purported tillites of Tremadoc age in the Andes of north-
western Argentina (Keidel 1943), which were demonstrated to
be coastal conglomerates (Bahlburg 1990), and conglomeratic
deposits in northern Chile allegedly linked to the Late Palaeozoic
Gondwana glaciation (Cecioni 1979, 1981), which were shown to
be of mass-flow origin in turbidite complexes (Charrier 1986;
Bahlburg & Breitkreuz 1993).
It is usually necessary to support one line of evidence with at
least a second, independent one in order to have confidence that
an interpretation is correct. This applies not only to controversial
sedimentological cases. The analysis of the CIA is a powerful
tool in studies of the palaeoclimatic record preserved in siliciclas-
tic sedimentary successions. However, as discussed previously,
numerous factors can influence the major element geochemical
composition of a siliciclastic rock. Thus, CIA values alone are
89
almost meaningless as climate indicators when not considered in
the context of the stratigraphic framework and facies of the ana-
lysed sedimentary rock. This is demonstrated very well in a
study of the Fiq Formation in Oman (Rieu et al. 2007; Fig. 6.6),
an association of alternating glacial diamictites, debris-flow depos-
its, turbidite sandstones, hemipelagic shales, and wave-rippled
shoreface deposits. The facies changes are connected to coincident
changes between low CIA values of 62 –70 for the glacial diamic-
tite matrix, and high ones between 80 and 85 (corrected after Fedo
et al. 1995) for shales, siltstones and fine-grained sandstones.
Facies changes and CIA data give evidence of variations in the
intensity of chemical weathering as a function of climate oscil-
lations between glacial and warm –humid interglacial conditions
(Rieu et al. 2007). Scheffler et al. (2003) presented similar data
in a study of the waning phase of the Late Palaeozoic glaciation
in South Africa recorded in the Dwyka Group. In the Neoprotero-
zoic as well as in the Late Palaeozoic, these climate oscillations
together with the facies of the respective deposits give evidence
of a functioning hydrological cycle and sediment dispersal
systems. In the Neoproterozoic case these data are in direct oppo-
sition to assumptions of a totally ice-covered Snowball Earth as
proposed by Hoffman & Schrag (2002).
These CIA values are compatible with a prominent influence of
physical weathering on the production of the diamictite detrital
silicate matrix. Together with sedimentological features including
the presence of dropstones in laminated facies, these values
strengthen interpretations of a glacially connected origin of the
Ghaub diamictites.
The interpretation of the Neoproterozoic Smalfjord and Mor-
tensnes formations of northern Norway is less controversial and
centres around the question whether the strata represent redepos-
ited detritus originally produced in a glacial environment
(Arnaud & Eyles 2002) or deposits of direct glacial action
(Edwards 2004; Rice 2004). Application of the CIA cannot solve
these differences in a fundamentally sedimentological contro-
versy. The uncorrected CIA values of the Smalfjord and Mor-
tensnes formations (Table 6.4) are uniformly low between 65
and 68, and between 64 and 74 (averages of 72 and 69, respect-
ively) if corrected according to Fedo et al. (1995; Fig. 6.4). It is
thus legitimate to infer at least a glacial origin for the detritus.
The Neoproterozoic lower and upper glacial diamictites of the
Yangtze platform represent associations of lodgement till, subgla-
cial melt-out deposits of grounded glaciers, sediments of concen-
trated density flows and glacial outwash, and turbidites and
laminated deposits containing ice-rafted debris (Dobrzinski &
Bahlburg, 2007). CIA values of both diamictite units are
between 50 and 90 (Fig. 6.4; Table 6.1; Dobrzinski et al. 2004).
Correction of values following Fedo et al. (1995) does not
change their range but leads to a shuffling of values within this
range. The corrected average value is 72 (Fig. 6.4). The combi-
nation of low values in the matrix of glacial diamictites and
higher values in associated glacial facies is interpreted as a result
of reworking of older, weathered material into the glacial deposits
(Dobrzinski et al. 2004). This is evident in some sections showing
a decrease in CIA values upsection (Fig. 6.2).
The non-glacial units of the China platform consisting of
preglacial siliciclastics, the shales and carbonates of the Datangpo
Formation intercalated between the glacial units, and the post-
glacial carbonates and shales of the Doushantuo Formation
(Table 6.1) have CIA values ranging between 59 and 97 with
abundant higher values. The corrected ranges are rather similar,
with a corrected average of 76 (Fig. 6.4). The higher values of
the non-glacial deposits relative to the diamictite successions are
taken to reflect more humid and warmer weathering and climate
conditions. Particularly noteworthy is the presence of K-feldspar
and plagioclase in the silicate fraction contained in the Marinoan
cap carbonates in South China indicating the minor influence
of chemical weathering on the silicate detritus in these strata
(Dobrzinski 2005).
90 H. BAHLBERG & N. DOBRZINSKI
Conclusions
The CIA (Nesbitt & Young 1982) is the most widely applied and
most indicative of the available weathering indices. When the
intricacies of the weathering systems and of applying the index
are appropriately considered (Fedo et al. 1995; Nesbitt & Young
1996; Nesbitt et al. 1997; Young 2001; and discussion of
Fig. 6.4), the index is a very valuable tool in the assessment of
past climate change as recorded by siliciclastic sedimentary
rocks, with one critical caveat: it needs to be applied in conjunction
with a comprehensive facies analysis. Concerning the Neoproter-
ozoic glacial periods, the combination of both data sets gives
strong evidence of (i) a functioning hydrological cycle, (ii) operat-
ive sediment routing systems, and (iii) variable climate conditions
oscillating between dry –cool and glacial, and warm –humid and
interglacial. These findings are incompatible with the hypothesis
of a totally ice-covered Snowball Earth.
H. Rice (Vienna, Austria) kindly supplied samples of the Smalfjord and Mor-
tensnes formations of northern Norway. We thank G. Young (London, Ontario,
Canada) for sending us sample powders of the Panahi & Young (1997) samples
of the Port Askaig Formation. This study was supported by the German Research
Foundation DFG (grants Ba 1011/23-1,2,3). This paper is a contribution to
IGCP project 512 ‘Neoproterozoic Ice Ages’. We thank A.R. Prave
(St. Andrews, Scotland), C. Augustsson and C. Reimann (Münster, Germany),
for commenting on earlier versions of the manuscript. We appreciate the very
constructive reviews by Hilmar v. Eynatten (Göttingen, Germany) and an anon-
ymous reviewer. B. Fister (Münster) kindly re-drafted Figure 6.6.
References
Aristizábal, E., Roser, B. & Yokota, S. 2005. Tropical chemical weath-
ering of hillslope deposits and bedrock source in the Aburrá Valley,
northern Colombian Andes. Engineering Geology, 81, 389– 406.
Arnaud, E. & Eyles, C. 2002. Glacial influence on Neoproterozoic sedi-
mentation: the Smalfjord Formation, northern Norway. Sedimentol-
ogy, 49, 765– 788.
Arnaud, E. & Eyles, C. 2006. Neoproterozoic environmental change
recorded in the Port Askaig Formation, Scotland: climatic vs tectonic
controls. Sedimentary Geology, 183, 99 –124.
Arnaud, E. & Fairchild, I. J. 2011. The Port Askaig Formation, Dalra-
dian Supergroup, Scotland. In: Arnaud, E., Halverson, G. P. &
Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic
Glaciations. Geological Society, London, Memoirs, 36, 635–642.
Bahlburg, H. 1990. The Ordovician basin in the Puna of NW Argentina
and N Chile: geodynamic evolution from back-arc to foreland basin.
Geotektonische Forschungen, 75, 1 –107.
Bahlburg, H. 1998. The geochemistry and provenance of Ordovician tur-
bidites in the Argentinian Puna. In: Pankhurst, R. J. & Rapela,
C. W. (eds) The Proto-Andean Margin of Gondwana. Geological
Society, London, Special Publication, 142, 127– 142.
Bahlburg, H. & Breitkreuz, C. 1993. Differential response of a Devo-
nian– Carboniferous platform-deeper basin system to sea-level
change and tectonics, N. Chilean Andes. Basin Research, 5, 21– 40.
Baum, S. K. & Crowley, T. J. 2003. The snow/ice instability as a mech-
anism for rapid climate change: a Neoproterozoic snowball Earth
model example. Geophysical Research Letters, 30, 2030,
doi: 10.1029/2003GL017333.
Benn, D. I. & Prave, A. R. 2006. Subglacial and proglacial glacitectonic
deformation in the Neoproterozoic Port Askaig Formation, Scotland.
Geomorphology, 75, 266– 280.
Bhatia, M. R. & Crook, K. A. W. 1986. Trace element characteristics of
graywackes and tectonic setting discrimination of sedimentary
basins. Contributions to Mineralogy and Petrology, 92, 181–193.
Bland, W. & Rolls, D. 1998. Weathering. An Introduction to the Scien-
tific Principles. Arnold Publishers, London.
Cecioni, G. 1979. Grupo El Toco, desierto de Atacama, Chile. Revista de
la Asociación Geológica Argentina, 34, 211–223.
Cecioni, G. , 1981. Triassic El Toco Group, Atacama Desert, Chile. In:
Hambrey, M. J. & Harland, W. B. (eds) Earth’s Pre-Pleistocene
Glacial Record. Cambridge University Press, Cambridge.
Charrier, R. 1986. The Gondwana glaciation in Chile: description of
alleged glacial deposits and paleogeographic conditions bearing on
the extension of the ice cover in Southern South America. Palaeogeo-
graphy, Palaeoclimatology, Palaeoecology, 56, 151–175.
Condie, K. C. 1993. Chemical composition and evolution of the upper
continental crust: contrasting results from surface samples and
shales. Chemical Geology, 104, 1– 37.
Condie, K. C., Des Marais, D. J. & Abbott, D. 2001. Precambrian super-
plumes and supercontinents: a record in black shales, carbon isotopes,
and paleoclimates? Precambrian Research 106, 239–260.
Crowell, J. C. 1999. Pre-Mesozoic ice ages: their bearing on understand-
ing the climate system. Geological Society of America Memoir, 192,
1 –106.
Crowley, T. J. & North, G. R. 1991. Paleoclimatology. Oxford Mono-
graphs on Geology and Geophysics, 18, 349.
Dobrzinski, N. 2005. Das Paradoxon äquatornah abgelagerter
glazialer Sedimentfolgen: Sedimentologische und geochemische
Klimaindizien von der neoproterozoischen Yangtze Plattform
(Südchina). Dissertation, Westfälische Wilhelms-Universität,
Münster, Germany.
Dobrzinski, N. & Bahlburg, H. 2007. Sedimentology and environ-
mental significance of the Cryogenian successions of the Yangtze
platform, South China block. Palaeogeography, Palaeoclimatology,
Palaeoecology, 254, 100–122.
Dobrzinski, N., Bahlburg, H., Strauss, H. & Zhang, Q. R. 2004.
Geochemical climate proxies applied to the Neoproterozoic glacial
succession on the Yangtze Platform, South China. In: Jenkins, G.,
McMenamin, M., McKay, C. P. & Sohl, L. (eds) The Extreme Pro-
terozoic: Geology, Geochemistry and Climate. American Geophysi-
cal Union Monograph Series, 146, 13– 32.
Duzgoren-Aydin, N. S., Aydin, A. & Malpas, J. 2002. Re-assessment of
chemical weathering indices: case study of pyroclastic rocks of Hong
Kong. Engineering Geology, 63, 99 –119.
Edwards, M. B. 2004. Glacial influence on Neoproterozoic sedimen-
tation: the Smalfjord Formation, northern Norway — discussion.
Sedimentology, 51, 1409– 1417.
Etienne, J. L., Allen, P. A., Rieu, R. & Le Gerroué, E. 2008. Neopro-
terozoic glaciated basins: a critical review of the Snowball Earth
hypothesis by comparison with Phanerozoic glaciations. Special
Publication of the International Association of Sedimentologists
Special Publication 39.
Evans, D. A. D. 2000. Stratigraphic, geochronological, and paleomagnetic
constraints upon the Neoproterozoic climatic paradox. American
Journal of Science, 300, 347– 433.
Eyles, N. 2008. Glacio-epochs and the supercontinent cycle after 3.0
Ga: Tectonic boundary conditions for glaciation. Palaeogeography,
Palaeoclimatology, Palaeoecology, 258, 89 – 125.
Eyles, N. & Januszczak, N. 2007. Syntectonic subaqueousmass flows of
the Neoproterozoic Otavi Group, Namibia: where is the evidence of
global glaciation? Basin Research, 19, 179– 198.
Fairchild, I. J. & Kennedy, M. J. 2007. Neoproterozoic glaciation in the
Earth system. Journal of the Geological Society, London, 164,
895– 921.
Fedo, C. M., Nesbitt, H. W. & Young, G. M. 1995. Unraveling the
effects of potassium metasomatism in sedimentary rocks and paleo-
sols, with implications for paleoweathering conditions and prove-
nance. Geology, 23, 921– 924.
Gallet, S., Jahn, B., Van Vliet Lano, B., Dia, A. & Rosello, E. 1998.
Loess geochemistry and its implications for particle origin and com-
position of the upper continental crust. Earth and Planetary Science
Letters, 156, 157– 177.
Gevers, T. W. 1931. An ancient tillite in South West Africa. Transactions
of the Geological Society of South Africa, 34, 1 –17.
Gutzmer, J. & Beukes, N. J. 1998. Earliest laterites and possible evi-
dence for terrestrial vegetation in the Early Proterozoic. Geology,
26, 263– 266.
Hoffman, P. F. 2011. Glaciogenic and associated strata of the Otavi car-
bonate platform and foreslope, northern Namibia: evidence for large
base-level and glacioeustatic changes. In: Arnaud, E., Halverson,
G. P. & Shields-Zhou, G. (eds) The Geological Record of the Neo-
proterozoic Glaciations. Geological Society, London, Memoirs, 36,
195– 209.
 A REVIEW OF THE CHEMICAL INDEX OF ALTERATION
Hoffman, P. F. & Schrag, D. P. 2002. The snowball Earth hypothesis:
testing the limits of global change. Terra Nova, 14, 129– 155.
Hoffman, P. F., Kaufman, A. J., Halverson, G. P. & Schrag, D. P.
1998. A Neoproterozoic snowball Earth. Science, 281, 1342–1346.
Hoffmann, K. H. & Prave, A. R. 1996. A preliminary note on a revised
subdivision and regional correlation of the Otavi group based on gla-
cigenic diamictites and associated cap dolostones. Communications
of the Geological Survey of Namibia, 11, 81 – 86.
IGME 1984. Mapa Geológica de España a escala 1:50.000, Hoja no. 107
(Barruelo de Santullan), 113.
ITGE 1994. Mapa Geológica de España a escala 1:50.000, Hoja no. 81
(Potes), 128.
Johnsson, M. J. 1993. The system controlling the composition of clastic
sediments. In: Johnsson, M. J. & Basu, A. (eds) Processes Control-
ling the Composition of Clastic Sediments. Geological Society of
America Special Paper, 285, 1 – 19.
Johnsson, M. J., Stallard, R. F. & Meade, R. H. 1988. First-cycle
quartz arenites in the Orinoco River Basin, Venezuela and Colombia.
Journal of Geology, 96, 263– 277.
Kahmann, J. A., Seaman, J. III & Driese, S. G. 2008. Evaluating trace
elements as paleoclimate indicators: multivariate statistical analysis
of Late Mississippian Pennington Formation paleosols, Kentucky,
U.S.A. Journal of Geology, 116, 254– 268.
Keidel, J. 1943. El Ordovı́cico inferior en los Andes del norte Argentino y
sus depósitos marino glaciales. Boletin de la Academı́a Nacional de
Ciencias de Córdoba, 36, 140– 229.
Keller, M., Bahlburg, H. & Reuther, C.-D. 2008. The transition from
passive to active margin sedimentation in the Cantabrian Mountains,
Northern Spain: Devonian or Carboniferous? Tectonophysics, doi:
10.1016/j.tecto.2008.06.022
Keller, M., Bahlburg, H., Reuther, C.-D. & Weh, A. 2007. Flexural to
broken foreland basin evolution as a result of Variscan collisional
events in northwestern Spain. In: Hatcher, R. D. Jr, Carlson, M.
P., McBride, J. H. & Martı́nez Catalán, J. R. (eds) The 4D Frame-
work of Continental Crust. Geological Society of America Memoir,
200, 489– 510.
Kirschvink, J. L. 1992. Late Proterozoic low-latitude global glacia-
tion: the snowball earth. In: Schopf, J. W & Klein, C. (eds) The
Proterozoic Biosphere. Cambridge University Press, Cambridge,
51 – 52.
Kramer, J. R. 1968. Mineral-water equilibria in silicate weathering. Inter-
national Geological Congress, 23rd session, Section 6, 149–160.
Martin, H. 1964. Beobachtungen zum Problem der jung-präkambrischen
glazialen Ablagerungen in Südwestafrika. Geologische Rundschau,
54, 115– 127.
Martin, H., Porada, H. & Walliser, O. H. 1985. Mixtite deposits of the
Damara sequence, Namibia: problem of interpretation. Palaeogeo-
graphy, Palaeoclimatology, Palaeoecology, 51, 159– 196.
McLennan, S. M. 1993. Weathering and global denudation. Journal of
Geology, 101, 295– 303.
McLennan, S. M. 2001. Relationships between the trace element compo-
sition of sedimentary rocks and upper continental crust. Geochemis-
try, Geophysics, Geosystems (G3), 2, doi:10.1029/2000GC000109.
McLennan, S. M., Hemming, S., McDaniel, D. K. & Hanson, G. N.
1993. Geochemical approaches to sedimentation, provenance and
tectonics. In: Johnsson, M. J. & Basu, A. (eds) Processes Control-
ling the Composition of Clastic Sediments. Geological Society of
America Special Paper, 285, 21 – 40.
Nedachi, Y., Nedachi, M., Bennett, G. & Ohmoto, H. 2005. Geochem-
istry and mineralogy of the 2.45 Ga Pronto paleosols, Ontario,
Canada. Chemical Geology, 214, 21 – 44.
Nesbitt, H. W. & Young, G. M. 1982. Early Proterozoic climates and
plate motions inferred from major element chemistry of lutites.
Nature, 199, 715– 717.
Nesbitt, H. W. & Young, G. M. 1984. Prediction of some weathering
trends of plutonic and volcanic rocks based on thermodynamic and
kinetic considerations. Geochimica et Cosmochimica Acta, 48,
1523– 1534.
Nesbitt, H. W. & Young, G. M. 1989. Formation and diagenesis of
weathering profiles. Journal of Geology, 97, 129–147.
Nesbitt, H. W. & Young, G. M. 1996. Petrogenesis of sediments
in the absence of chemical weathering: effects of abrasion and
91
sorting on bulk composition and mineralogy. Sedimentology, 42,
341– 358.
Nesbitt, H. W., Fedo, C. M. & Young, G. M. 1997. Quartz and
feldspar stability, steady and non-steady-state weathering, and petro-
genesis of siliciclastic sands and muds. Journal of Geology, 105,
173– 191.
Nesbitt, H. W., Young, G. M., McLennan, S. M. & Keays, R. R. 1996.
Effects of chemical weathering and sorting on the petrogenesis of sili-
ciclastic sediments, with implications for provenance studies. Journal
of Geology, 104, 525–542.
Ohta, T. & Arai, H. 2007. Statistical empirical index of chemical weath-
ering in igneous rocks: a new tool for evaluating the degree of weath-
ering. Chemical Geology, 240, 280– 297.
Panahi, A. & Young, G. M. 1997. A geochemical investigation into the
provenance of the Neoproterozoic Port Askaig Tillite, Dalradian
Supergroup, western Scotland. Precambrian Research, 85, 81 –96.
Parker, A. 1970. An index of weathering for silicate rocks. Geological
Magazine, 107, 501– 504.
Pettijohn, F. J., Potter, P. E. & Siever, R. 1987. Sand and Sandstone.
Springer, New York.
Rice, A. H. N. 2004. Glacial influence on Neoproterozoic sedimentation:
the Smalfjord Formation, northern Norway – discussion. Sedimentol-
ogy, 51, 1419–1422.
Rice, A. H. N. & Hofmann, C. C. 2001. The transition from
Neoproterozoic glacial to interglacial sedimentation near Ham-
marnes, East Finnmark, North Norway. Norsk Geologisk Tidsskrift,
81, 257– 262.
Rice, A. H. N., Edwards, M. B., Hansen, T. A., Arnaud, E. & Halver-
son, G. P. 2011. Glaciogenic rocks of the Neoproterozoic Smalfjord
and Mortensnes Formations, Vestertana Group, E. Finnmark, Norway.
In: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds)
The Geological Record of Neoproterozoic Glaciations. Geological
Society, London, Memoirs, 36, 593– 602.
Rieu, R., Allen, P. A., Plotze, M. & Pettke, T. 2007. Compositional
and mineralogical variations in a Neoproterozoic glacially influenced
succession, Mirbat area, south Oman: Implications for paleoweather-
ing conditions. Precambrian Research, 154, 248– 265.
Ruxton, B. P. 1968. Measures of the degree of chemical weathering of
rocks. Journal of Geology 76, 518–527.
Scheffler, K., Hoernes, S. & Schwark, L. 2003. Global changes during
Carboniferous–Permian glaciation of Gondwana: Linking polar and
equatorial climate evolution by geochemical proxies. Geology, 31,
605– 608.
Schermerhorn, L. J. G. 1974. Late Precambrian mixtites: glacial and/or
non-glacial? American Journal of Science, 274, 673– 824.
Schrag, D. P., Berner, R. A., Hoffman, P. F. & Halverson, G. P. 2002.
On the initiation of Snowball Earth. Geochemistry, Geophysics, Geo-
systems, 3, doi:10.1029/2001GC000219.
Shields, G. A. 2008. Palaeoclimate – Marinoan meltdown. Nature
Geoscience, 1, 351– 353.
Taylor, S. R. & McLennan, S. M. 1985. The Continental Crust: Its
Composition and Evolution. Blackwell, Oxford.
von Eynatten, H. 2004. Statistical modelling of compositional trends in
sediments. Sedimentary Geology, 171, 79– 89.
von Eynatten, H., Barceló-Vidal, C. & Pawlowsky-Glahn, V.
2003. Modelling compositional change: the example of chemi-
cal weathering of granitoid rocks. Mathematical Geology, 35,
231– 251.
Wagner, R. H. & Wagner-Gentis, C. H. T. 1963. Summary of the stra-
tigraphy of Upper Paleozoic rocks in NE. Palencia, Spain. Proceed-
ings Kongelige Nederlandse Akademie Wetenschappen (B) LXVI, 3,
149– 163.
Weil, A. B., Van der Voo, R. & Van der Pluijm, B. A. 2001. Oroclinal
bending and evidence against the Pangea megashear; the Cantabria –
Asturias Arc (northern Spain). Geology, 29, 991– 994.
Wintsch, R. P. & Kvale, C. M. 1994. Differential mobility of elements in
burial diagenesis of siliciclastic rocks. Journal of Sedimentary
Research, A64, 349– 361.
Young, G. M. 2001. Comparative geochemistry of Pleistocene and Paleo-
proterozoic (Huronian) glaciogenic laminated deposits: relevance to
crustal and atmospheric composition in the last 2.3 Ga. Journal of
Geology, 109, 463– 477.
92 H. BAHLBERG & N. DOBRZINSKI

Young, G. M. & Nesbitt, H. W. 1999. Paleoclimatology and provenance
of the glaciogenic Gowganda Formation (Paleoproterozoic), Ontario,
Canada: a chemostratigraphic approach. Geological Society of
America Bulletin, 111, 264–274.
Young, G. M., Minter, W. E. L. & Theron, J. N. 2004. Geochemistry
and palaeogeography of upper Ordovician glaciogenic sedimentary
rocks in the Table Mountain Group, South Africa. Palaeogeography,
Palaeoclimatology, Palaeoecology, 214, 323– 345.
Zhang, Q.-R., Chu, X.-L. & Feng, L.-J. 2011. Neoproterozoic glacial
records in the Yangtze Region, China. In: Arnaud, E., Halverson,
G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproter-
ozoic Glaciations. Geological Society, London, Memoirs, 36, 357–366.
 Chapter 7

Neoproterozoic glacial palaeolatitudes: a global update

D. A. D. EVANS1* & T. D. RAUB2

1
Department of Geology & Geophysics, Yale University, 210 Whitney Avenue, New Haven, CT 06520-8109, USA
2
Division of Geological and Planetary Sciences, 100-23 Caltech, Pasadena CA 91125, USA
*Corresponding author (e-mail: dai.evans@yale.edu)

Abstract: New stratigraphic, geochronological and palaeomagnetic constraints allow updates to be made to a synthesis of Neoprotero-
zoic glacial palaeolatitudes, including modifications to some reliability estimates. The overall pattern of a Neoproterozoic climatic
paradox persists: there is an abundance of tropical palaeolatitudes and near to complete absence of glaciogenic deposits demonstrably
laid down between latitudes of 608 and 908. In addition to 12 units with palaeolatitude estimates that are somewhat reliable, estimates
with moderate to high reliability now include Konnarock (less than 108 from the palaeo-equator), Elatina, Rapitan, Mechum River, Grand
Conglomerat (10– 208), Upper Tindir, Puga (20– 308), Nantuo, Gaskiers (30–408) and Walsh (40– 508). Among these, Elatina, Upper
Tindir and Nantuo are considered to have the highest reliability, all with estimates of low to moderate palaeolatitude. The Elatina
result stems from sedimentary rocks with quantitative correction of inclination-shallowing effects, and the Upper Tindir result stems
from data collected from igneous rocks that are precisely coeval with the glacial deposits. Despite continuing debate on the global char-
acter of Neoproterozoic ice ages, their pan-glacial extent (ice extending to low latitude in a low-obliquity world) is well demonstrated.

Palaeomagnetism of glaciogenic deposits has provided a quantita-
tive basis for hypotheses of extreme climatic shifts in the late
Neoproterozoic Era. The most recent global compilations of
palaeomagnetic depositional latitudes for Proterozoic ice ages
indicate a dominant mode near the palaeo-equator (Evans 2000,
2003), with no robust palaeo-polar deposits yet discovered. Such
results could therefore support either the Snowball Earth (Kirsch-
vink 1992) or the high-obliquity (Williams 1993) hypotheses for
Precambrian ice ages, but would appear to reject the uniformitarian
comparison to polar/temperate-restricted Phanerozoic glaciogenic
deposits (Evans 2000). Hoffman (2009) has suggested that
Neoproterozoic ice ages represent a globally all-encompassing
‘pan-glacial’ state of the Earth’s climate system, fundamentally
distinct from either partially ice-covered ‘glacial –interglacial’
or ice-free ‘nonglacial’ palaeoclimates experienced during the
Phanerozoic Eon.
Several reviews of stratigraphic (Halverson et al. 2005, 2007),
sedimentological (Hoffman & Schrag 2002; Eyles & Januszczak
2004; Fairchild & Kennedy 2007; Hoffman et al. 2007; Allen &
Etienne 2008; Hoffman 2009) and palaeomagnetic (Trindade &
Macouin 2007; Hoffman & Li 2009) data sets pertaining to
Neoproterozoic ice ages have appeared recently. These discussions
of glacial deposits owe much to the pioneering synthesis of
Hambrey & Harland (1981), but the recent reviews arrive at differ-
ing conclusions regarding the extent and severity of Neoprotero-
zoic ice ages. In particular, the study by Eyles & Januszczak
(2004) is commonly cited as a palaeogeographic alternative that
avoids the need for nonuniformitarian processes to account for
the advance of widespread continental ice sheets into tropical
palaeolatitudes. It becomes useful, then, to review the global evi-
dence for or against low-latitude glaciation. Herein, we reassess
Neoproterozoic glacial palaeolatitudes in light of new strati-
graphic, geochronological and palaeomagnetic data obtained
within the last decade, providing the first comprehensive update
and revision of the palaeographic analyses of Evans (2000).
Methods
Table 7.1 lists the known or alleged Neoproterozoic glaciogenic
deposits, revising the unit-identifying numbering scheme intro-
duced by Evans (2000), but with cross-references to that study,
as well as those of Hambrey & Harland (1981), Eyles &
Januszczak (2004, table 1), Trindade & Macouin (2007) and
Hoffman & Li (2009). Deposits are numbered by present geo-
graphical location, differentiating units that are separated from
each other on the scale of 100 km or more, or in some cases dis-
tinguishing units of uncertain relative correlation that are presently
adjacent via tectonic stacking in orogenic belts. Within each
numbered location, the units are denoted (a, b, c, etc.) in ascending
stratigraphic order. Many global chronostratigraphic schemes
have been proposed, that assign ages to undated deposits via cor-
relation, but herein, we adopt the approach used by Evans (2000)
to consider each unit’s age constraints in isolation. This splitting,
rather than lumping, approach serves to illustrate how few of the
deposits are precisely dated or studied palaeomagnetically. Many
of the deposits have been classified using the loose terms
‘Sturtian’, ‘Marinoan’, or ‘Ediacaran’ based on the lithographic
character of either diamictites or their overlying cap carbonates
(e.g. Hoffman & Li 2009). Recent U – Pb geochronology has
thus far permitted many so-called Marinoan deposits to record
the synchronous end of a widespread ice age of unknown duration,
which ended at 635 Ma (Hoffmann et al. 2004; Condon et al. 2005;
Zhang et al. 2008), and a mid-Ediacaran ice age at c. 580 Ma
(Bowring et al. 2002, abstract only). U –Pb ages of so-called
Sturtian deposits range from c. 765 Ma (Key et al. 2001) to poss-
ibly as young as c. 660 Ma (Fanning & Link 2008; from the type
area in South Australia). Use of the lithological characteristics of
cap carbonates as global correlation tools appears successful in
some instances, but may be problematic in others (Corsetti &
Lorentz 2006; Kendall et al. 2009).
Finally, several of the units discussed by Evans (2000) are now
considered irrelevant to discussions on Neoproterozoic ice ages, as
they are likely nonglacial, or demonstrably Cambrian or younger
in age. Those units are included in Table 7.1, but are now stripped
of their numerical codes. In a few cases, questions remain about
these issues, and the deposits retain their numerical status until
documented otherwise.
Table 7.1 also lists our preferred interpretation on the reliability
of palaeolatitude determination for the deposits, using the three-
level qualitative scale after Evans (2000). Although quantitative
measures of palaeomagnetic pole reliability exist (e.g. Van der
Voo 1990), they have generally been tailored towards plate recon-
structions rather than palaeoclimatic problems. Herein, we regard

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 93– 112. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.7
 94
Table 7.1 Published assessments of glacial influence and depositional palaeolatitude for Neoproterozoic strata

# Deposit name Lat (88 N) Long (88 E) HH81 E00 EJ04 TM07 HL09 This study

Laurentia and environs

1a Hula Hula 69.5 215.5 – – – – Hu(S) 0 –15 No pmag data
1b Katakturuk unit 2 69.5 214.5 – – – – – No pmag data
2a Upper Tindir: unit 2 65 222 Yes (F10) No pmag data (#3) – – Ti(S) 0 –15 *** 21 +3
2b Upper Tindir: unit 3b 65 219 – – – – – No pmag data
3a Rapitan 64 230 Yes (F11) ** 06 þ8/27 (#1) Yes Q ¼ 4; 06 +4 Ra(S) 0 –15 ** 18 +3
3b Ice Brook (Stelfox) 63.5 231.5 – Unreliable pmag (#2) ?? – IB(M) 0 –15 Unreliable pmag
4 Mt Lloyd George 58 235 Yes (F12) No pmag data (#4) – – – No pmag data
5 Deserters Range 57 235.5 – No pmag data (#4) – – – No pmag data
6 Mount Vreeland 54.5 239 Uncertain (F12) No pmag data (#5) – – Vr(M) 0–15 No pmag data
7 Toby 49.5 243 Yes (F13) * 08 +4 (#6) – – To(S) 0 –15 No pmag data
8 Edwardsburg 45 244.5 – No pmag data (#7) – No pmag data – No pmag data

D. A. D. EVANS & T. D. RAUB

9 Pocatello: Scout Mtn 42.5 248 Yes (F14) No pmag data (#7) – Q ¼ 6; 08 +3 Po(S) 0– 15 No pmag data
10 Mineral Fork/Dutch Pk 40.5 247.5 Yes (F15) No pmag data (#7) – – – No pmag data
11a Kingston Peak: Surprise 36 243.5 Yes (F16) No pmag data (#8) No – Su(S) 0– 15 No pmag data
11b Kingston Peak: Wildrose 36 243.5 – No pmag data (#8) No – Wr(M) 0– 15 No pmag data
11c Ibex 36 243.5 – – – – – No pmag data
11d Johnnie Rainstorm 36 244 – * 01 +4 (#9) – – – * 01 +4
12 Cerro Las Bolas 29 250.5 – – – – – No pmag data
– Florida Mtns 32 252.5 Yes (F17) * 03 +8 (#10) – – – Too young
13 Konnarock/Grandf. Mtn 36.5 278.5 Yes (F18) See Mechum R (#11) No – Kn(S) 15–30 ** 06 +5
14a Mechum River 38.5 281.5 – ** 20–21 +4 (#11) – – – ** 16 +3
14b Fauquier 39 282.5 – – – – – * 78 +12
15a Port Askaig 55.5 353.5 Yes (E18 –E20) Unreliable pmag (#12) Yes No pmag data Pt(S) 30– 45 Unreliable pmag
15b Stralinchy (Reelan) 55.5 353.5 – – – – Re(M) 30– 45 No pmag data
15c Loch na Cille 55.5 353.5 – No pmag data (#12) – – Lo(E) 75–90 Unreliable pmag
16 Gåseland/Charcot Land 71 330.5 Yes (F23,F24) No pmag data (#14) – – – No pmag data
17a Tillite Gp: Ulvesø 73 336 Yes (F25) Unreliable pmag (#15) Yes – Ul(S) 15– 30 Unreliable pmag
17b Tillite Gp: Storeelv 73 336 Yes (F25) Unreliable pmag (#15) Yes – St(M) 30– 45 Unreliable pmag
18a Elbobreen (Petrovbreen) 79.5 18 Yes (E10) No pmag data (#16) – – Pb(S) 15– 30 No pmag data
18b Wilsonbreen 79.5 18 Yes (E10) No pmag data (#16) – – Wb(M) 30–45 No pmag data
19 West Spitsbergen 78 13 Yes (E10) No pmag data (#17) – – – No pmag data
20 Moraenesø 82 326 Yes (F26) No pmag data (#18) – – – No pmag data
21 Pearya 82.5 281.5 – No pmag data (#19) – – – No pmag data
Baltica
22a Vestertana: Smalfjord 70 28 Yes (E12) ** 33 þ14/212 (#20) Yes (a or b) Q ¼ 3; 33 +9 Sm(M) 45–60 * 33 þ14/212
22b Vestertana: Mortensnes 70 28 Yes (E12) ** 33 þ14/212 (#20) Yes (a or b) Q ¼ 3; 33 +9 Mt(E) 45– 60 No pmag data
– Sito/Vakkejokk 68 19 Uncertain (E13) Too young (#21) – – – Too young
23 Långmarkberg/Lillfjället 64 14.5 Yes (E14/E15) No pmag data (#22) – – – No pmag data
24 Moelv 61.5 11 Yes (E16) Unreliable pmag (#23) Yes No pmag data Mo(E) 45– 60 Unreliable pmag
25 Vilchitsy/Blon 53 32 Yes (E24 –E28) Unreliable pmag (#24) – – Vi(E) 45–60 Unreliable pmag
26 N Urals: Churochnaya 60 57.5 Yes (E30) No pmag data (#25) – – Cn(E) 30– 45 No pmag data
27a C Urals: Tany 58.5 59 Yes (E31) No pmag data (#26) – – Ty(S) 45– 60 No pmag data
27b C Urals: Koyva 58.5 59 Yes (E31) No pmag data (#26) – – – No pmag data
27c C Urals: Staryye Pechi 58.5 59 Yes (E31) No pmag data (#26) – – – No pmag data
28 S Urals: Kurgashlya 53.5 57.5 Yes (E32) Unreliable pmag (#27) – – – Unreliable pmag
Altaids and Siberia
29a West Altaids (Satan/Dzhetym) 44 72 Yes (C27) No pmag data (#31) – – – No pmag data
29b West Altaids (Baykonur) 44 72 Variable (C22– C28) No pmag data (#31) – – Br(S) 45– 60 No pmag data
30a Tsagaan Oloom (Maikhan Ul) 47 96 – No pmag data (#32) Blank – Mk(S) 0–15 Unreliable pmag
30b Tsagaan Oloom (Khongoryn) 47 96 – – – – Kg(M) 0– 15 * 03 + 11
31 East Sayan/Khubsugul: Zabit 51 100 – No pmag data (#30) – – – Unreliable pmag
32a Chivida 61 92 Nonglacial (C30) Unreliable pmag (#28) – – Cv(S) 15–30 Unreliable pmag
32b Pod’em 61 92 – – – – Pd(M) 0–15 No pmag data
33 Marnya 54.5 99 – – – Ma(M) 0 –15 No pmag data
34a Patom: Kharluktakh 59 115 – – – – Kh(S) 0 –15 No pmag data
34b Patom: Dzhemkukan 59 115 Uncertain (C31) No pmag data (#29) – – Dz(M) 0– 15 No pmag data
China cratons
35a Aksu: Qiaoenbrak 41 79.5 Yes (C33) * 08 +8 (#33) – – – No pmag data

GLOBAL NEOPROTEROZOIC GLACIAL PALAEOLATITUDES

35b Aksu: Yuermeinbrak 41 79.5 Yes (C33) * 08 +8 (#33) – – – * 27 + 9
36a Quruqtagh: Bayisi 41.5 87.5 Yes (C33) * 08 +8 (#33) – Q ¼ 4; 01 +3 By(S*) 30–45 * 01 þ4/22
36b Quruqtagh: Tereeken 41.5 87.5 Yes (C33) * 08 +8 (#33) – – Te(M) 15– 30 No pmag data
36c Quruqtagh: Hankalchough 41.5 87.5 Yes (C33) * 08 +8 (#33) – – Ha(E) 15– 30 No pmag data
37 Qaidam: Hongtiegou 37.5 96 Yes (C33) – – – – No pmag data
38 Luoquan 34 115 Yes (C33,C34) Unreliable pmag (#34) Yes – Lq(E) 0 –15 Unreliable pmag
39a Chang’an/Tiesiao/Jiangkou 27 111 Yes (C33,C35) *** 30–40 +12 (#35) Yes (a or b) – Ji(S) 15–30 No pmag data
39b Nantuo 27 111 Yes (C33,C35) *** 30–40 +12 (#35) Yes (a or b) Q ¼ 6; 04 +4 Na(M) 15– 30 *** 37 +9
India to Nubia
40 Blaini 30 78.5 Yes (C14) Unreliable pmag (#37) – – Bl(M) 0– 15 Unreliable pmag
– Penganga 19.5 75 – No pmag data (#38) – – – Nonglacial
41 Rizu 31 56 – – – – Ri(M) 0– 15 Too young?
42a Ghubrah 23 58 – Unreliable pmag (#36) – No pmag data Gu(S) 15–30 No pmag data
42b Fiq 23 58 – No pmag data (#36) Yes (’Shuram’) Q ¼ 6; 15 +4 Fi(M) 30– 45 * 13 +7
43a Ayn 17.5 55 – Unreliable pmag (#36) – – Ay(S) 15–30 Unreliable pmag
43b Shareef 17.5 55 – No pmag data (#36) – – Sh(M) 30– 45 * 18 +7
44 Tambien 14 39 – – – – Ta(S) 15–30 No pmag data
45 Atud 26 35 – – – – – No pmag data
Australia and Mawsonland
46 Walsh 217 126 Yes (D16) ** 45 þ14/212 (#42) – Q ¼ 5; 45 +5 – ** 45 þ14/212
47a Landrigan 218 126.5 Yes (D16) No pmag data (#44) – – La(M) 0– 15 No pmag data
47b Egan 218.5 126.5 Yes (D16) * 21 +8 (#43) – – Eg(E) 15–30 * 21 +8
48a Fargoo 217 129 Yes (D16) No pmag data (#44) – – – No pmag data
48b Moonlight Valley 217 129 Yes (D16) No pmag data (#44) Yes – – No pmag data
49 Little Burke 221.5 140 Yes (D19) Unreliable pmag (#40) – – – Unreliable pmag
50a Areyonga/Naburula/Yardida 223 134 Yes (D17,D18) No pmag data (#39) – – Ar(S) 15– 30 No pmag data
50b Olympic/Mount Doreen 223 135 Yes (D17) No pmag data (#40) – – Ol(M) 0 –15 No pmag data

(Continued)

95
 96
Table 7.1 Continued

# Deposit name Lat (88 N) Long (88 E) HH81 E00 EJ04 TM07 HL09 This study

51 Boondawari/Wahlgu/Turkey Hill 225 124 – No pmag data (#40) – – – Unreliable pmag

52 Chambers Bluff 227 134 – No pmag data (#39) – – – No pmag data
53a Sturtian (type area) 232 139 Yes (D21) Unreliable pmag Yes No pmag data St(S) 0– 15 Unreliable pmag
53b Elatina 232 139 Yes (D21) *** 03–09 +4 (#40) Blank Q ¼ 4,6; 05 +6 El(M) 0–15 *** 10– 14
54a Yancowinna 231.5 141.5 Yes (D20) No pmag data (#39) – – – No pmag data
54b Teamsters Creek 231.5 141.5 Yes (D20) No pmag data (#40) – – – No pmag data
55 Cottons 240 144 Uncertain (D22) Unreliable pmag (#41) – No pmag data Co(M) 0– 15 Unreliable pmag
56a Julius River 241 145 Uncertain (D22) No pmag data (#41) – No pmag data Ju(S) 0 –15 No pmag data
56b Croles Hill 240.5 145 – – – No pmag data Cr(E) 0 –15 No pmag data

D. A. D. EVANS & T. D. RAUB

57 Goldie 283 160 – No pmag data (#45) – – – No pmag data
Kalahari and environs
58a Blaubeker/Court 223.5 17.5 Yes (A29) Unreliable pmag (#46) – – Bb(S) 0– 15 Unreliable pmag
58b Blasskrans/Naos 223.5 17.5 Map only (A29) No pmag data (#47) – – Bk(M) 0– 15 No pmag data
59a Kaigas 228.5 16.5 Nonglacial (A29) No pmag data (#48) – No pmag data Ka(S*) 0 –15 No pmag data
59b Numees 228.5 16.5 Yes (A29) No pmag data (#49) – – Nu(M) 0– 15 No pmag data
59c Namaskluft 228.5 16.5 Yes (A29) No pmag data (#49) – – 0–15 No pmag data
59d Schwarzrand 227 18 Yes (A29) * 38 +3 (#51) – – – * 38 +3
60a Karoetjes Kop 231.5 18.5 – – – – – No pmag data
60b Aties 232 18.5 Nonglacial (A29) No pmag data (#50) – – – No pmag data
61 Dernburg 228 15.5 – – – – – No pmag data
Congo-São Francisco
62a Grand Conglomérat 211 27 Yes (A28) No pmag data (#52) – Q ¼ 5; 10 +6 Gr(S*) 15–30 ** 10 +5
62b Petit Conglomérat 211 27 Uncertain (A28) No pmag data (#53) – – Pe(M) 0 –15 No pmag data
63 Geci 213 35 – – – – – No pmag data
64 Tshibangu 22 29 – No pmag data (#54) – – – No pmag data
65 Bunyoro 1.5 31.5 Yes (A25) No pmag data (#55) – – – No pmag data
66a Akwokwo/Bandja 3 24 – No pmag data (#55) – – Ak(S) 30–45 No pmag data
66b Bondo 5 20 – No pmag data (#56) – – Bo(M) 15– 30 No pmag data
67 Mintom 3 13 – – – – – No pmag data
68a Sergipe: Juetê/Ribeirópolis 211 322 – No pmag data (#56) – – – No pmag data
68b Sergipe: Palestina 211 322 – No pmag data (#56) – – Pa(M) 30–45 No pmag data
69 Bebedouro 211 319 Uncertain (G28) No pmag data (#58) – – – Unreliable pmag
70 Rio Preto: Canabravinha 211 314 – – – – – No pmag data
71 Brasilia int: Ibiá/Cristalina 218 313 Uncertain (G27) No pmag data (#59) – – – No pmag data
72a Vazante: St. Antônio do Bonito 217.5 313 – – – – – No pmag data
72b Vazante: Lapa 217.5 313 – – – – – No pmag data
73 Carandaı́ 221 316 Uncertain (G30) No pmag data (#60) – – – No pmag data
74a Bambui: Jequitaı́/Macaúbas 217 316 Yes (G31) Unreliable pmag (#57) No – Je(S) 30–45 Unreliable pmag
74b Bambui: Inhaúma 219.5 315.5 – – – – – No pmag data
74c Bambui: Lagoa Formosa 218.5 313.5 – – – – – No pmag data
75 Rio Pardo: Salobro 215.5 320.5 Uncertain (G29) Uncertain glacial (#57) – – – Nonglacial
76a West Congo Inférieure 25 14 Nonglacial (A26,A27) No pmag data (#61) – – In(S) 30–45 No pmag data
76b West Congo Supérieure 25 14 Nonglacial (A26,A27) No pmag data (#62) – – Sp(M) 15– 30 No pmag data
77a Chuos/Varianto 221 16 Variable (A29) * 10 +5 (#63) No – Ch(S) 15–30 * 09 +5
77b Ghaub 221 16 Uncertain (A29) Unreliable pmag (#64) No No pmag data Gh(M) 0– 15 Unreliable pmag
West Africa and Hoggar
78a Basal Atar Gp 22 352 – Unreliable pmag (#66) – – – Nonglacial?
78b Jbéliat/Bthaat Ergil (triad) 22 352 Yes (A19) * 30– 70 (#65) Yes – Jb(M) 45–60 No pmag data
79 Mali Gp/Bakoye 15 351 Yes (A19) Same as #65 (#67) – – Ba(M) 30– 45 No pmag data
80 Rokel River 8.5 348 Yes (A20) No pmag data (#68) – – – No pmag data
81a Kodjari 10 0 Yes (A21) No pmag data (#69) – – Ko(M) 30–45 No pmag data
81b Tamale/Obosum 8 0 Yes (A21) No pmag data (#70) – – – No pmag data
82a Série Verte (Tafeliant) 21 2 Yes (A23) Unreliable pmag (#71) – – – Unreliable pmag
82b Série Pourprée 24 2 Yes (A22) Same as #65 (#72) – – – No pmag data
83 Série Tirririne 21 8.5 Yes (A23) Unreliable pmag (#71) Unreliable pmag

GLOBAL NEOPROTEROZOIC GLACIAL PALAEOLATITUDES

84a Sarhro 31 352.5 – No pmag data (#73) – – – No pmag data
84b Tiddiline 30 352 Nonglacial (A18) No pmag data (#73) – – – Unreliable pmag
Avalonia and Cadomia
85 Squantum 42.5 289 Uncertain (F19) * 55 þ8/27 (#74) No Q ¼ 3; 55 +3 Sq(E) 75– 90 * 55 þ8/27
86 Gaskiers 47.5 307 Yes (F20) * 31 þ10/28 (#75) Yes No pmag data Ga(E) 75– 90 ** 34 þ8/27
87 Gwna/Mona 53 355.5 – – – – – No pmag data
88 Brioverian (Granville) 48.5 358 Nonglacial (E21) No pmag data (#76) No – Gr(E) 45– 60 No pmag data
89 Clanzschwitz/Weesenstein 51 13.5 Correlative to (E22)? – – – – No pmag data
Amazonia and environs
90a Puga 215.5 303 Yes (G26) No pmag data (#77) – Q ¼ 5; 22 +6 Pu(M) 30– 45 ** 22 þ6/25
90b Serra Azul 215.5 303 – – – – Az(E) 60– 75 No pmag data
91 Chiquerı́o 215.5 285 – No pmag data (13) – – Cq(S) 15–30 No pmag data
Rio Plata and environs
92a Zanja del Tigre 234.5 305 – – – – – Nonglacial?
92b Las Ventanas/Playa Hermosa 234.5 305 – – – – – Unreliable pmag
93 Picada das Graças 231 305.5 – Unreliable pmag (#78) – – – Unreliable pmag
94 Iporanga 224.5 311.5 – – – – – No pmag data
95 Sierra del Volcán 238 302 – – – Q ¼ 5; 48 +7 – Too young?

Deposits are numbered by geographical area; within each region, they are lettered by ascending stratigraphic order. Glaciogenic influence assessments: HH81, Hambrey & Harland (1981); EJ04, Eyles & Januszczak (2004, table 1).
The terms ‘yes’, ‘no’ and ‘uncertain’ refer to the question of whether a glaciogenic influence has been demonstrated according to each compilation. ‘Blank’ refers to blank entries in Eyles & Januszczak (2004, table 1).
Palaeomagnetic constraint assessments: E00, Evans (2000); TM07, Trindade & Macouin (2007); HL09, Hoffman & Li (2009). Correlations by Hoffman & Li (2009) presented in parentheses, follow the deposit abbreviations
used in their work: E, Ediacaran; M, Marinoan; S, Sturtian; S*, nominally Sturtian but with possibly older ages than the deposits labelled ‘S’. Asterisks in the columns for E00 and this study indicate the relative reliability
(*, low; **, moderate; ***, high) of glacial palaeolatitudes assessed by Evans (2000) or herein. Q values use the reliability scheme of Van der Voo (1990). Throughout the table, dashed entries indicate that the deposit was not
mentioned by the particular compilation.

97
98 D. A. D. EVANS & T. D. RAUB

reliability of glacial palaeolatitudes to depend mainly on the

(a)
quality of palaeomagnetic data, but also on the confidence in Shareef Somewhat reliable
chrono/lithostratigraphic correlations, if necessary for palaeolati-
tude estimation, and to a lesser extent the amount of consensus on Moderately reliable
Chuos Fiq
the glaciogenicity of the deposits. In addition, some of our assess-
ments in Table 7.1 are labelled ‘unreliable pmag [palaeomagnetic Very reliable
data]’, which merely indicates that the data are not applicable for Bayisi Grand Egan Schwarz-
Uniform distribution (n=22)
Congl. rand
reliable estimates of glacial palaeolatitude, although in some cases
they could be useful for other purposes such as tectonic reconstruc- Khon- Yuer- Smal-
tions. We do not provide references to all the stratigraphic, geo- goryn Mechum meinbrak
fjord
chronological, and palaeomagnetic constraints on the deposits,
but rather only discuss those instances in which our current assess- Johnnie Rapitan Puga Gaskiers
ment differs from previously published global compilations. Our
global palaeolatitude estimates for Neoproterozoic ice ages,
Konna- U.Tindir Squan-
using this quality filtering method, are summarized graphically rock Elatina unit 2
Nantuo Walsh tum Fauquier
in Fig. 7.1.
We begin with the assumption that Earth’s time-averaged (.103 0 10 20 30 40 50 60 70 80 90

years) magnetic field has the geometry of a geocentric-axial dipole
(GAD). This assumption is justified to first order by the compari-
son of evaporite palaeolatitudes in the palaeomagnetic reference
frame, relative to their expected palaeolatitudes according to inter-
tropical convergence of zonal atmospheric circulation. Agreement
between these two reference frames for most Proterozoic evaporite
deposits implies that the GAD hypothesis is tenable for most of the
last two billion years (Evans 2006). Departures from this model are
possible, due to data gaps, over intervals as long as 100–
200 million years; notably, the Neoproterozoic glacial interval
does indeed lack voluminous evaporite deposits. Ephemeral, dom-
inantly non-GAD fields (on the order of 10 million years duration
or shorter) are also permissible, as will be discussed further below,
but in light of the broader record of a uniformitarian geodynamo
(see also Swanson-Hysell et al. 2009), a burden of proof must lie
with these alternatives for any specific age.
Palaeomagnetic constraints on Neoproterozoic glacial
palaeolatitudes
Laurentia and environs
Macdonald et al. (2010a) have produced direct, high-precision,
U –Pb age constraints on the Upper Mount Harper Group
diamictite-bearing succession, at 716.5 Ma, which is precisely
coeval with the immediately underlying Mount Harper Volcanic
Complex and the transcontinental Franklin large igneous province.
The latter has yielded a high-quality palaeomagnetic pole repre-
senting data measured in several laboratories and spanning the
width of Arctic Canada plus northwestern Greenland (Denyszyn
et al. 2009). The Mount Harper and underyling Fifteenmile
groups are well correlated, lithostratigraphically, to the Tindir
Group across the length of the Oglivie Mountains, and the Franklin
palaeomagnetic pole implies a depositional palaeolatitude of
21+38 for the Tindir/Mount Harper glaciogenic succession. We
assign this result to the highest reliability in the present analysis,
and it is noteworthy by the fact that it derives from igneous
rocks that are immune from any possible effects of systematic
inclination shallowing. We assign only slightly less reliability to
the application of the Franklin mean palaeomagnetic pole to the
Rapitan glaciogenic deposits in the adjacent Mackenzie Mountains
(implied depositional palaeolatitude of 18 + 38) because that suc-
cession lacks the volcanic manifestation of the Franklin event.
Correlation among the Upper Tindir, Upper Mount Harper and
Rapitan diamictite successions is moderately strong, because of
their close associations with Fe formation, similarity of overlying
shale units, and reasonable matches of chemostratigraphic profiles
from underlying units (Macdonald et al. 2010a). The palaeolati-
tudes determined herein for these deposits are slightly higher
(c. 208) than the previous estimate (06 þ8/–78; Evans 2000),
Palaeolatitude (°)
Shareef (b)
Chuos Fiq
Area-normalized distribution
Schwarz.
Bayisi Grand C. Egan
Smalfj.
Khong. Mechum Yuerm.
Fauquier
Johnnie Rapitan Puga Gaskiers
Konnar. Elatina U.Tindir Nantuo Walsh Squant.
0 10 20 30 40 50 60 70 80 90
Fig. 7.1. Reliability of palaeomagnetic depostional latitudes for Neoproterzoic
glacial deposits. (a) Unit-weight given to each estimate, following Evans
(2000). Thick lines indicate the normalized probability at each latitude band, for
a uniform (random) distribution over the surface of the Earth. (b) Latitude
band-area normalization of the palaeolatitude estimates.
which was determined by direct measurements of the Rapitan suc-
cession that apparently were substantially affected by sedimentary
inclination shallowing (implied flattening factor, f  0.3 calcu-
lated here; see Tauxe et al. 2008). Nonetheless, general concor-
dance of 750– 700 Ma palaeomagnetic results from Laurentia
(Fig. 7.2) indicates that the Earth’s geomagnetic field was stable
at that time. Distinctly higher in the Tindir stratigraphy, a second
diamictite level (unit 3b) and overlying cap carbonate bear striking
resemblance to the Stelfox/Ravensthroat diamictite/cap carbon-
ate couplet in the nearby Mackenzie Mountains (Macdonald et al.
2010b). Neither of the younger diamictite/cap carbonate pairs at
the two localities have palaeomagnetic constraints on depositional
palaeolatitude.
Inliers on Alaska’s North Slope belong to an Arctic Alaska –
Chukotka microplate of questionable Laurentian or Siberian affi-
nity. Regionally, diamictite (informally named Hula Hula) lies
above post-760 Ma volcanic rocks, and locally, it is interbedded
with basaltic flows (Macdonald et al. 2009a). The overlying
Katakturuk dolomite contains a diagnostic unit (Katakturuk 2)
considered correlative to classic Marinoan cap carbonates else-
where in the world. Although the Hula Hula-associated flows
might correlate to the Franklin event and merit application of
that palaeomagnetic pole, the allochthonous nature of the terrane
and uncertain timing and style of its accretion to Alaska render
these deposits palaeomagnetically unconstrained.
New U –Pb ages on zircon from Neoproterozoic successions in
Idaho support several distinct pulses of glaciation in that region. In
the Pocatello region, Fanning & Link (2004) produced U –Pb
SHRIMP ages constraining the Scout Mountain Member (mainly
diamictite) of the Pocatello Formation, to slightly younger than
709 + 5 Ma and substantially older than 667 + 5 Ma. The
younger age was determined using a zircon subpopulation from
a channelized tuff horizon that lies closely beneath a crystal fan-
bearing marble. The fans resemble those in post-glacial cap
carbonates. It was suggested that the Scout Mountain Member rep-
resents a distinct glacial event from the Edwardsburg metadiamic-
tites dated by Lund et al. (2003) at c. 685 Ma. More recently,
however, the older age has been ‘corrected’ to 686 + 4 Ma by
 GLOBAL NEOPROTEROZOIC GLACIAL PALAEOLATITUDES
17a/
18a
1a
2a
15a
3a
7 however, we distinguish the Grandfather Mountain and Konnarock
9/10
14a
11a
13
Fig. 7.2. Palaeoequators plotted across Laurentia, according to the named
palaeomagnetic poles from 750 to 700 Ma, as discussed in the text and in Evans
(2000). Numbered stars correspond to glaciogenic deposits keyed to Table 7.1.
Star colours are keyed to the reliability scheme of Figure 7.1, or if uncoloured,
stars represent deposits typically correlated with the Rapitan glaciation.
References to palaeoequators given in Evans (2000) except for the recently
updated Franklin LIP mean (Denyszyn et al. 2009).
the same authors (Fanning & Link 2008). These dates strengthen
the evidence for glaciogenic deposits in the Cordilleran rift
basins that are younger than 700 Ma. Correlations across the Cor-
dilleran margin of Laurentia have emphasized the close association
of allegedly glaciogenic diamictites and volcanic rocks in that time
frame. Evans (2000) suggested that all of these deposits could be
correlative, with an age of c. 740–720 Ma. In that case, palaeo-
magnetic poles from the c. 720 Ma Franklin event could be used
to infer palaeolatitudes of the Cordillera (applying the GAD field
assumption and a rigid Laurentian plate). However, new data
suggest some of the deposits of this diamictite –volcanic associ-
ation are demonstrably younger, and therefore applications of
the Franklin palaeomagnetic pole to undated Cordilleran deposits
are tenuous. For these reasons, the Toby Formation’s deep-tropical
palaeolatitude estimate (Evans 2000) is no longer considered to
be valid.
Elsewhere in western Laurentia, the Johnnie Rainstorm
Member’s incised canyons and features reminiscent of postglacial
cap carbonates (Corsetti & Kaufman 2003) still retain the near-
equatorial palaeolatitude of 01 + 48 as assessed by Evans
(2000). Stratigraphically below that unit, but above the Kingston
Peak glaciogenic unit, the Ibex Formation contains a polymictic
conglomerate and overlying pink dolostone with negative carbon
isotope values similar to many postglacial cap carbonates (Corsetti
& Kaufman 2005). However, it is impossible to apply palaeomag-
netic constraints from better dated units elsewhere in Laurentia
without precise age constraints. A newly recognized glacial/cap-
carbonate succession has been described in Sonora, Mexico (Cor-
setti et al. 2007), lacking both a precise age and a palaeomagnetic
estimate of depositional palaeolatitude. Finally, although Evans
(2000) included the Florida Mountains (New Mexico) diamictite
99
in his summary figure on glacial palaeolatitudes, that unit is
demonstrably early Palaeozoic in age and is therefore excluded
from the present analysis.
In the central Appalachians of eastern Laurentia, pebbly mud-
stone diamictites of the Grandfather Mountain Formation lie
immediately above rhyolitic flows dated at 742 + 2 Ma; this suc-
cession has been correlated with the Konnarock Formation diamic-
tites and rhythmites with dropstones. In the Blue Ridge Mountains
of Virginia, the c. 730–700 Ma Mechum River succession also
contains glaciogenic features (Bailey et al. 2007). Evans (2000)
correlated all three successions for the sake of simplicity, and esti-
mated depositional palaeolatitudes based on a palaeomagnetic pole
from the Franklin LIP at c. 720 Ma. In the present compilation,
formations from the Mechum River succession, based on slightly
different ages as reviewed above. For the former units, we apply
a new palaeomagnetic pole from the Kwagunt Formation in the
Grand Canyon (Weil et al. 2004), which is precisely coeval at
742 + 6 Ma (Karlstrom et al. 2000). Assuming negligible rotation
of the Colorado Plateau relative to cratonic North America, as con-
sidered by Weil et al. (2004), this pole implies a palaeolatitude of
06 + 58 for the Konnarock/Grandfather Mountain region. Alter-
natively, a near-maximum estimate of Colorado Plateau rotation
of 128 anticlockwise would imply a palaeolatitude for Konnar-
ock/Grandfather Mountain at 04 + 58, within the error of the
no-rotation model. This estimate is rated as only moderately
reliable, because the dated Grandfather Mountain diamictites are
not yet conclusively demonstrated as glacially influenced, and
the presumed correlative Konnarock glaciogenic strata have
more lax age constraints.
The Mechum River succession remains within the age range of
the c. 720 Ma Franklin LIP, and the new grand-mean pole for that
event (Denyszyn et al. 2009), implies a depositional palaeolatitude
of 16 + 38 (assuming an internally rigid Laurentian plate and a
geocentric axial dipole geomagnetic field). Stratigraphically
higher in the succession, the Fauquier Formation contains a
basal diamictite member, followed by a thick sandstone interval
that is capped by a thin carbonate unit with a negative d13C excur-
sion; in turn, this unit is followed conformably by basalts of the
c. 570 Ma Catoctin Formation (Hebert et al. 2010). Palaeomag-
netic results from the latter volcanic succession include two rema-
nence components of greatly differing inclination, hence, implied
palaeolatitude (Meert et al. 1994). The coexistence of both high-
and low-inclination directions in these rocks is symptomatic of
many broadly coeval igneous formations in eastern North America,
leading to a fundamental debate on their general interpretation
(e.g. Meert & Van der Voo 2001; Pisarevsky et al. 2001). The high-
palaeolatitude Catoctin component is corroborated by other data
of generally higher quality than the low-palaeolatitude data from
the Iapetus margin of Laurentia (McCausland et al. 2007), so we
prefer its implied palaeolatitude of 78 + 128 for the conformably
underlying Fauquier Formation. This is the first near-polar esti-
mate for a Neoproterozoic glacial deposit among recent palaeo-
geographic syntheses, but it is only somewhat reliable for the
following reasons: (i) the Catoctin palaeomagnetic data remain
ambiguous with respect to the two remanence components, (ii)
the corroborating data from elsewhere in eastern Laurentia are
also not of the highest quality, (iii) although the contact between
the Fauquier Formation and the overlying basalts appears to be
conformable, the basal Fauquier diamictites lie as much as hun-
dreds of metres lower in the stratigraphy, and (iv) the 580–
570 Ma interval appears to harbour very rapid apparent polar
wander for Laurentia (McCausland et al. 2007), so absolute age
uncertainties of only a few million years can permit a wide
range of possible palaeolatitudes (tens of degrees) for sedimentary
basins on that continent.
In Scotland and stratigraphically correlative areas of northern
Ireland, the Port Askaig glaciogenic level is now generally corre-
lated to the Sturtian ice ages (Prave 1999; Condon & Prave 2000;
100 D. A. D. EVANS & T. D. RAUB
McCay et al. 2006), but with the uncertainty surrounding the
numerical age of Sturtian deposits, it is not possible to apply extra-
basinal palaeomagnetic data to address the palaeolatitude of Port
Askaig deposition. Older palaeomagnetic data obtained directly
from the Port Askaig succession were summarized by Evans
(2000) as unreliable, a conclusion followed here.
McCay et al. (2006) also identified a so-called Marinoan equiv-
alent in northern Ireland, comprising the Stralinchy and Reelan
Formations, based on the presence of diamictite, exotic clasts
interpreted as dropstones, and a cap carbonate with a negative
d13C anomaly. As with other putative Laurentian glaciogenic
deposits of this age (assumed to be c. 635 Ma by correlation; see
above), there are no available palaeomagnetic data from Laurentia
that can be applied either directly from the relevant sedimentary
basin or from elsewhere across the palaeocontinent.
The youngest purported Neoproterozoic glaciogenic unit in the
Scottish and northern Irish Dalradian stratigraphic succession is
the Loch na Cille boulder bed (and its correlative units). It
closely overlies the Tayvallich Volcanics that have been recently
re-dated at 601 + 4 Ma (Dempster et al. 2002). The two relevant
palaeomagnetic poles from Laurentia that approximate this age
are from the Long Range dykes of Newfoundland and Labrador,
dated by U – Pb at 615+2 and 614 þ 6/–4 Ma (Kamo et al.
1989; Kamo & Gower 1994) and the Grenville Dykes of southeast-
ern Ontario, dated at 590 þ 2/ –1 Ma (Kamo et al. 1995). Both
data sets contain an unusually large spread of inclinations and
hence implied palaeolatitudes of emplacement, confounding sim-
ple interpretations (reviewed by Hodych & Cox 2007; McCaus-
land et al. 2007). Low to moderate palaeolatitudes for the Long
Range dykes (Murthy et al. 1992) are supported by subsequent
work (McCausland et al. 2009, abstract only). For the younger,
Grenville Dykes, a positive baked-contact test demonstrates that
the steep palaeomagnetic B remanence found in some of the
dykes is primary (Murthy 1971; Hyodo & Dunlop 1993). If these
results withstand further scrutiny, then the Loch na Cille boulder
bed would have been deposited during a migration of Laurentia
from low to high latitudes.
For the East Greenland Neoproterozoic succession, Mac
Niocaill et al. (2008) report in abstract some preliminary palaeo-
magnetic data that would appear to indicate rapid motion of
Laurentia across palaeolatitudes. The upper Tillite Group glacio-
genic unit bears a high-palaeolatitude magnetic remanence direc-
tion, but further sample analysis is needed to confirm that result.
Baltica
Several new geochronological and palaeomagnetic constraints have
forced revision of Evans’s (2000) estimate of depositional palaeola-
titude of Neoproterozoic glaciogenic deposits on Baltica. First,
the Vestertana Group, in which the palaeomagnetically studied
Nyborg Formation lies between diamictite units of the underlying
Smalfjord Formation and overlying Mortensnes Formation, has an
imprecise palaeolatitude of 33 þ 14/– 128 (Torsvik et al. 1995).
The reliability of this estimate has been downgraded because recent
syntheses of Baltica’s apparent polar wander path (Iglesia-Llanos
et al. 2005) show the Nyborg pole to coincide with a c. 510 Ma
overprint pole from elsewhere in Baltica. Because Middle Cam-
brian Finnmarkian deformation is prevalent in the northernmost
Caledonides (Roberts 2003), there is a distinct possibility that
the Nyborg palaeomagnetic remanence is an overprint acquired
shortly before Finnmarkian folding, and thus not indicative of
depositional palaeolatitudes. In addition, given the possibility of
substantial age difference between the Smalfjord and Mortensnes
Formations (e.g. correlations by Hoffman & Li 2009), the
Nyborg pole is now considered as only applicable (if primary) to
the conformably underlying Smalfjord Formation.
Elsewhere in Baltica, depositional palaeolatitudes on Ediacaran
glacial deposits are similarly or even more poorly constrained.
New ages exist for the Moelv diamictites of the southern
Norwegian sparagmite basins: younger than 620 + 15 Ma detrital
zircons in a distally underlying quartzite (Bingen et al. 2005) or
perhaps even younger than 561 + 4 Ma based on Re – Os dating
of the immediately underlying Biri black shale (Hannah et al.
2007, abstract only). Nonetheless, the mid-Ediacaran apparent
polar wander path for Baltica is highly uncertain (Iglesia-Llanos
et al. 2005; Meert et al. 2007). Where Vilchitsy and Blon diamictite
units are in close stratigraphic proximity to c. 551 Ma Volhyn volca-
nic rocks of the southwestern Baltic craton, palaeomagnetic results
from the latter units imply a depositional palaeolatitude of 50 + 38
(A component of Nawrocki et al. 2004); however, nearly coeval
sedimentary rocks of the White Sea region (555.3 + 0.3 Ma;
Martin et al. 2000) yield a more convincingly primary palaeomag-
netic pole (Popov et al. 2002, 2005; Iglesia-Llanos et al. 2005) that
implies a palaeolatitude of 28 + 38 for the Vilchitsy/Blon region.
Any of these younger poles could also apply to the Moelv diamic-
tite and correlative units in the Caledonides, now dated as late
Ediacaran as noted above.
Ediacaran palaeogeography of Baltica is usually paired with
eastern Laurentia, but both cratons’ palaeomagnetic data appear
to indicate rapid motions that challenge uniformitarian plate-
tectonic interpretations (Abrajevitch & Van der Voo 2010;
McCausland et al. 2011). Amid these uncertainties, the mid –late
Ediacaran glacial palaeolatitudes of Baltica must be considered
unresolved at present.
Altaids and Siberia
Chumakov (2009) has summarized the terminal Proterozoic to
Cambrian stratigraphy across Kazakhstan and Kyrgyzstan, dis-
tinguishing between two prominent glacial levels (Satan, Rang,
or Dzhetym Formations below, and Baykonur above). Both
levels are younger than the c. 700 Ma volcanic complexes lying
unconformably below. He also has described the Zabit Formation,
in the East Sayan region, as glaciogenic and correlative with the
Baykonur level, which is in turn correlated with the uppermost
glacial level in Tarim (Hankalchough Formation; see below). A
complex palaeomagnetic data set was obtained from the succes-
sion c. 1000 m or more above the Zabit diamictites into Cambrian
strata (Kravchinsky et al. 2010), which may be useful for tectonic
reconstructions but shed no direct light on glacial palaeolatitudes.
Two glaciogenic levels are also now recognized in the Dzhab-
khan region of western Mongolia (Macdonald et al. 2009b): the
basal, Maikhan Ul Member of the Tsagaan Oloom Formation,
and the Khongoryn Member in the middle of the same formation.
Chemostratigraphy of carbon isotopes in this succession supports
correlation with the mid-Cryogenian (so-called Sturtian) and late
Cryogenian (Marinoan) ice ages worldwide. Kravchinsky et al.
(2001) found several components of magnetization through
mainly the middle and upper parts of the Tsagaan Oloom
Formation, and preferred a high-unblocking-temperature, low-
palaeolatitude component as tentatively primary. That component,
although of dual polarity, is not supported by field stability or
reversals tests on the magnetization age, so is only somewhat
reliable. Levashova et al. (2010) found a high-unblocking-
temperature component of magnetization in the underlying
Dzabkhan volcanic rocks, with U –Pb laser ablation ages of
c. 800– 770 Ma, with moderate palaeolatitudes of igneous
emplacement. This estimate, however, would precede Maikhan
Ul glaciogenic deposition by nearly 100 million years if the
latter is correlated to the Rapitan ice age as discussed above
(Macdonald et al. 2010a).
New recognition of late Neoproterozoic glaciation in southern
Siberia has focused on the Yenisei ridge and the pre-Sayan
region NW of Irkutsk. Sovetov (2002) and Sovetov & Komlev
(2005) describe regional stratigraphic correlations around two dis-
tinct levels of diamictites with various glaciogenic features. The
older Chivida unit of the Chingasan Group is more limited in geo-
graphic distribution in the Teya-Chapa trough (Yenisei region)
 GLOBAL NEOPROTEROZOIC GLACIAL PALAEOLATITUDES
relative to the younger Pod’em Formation that is correlated to the
Marnya Formation best exposed along the Biryusa and Uda Rivers
(pre-Sayan). The Chivida diamictites are closely associated with
volcanic rocks and are traditionally associated with an age of
c. 700– 750 Ma (Khomentovsky 1986), but they are not directly
dated with modern methods. The Marnya Formation contains
clasts that are speculated to derive from the Nersa Complex, a
nearby mafic intrusive suite, according to Sovetov & Komlev
(2005). Those authors quoted an Ar/Ar age of 611 + 3 Ma for
the Nersa Complex, but that age (now revised to 612 + 6 Ma),
in fact, derives from other mafic intrusive suites in southern
Siberia (Gladkochub et al. 2006); thus the Marnya Formation dia-
mictites lack any precise age constraints.
In the Patom region NE of Lake Baikal, at least two widespread
glaciogenic levels have been reported: the older Kharlaktakh unit
of the Ballaganakh Group and the younger Dzhemkukan or
Bol’shoy Patom Formation of the Dal’nyaya Taiga Group (Pok-
rovskii et al. 2006; Sovetov 2008). No results from these succes-
sions are available in the global palaeomagnetic database.
The late Neoproterozoic to Early Cambrian palaeomagnetic
apparent polar wander path for Siberia is highly contentious
with a wide scatter of poles that universally imply low palaeolati-
tude for the craton; and if one includes the full range of data
(Kirschvink & Rozanov 1984; Kravchinsky et al. 2001; Gallet
et al. 2003; Metelkin et al. 2005) rather than a simplified selection
(Cocks & Torsvik 2007), the problem is compounded further.
China cratons
Recent work has helped refine the chronology and palaeomagnet-
ism of Neoproterozoic volcanic-sedimentary successions on the
margin of the Tarim block in NW China. Two principal regions
contain diamictites interpreted as glaciogenic deposits: Aksu in
the west, and Quruqtagh in the east. The Aksu succession contains
two diamictite levels, and a coarse magnetostratigraphic study by
Li et al. (1991) yielded near-equatorial palaeolatitudes throughout
the succession. More recently, Zhan et al. (2007) conducted a more
detailed palaeomagnetic investigation of only the Sugetbrak For-
mation, which rests with disconformity upon the upper diamictite
level (Yuermeinbrak Formation). Although Zhan et al. (2007)
suggested that their high-stability, dual-polarity direction for the
upper Aksu succession, which was acquired prior to folding of
unspecified age, implies a palaeolatitude of c. 278 for the
Yuermeinbrak ice age, there is lingering concern over why the
palaeomagnetic results by Li et al. (1991) and Zhan et al.
(2007), in part from the same sampling region on the same strati-
graphic units, are so different. We tentatively accept the new con-
straint as more reliable, because the pole by Li et al. (1991) is
similar to Cretaceous poles from Tarim and could represent an over-
print of that age.
Farther east in the Tarim block, the Quruqtagh region contains
three Neoproterozoic diamictite levels: Bayisi, Altungol-Tereeken
and Hankalchough. The oldest level lies immediately above
bimodal volcanic rocks dated by U –Pb SHRIMP at 740 + 7 Ma
(previously 755 + 15 Ma from the same sample; Xu et al.
2005), and below volcanic rocks dated at 725 + 10 Ma (Xu
et al. 2009). The proximity of ages between volcanism and diamic-
tite deposition allows the use of palaeomagnetic data from the
former rocks (Huang et al. 2005) to constrain the latitude of glacia-
tion, assuming the Bayisi diamictites are glaciogenic. That palaeo-
magnetic work found a characteristic remanence component of
high consistency but no field tests were performed on its stability;
it is therefore unclear whether its near-equatorial palaeolatitude
(01 þ 4/ –28) is primary or secondary. It is rated here as only
somewhat reliable. The Altungol-Tereeken diamictites, overlain
by a notable pink cap carbonate, lie above the 725 Ma volcanic
rocks and below another volcanic horizon dated at 615 + 6 Ma
(Xu et al. 2009). The Altungol-Tereeken interval has not been
studied palaeomagnetically. In summary, both of the two recent
101
palaeomagnetic studies from Tarim demand further work to
verify their applicability to Neoproterozoic glacial palaeolatitudes.
In the Qaidam terrane near the boundary between the Tarim and
North China cratons, the Hongtiegou diamictite and overlying cap
carbonate have been correlated, via stable isotope stratigraphy, to
the uppermost Ediacaran (Sinian) successions in South China
(Shen et al. 2010). Palaeomagnetic data are not available from
the Neoproterozoic of Qaidam.
Chronology of the Sinian sedimentary basins in South China has
improved vastly during the past few years. Two glacial levels are
described with various local names, but generally referred to as the
older Chang’an (or Tiesiao, or Jiangkou) episode and the younger
Nantuo ice age. The Chang’an level is now bracketed in age
between 725 + 10 Ma (youngest U – Pb SHRIMP population
from detrital zircons in the underlying Danzhou Group; Zhang
et al. 2008) and 663 + 4 Ma (Zhou et al. 2004). The Nantuo
diamictite is younger than 654.5 + 3.8 Ma and contains a basal
tuff layer dated at 636.3 + 4.9 Ma (Zhang et al. 2008), with a
cap carbonate dated at 635.2 + 0.6 Ma (Condon et al. 2005).
Prior to the acquisition of these ages, Evans (2000) assigned a
moderate depositional palaeolatitude to the Nantuo Formation
based on results from the immediately underlying Liantuo
redbeds (Evans et al. 2000). However, given that the Liantuo For-
mation contains an ash bed dated at 748 + 12 Ma (Evans et al.
2000), the new dates from South China now preclude use of the
Liantuo pole to constrain the significantly younger ice ages
(Zhang et al. 2009).
Macouin et al. (2004) conducted a palaeomagnetic study on the
Doushantuo Formation, which conformably overlies the Nantuo
diamictites across most of South China. They obtained consistent
results throughout the entire formation, implying near-equatorial
palaeolatitudes, except for the postglacial cap carbonate at its
base, which was palaeomagnetically unstable. Given that the
Doushantuo Formation is now constrained to span more than 80
million years of time (Condon et al. 2005), and that the direction
obtained by Macouin et al. (2004) is of single geomagnetic
polarity, we suspect that it is a secondary overprint. The positive
fold test only constrains the age of magnetization to pre-Mesozoic,
and the direction is similar to both Early Cambrian and Silurian
directions for South China. Zhang et al. (2009) correct numerous
misconceptions about the stratigraphy and palaeomagnetic data-
base for the Sinian of South China, and the best estimate for
depositional palaeolatitude of the Nantuo diamictite reverts to
the value obtained by Zhang & Piper (1997) from Yunnan Pro-
vince, at 37 + 98. An identical result has been obtained by
Zhang et al. (2006, abstract only) from Guizhou Province, support-
ing this conclusion. The soft-sediment fold test reported in the
earlier study (Zhang & Piper 1997) imparts a high level of confi-
dence to the moderate-palaeolatitude determination.
India to Nubia
Although Evans (2000) speculated that conglomerates within
the Penganga Group might be glaciogenic, more thorough study
of that region suggests no demonstrable glacial influence, as is
evident, for example, in the discussions by Mukhopadhyay &
Chaudhuri (2003) and Chakraborty et al. (2010). Other Neoproter-
ozoic glaciogenic deposits in southern Asia are listed in Table 7.1.
In Iran, the so-called Infracambrian stratigraphic succession,
between Neoproterozoic metamorphic basement and overlying
nonmetamorphosed Cambrian sedimentary rocks, is variable in
lithology but contains diamictite and lonestone-bearing units in
its lower part, named the Rizu Formation (Huckriede et al. 1962;
Hamdi 1992). The upper part of the succession is equally litho-
logically variable, as currently correlated across the country
(Alsharhan & Nairn 1997), but contains rhyolites thought to be
consanguineous with c. 530 Ma intrusions (Ramezani & Tucker
2003). No reliable palaeomagnetic data have yet been published
on the Rizu strata or correlative rocks.
102 D. A. D. EVANS & T. D. RAUB
Despite a new, precise U–Pb TIMS age of 711.5+ 1.1 Ma for
the Ghubrah diamictite in the Jebel Akhdar region of northeastern
Oman (Bowring et al. 2007; all sources of error included), its palaeo-
latitude remains poorly constrained. However, unmetamorphosed
deposits of the possibly Ghubrah-correlative, lower Mirbat sand-
stone/diamictite assemblage, now renamed as the Ayn Formation
(Rieu et al. 2006, 2007), yield a low-reliability palaeomagnetic
direction (Kempf et al. 2000). This under-defined magnetization
(n ¼ 10 samples from two sites) would imply a palaeolatitude of
09 + 48 for the Ayn Formation, but we cannot include it in our
final compilation due to the severely limited sample size.
On the younger and less precisely dated Fiq diamictite and its
overlying Hadash cap carbonate, a two-polarity, pre-fold palaeo-
magnetic remanence implies a palaeolatitude of 13 + 78 for the
Jebel Akhdar region, and 18 + 78 for the Mirbat region (Kilner
et al. 2005), the latter area hosting the erosional remnants of an
upper diamictite unit (Shareef Formation) that is correlated with
the Fiq (Rieu et al. 2007). The folding is probably Palaeozoic in
age, suggesting that the remanence could be primary. However,
Rowan & Tait (2010, abstract only) have undertaken a more
detailed magnetostratigraphic study of the Hadash cap carbonate
in two sections with greatly differing structural attitudes, and
have found the same remanence direction to be definitively post-
folding, and possibly associated with regional remagnetization
during emplacement of the Semail ophiolite. Given that only
about 20% of the samples in the study by Kilner et al. (2005)
yielded the preferred remanence direction, and only a few of
their sites contributed significantly to their fold test, the low-
palaeolatitude estimates for Fiq and Shareef diamictites can only
be considered as somewhat reliable. Allen (2007) points out
additional problems with the published data from Oman, in
terms of their stratigraphic correlations and tectonic implications.
In the northern East African Orogen of Ethiopia, the Tambien
Group contains a glaciogenic/cap-carbonate succession that
has only recently been described in detail (Beyth et al. 2003;
Miller et al. 2003). The Tambien Group is tightly folded into struc-
tures that are crosscut by the Mereb granite suite, dated by zircon
Pb-evaporation on two plutons at 606.0 + 0.9 and 613.4 + 0.9
Ma (Miller et al. 2003), and by zircon U –Pb SHRIMP at
612.3 + 7.5 Ma (Avigad et al. 2007). Detrital zircons in the
Negash diamictite have yielded near-concordant U – Pb SHRIMP
analyses as young as c. 750 Ma (Avigad et al. 2007; all younger
diamictite detrital ages quoted in that work are more than 10%
discordant). No palaeomagnetic data are available for the
Tambien Group. Farther to the north in the Arabian –Nubian
Shield, widespread banded Fe-formation is associated with dia-
mictites of the Atud and Nuwaybah Formations (Stern et al.
2006; Ali et al. 2009).
Australia and Mawsonland
The palaeolatitude assessments by Evans (2000) remain valid for
most of the many glaciogenic deposits scattered across the late
Neoproterozoic glacial record of Australia. Ediacaran stratigraphic
correlations in Australia are now under intense debate, following
the recognition of the possibly glaciogenic Croles Hill Diamictite
in Tasmania coincident with 582 + 4 Ma rhyolite volcanism
(Calver et al. 2004). Whether those diamictites correlate to the
Elatina Formation in South Australia (by way of the Cottons
Breccia on King Island), or whether they are substantially
younger than Elatina, carries implications as profound as a factor-
of-two uncertainty in the duration of the Ediacaran Period that
begins, as formally defined, with deposition of the post-Elatina,
Nuccaleena cap carbonate. No published palaeomagnetic data
are available for the Tasmanian diamictites, but data are reported
in abstract for the Cottons Breccia, with an implied palaeolatitude
of 21 þ 9/ –88; and the subsequent igneous activity dated at
c. 580 Ma (Meffre et al. 2004), with implied palaeolatitudes
in the range of c. 12 –218 (McWilliams & Schmidt 2003) that
could be representative of Croles Hill deposition. As these data
are reported in abstract only, they are omitted from our final
compilation.
As for the Elatina glacial deposits and their continent-wide cor-
relative units, new palaeomagnetic results add to an already long
history of palaeomagnetic studies of this unit (reviewed by
Evans 2000; Williams et al. 2008). Schmidt et al. (2009) presented
a detailed analysis of palaeomagnetic data from both the Elatina
and Nuccaleena formations in the Flinders Ranges, combining
new results with previously published work. Based on moderate
differences in remanent inclination between the two formations,
they concluded that deglaciation coincided with rapid drift of the
proto-Australian continent during a hiatus in deposition between
the two units. The observed mean remanent inclination of 12.98
from the Elatina Formation would correspond to a depositional
palaeolatitude of 6.58 if the geomagnetic field was dipolar and
sedimentary inclination shallowing was negligible. For the Nucca-
leena Formation, the mean inclination of 34.98 would correspond
to a palaeolatitude of 19.28. Schmidt et al. (2009) also reported
results from four measurements designed to quantify the effects
of remanence shallowing in the sedimentary rocks. In the first
measurement, a correction for their measured values of remanence
anisotropy would increase the Elatina inclination to 14.08;
however, Tauxe et al. (2008) showed that this method commonly
underestimates the total amount of inclination shallowing. Indeed,
two other measurements by Schmidt et al. (2009) appear to justify
such an assessment. Their second test, elongation/inclination
analysis on the distribution of data around the mean direction
(using the assumed TK03 model of geomagnetic secular vari-
ation), would correct the Elatina inclination to 198 (bootstrap-
generated uncertainties were not shown but are typically large
for such tests). The third analysis, a dip test on crossbedded sand-
stones in the succession, would correct the Elatina inclination to
between 18.4 and 20.28. Because sandstones should be the least
compacted of all the Elatina siliciclastic facies, this correction of
the aggregate palaeomagnetic results include data from both sand-
stones and siltstones. The fourth test considered differences in
inclination between carbonates and mudstones in the Nuccaleena
Formation, the latter of which are expected to be more compacted
than the former. The carbonates’ mean inclination was 33.68
(a95 ¼ 7.78), substantially greater than the mudstones’ mean incli-
nation of 21.18 (a95 ¼ 7.88). Schmidt et al. (2009) claimed that
these two means (taking into account both declination and incli-
nation) were indistinguishable, using a test from McFadden &
Lowes (1981). However, the test they used from that paper
(equation 25) assumes an equivalency in concentration parameters
between the two sample sets, which for the Nuccaleena data can be
rejected with .99% confidence (Fisher et al. 1987, p. 219). If the
correct test is used (the more general equation 23 from the same
paper), the Nuccaleena data-derived statistic increases to 4.40,
exceeding the p ¼ 0.05 F[2,76] value of 3.12; thus, the carbonates’
and mudstones’ inclinations are significantly distinct at more than
95% confidence, and the Nuccaleena data therefore do support
inclination shallowing having affected any fine-grained sedimen-
tary rocks contributing to the mean directions in the sampling
area (contra Schmidt et al. 2009).
In summary, the various inclination corrections to the Elatina
palaeomagnetic data, as reported by Schmidt et al. (2009), imply
a ‘true’ inclination value around 208, with an implied palaeolati-
tude of approximately 108 for the sampled areas in the Flinders
Ranges. Nuccaleena inclinations are steeper, due to the less-
compacted carbonate lithology and perhaps also partly due to Aus-
tralian plate motion between the ice age and the deglacial interval.
Raub (2008) found a mean inclination of 278 (a95 ¼ 3.58) for the
Nuccaleena Formation, with several correlatable magnetozones
spanning geodynamo reversals during deposition. That inclination
value indicates a primary palaeolatitude of 14 + 28, between the
estimates by Schmidt et al. (2009) for their compaction-corrected
 GLOBAL NEOPROTEROZOIC GLACIAL PALAEOLATITUDES
Elatina and Nuccaleena data. Rare stratigraphic sections preserve
sedimentological conformity (mixed transition) between siliciclas-
tic rocks indistinguishable from Elatina Formation sandstones and
Nuccaleena Formation cap-carbonate beds (Raub et al. 2007a),
and the onlap relationships between the two formations described
by Schmidt et al. (2009) can be accounted for by sediment draping
of an irregular post-glacial topography, perhaps with diachronous
onset of cap-carbonate deposition, without a substantial lacuna
between the formations (Rose & Maloof 2010). For this reason,
we propose that 10–148 of palaeolatitude best characterizes
the Elatina/Nuccaleena glacial/postglacial palaeogeography.
The 10–208 palaeolatitude range (as indicated in Fig. 7.1) also
well represents the distribution of likely Elatina-correlative depos-
its across the Australian –Mawsonland palaeocontinent (Fig. 7.3).
The compaction-corrected Elatina/Nuccaleena result is con-
sidered here to be one of the two most reliable palaeolatitude deter-
minations among all Neoproterozoic glaciogenic deposits.
Grey et al. (2005) produced a landmark correlation of Neopro-
terozoic strata among deep boreholes throughout the Officer Basin
of Western Australia, using diamictites and other indicators. New
palaeomagnetic data from the Wahlgu diamictite level (correlated
with the Elatina Formation) in the Lancer-1 borehole (Pisarevsky
et al. 2007) generate a palaeolatitude of 7 + 178, but the data
are highly scattered and merely consistent with the range of pre-
vious results from the Elatina Formation. Haines et al. (2008)
revise the correlation of one diamictite and dropstone-bearing
unit (therein named the Pirrilyungka Formation) to an age sub-
stantially older than the Wahlgu and Turkey Hill Formations,
perhaps correlating with the Chambers Bluff diamictite in South
Australia.
In the Kimberley region of northern Australia, the cap carbonate
of the Walsh Tillite (western Kimberley) has yielded a moderate-
palaeolatitude magnetic remanence supported by a tectonic fold
test and two polarities of magnetization (Li 2000). Although the
Walsh Tillite is variously correlated with either the Sturtian or
Marinoan glacial level across Australia (see Corkeron 2007 for a
summary), that issue is irrelevant for the reliability of the palaeo-
magnetic pole and its implications for deglaciation of the Walsh
ice age. We accept the correlation by Corkeron & George (2001)
of the Egan glaciogenic deposit (central Kimberley) with the
Boonall dolomite in eastern Kimberley, and by further correlation
using stromatolites, the mid-Ediacaran Julie Formation in central
Australia. A palaeomagnetic pole from the latter region (Kir-
schvink 1978) implies a depositional palaeolatitude of 21 + 88
for the Egan, assuming that the pole is primary and the correlations
are correct.
103
Figure 7.3 shows two possible palaeogeographies of Australia
during the time of terminal Cryogenian glaciation. The first rep-
resentation (Fig. 7.3a) retains coherence of cratonic Australia in
present-day coordinates, with palaeoequators according to the
various palaeomagnetic studies of the Elatina or Nuccaleena for-
mations. The second (Fig. 7.3b) incorporates a new model of Pro-
terozoic Australia, with the northern part of the continent displaced
from the western and southern parts by a local Euler rotation angle
of 408 (Li & Evans 2011). In this model, the palaeoequator deter-
mined from the Walsh Tillite cap carbonate (Li 2000) returns to a
position that is much closer to the palaeoequator attributed to the
Nuccaleena cap carbonate in South Australia.
Southern Australia may also have been glaciated in the mid-
Ediacaran interval. Gostin et al. (2011) describe possible ice-
rafting features in the Bunyeroo Formation and equivalent units,
which are demonstrated to have a pre-folding palaeomagnetic
remanence that was acquired about 158 from the equator. In
addition, Jenkins (2011) proposes a glacial influence on strata of
the Billy Springs Formation higher in the stratigraphy.
Most tectonic syntheses of the region formerly known as East
Gondwanaland now subdivide the cratonic areas of Antarctica
into (at least) three fragments that collided during Cambrian
time (e.g. Fitzsimons 2000). The Mawson Continent, or Mawson-
land, comprises South Australia and adjacent regions of Antarctica
in a pre-breakup Gondwanaland fit; those regions extend much of
the length of the Transantarctic Mountains, from Victoria Land
toward the South Pole. The Goldie Formation (redefined by
Myrow et al. 2002) contains diamictites of uncertain depositional
setting. Although no estimates of palaeolatitude are available from
the Goldie Formation in the Transantarctic Mountains, its age is
now tightly constrained by U –Pb on zircon, 668 + 1 Ma, from
interleaved mafic (meta)volcanic rocks (Goodge et al. 2002).
Kalahari and environs
Neoproterozoic glaciogenic deposits on the Kalahari craton are
restricted to its present western margin, where as many as two
diamictite levels are preserved in any single succession. In
addition, several levels of canyon incision with diamictic infill
are present within the Schwarzrand Subgroup of the Nama
Group, spanning the Ediacaran –Cambrian boundary (reviewed
by Evans 2000). No new palaeomagnetic constraints are available
for any of the pre-Schwarzrand glacial levels, and the Schwarzrand
palaeolatitude of 38 + 38 (implied by the pole from the
547 + 4 Ma Sinyai metadolerite on the Congo craton; Meert &
Fig. 7.3. (a) Palaeoequators plotted across
Australia and Mawsonland, according to
various palaeomagnetic studies of the
Elatina Formation, or, as indicated, of the
Nuccaleena Formation (Nucc). (b)
Restoration of Ediacaran tectonic rotation
between the northern and western/southern
portions of cratonic Australia, following Li
& Evans (2011). For star legend, see
caption to Figure 7.2 (uncoloured stars
representing deposits typically correlated
with the Elatina glaciation. Palaeoequators
are referenced as follows: EW86, Embleton
& Williams (1986); SWE91, Schmidt et al.
(1991); SW95, Schmidt & Williams (1995);
SCBK, Sohl et al. (1999); R08, Raub
(2008); SWM09, Schmidt et al. (2009);
Walsh, Li (2000).
104 D. A. D. EVANS & T. D. RAUB
Van der Voo 1996) is assigned only low reliability because, despite
excellent age constraints and the consistency of earliest Cambrian
Gondwanaland palaeomagnetic poles from a variety of areas
within the megacontinent (Mitchell et al. 2010), it remains
unclear whether the Schwarzrand succession is truly glaciogenic.
Among older glaciogenic strata of the Gariep belt, Macdonald
et al. (2010c) have revised correlations to distinguish two separate
stratigraphic levels among outcrops previously lumped together
as the Numees Formation. In this new correlation, the name
‘Numees’ is retained for an older level associated with an organic-
rich cap carbonate typical of ’Sturtian’ strata worldwide, whereas a
new name, ‘Namaskluft’, is introduced for a younger level asso-
ciated with a light-coloured peloidal cap carbonate typical of
‘Marinoan’ strata in other regions. Kaufman et al. (2009) review
these successions and present an unconventional set of correlations
with the global stratigraphic record, but it is not clear how such
correlations might change according to the proposed distinction
between Numees and Namuskluft glacial horizons. Farther south
in the parautochthonous Gariep belt stratigraphy, the Aties For-
mation as listed by Evans (2000) has been subdivided into
several new units, which include two glaciogenic levels recog-
nized by Frimmel (2011): the lower Karoetjes Kop Formation,
and the upper Bloupoort Formation. The former is correlated
with the Kaigas diamictites in the Gariep belt to the north,
whereas the latter is correlated to the Numees diamictites (s.l.),
although further work is required to discern whether Bloupoort
would correspond to Numees or Namuskluft, as distinguished by
Macdonald et al. (2010c).
In the allochthonous Marmora terrane of the Gariep Belt,
Namibia, diamictites and dropstone-bearing stratigraphic units
are identified within the Dernburg Formation. These deposits are
interpreted as having been deposited on a seamount in the Adamas-
tor Ocean that lay between the Kalahari and Rio de la Plata cratons
prior to their collision (Frimmel & Jiang 2001; Frimmel et al.
2002). No palaeomagnetic data are available for the Marmora
terrane, although extensive carbonates and evaporites in the suc-
cession suggest low to moderate palaeolatitudes of deposition.
Gaucher et al. (2009) included the Marmora terrane within a
crustal block they named ‘Arachania’, inferred to have transferred
from the Rio de la Plata craton to the Kalahari craton between
c. 630 and 550 Ma.
Congo – São Francisco
The Katangan succession in east-central Africa contains two pro-
minent diamictite horizons with overlying cap carbonates,
named the Grand Conglomerat and Petit Conglomerat, in strati-
graphic order. U– Pb SHRIMP ages of 765–735 Ma from volca-
nic strata, interbedded with diamictites that are correlated with
the Grand Conglomerat (Key et al. 2001), overlap with a
748 + 6 Ma date on the syenitic phase of the Mbozi complex
from the same craton (Mbede et al. 2004, abstract only). This
allows the Grand Conglomerat’s depositional latitude to be esti-
mated indirectly at 10 + 58, using the near-coeval Mbozi palaeo-
magnetic pole (Meert et al. 1995). This estimate is considered
more reliable than the same pole applied for the Chuos Formation
in northern Namibia, which might be substantially younger than
the c. 750 Ma Mbozi pole. Both of these units are included separ-
ately in our new compilation because of the c. 1000 km lateral dis-
tance between their exposures, as well as the possibility that they
are not precisely coeval. Our new estimate of Chuos depositional
palaeolatitude differs slightly from that of Evans (2000) due to a
refined location for the centre of Chuos-correlated strata in the
northern and central Damara belt. As for the younger portion of
these successions, there is still no reliable palaeomagnetic con-
straint on that unit, despite the recent U –Pb TIMS age constraint
of 635.5 + 1.2 Ma on a likely correlative of the Ghaub Formation
in the internal sector of the Damara orogen (Hoffmann et al. 2004;
but see also Kaufman et al. 2009).
Several recent papers provide additional insight on correlations
of diamictite units across the present eastern and northern edges
of the Congo – São Francisco craton. Melezhik et al. (2006)
provide stable isotopic data bearing on regional and global strati-
graphic correlations of the Geci metacarbonates in northern
Mozambique, which are associated with metadiamictites that may
be glaciogenic (Pinna et al. 1993). To the NW, Poidevin (2007)
reviews stratigraphic data from the Oubanguide belt in central
Africa, whereas Caron et al. (2010) report the discovery of a
new diamictite cap-carbonate succession in southern Cameroon.
Lack of precise ages on all of these units, as well as a general
dearth of late Neoproterozoic palaeomagnetic data from the
Congo craton, prohibit assignment of depositional palaeolatitudes.
Continuing around the São Francisco side of the craton,
in Brazil, the various lithostratigraphic names for late Neoprotero-
zoic diamictites are well summarized by Misi et al. (2007) and Sial
et al. (2009). As many as two or perhaps three glaciogenic levels
are known from each area, with few precise age constraints and
thus, many possible correlations across the region (see also
Kaufman et al. 2009). A comprehensive summary of all these poss-
ible relationships is beyond the scope of the present paper, but
Table 7.1 splits the deposits into many distinct geographical and
tectonostratigraphic subdivisions, which later can be combined
as correlations gain reliability. In the cratonic autochthon, the
basal Bambuı́ cap carbonate (Sete Lagoas Formation) has
yielded a Pb/Pb isochron age of 740 + 22 Ma (Babinski et al.
2007), and the authors of that study infer a depositional palaeola-
titude according to c. 750 Ma palaeomagnetic data from Africa;
however, we consider the Sete Lagoas age constraints lax
enough to render its palaeolatitude assignment too tentative.
Another age constraint of particular note is an astounding Re –
Os black shale isochron in the 1100–1000 Ma range (Azmy
et al. 2008) for the Lapa Formation (uppermost formation in the
diamictite-bearing Vazante Group) that, if correct, would represent
the first discovery of a Mesoproterozoic ice age, anywhere in the
world. However, Rodrigues et al. (2008, extended abstract) find
,1000 Ma detrital zircons in all Vazante units, in direct contrast
to the Re –Os ages. As for palaeomagnetic constraints, the
Salitre cap carbonate of the Bebedouro Formation has only
yielded a Cambrian remagnetization (Trindade et al. 2004). No
reliable palaeomagnetic data are available for the other diamictites
of the São Francisco craton.
West Africa and Hoggar
Although Evans (2000), following earlier syntheses across the
Taoudeni basin, considered the possibility that the glaciogenic
‘triad’ succession is late Ediacaran to earliest Cambrian in age, a
more detailed study of that succession in the Volta basin (Porter
et al. 2004) provides compelling chemostratigraphic comparisons
with basal Ediacaran cap carbonates elsewhere in the world that
are dated at 635 Ma (South China and Namibia, see above). Fur-
ther, the assigned Early Cambrian age of the glaciogenic Hassanah
Diallo Formation in Senegal, based on a small shelly fossil assem-
blage in dolomite resembling cap carbonate, is now retracted with
the interpretation that the fossiliferous sample was collected from a
loose boulder fallen from overlying strata (Shields et al. 2007). For
these reasons, the indirect estimation of triad palaeolatitudes from
an Early Cambrian aggregate Gondwana Land apparent polar
wander path (Evans, 2000) now appears inappropriate. As dis-
cussed in Evans (2000), previous palaeomagnetic data from the
Taoudeni sedimentary succession are of questionable reliability.
However, a new result from one site of the ‘triad’ cap carbonate
in the Gourma region indicates a magnetic remanence that was
acquired prior to Pan-African folding in the area (Boudzoumou
et al. 2011). If primary, the result would imply a depositional
palaeolatitude of 08 + 78, but because of the small sample size,
the result warrants further study before it can be considered as
reliable as the other determinations shown in Figure 7.1.
 GLOBAL NEOPROTEROZOIC GLACIAL PALAEOLATITUDES
Alleged periglacial features from the base of the Atar Group
(Trompette 1994), which were discussed by Evans (2000) with
additional references, must now be considered as Mesoproterozoic,
given multi-sample, black shale Re –Os isochron ages of
c. 1105 Ma (Rooney et al. 2010) from somewhat higher levels
in the succession. The ages are reminiscent of those from the
Vazante Group and its diamictites, marginal to the São Francisco
craton (see above), but the Atar Re –Os isochrons have lower
mean square of weighted deviates values and thus, are more
reliable. The alleged periglacial influence on the basal Atar sedi-
ments should be tested by further study.
Despite new U –Pb ages from the Anti-Atlas Mountains of
Morocco, late Neoproterozoic glaciogenic deposits there remain
unconstrained in depositional palaeolatitude. The older Sarhro dia-
mictites in the Sirwa region were deformed and metamorphosed
c. 660 Ma (Thomas et al. 2002, 2004). Although Thomas et al.
(2004) advocate the traditional correlation of the Sarhro Group
with the Tiddiline succession in the Bou Azzer –El Graara inlier,
the latter diamictite-bearing succession contains boulder-sized
clasts of the Bleı̈da granodiorite now dated at 579.4 + 1.2 Ma
(Inglis et al. 2004). Therefore, the new stratigraphic constraints
appear to distinguish the two successions in age by c. 100
million years. No reliable palaeomagnetic data exist for the older
age interval, but the Tiddiline succession depositional palaeolati-
tude could be estimated by the immediately overlying, and pene-
contemporaneous (c. 575 –560 Ma; Thomas et al. 2002; Maloof
et al. 2005), Ouarzazate volcanic group. Unfortunately, the exist-
ing palaeomagnetic results from those rocks are of near-zero
reliability, as summarized by Tohver et al. (2006).
Avalonia and Cadomia
The Gaskiers glaciation is now dated by U –Pb TIMS to c. 580 Ma
(Bowring et al. 2002, abstract only), and serendipitously, the age of
the Marystown volcanic rocks (used by Evans 2000 to estimate
Gaskiers depositional palaeolatitudes) has also been revised to
c. 585–575 Ma (McNamara et al. 2001). The concordance of
these ages increases the likelihood that the Marystown palaeolati-
tude of 34 þ 8/– 78 applies to the Gaskiers deposit. It is considered
here as only moderately reliable as an indicator for global palaeo-
climate models, because the Gaskiers unit is geographically
limited and of short-lived duration (Bowring et al. 2002), and
thus, it may represent only local, alpine glaciation in an active-
margin tectonic setting.
In the Boston basin, the Roxbury Conglomerate has yielded a
magnetization in nonglacial red slates bracketing the glaciogenic
Squantum Member, which is consistent with that of unconform-
ably underlying volcanic rocks, and partially supported by a posi-
tive conglomerate test on those volcanic rocks (Fang et al. 1986).
Further palaeomagnetism and geochronology of this same succes-
sion (Thompson et al. 2007) has found that the c. 600– 595 Ma,
pre-Squantum, Mattapan volcanic rocks yielded a distinct magne-
tization that had not been recovered previously, and the earlier
characteristic magnetization of Fang et al. (1986) was not repro-
duced. Thompson et al. (2007) reported a possibly primary com-
ponent from only one site of the Squantum unit, and did not
discuss the discrepancy between their results and those of the
earlier study. Because the Squantum Member is younger than
593 Ma (U– Pb on detrital zircons; Thompson & Bowring 2000)
but otherwise poorly constrained in age, it is possible that the
difference is due to plate motions and rotations of the Avalonian
arc, between deposition of the volcanic rocks and the overlying
glaciogenic succession. We continue to use the directly obtained
result of Fang et al. (1986) in our analysis, implying a depositional
palaeolatitude of 55 þ 8/– 78.
In the Anglesey region of Wales, an East Avalonian accre-
tionary ‘Mona’ complex includes heterolithic, cobble-supporting,
volcaniclastic mudstone (Gwna Group) of interpreted ophio-
litic affinity, plausibly representing ice-rafted debris of
105
Gaskiers-correlative age in an open ocean setting between 595
and 550 Ma (Kawai et al. 2008). The authors of the study inferred
a moderate depositional palaeolatitude from the presumably
adjacent Avalonian arc: the 580– 575 Ma Marystown Volcanics
(see above) and the c. 603 Ma Caldecote Volcanics plus related
intrusions (Vizan et al. 2003). We suspect that this estimate is
unreliable, given the combined uncertainties in glacial influence
of the Gwna beds, palaeogeographic affinity of the Mona
complex relative to the Avalonian arc, and age of the succession.
In the Cadomian terrane, Bohemian massif, marine diamictite
lenses associated with inferred, glacially influenced, base level
changes are found in the Clanzschwitz and Weesenstein Groups.
These lenses contain detrital zircons as young as c. 570 Ma, and
therefore, their deposition is also approximately the same age (Lin-
nemann et al. 2007). These formations, in the Saxo-Thuringian
Zone, were likely proximal to West Africa during Ediacaran to
early Palaeozoic time; however, lack of reliable Ediacaran palaeo-
magnetic data from either the Cadomian Belt itself or the West
African craton prohibit assignment of depositional palaeolatitudes
for the Clanzschwitz and Weesenstein units.
Amazonia and environs
In the Paraguay belt at the southeastern margin of the Amazon
craton, the Puga diamictite is overlain by the Mirassol d’Oeste
cap carbonate. A characteristic component of magnetization has
been found in the cap dolostone (Trindade et al. 2003), carried
by both magnetite and haematite with high rock-magnetic stability
(Font et al. 2005) and multiple reversals were observed through
the stratigraphic section (Font et al. 2010) exposed in a single
rock quarry. If primary, that magnetic remanence would indicate
a depositional palaeolatitude of 22 þ 6/– 58 for Puga deglaciation.
The direction, however, is coincident with the Mesozoic to recent
expected directions at the sampling sites. Although the presence of
two magnetic polarities is encouraging, remagnetizations of that
nature are well known in the palaeomagnetic literature (e.g. Kent
& Dupuis 2003), and the Mirassol d’Oeste characteristic rema-
nence is so far observed at only a single sampling locality with
no field stability tests on its age. For this reason, it is considered
here to be only moderately reliable.
Higher in the stratigraphy, the Serra Azul Formation is now
recognized as a second glaciogenic level in the Paraguay belt
(summarized by Alvarenga et al. 2009). Age constraints are lax,
and its depositional palaeolatitude is unknown.
Along coastal Peru, the Arequipa massif contains a diamictite
unit, the Chiquerı́o Formation, which has been correlated to glacio-
genic deposits in eastern Laurentia (see references in Evans 2000).
Recent U –Pb dating of its host terrane, the Arequipa-Antofalla
block (AAB), however, has identified more compelling tectonic
affinities to the Amazon craton (Chew et al. 2007; Casquet et al.
2010). In either case, without a precise age, the Chiquerı́o diamic-
tite’s depositional palaeolatitude remains unconstrained.
Rio de la Plata craton and environs
The Las Ventanas and Playa Hermosa Formations, near the eastern
edge of the Rio de la Plata craton in southern Uruguay, contain the
most convincingly glaciogenic features in the entire region (Pazos
et al. 2003, 2008; Pecoits et al. 2008). Their age constraints are not
well defined within the Ediacaran Period. Palaeomagnetic results
obtained directly from the Playa Hermosa Formation suggest a tro-
pical palaeolatitude of 13 þ 9/ –88 (Sánchez-Bettucci & Rapalini
2002), but we exclude this result from our final analysis because of
very limited sampling in that preliminary study.
Lower in the stratigraphy, within the same tectonic unit (Nico
Pérez terrane and adjacent portion of the Dom Feliciano belt),
the Zanja del Tigre Formation is reported to contain a thin meta-
diamictite horizon as part of a succession (Lavalleja Group) that
106 D. A. D. EVANS & T. D. RAUB
metamorphosed at c. 630 Ma (Pazos et al. 2008, fig. 4). A glacio-
genic influence is disputed for these metamorphic rocks (Gaucher
& Poiré 2009; Pecoits & Aubet, pers. comm. 2009) and, regardless,
depositional palaeolatitudes for this unit are unconstrained.
Within the southern Brazilian portion of the Dom Feliciano belt,
Eerola (2006) identified isolated cobbles within schists of the
Brusque Group, which he correlated with the c. 750 Ma Chuos For-
mation in Namibia. However, a much older age for the Brusque
succession is possible, as it contains no detrital zircons younger
than 2000 Ma (Hartmann et al. 2003). A glaciogenic origin for
this unit is not yet well demonstrated, but interestingly, Pazos
et al. (2008) suggest correlation of the Brusque Group with the
Lavalleja Group. This correlation suggests that scattered remnants
of a pre-630 Ma ice age could be present throughout the region.
However, the name Lavalleja Group has been applied to various
metamorphic rocks with protolith ages now found to range from
Archaean to Ediacaran (Bossi & Cingolani 2009); thus, all corre-
lations should be treated with caution.
Eerola (1995) also described diamictites and dropstone-bearing
laminated mudstones in the Camaquã basin, referred to as the
Passo da Areia Formation (see also Pazos et al. 2008) or, alterna-
tively, the Picada das Graças Formation (Eerola 2006; Janikian
et al. 2008). Any proposed glacial influence for that unit was dis-
counted by the latter study (Janikian et al. 2008), which provided
a direct U –Pb SHRIMP zircon age of 580.0 + 3.6 Ma from an
interbedded tuff. Janikian et al. (2008) cite a palaeolatitude of
23 þ 10/– 78 from the slightly older Hilario Formation volcanic
rocks, but as reviewed by Tohver et al. (2006), that study lacks
field tests and is only of moderate reliability. Nonetheless,
farther north in the Ribeira belt, Campanha et al. (2008) have
determined a maximum age limit of c. 590– 580 Ma for diamictites
of the highly deformed Iporanga Formation, which contain exotic
clasts and thus, could be of glaciogenic origin. Given the more con-
vincing glacial influence on the Las Ventanas and Playa Hermosa
formations of the same general age range on the proximal Rio de la
Plata craton, the Iporanga strata should be investigated further to
confirm or refute the influence of ice during sedimentation.
Trindade & Macouin (2007) refer to glaciation at moderate to
high palaeolatitudes in the Rio de la Plata craton, constrained by
the ‘La Tinta’ palaeomagnetic pole (see below), and depicted as
such in a 620 Ma global reconstruction. That citation probably
refers to the Sierra del Volcán Formation, recognized as bearing
glaciogenic features with a possible Sturtian age (Pazos et al.
2003). However, the unit lacks precise geochronological data
(Pazos et al. 2008) and in fact could be as young as Ordovician
(unpublished detrital zircon ages cited by Poiré & Gaucher
2009). Regardless, the La Tinta pole is now considered to represent
a probable Cambrian overprint (Rapalini & Sánchez-Bettucci
2008), leaving these deposits completely unconstrained in deposi-
tional palaeolatitude.
Discussion
Figure 7.1 summarizes the results of our updated synthesis.
Although several of the assessments of glacial palaeolatitudes
have changed in detail, owing to new geochronologic or palaeo-
magnetic data, the general result is essentially the same as that
determined by Evans (2000): low and moderate palaeolatitudes
are abundant, at all reliability levels, whereas few if any near-polar
(.608) glacial deposits have thus far been identified in the Neo-
proterozoic rock record. From a total of 137 stratigraphic units
distinguished in the present analysis, only 22 deposits have deposi-
tional palaeolatitude constraints that are at least ‘somewhat’
reliable, and among these, merely 10 are constrained moderately
well. Only three deposits (Upper Tindir, Nantuo and Elatina) are
considered to have the highest degree of reliability, and all were
laid down in low to moderate palaeolatitudes. The lowest of
these, Elatina, is supported by numerous field stability tests to
verify the primary age of magnetization; reproducibility by four
different palaeomagnetic laboratories through 15 years of study;
and several independent, quantitative methods to correct for
modest amounts of sedimentary inclination shallowing. Several
lines of evidence demonstrate the effects of regionally cold cli-
mates near sea level during the Elatina ice age (Williams et al.
2008). The conclusion appears inescapable: Neoproterozoic conti-
nental ice sheets extended deep within the tropics.
Among the other high- to moderate-reliability data shown in
Figure 7.1, the only other results obtained from sedimentary
rocks are the Puga, Nantuo and Walsh. Puga and Walsh magneti-
zations are dominated by carbonate-associated haematite, so these
results also may be less prone to inclination bias. The Nantuo result
represents siliciclastic rocks, so its remanence could be slightly
shallowed in inclination and hence, shallowed in palaeolatitude.
However, estimates for the Konnarock, Rapitan, Mechum River,
Grand Conglomerat, Upper Tindir and Gaskiers deposits are calcu-
lated indirectly from coeval igneous rocks on the same cratons.
Those results are not affected by inclination shallowing.
Concerns that the Neoproterozoic geomagnetic field was
either nonaxial or nondipolar are valid in principle, but many
palaeomagnetic results, including those from the Elatina-
Nuccaleena succession, show typical patterns of a sporadically
reversing field with circularly symmetric secular variation about
the time-averaged mean (e.g. Raub 2008; Schmidt et al. 2009;
Font et al. 2010). It should therefore be emphasized that a non-
actualistic geodynamo invoked to produce strictly polar or temper-
ate glacial palaeolatitudes from the available palaeomagnetic
database is as shockingly nonuniformitarian to geophysicists as
near-equatorial glaciation is to palaeoclimatologists. Similarly,
hypotheses invoking rapid palaeolatitude shifts of continents to
generate erroneous palaeomagnetic latitudes, due to lax age con-
straints in some instances, would require such motions at rates
beyond what is normally considered reasonable for plate tectonics.
Such shifts would need to be consistently biased to produce wholly
low- to mid-palaeolatitudes at the expense of any polar results.
Evans (2006) showed that Proterozoic evaporites produce sub-
tropical palaeomagnetic latitudes, as expected for a uniformitarian
geodynamo, as well as low planetary obliquity. Williams (2008)
attempted to discredit the latter conclusion by appealing to uncer-
tainties in geomagnetic polarity, to generate an equatorial mean for
the evaporites, as predicted by numerical simulations of high-
obliquity palaeoclimate. However, that issue was already foreseen
and refuted by Evans (2006), who wrote (p. 53) ‘A zero mean could
be obtained if half of the palaeolatitudes were considered to rep-
resent the opposite hemisphere (geomagnetic polarity being gener-
ally unknown in the Precambrian), but even then the total
distribution would be significantly bimodal, in contrast with the
high-obliquity model predictions.’
Our summary analysis (Fig. 7.1) differs from the similar figure
in Evans (2000) in several minor ways; most importantly (i) a
slight increase in the estimated depositional palaeolatitude of the
Rapitan deposits, based on new geochronology that allows
precise application of the higher-reliability Franklin large
igneous province pole that is immune to sedimentary inclination
shallowing, (ii) a slight increase in the estimated depositional
palaeolatitude of the Elatina deposits, based on new studies that
quantify such inclination shallowing by several independent
methods, (iii) exclusion of the Jbeliat palaeolatitude estimate due
to its more conclusively pre-Cambrian age and thus inappropriate-
ness of applying Early Cambrian palaeolatitudes from the Gond-
wana Land apparent polar wander path, and (iv) inclusion of the
somewhat reliable, near-polar palaeolatitude for the Fauquier For-
mation as determined by palaeomagnetism on the overlying Catoc-
tin basalts and related intrusions. This last entry needs to be
confirmed, but if so, it has the potential to strike down a pillar of
the Precambrian high-obliquity Earth model (Williams 1993,
2008) – that Precambrian glaciogenic deposits are only known
from the palaeo-tropics, rather than the palaeo-poles. In general,
 GLOBAL NEOPROTEROZOIC GLACIAL PALAEOLATITUDES
however, our update does not significantly affect the basic con-
clusion by Evans (2000) of a preponderance of low- to mid-
latitudes for Neoproterozoic glaciation.
Our analysis is the most detailed of its kind in the published lit-
erature. Although our conclusions are nearly opposite to those of
Eyles & Januszczak (2004), we note that their analysis included
less than a fifth of the number of deposits considered herein.
Among the deposits accepted by them as having ‘demonstrated’
glacial influence, five have palaeolatitudes of at least somewhat
reliable quality ranging between 13 and 378. One of the deposits
(Rapitan) has moderate palaeomagnetic reliability indicating tropi-
cal palaeolatitudes. Although we did not assess the evidence for or
against glacial influence on sedimentation in our present analysis,
we anticipate that future efforts in that regard will consider the full
extent of units listed in Table 7.1.
Trindade & Macouin (2007) also considered only a small
fraction of the deposits listed in Table 7.1, although their con-
clusion is similar to ours. Considering only the subset of deposits
discussed by Trindade & Macouin (2007), the most striking differ-
ences between their conclusions and ours are as follows: (i) our
exclusion of the Scout Mountain palaeolatitude based on an
updated age for that unit, which differs from the palaeomagneti-
cally studied rock units elsewhere in Laurentia, (ii) our inclusion
of a Gaskiers palaeolatitude applied indirectly from a palaeomag-
netic study of precisely coeval igneous rocks in Avalon, (iii) our
estimate of mid-latitude deposition for the Nantuo ice age, rather
than near-equatorial, based on our selection of the direct Nantuo
constraint, rather than the (in our opinion) lower-reliability Doush-
antuo result, and (iv) our estimate of low-latitude deposition of the
Playa Hermosa Formation, preferred over the higher latitudes
implied by the now-obsolete La Tinta Formation.
The most interesting comparison to earlier palaeogeographic
syntheses is between our present analysis and that of Hoffman &
Li (2009). The essence of our conclusions is similar: only low-
and mid-latitudes have been moderately or highly reliably deter-
mined for most Neoproterozoic glacial deposits. The principal
exception to this rule stems from Ediacaran deposits such as
Loch na Cille, Squantum, Gaskiers and Serra Azul, all of which
are estimated at .608 palaeolatitude by Hoffman & Li (2009).
Among those units, we only consider Squantum and Gaskiers as
having direct palaeomagnetic constraints classified as even some-
what reliable, and the Gaskiers palaeolatitude, by our estimation, is
remarkably different at 31 þ 10/– 88. These discrepancies can be
attributed to the very distinct approaches between the two
studies. We took a bottom-up approach that considered ages and
palaeolatitudes unconstrained unless demanded by strictly relevant
data, whereas Hoffman & Li (2009) adopted a top-down approach
that assumed known glacial ages within a specified global plate-
kinematic model. That model was based on a mostly independent
palaeomagnetic database obtained primarily from well-dated
igneous rocks with little relationship to glacial units. So, although
both studies derive from the palaeomagnetic method, they are quite
distinct in the assumptions and selection of data. The principal dis-
cordance between our work and that of Hoffman & Li (2009)
occurs during the mid-Ediacaran interval, when the global palaeo-
geographic models are most uncertain. This period is associated
with the fastest continental motions in Earth history, which may
be attributed, in part, to global tectonic reorganization (Hoffman
1999), rapid true polar wander (Raub et al. 2007b) and an isolated
episode of nonuniformitarian geodynamo behaviour (Abrajevitch
& Van der Voo 2010). If such an isolated episode occurred, and
it is far from proven, then it would appear to be limited to the mid-
Ediacaran interval rather than the late Cryogenian, Elatina or Puga
ice ages (with uniformitarian symmetry in two-polarity palaeo-
magnetic directions), or the older Cryogenian, Upper Tindir/
Rapitan ice age (as judged by the highly stable Franklin palaeo-
magnetic pole).
In summary, the palaeomagnetic case for low-latitude Neopro-
terozoic glaciation is strong, and Snowball Earth remains an
107
attractive model for evaluating the enigmatic geological record
of icehouse and hothouse environmental conditions just prior to
the dawn of animal life on our planet.
We thank the following individuals for constructive criticisms on drafts of this
manuscript: N. Aubet, M. Babinski, C. Gaucher, Z.-X. Li, A. Maloof, P. Pazos,
E. Pecoits, S. Pisarevsky, P. Schmidt, R. Van der Voo, and G. Williams. We
would particularly like to thank E. Arnaud for her remarkable patience throughout
the submission and revision process. Our research on Neoproterozoic ice ages has
been supported by the David and Lucile Packard Foundation, Agouron Institute
for Geobiology, National Science Foundation and Australian Research Council’s
Tectonics Special Research Centre. This represents a contribution of the IUGS-
and UNESCO-funded IGCP (International Geoscience Programme) Project #512.
References
Abrajevitch, A. & Van der Voo, R. 2010. Incompatible Ediacaran
paleomagnetic directions suggest an equatorial geomagnetic dipole
hypothesis. Earth and Planetary Science Letters, 293, 164– 170.
Ali, K. A., Stern, R. J., Manton, W. I., Johnson, P. R. & Mukherjee,
S. K. 2009. Neoproterozoic diamictite in the Eastern Desert of Egypt
and Northern Saudi Arabia: evidence of 750 Ma glaciation in the
Arabian– Nubian Shield? International Journal of Earth Sciences,
doi: 10.1007/s00531-009-0427-3.
Allen, P. A. 2007. The Huqf Supergroup of Oman: basin development
and context for Neoproterozoic glaciation. Earth-Science Reviews,
84, 139– 185.
Allen, P. A. & Etienne, J. L. 2008. Sedimentary challenge to Snowball
Earth. Nature Geoscience, 1, 817– 825.
Alsharhan, A. S. & Nairn, A. E. M. 1997. Chapter 3: Infracambrian
of the Middle East. In: Alsharhan, A. S. & Nairn, A. E. M. (eds)
Sedimentary Basins and Petroleum Geology of the Middle East.
Elsevier, Amsterdam, 65 – 86.
Alvarenga, C. J. S. de, Boggiani, P. C., Babinski, M., Dardenne,
M. A., Figueiredo, M., Santos, R. V. & Dantas, E. L. 2009.
The Amazonian palaeocontinent. In: Gaucher, C., Sial, A. N., Hal-
verson, G. P. & Frimmel, H. E. (eds) Neoproterozoic– Cambrian
Tectonics, Global Change and Evolution. Elsevier, Amsterdam,
Developments in Precambrian Geology, 16, 15 –28.
Avigad, D., Stern, R. J., Beyth, M., Miller, N. & McWilliams, M. O.
2007. Detrital zircon U– Pb geochronology of Cryogenian diamictites
and Lower Paleozoic sandstone in Ethiopia (Tigrai): age constraints
on Neoproterozoic glaciation and crustal evolution of the southern
Arabian– Nubian Shield. Precambrian Research, 154, 88– 106.
Azmy, K., Kendall, B., Creaser, R. A., Heaman, L. & de Oliveira,
T. F. 2008. Global correlation of the Vazante Group, São Francisco
Basin, Brazil: Re– Os and U– Pb radiometric age constraints. Pre-
cambrian Research, 164, 160– 172.
Babinski, M., Vieira, L. C. & Trindade, R. I. F. 2007. Direct dating of the
Sete Lagoas cap carbonate (Bambuı́ Group, Brazil) and implications
for the Neoproterozoic glacial events. Terra Nova, 19, 401– 406.
Bailey, C. M., Peters, S. E., Morton, J. & Shotwell, N. L. 2007. The
Mechum River Formation, Virginia Blue Ridge: a record of Neopro-
terozoic and Paleozoic tectonics in southeastern Laurentia. American
Journal of Science, 307, 1– 22.
Beyth, M., Avigod, D., Wetzel, H.-U., Matthews, A. & Berhe, S. M.
2003. Crustal exhumation and indications for Snowball Earth in the
East African Orogen: north Ethiopia and east Eritrea. Precambrian
Research, 123, 187–201.
Bingen, B., Griffin, W. L., Torsvik, T. H. & Saeed, A. 2005. Timing of
Late Neoproterozoic glaciation on Baltica constrained by detrital
zircon geochronology in the Hedmark Group, south-east Norway.
Terra Nova, 17, 250–258.
Bossi, J. & Cingolani, C. 2009. Extension and general evolution of the
Rı́o de la Plata Craton. In: Gaucher, C., Sial, A. N., Halverson,
G. P. & Frimmel, H. E. (eds) Neoproterozoic –Cambrian Tectonics,
Global Change and Evolution. Elsevier, Amsterdam, Developments
in Precambrian Geology, 16, 73– 85.
Boudzoumou, F., Vandamme, D., Affaton, P. & Gattacceca, J. 2011.
Neoproterozoic paleomagnetic poles in the Taoudeni basin (West
Africa). Comptes Rendus Geoscience, 343, 284– 294.
108 D. A. D. EVANS & T. D. RAUB
Bowring, S. A., Landing, E., Myrow, P. & Ramezani, J. 2002. Abstract
#13045-Geochronological constraints on terminal Neoproterozoic
events and the rise of metazoans. Astrobiology, 2, 457– 458.
Bowring, S. A., Grotzinger, J. P., Condon, D. J., Ramezani, J.,
Newall, M. J. & Allen, P. A. 2007. Geochronologic constraints
on the chronostratigraphic framework of the Neoproterozoic Huqf
Supergroup, Sultanate of Oman. American Journal of Science, 307,
1097– 1145.
Calver, C. R., Black, L. P., Everard, J. L. & Seymour, D. B. 2004.
U –Pb zircon age constraints on late Neoproterozoic glaciation in
Tasmania. Geology, 32, 893–896.
Campanha, G. A. C., Basei, M. S., Tassinari, C. C. G., Nutman, A. P. &
Faleiros, F. M. 2008. Constraining the age of the Iporanga For-
mation with SHRIMP U– Pb zircon: implications for possible Edia-
caran glaciation in the Ribeira Belt, SE Brazil. Gondwana
Research, 13, 117– 125.
Caron, V., Ekomane, E., Mahieux, G., Moussango, P. & Ndjeng, E.
2010. The Mintom Formation (new): sedimentology and geochemis-
try of a Neoproterozoic, Paralic succession in south-east Cameroon.
Journal of African Earth Sciences, 57, 367– 385.
Casquet, C., Fanning, C. M., Galindo, C., Pankhurst, R. J., Rapela,
C. W. & Torres, P. 2010. The Arequipa Massif of Peru: new
SHRIMP and isotope constraints on a Paleoproterozoic inlier in the
Grenvillian orogen. Journal of South American Earth Sciences, 29,
128– 142.
Chakraborty, P. P., Dey, S. & Mohanty, S. P. 2010. Proterozoic plat-
form sequences of Peninsular India: implications towards basin evol-
ution and supercontinent assembly. Journal of Asian Earth Sciences,
39, 589– 607.
Chew, D., Kirkland, C., Schaltegger, U. & Goodhue, R. 2007. Neo-
proterozoic glaciation in the Proto-Andes: tectonic implications and
global correlation. Geology, 35, 1095–1098.
Chumakov, N. M. 2009. The Baykonurian glaciohorizon of the Late
Vendian. Stratigraphy and Geological Correlation, 17, 373– 381.
Cocks, L. R. M. & Torsvik, T. H. 2007. Siberia, the wandering northern
terrane, and its changing geography through the Palaeozoic. Earth-
Science Reviews, 82, 29 –74.
Condon, D. J. & Prave, A. R. 2000. Two from Donegal: Neoproterozoic
glacial episodes on the northeast margin of Laurentia. Geology, 28,
951– 954.
Condon, D., Zhu, M., Bowring, S., Wang, W., Yang, A. & Jin, Y. 2005.
U –Pb ages from the Neoproterozoic Doushantuo Formation, China.
Science, 308, 95 –98.
Corkeron, M. 2007. ‘Cap carbonates’ and Neoproterozoic glacigenic
successions from the Kimberley region, north-west Australia. Sedi-
mentology, 54, 871– 903.
Corkeron, M. L. & George, A. D. 2001. Glacial incursion on a
Neoproterozoic carbonate platform in the Kimberley region, Austra-
lia. Geological Society of America Bulletin, 113, 1121– 1132.
Corsetti, F. A. & Kaufman, A. J. 2003. Stratigraphic investigations of
carbon isotope anomalies and Neoproterozoic ice ages in Death
Valley, California. Geological Society of America Bulletin, 115,
916– 932.
Corsetti, F. A. & Kaufman, A. J. 2005. The relationship between the
Neoproterozoic Noonday Dolomite and the Ibex Formation: new
observations and their bearing on ‘Snowball Earth’. Earth-Science
Reviews, 73, 63– 78.
Corsetti, F. A. & Lorentz, N. J. 2006. On Neoproterozoic cap carbon-
ates as chronostratigraphic markers. In: Xiao, S. & Kaufman, A. J.
(eds) Neoproterozoic Geobiology and Paleobiology. Springer,
New York, 273– 294.
Corsetti, F. A., Stewart, J. H. & Hagadorn, J. W. 2007. Neopro-
terozoic diamictite-cap carbonate succession and d13C chemo-
stratigraphy from eastern Sonora, Mexico. Chemical Geology, 237,
129– 142.
Dempster, T. J., Rogers, G. et al. 2002. Timing of deposition, orogen-
esis and glaciation within the Dalradian rocks of Scotland: constraints
from U– Pb zircon ages. Journal of the Geological Society of London,
159, 83 – 94.
Denyszyn, S. W., Halls, H. C., Davis, D. W. & Evans, D. A. D. 2009.
Paleomagnetism and U– Pb geochronology of Franklin dykes in High
Arctic Canada and Greenland: a revised age and paleomagnetic pole
constraining block rotations in the Nares Strait region. Canadian
Journal of Earth Sciences, 46, 689–705.
Eerola, T. 2006. Myöhäisproterotsooiset ilmastonmuutokset Tutkimuk-
sia Etelä-Brasiliassa [Neoproterozoic climate changes; research on
southern Brazil; Finnish with English abstract and bilingual figure
captions]. Geologi, 58, 164– 174.
Embleton, B. J. J. & Williams, G. E. 1986. Low palaeolatitude of depo-
sition for late Precambrian periglacial varvites in South Australia:
implications for palaeoclimatology. Earth and Planetary Science
Letters, 79, 419– 430.
Evans, D. A. D. 2000. Stratigraphic, geochronological, and paleomagnetic
constraints upon the Neoproterozoic climatic paradox. American
Journal of Science, 300, 347– 433.
Evans, D. A. D. 2003. A fundamental Precambrian – Phanerozoic shift in
Earth’s glacial style? Tectonophysics, 375, 353– 385.
Evans, D. A. D. 2006. Proterozoic low orbital obliquity and axial-dipolar
geomagnetic field from evaporite palaeolatitudes. Nature, 444,
51 – 55.
Evans, D. A. D., Li, Z. X., Kirschvink, J. L. & Wingate, M. T. D. 2000.
A high-quality mid-Neoproterozoic paleomagnetic pole from the
South China block, with implications for ice ages and the breakup
configuration of Rodinia. Precambrian Research, 100, 313– 335.
Eyles, N. & Januszczak, N. 2004. ‘Zipper-rift’: a tectonic model for
Neoproterozoic glaciations during the breakup of Rodinia after 750
Ma. Earth-Science Reviews, 65, 1– 73.
Fairchild, I. J. & Kennedy, M. J. 2007. Neoproterozoic glaciation in the
Earth System. Journal of the Geological Society of London, 164,
895–921.
Fang, W., Van der Voo, R. & Johnson, R. J. E. 1986. Eocambrian paleo-
magnetism of the Boston basin: evidence for displaced terrane. Geo-
physical Research Letters, 13, 1450–1453.
Fanning, C. M. & Link, P. K. 2004. U –Pb SHRIMP ages of Neoproter-
ozoic (Sturtian) glaciogenic Pocatello Formation, southeastern Idaho.
Geology, 32, 881–884.
Fanning, C. M. & Link, P. K. 2008. Age constraints for the Sturtian Gla-
ciation; data from the Adelaide Geosyncline, South Australia and
Pocatello Formation, Idaho, USA. Geological Society of Australia
Abstracts, 91, 57 – 62.
Fisher, N. I., Lewis, T. & Embleton, B. J. J. 1987. Statistical Analysis of
Spherical Data. Cambridge, Cambridge University Press.
Fitzsimons, I. C. W. 2000. Grenville-age basement provinces in East Ant-
arctica: evidence for three separate collisional orogens. Geology, 28,
879– 882.
Font, E., Trindade, R. I. F. & Nédélec, A. 2005. Detrital remanent mag-
netization in haematite-bearing Neoproterozoic Puga cap dolostone,
Amazon craton: a rock magnetic and SEM study. Geophysical
Journal International, 163, 491– 500.
Font, E., Nédélec, A., Trindade, R. I. F. & Moreau, C. 2010. Fast or
slow melting of the Marinoan Snowball Earth? The cap dolostone
record. Palaeogeography, Palaeoclimatology, Palaeoecology, 295,
215– 225.
Frimmel, H. E. 2011. The Karoetjes Kop and Bloupoort Formations,
Gifberg Group, South Africa, In: Arnaud, E., Halverson, G. P. &
Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic
Glaciations. Geological Society, London, Memoirs, 36, 233 –237.
Frimmel, H. E. & Jiang, S.-Y. 2001. Marine evaporites from an oceanic
island in the Neoproterozoic Adamastor ocean. Precambrian
Research, 105, 57– 71.
Frimmel, H. E., Fölling, P. G. & Eriksson, P. G. 2002. Neoproterozoic
tectonic and climatic evolution recorded in the Gariep Belt, Namibia
and South Africa. Basin Research, 14, 55 –67.
Gallet, Y., Pavlov, V. & Courtillot, V. 2003. Magnetic reversal fre-
quency and apparent polar wander of the Siberian platform in the ear-
liest Palaeozoic, inferred from the Khorbusuonka river section
(northeastern Siberia). Geophysical Journal International, 154,
829– 840.
Gaucher, C. & Poiré, D. G. 2009. Palaeoclimatic events. Neoprotero-
zoic– Cambrian evolution of the Rı́o de la Plata Palaeocontinent.
In: Gaucher, C., Sial, A. N., Halverson, G. P. & Frimmel,
H. E. (eds) Neoproterozoic– Cambrian Tectonics, Global Change
and Evolution. Elsevier, Amsterdam, Developments in Precambrian
Geology, 16, 123–130.
 GLOBAL NEOPROTEROZOIC GLACIAL PALAEOLATITUDES
Gaucher, C., Frimmel, H. E. & Germs, G. J. B. 2009. Tectonic events
and palaeogeographic evolution of southwestern Gondwana in the
Neoproterozoic and Cambrian. In: Gaucher, C., Sial, A. N., Hal-
verson, G. P. & Frimmel, H. E. (eds) Neoproterozoic–Cambrian
Tectonics, Global Change and Evolution. Elsevier, Amsterdam,
Developments in Precambrian Geology, 16, 295–316.
Gladkochub, D. P., Wingate, M. T. D., Pisarevsky, S. A., Donskaya,
T. V., Mazukabzov, A. M., Ponomarchuk, V. A. & Stanevich, A.
M. 2006. Mafic intrusions in southwestern Siberia and implications
for a Neoproterozoic connection with Laurentia. Precambrian
Research, 147, 260– 278.
Goodge, J. W., Myrow, P., Williams, I. S. & Bowring, S. A. 2002.
Age and provenance of the Beardmore Group, Antarctica: constraints
on Rodinia supercontinent breakup. Journal of Geology, 110,
393– 406.
Gostin, V. A., McKirdy, D. M., Webster, L. J. & Williams, G. E. 2011.
Mid-Ediacaran ice-rafting in the Adelaide Geosyncline and Officer
Basin, South Australia. In: Arnaud, E., Halverson, G. P. &
Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic
Glaciations. Geological Society, London, Memoirs, 36, 673– 676.
Grey, K., Hocking, R. M. et al. 2005. Lithostratigraphic nomenclature
of the Officer Basin and correlative parts of the Paterson Orogen,
Western Australia. Western Australia Geological Survey, Report,
93, 89.
Haines, P. W., Hocking, R. M., Grey, K. & Stevens, M. K. 2008. Vines
1 revisited: are older Neoproterozoic glacial deposits preserved in
Western Australia? Australian Journal of Earth Sciences, 55,
397– 406.
Halverson, G. P., Hoffman, P. F., Schrag, D. P., Maloof, A. C. & Rice,
A. H. N. 2005. Toward a Neoproterozoic composite carbon-isotope
record. Geological Society of America Bulletin, 117, 1181–1207.
Halverson, G. P., Dudás, F. Ö., Maloof, A. C. & Bowring, S. A.
2007. Evolution of the 87Sr/86Sr composition of Neoproterozoic
seawater. Palaeogeography, Palaeoclimatology, Palaeoecology,
256, 103– 129.
Hambrey, M. J. & Harland, W. B. (eds) 1981. Earth’s Pre-Pleistocene
Glacial Record. Cambridge University Press, Cambridge, 1004.
Hamdi, B. 1992. Late Precambrian glacial deposits in central Iran. 29th
International Geological Congress, Abstracts, 2, 232.
Hannah, J., Yang, G., Bingen, B., Stein, H. & Zimmerman, A. 2007.
560 and 300 Ma Re– Os ages constrain Neoproterozoic glaciation
and record Variscan hydrocarbon migration on extension of Oslo rift.
In: Redfield, T., Buiter, S. J. H. & Smethurst, M. A. (eds) geo-
dynamics, Geomagnetism and Paleogeography: A 50 Year Celebra-
tion. NGU Report 2007.057 (Trondheim: Geological Survey of
Norway), 50.
Hartmann, L. A., Bitencourt, M. F., Santos, J. O. S., McNaughton,
N. J., Rivera, C. B. & Betiollo, L. 2003. Prolonged Paleoprotero-
zoic magmatic participation in the Neoproterozoic Dom Feliciano
belt, Santa Catarina, Brazil, based on zircon U– Pb SHRIMP geochro-
nology. Journal of South American Earth Sciences, 16, 477–492.
Hebert, C. L., Kaufman, A. J., Penniston-Dorland, S. C. & Martin,
A. J. 2010. Radiometric and stratigraphic constraints on terminal
Ediacaran (post-Gaskiers) glaciation and metazoan evolution. Pre-
cambrian Research, 182, 402–412.
Hodych, J. P. & Cox, R. A. 2007. Ediacaran U– Pb zircon dates for the
Lac Matapédia and Mt. St.-Anselme basalts of the Quebec Appala-
chians: support for a long-lived mantle plume during the rifting
phase of Iapetus opening. Canadian Journal of Earth Sciences, 44,
565– 581.
Hoffman, P. F. 1999. The break-up of Rodinia, birth of Gondwana, true
polar wander and the Snowball Earth. Journal of African Earth
Sciences, 28, 17– 33.
Hoffman, P. F. 2009. Pan-glacial — a third state in the climate system.
Geology Today, 25, 107–114.
Hoffman, P. F. & Schrag, D. P. 2002. The Snowball Earth hypothesis:
testing the limits of global change. Terra Nova, 14, 129– 155.
Hoffman, P. F. & Li, Z.-X. 2009. A palaeogeographic context for Neopro-
terozoic glaciation. Palaeogeography, Palaeoclimatology, Palaeo-
ecology, 277, 158– 172.
Hoffman, P. F., Halverson, G. P., Domack, E. W., Husson, J. M.,
Higgins, J. A. & Schrag, D. P. 2007. Are basal Ediacaran (635
109
Ma) post-glacial ‘cap dolostones’ diachronous? Earth and Planetary
Science Letters, 258, 114–131.
Hoffmann, K.-H., Condon, D. J., Bowring, S. A. & Crowley,
J. L. 2004. U– Pb zircon date from the Neoproterozoic Ghaub Forma-
tion, Namibia: constraints on Marinoan glaciation. Geology, 32,
817– 820.
Huang, B., Xu, B., Zhang, C., Li, Y. & Zhu, R. 2005. Paleomagnetism of
the Baiyisi volcanic rocks (ca. 740 Ma) of Tarim, Northwest China: a
continental fragment of Neoproterozoic Western Australia? Precam-
brian Research, 142, 83 –92.
Huckriede, R., Kürsten, M. & Venzlaff, H. 1962. Zur Geologie des
Gebiets zwischen Kerman und Saghand (Iran). Beihefte zum Geolo-
gischen Jahrbuch, 51, 197.
Hyodo, H. & Dunlop, D. J. 1993. Effect of anisotropy on the paleomag-
netic contact test for a Grenville dike. Journal of Geophysical
Research, 98, 7997–8017.
Iglesia-Llanos, M. P., Tait, J. A., Popov, V. & Abalmassova, A. 2005.
Palaeomagnetic data from Ediacaran (Vendian) sediments of the
Arkhangelsk region, NW Russia: an alternative apparent polar
wander path of Baltica for the Late Proterozoic – Early Palaeozoic.
Earth and Planetary Science Letters, 240, 732– 747.
Inglis, J. D., MacLean, J. S., Samson, S. D., D’Lemos, R. S., Admou, H.
& Hefferan, K. 2004. A precise U –Pb zircon age for the Bleı̈da
granodiorite, Anti-Atlas, Morocco: implications for the timing of
deformation and terrane assembly in the eastern Anti-Atlas.
Journal of African Earth Sciences, 39, 277–283.
Janikian, L., Almeida, R. P. et al. 2008. The continental record of Edia-
caran volcano-sedimentary successions in southern Brazil and their
global implications. Terra Nova, 20, 259– 266.
Jenkins, R. J. F. 2011. Billy Springs glaciation, South Australia. In:
Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geo-
logical Record of Neoproterozoic Glaciations. Geological Society,
London, Memoirs, 36 693– 699.
Kamo, S. L. & Gower, C. F. 1994. U –Pb baddeleyite dating clarifies age
of characteristic paleomagnetic remanence of Long Range dykes,
southeastern Labrador. Atlantic Geology, 30, 259–262.
Kamo, S. L., Gower, C. F. & Krogh, T. E. 1989. Birthdate for the Iapetus
Ocean? A precise U –Pb zircon and baddeleyite age for the Long
Range dikes, southeast Labrador. Geology, 17, 602–605.
Kamo, S. L., Krogh, T. E. & Kumarapeli, P. S. 1995. Age of the
Grenville dyke swarm, Ontario-Quebec: implications for the timing
of Iapetus rifting. Canadian Journal of Earth Sciences, 32, 273– 280.
Karlstrom, K. E., Bowring, S. A. et al. 2000. Chuar Group of the
Grand Canyon: record of breakup of Rodinia, associated change in
the global carbon cycle, and ecosystem expansion by 740 Ma.
Geology, 28, 619– 622.
Kaufman, A. J., Sial, A. N., Frimmel, H. E. & Misi, A. 2009. Neoproter-
ozoic to Cambrian palaeoclimatic events in southwestern Gondwana.
In: Gaucher, C., Sial, A. N., Halverson, G. P. & Frimmel, H. E.
(eds) Neoproterozoic –Cambrian Tectonics, Global Change and
Evolution. Elsevier, Amsterdam, Developments in Precambrian
Geology, 16, 369– 388.
Kawai, T., Windley, B. F., Terabayashi, M., Yamamoto, H., Isozaki,
Y. & Maruyama, S. 2008. Neoproterozoic glaciations in the mid-
oceanic realm: an example from hemi-pelagic mudstones on Llandd-
wyn Island, Anglesey, U.K. Gondwana Research, 14, 105–114.
Kempf, O., Kellerhals, P., Lowrie, W. & Matter, A. 2000. Paleomag-
netic directions in Late Precambrian glaciomarine sediments of the
Mirbat sandstone formation, Oman. Earth and Planetary Science
Letters, 175, 181– 190.
Kendall, B., Creaser, R. A., Calver, C. R., Raub, T. D. & Evans,
D. A. D. 2009. Correlation of Sturtian diamictite successions in
southern Australia and northwestern Tasmania by Re– Os black
shale geochronology and the ambiguity of ‘Sturtian’-type diamic-
tite– cap carbonate pairs as chronostratigraphic marker horizons.
Precambrian Research, 172, 301– 310.
Kent, D. V. & Dupuis, C. 2003. Paleomagnetic study of the Paleocene–
Eocene Tarawan chalk and Esna shale: dual polarity remagnetizations
of Cenozoic sediments in the Nile valley (Egypt). Micropaleontology,
49, 139– 146.
Key, R. M., Liyungu, A. K., Njamu, F. M., Somwe, V. & Banda, J.
2001. The western arm of the Lufilian Arc in NW Zambia and its
110 D. A. D. EVANS & T. D. RAUB
potential for copper mineralization. Journal of African Earth
Sciences, 33, 503– 528.
Khomentovsky, V. V. 1986. The Vendian System of Siberia and a stan-
dard stratigraphic scale. Geological Magazine, 123, 333– 348.
Kilner, B., Mac Niocaill, C. & Brasier, M. 2005. Low-latitude glacia-
tion in the Neoproterozoic of Oman. Geology, 33, 413– 416.
Kirschvink, J. L. 1978. The Precambrian-Cambrian boundary problem:
paleomagnetic directions from the Amadeus Basin, Central Australia.
Earth and Planetary Science Letters, 40, 91– 100.
Kirschvink, J. L. 1992. Late Proterozoic low-latitude global glaciation:
the Snowball Earth. In: Schopf, J. W. & Klein, C. C. (eds) The Pro-
terozoic Biosphere: A Multidisciplinary Study. Cambridge University
Press, Cambridge, 51 –52.
Kirschvink, J. L. & Rozanov, A. Yu. 1984. Magnetostratigraphy of
Lower Cambrian strata from the Siberian platform: a paleomagnetic
pole and a preliminary polarity time-scale. Geological Magazine,
121, 189– 203.
Kravchinsky, V. A., Konstantinov, K. M. & Cogné, J.-P. 2001.
Palaeomagnetic study of Vendian and Early Cambrian rocks of
South Siberia and Central Mongolia: was the Siberian platform
assembled at this time? Precambrian Research, 110, 61– 92.
Kravchinsky, V. A., Sklyarov, E. V., Gladkochub, D. P. & Harbert,
W. P. 2010. Paleomagnetism of the Precambrian Eastern Sayan rocks:
implications for the Ediacaran –Early Cambrian paleogeography of
the Tuva-Mongolian composite terrane. Tectonophysics, 486, 65 – 80.
Levashova, N. M., Kalugin, V. M., Gibsher, A. S., Yff, J., Ryabinin,
A. B., Meert, J. G. & Malone, S. J. 2010. The origin of the Baydaric
microcontinent, Mongolia: constraints from paleomagnetism and
geochronology. Tectonophysics, 485, 306–320.
Li, Y., Li, Y., Sharps, R., McWilliams, M. & Gao, Z. 1991. Sinian paleo-
magnetic results from the Tarim block, western China. Precambrian
Research, 49, 61 –71.
Li, Z.-X. 2000. New palaeomagnetic results from the ‘cap dolomite’ of the
Neoproterozoic Walsh Tillite, northwestern Australia. Precambrian
Research, 100, 359– 370.
Li, Z.-X. & Evans, D. A. D. 2011. Late Neoproterozoic 408 intraplate
rotation within Australia allows for a tighter-fitting and longer-lasting
Rodinia. Geology, 39, 39– 42.
Linnemann, U., Gerdes, A., Drost, K. & Buschmann, B. 2007. The
continuum between Cadomian orogenesis and opening of the Rheic
Ocean: constraints from LA– ICP– MS U–Pb zircon dating and
analysis of plate-tectonic setting (Saxo-Thuringian zone, northeastern
Bohemian Massif, Germany. In: Linnemann, U., Nance, R. D.,
Kraft, P. & Zulauf, G. (eds) The Evolution of the Rheic
Ocean: From Avalonian –Cadomian Active Margin to Alleghe-
nian– Variscan Collision. Geological Society of America, Special
Paper, 423, 61 –96.
Lund, K., Aleinikoff, J. N. & Evans, K. V. 2003. SHRIMP U –Pb geo-
chronology of Neoproterozoic Windermere Supergroup, central
Idaho: implications for rifting of western Laurentia and synchroneity
of Sturtian glacial deposits. Geological Society of America Bulletin,
115, 349– 372.
Macouin, M., Besse, J., Ader, M., Gilder, S., Yang, Z., Sun, Z. &
Agrinier, P. 2004. Combined paleomagnetic and isotopic data
from the Doushantuo carbonates, South China: implications for the
“Snowball Earth” hypothesis. Earth and Planetary Science Letters,
224, 387– 398.
Macdonald, F. A., McClelland, W. C., Schrag, D. P. & Macdonald,
W. P. 2009a. Neoproterozoic glaciations on a carbonate platform
margin in Arctic Alaska and the origin of the North Slope subterrane.
Geological Society of America Bulletin, 121, 448–473.
Macdonald, F. A., Jones, D. S. & Schrag, D. P. 2009b. Stratigraphic
and tectonic implications of a newly discovered glacial
diamictite– cap carbonate couplet in southwestern Mongolia.
Geology, 37, 123–126.
Macdonald, F. A., Schmitz, M. D. et al. 2010a. Calibrating the Cryo-
genian. Science, 327, 1241–1243.
Macdonald, F. A., Cohen, P. A., Dudás, F. Ö. & Schrag, D. P. 2010b.
Early Neoproterozoic scale microfossils in the Lower Tindir Group of
Alaska and the Yukon Territory. Geology, 38, 143–146.
Macdonald, F. A., Strauss, J. V., Rose, C. V., Dudás, F. Õ. & Schrag,
D. P. 2010c. Stratigraphy of the Port Nolloth Group of Namibia and
South Africa and implications for the age of Neoproterozoic iron for-
mations. American Journal of Science, 310, 862– 888.
Mac Niocaill, C., Kilner, B., Stouge, S., Knudsen, M. F., Harper,
D. A. T. & Christiansen, J. L. 2008. The Neoproterozoic drift
history of Laurentia: a critical evaluation and new palaeomagnetic
data from Northern and Eastern Greenland. EOS Transactions of
the American Geophysical Union, 89, S514–05.
Maloof, A. C., Schrag, D. P., Crowley, J. L. & Bowring, S. A. 2005.
An expanded record of Early Cambrian carbon cycling from the Anti-
Atlas Margin, Morocco. Canadian Journal of Earth Sciences, 42,
2195– 2216.
Martin, M. W., Grazhdankin, D. V., Bowring, S. A., Evans, D. A. D.,
Fedonkin, M. A. & Kirschvink, J. L. 2000. Age of Neoproterozoic
bilaterian body and trace fossils, White Sea, Russia: implications for
metazoan evolution. Science, 288, 841– 845.
Mbede, E. I., Kampunzu, A. B. & Armstrong, R. A. 2004. Neoproter-
ozoic inheritance during Cainozoic rifting in the western and south-
western branches of the East African rift system: evidence from
carbonatite and alkaline intrusions. Conference abstract, The East
African Rift System: Development, Evolution and Resources,
Addis Ababa, Ethiopia, June 20– 24, 2004.
McCausland, P. J. A., Van der Voo, R. & Hall, C. M. 2007.
Circum-Iapetus paleogeography of the Precambrian – Cambrian tran-
sition with a new paleomagnetic constraint from Laurentia. Precam-
brian Research, 156, 125–152.
McCausland, P. J. A., Smirnov, A., Evans, D. A. D., Izard, C. & Raub,
T. 2009. Low-latitude Laurentia at 615 Ma: Paleomagnetism of the
Long Range Dykes and Coeval Lighthouse Cove Formation, Northern
Newfoundland and SE Labrador. Abstracts, AGU-GAC/MAC-CGU
Joint Assembly, Toronto, May 2009.
McCausland, P. J. A., Hankard, F., Van der Voo, R. & Hall, C. M.
2011. Ediacaran paleogeography of Laurentia: paleomagnetism and
40
Ar-39Ar geochronology of the 583 Ma Baie des Moutons syenite,
Quebec. Precambrian Research, 187, 58– 78.
McCay, G. A., Prave, A. R., Alsop, G. I. & Fallick, A. E. 2006. Glacial
trinity: Neoproterzoic Earth history within the British-Irish Caledo-
nides. Geology, 34, 909–912.
McFadden, P. L. & Lowes, F. J. 1981. The discrimination of mean direc-
tions drawn from Fisher distributions. Geophysical Journal of the
Royal Astronomical Society, 67, 19– 33.
McNamara, A. K., Mac Niocaill, C., Van der Pluijm, B. A. & Van der
Voo, R. 2001. West African proximity of the Avalon terrane in the
latest Precambrian. Geological Society of America Bulletin, 113,
1161– 1170.
McWilliams, M. & Schmidt, P. W. 2003. Paleomagnetism of Grassy
Group rocks, King Island, Tasmania. European Geophysical
Society, Geophysical Research Abstracts, 5, no. 12406.
Meert, J. G. & Van der Voo, R. 1996. Paleomagnetic and 40Ar/39Ar
study of the Sinyai Dolerite, Kenya: implications for Gondwana
assembly. Journal of Geology, 104, 131– 142.
Meert, J. G. & Van der Voo, R. 2001. Comment on ‘New palaeomag-
netic result from Vendian red sediments in Cisbaikalia and the
problem of the relationship of Siberia and Laurentia in the
Vendian’ by S. A. Pisarevsky, R. A. Komissarova and A. N.
Khramov. Geophysical Journal International, 146, 867– 870.
Meert, J. G., Van der Voo, R. & Payne, T. W. 1994. Paleomagnetism of
the Catoctin volcanic province: a new Vendian –Cambrian apparent
polar wander path for North America. Journal of Geophysical
Research, 99, 4625– 4641.
Meert, J. G., Van der Voo, R. & Ayub, S. 1995. Paleomagnetic investi-
gation of the Neoproterozoic Gagwe lavas and Mbozi complex, Tan-
zania and the assembly of Gondwana. Precambrian Research, 74,
225– 244.
Meert, J. G., Walderhaug, H. J., Torsvik, T. H. & Hendriks, B. W. H.
2007. Age and paleomagnetic signature of the Alnø
carbonatite complex (NE Sweden): additional controversy for the
Neoproterozoic paleoposition of Baltica. Precambrian Research,
154, 159– 174.
Meffre, S., Direen, N. G., Crawford, A. J. & Kamenetsky, V. 2004.
Mafic volcanic rocks on King Island, Tasmania: evidence for
579 Ma break-up in east Gondwana. Precambrian Research, 135,
177– 191.
 GLOBAL NEOPROTEROZOIC GLACIAL PALAEOLATITUDES
Melezhik, V. A., Kuznetsov, A. B., Fallick, A. F., Smith, R. A., Gor-
okhov, I. M., Jamal, D. & Catuane, F. 2006. Depositional environ-
ments and an apparent age for the Geci meta-limestones: constraints
on the geological history of northern Mozambique. Precambrian
Research, 148, 19– 31.
Metelkin, D. V., Belonosov, I. V., Gladkochub, D. P., Donskaya,
T. V., Mazukabzov, A. M. & Stanevich, A. M. 2005. Paleomag-
netic directions from Nersa intrusions of the Biryusa terrane, Siberian
Craton, as a reflection of tectonic events in the Neoproterozoic.
Russian Geology and Geophysics, 46, 398– 413.
Miller, N. R., Alene, M., Sacchi, R., Stern, R. J., Conti, A., Kröner,
A. & Zuppi, G. 2003. Significance of the Tambien Group (Tigrai,
N. Ethiopia) for Snowball Earth events in the Arabian– Nubian
Shield. Precambrian Research, 121, 263– 283.
Misi, A., Kaufman, A. J. et al. 2007. Chemostratigraphic correlation of
Neoproterozoic successions in South America. Chemical Geology,
237, 143– 167.
Mitchell, R. N., Evans, D. A. D. & Kilian, T. M. 2010. Rapid Early
Cambrian rotation of Gondwana. Geology, 38, 755–758.
Mukhopadhyay, J. & Chaudhuri, A. K. 2003. Shallow to deep-water
deposition in a Cratonic basin: an example from the Proterozoic Pen-
ganga Group, Pranhita– Godavari Valley, India. Journal of Asian
Earth Sciences, 21, 613–622.
Murthy, G. S. 1971. The paleomagnetism of diabase dikes from the
Grenville Province. Canadian Journal of Earth Sciences, 8,
802– 812.
Murthy, G., Gower, C., Tubrett, M. & Pätzold, R. 1992. Paleomag-
netism of Eocambrian Long Range dykes and Double Mer Formaa-
tion from Labrador, Canada. Canadian Journal of Earth Sciences,
29, 1224– 1234.
Myrow, P. M., Pope, M. C., Goodge, J. W., Fischer, W. & Palmer,
A. R. 2002. Depositional history of pre-Devonian strata and timing
of Ross orogenic tectonism in the central Transantarctic Mountains,
Antarctica. Geological Society of America Bulletin, 114, 1070– 1088.
Nawrocki, J., Boguckij, A. & Katinas, V. 2004. New Late Vendian
palaeogeography of Baltica and the TESZ. Geological Quarterly,
48, 309– 316.
Pazos, P. J., Bettucci, L. S. & Loureiro, J. 2008. The Neoproterozoic
glacial record in the Rı́o de la Plata Craton: a critical reappraisal.
In: Pankhurst, R. J., Truow, R. A. J., Brito Neves, B. B. & de
Wit, M. J. (eds) West Gondwana: Pre-Cenozoic Correlations
Across the South Atlantic Region. Geological Society, London,
Special Publication, 294, 343– 364.
Pazos, P. J., Sánchez-Bettucci, L. & Tofalo, O. R. 2003. The record
of the Varanger glaciation at the Rı́o de la Plata craton, Vendian-
Cambrian of Uruguay. Gondwana Research, 6, 65 –77.
Pecoits, E., Gingras, M., Aubet, N. & Konhauser, K. 2008. Ediacaran
in Uruguay: palaeoclimatic and palaeobiological implications. Sedi-
mentology, 55, 689– 719.
Pinna, P., Jourde, G., Calvez, J. Y., Mroz, J. P. & Marques, J. M.
1993. The Mozambique Belt in northern Mozambique: Neoprotero-
zoic (1100 –850 Ma) crustal growth and tectogenesis, and superim-
posed Pan-African (800– 550 Ma) tectonism. Precambrian
Research, 62, 1 – 59.
Pisarevsky, S. A., Komissarova, R. A. & Khramov, A. N. 2001. Reply
to comment by J. G. Meert and R. Van der Voo on ‘New palaeomag-
netic result from Vendian red sediments in Cisbaikalia and the
problem of the relationship of Siberia and Laurentia in the
Vendian’. Geophysical Journal International, 146, 871– 873.
Pisarevsky, S. A., Wingate, M. T. D., Stevens, M. K. & Haines,
P. W. 2007. Palaeomagnetic results from the Lancer 1 stratigraphic
drillhole, Officer Basin, Western Australia, and implications for
Rodinia reconstructions. Australian Journal of Earth Sciences, 54,
561– 572.
Poidevin, J.-L. 2007. Stratigraphie isotopique du strontium et datation
des formations carbonatées et glaciogéniques néoprotérozoiques du
Nord et de l’Ouest du craton du Congo. Comptes Rendus Geoscience,
339, 259– 273.
Poiré, D. G. & Gaucher, C. 2009. Lithostratigraphy. Neoproterozoic –
Cambrian evolution of the Rı́o de la Plata Palaeocontinent. In:
Gaucher, C., Sial, A. N., Halverson, G. P. & Frimmel, H. E.
(eds) Neoproterozoic– Cambrian Tectonics, global Change and
111
Evolution. Elsevier, Amsterdam, Developments in Precambrian
Geology, 16, 87 –101.
Pokrovskii, B. G., Melezhik, V. A. & Bujakaite, M. I. 2006. Carbon,
oxygen, strontium, and sulfur isotopic compositions in late Precam-
brian rocks of the Patom Complex, central Siberia: communication
1. results, isotope stratigraphy, and dating problems. Lithology and
Mineral Resources, 41, 450– 474.
Popov, V., Iosifidi, A., Khramov, A., Tait, J. & Bachtadse, V. 2002.
Paleomagnetism of Upper Vendian sediments from the Winter
Coast, White Sea region, Russia: implications for the paleogeography
of Baltica during Neoproterozoic times. Journal of Geophysical
Research, 107, doi: 10.1029/2001JB001607.
Popov, V. V., Khramov, A. N. & Bachtadse, V. 2005. Palaeomagnet-
ism, magnetic stratigraphy, and petromagnetism of the Upper
Vendian sedimentary rocks in the sections of the Zolotitsa River
and in the Verkhotina Hole, Winter Coast of the White Sea, Russia.
Russian Journal of Earth Sciences, 7, 1– 29.
Porter, S. M., Knoll, A. H. & Affaton, P. 2004. Chemostratigraphy of
Neoproterozoic cap carbonates from the Volta Basin, West Africa.
Precambrian Research, 130, 99– 112.
Prave, A. R. 1999. The Neoproterozoic Dalradian Supergroup of Scot-
land: an alternative hypothesis. Geological Magazine, 136, 609– 617.
Ramezani, J. & Tucker, R. D. 2003. The Saghand region, central Iran:
U–Pb geochronology, petrogenesis and implications for Gondwana
tectonics. American Journal of Science, 303, 622–665.
Rapalini, A. E. & Sánchez-Bettucci, L. 2008. Widespread remagneti-
zation of late Proterozoic sedimentary units of Uruguay and the
apparent polar wander path for the Rio de La Plata craton. Geophysi-
cal Journal International, 174, 55 –74.
Raub, T. D. 2008. Prolonged deglaciation of Snowball Earth. PhD thesis,
Yale University.
Raub, T. D., Evans, D. A. D. & Smirnov, A. V. 2007a. Siliciclastic
prelude to Elatina deglaciation: lithostratigraphy and rock magnetism
of the base of the Ediacaran System. In: Vickers-Rich, P. & Komar-
ower, P. (eds) The Rise and Fall of the Ediacaran Biota. Geological
Society, London, Special Publication, 286, 53– 76.
Raub, T. D., Kirschvink, J. L. & Evans, D. A. D. 2007b. True polar
wander: linking deep and shallow geodynamics to hydro- and bio-
spheric hypotheses. In: Kono, M. (ed.) Treatise on Geophysics
Volume 5: Geomagnetism. Elsevier, Amsterdam, 565–589.
Rieu, R., Allen, P. A., Cozzi, A., Kosler, J. & Bussy, F. 2007. A
composite stratigraphy for the Neoproterozoic Huqf Supergroup of
Oman: integrating new litho-, chemo- and chronostratigraphic data
of the Mirbat area, southern Oman. Journal of the Geological
Society of London, 164, 997– 1009.
Rieu, R., Allen, P. A., Etienne, J. L., Cozzi, A. & Wiechert, U.
2006. A Neoproterozoic glacially influenced basin margin succession
and ‘atypical’ cap carbonate associated with bedrock palaeovalleys,
Mirbat area, southern Oman. Basin Research, 18, 471–496.
Roberts, D. 2003. The Scandinavian Caledonides: event chronology,
palaeogeographic settings and likely modern analogues. Tectonophy-
sics, 365, 283–299.
Rodrigues, J. B., Pimentel, M. M., Buhn, B., Dardenne, M. A. &
Alvarenga, C. J. S. 2008. Provenance of Vazante Group — Prelimi-
nary data. VI South American Symposium on Isotope Geology, San
Carlos de Bariloche, Argentina, 4.
Rooney, A. D., Selby, D., Houzay, J.-P. & Renne, P. R. 2010. Re– Os
geochronology of a Mesoproterozoic sedimentary succession, Taou-
deni basin, Mauritania: implications for basin-wide correlations and
Re– Os organic-rich sediments systematics. Earth and Planetary
Science Letters, 289, 486–496.
Rose, C. V. & Maloof, A. C. 2010. Testing models for post-glacial ‘cap
dolostone’ deposition: Nuccaleena Formation, South Australia. Earth
and Planetary Science Letters, 296, 165–180.
Rowan, C. J. & Tait, J. 2010. Oman’s Low Latitude ‘Snowball Earth’
Pole Revisited: Late Cretaceous remagnetisation of Late Neoproter-
ozoic Carbonates in Northern Oman. American Geophysical Union,
Fall Meeting Abstract GP33C-0959.
Sánchez-Bettucci, L. & Rapalini, A. E. 2002. Paleomagnetism of the
Sierra de Las Animas Complex, southern Uruguay: its implications in
the assembly of western Gondwana. Precambrian Research, 118,
243– 265.
112 D. A. D. EVANS & T. D. RAUB
Schmidt, P. W. & Williams, G. E. 1995. The Neoproterozoic climatic
paradox: equatorial palaeolatitude for Marinoan glaciation near sea
level in South Australia. Earth and Planetary Science Letters, 134,
107– 124.
Schmidt, P. W., Williams, G. E. & Embleton, B. J. J. 1991. Low palaeo-
latitude of Late Proterozoic glaciation: early timing of remanence in
haematite of the Elatina Formation, South Australia. Earth and Pla-
netary Science Letters, 105, 355–367.
Schmidt, P. W., Williams, G. E. & McWilliams, M. O. 2009. Palaeo-
magnetism and magnetic anisotropy of late Neoproterozoic strata,
South Australia: implications for the palaeolatitude of late Cryogen-
ian glaciation, cap carbonate and the Ediacaran System. Precambrian
Research, 174, 35 – 52.
Shen, B., Xiao, S., Zhou, C., Kaufman, A. J. & Yuan, X. 2010. Carbon
and sulfur isotope chemostratigraphy of the Neoproterozoic Quanji
Group of the Chaidam Basin, NW China: basin stratification in the
aftermath of an Ediacaran glaciation postdating the Shuram event?
Precambrian Research, 177, 241– 252.
Shields, G. A., Deynoux, M., Culver, S. J., Brasier, M. D., Affaton,
P. & Vandamme, D. 2007. Neoproterozoic glaciomarine and cap
dolostone facies of the southwestern Taoudéni Basin (Walidiala
Valley, Senegal/Guinea, NW Africa). Comptes Rendus Geoscience,
339, 186– 199.
Sial, A. N., Dardenne, M. A. et al. 2009. The São Francisco palaeocon-
tinent. In: Gaucher, C., Sial, A. N., Halverson, G. P. & Frimmel,
H. E. (eds) Neoproterozoic– Cambrian Tectonics, Global Change
and Evolution. Elsevier, Amsterdam, Developments in Precambrian
Geology, 16, 31 –69.
Sohl, L. E., Christie-Blick, N. & Kent, D. V. 1999. Paleomagnetic
polarity reversals in Marinoan (ca. 600 Ma) glacial deposits of Aus-
tralia: implications for the duration of low-latitude glaciation in Neo-
proterozoic time. Geological Society of America Bulletin, 111,
1120– 1139.
Sovetov, J. K. 2002. Vendian foreland basin of the Siberian cratonic
margin: Paleopangean accretionary phases. Russian Journal of
Earth Sciences, 4, 363– 387.
Sovetov, J. 2008. Marinoan Glaciation in the Siberian Craton: Locality,
Erosional Forms, Deposits and Constraints to Age. 33rd International
Geological Congress, Abstracts, Session CGC-04.
Sovetov, Yu. K. & Komlev, D. A. 2005. Tillites at the base of the Oselok
Group, foothills of the Sayan Mountains, and the Vendian lower
boundary in the southwestern Siberian Platform. Stratigraphy and
Geological Correlation, 13, 337– 366.
Stern, R. J., Avigad, D., Miller, N. R. & Beyth, M. 2006. Evidence for
the Snowball Earth hypothesis in the Arabian – Nubian Shield and the
East African Orogen. Journal of African Earth Sciences, 44, 1– 20.
Swanson-Hysell, N. L., Maloof, A. C., Weiss, B. P. & & Evans,
D. A. D. 2009. No asymmetry in geomagnetic reversals recorded
by 1.1-billion-year-old Keweenawan basalts. Nature Geoscience, 2,
713– 717.
Tauxe, L., Kodama, K. P. & Kent, D. V. 2008. Testing corrections for
paleomagnetic inclination error in sedimentary rocks: a comparative
approach. Physics of the Earth and Planetary Interiors, 169,
152– 165.
Thomas, R. J., Chevallier, L. C. et al. 2002. Precambrian evolution of
the Sirwa Window, Anti-Atlas Orogen, Morocco. Precambrian
Research, 118, 1 –57.
Thomas, R. J., Fekkak, A. et al. 2004. A new lithostratigraphic frame-
work for the Anti-Atlas Orogen, Morocco. Journal of African Earth
Sciences, 39, 217– 226.
Thompson, M. D. & Bowring, S. A. 2000. Age of the Squantum ‘Tillite’,
Boston Basin, Massachusetts: U– Pb zircon constraints on terminal
Neoproterozoic glaciation. American Journal of Science, 300,
630– 655.
Thompson, M. D., Grunow, A. M. & Ramezani, J. 2007. Late Neopro-
terozoic paleogeography of the southeastern New England Avalon
zone: insights from U– Pb geochronology and paleomagnetism. Geo-
logical Society of America Bulletin, 119, 681–696.
Tohver, E., D’Agrella-Filho, M. S. & Trindade, R. I. F. 2006. Paleo-
magnetic record of Africa and South America for the 1200–500 Ma
interval, and evaluation of Rodinia and Gondwana assemblies. Pre-
cambrian Research, 147, 193–222.
Torsvik, T. H., Lohmann, K. C. & Sturt, B. A. 1995. Vendian glacia-
tions and their relation to the dispersal of Rodinia: Paleomagnetic
constraints. Geology, 23, 727–730.
Trindade, R. I. F. & Macouin, M. 2007. Palaeolatitude of glacial depos-
its and palaeogeography of Neoproterozoic ice ages. Comptes Rendus
Geoscience, 339, 200–211.
Trindade, R. I. F., D’Agrella-Filho, M. S., Babinski, M., Font, E. &
Brito Neves, B. B. 2004. Paleomagnetism and geochronology of the
Bebedouro cap carbonate: evidence for continental-scale Cambrian
remagnetization in the São Francisco craton, Brazil. Precambrian
Research, 128, 83– 103.
Trindade, R. I. F., Font, E., D’Agrella-Filho, M. S., Nogueira,
A. C. R. & Riccomini, C. 2003. Low-latitude and multiple geomag-
netic reversals in the Neoproterozoic Puga cap carbonate, Amazon
craton. Terra Nova, 15, 441–446.
Trompette, R. 1994. Geology of Western Gondwana (2000– 500 Ma):
Pan-African –Brasiliano Aggregation of South America and Africa.
Balkema, Rotterdam.
Van der Voo, R. 1990. The reliability of paleomagnetic data. Tectonophy-
sics, 184, 1– 9.
Vizan, H., Carney, J. N., Turner, P., Ixer, R. A., Tomasso, M.,
Mullen, R. P. & Clarke, P. 2003. Late Neoproterozoic to Early
Palaeozoic palaeogeography of Avalonia: some palaeomagnetic con-
straints from Nuneaton, central England. Geological Magazine, 140,
685– 705.
Weil, A. B., Geissman, J. W. & Van der Voo, R. 2004. Paleomagnetism
of the Neoproterozoic Chuar Group, Grand Canyon Supergroup,
Arizona: implications for Laurentia’s Neoproterozoic APWP and
Rodinia break-up. Precambrian Research, 129, 71 –92.
Williams, G. E. 1993. History of the Earth’s obliquity. Earth-Science
Reviews, 34, 1– 45.
Williams, G. E. 2008. Proterozoic (pre-Ediacaran) glaciation and the high
obliquity, low-latitude ice, strong seasonality (HOLIST) hypothesis:
principles and tests. Earth-Science Reviews, 87, 61 –93.
Williams, G. E., Gostin, V. A., McKirdy, D. M. & Preiss, W. V. 2008.
The Elatina glaciation, late Cryogenian (Marinoan Epoch), South
Australia: sedimentary facies and palaeoenvironments. Precambrian
Research, 163, 307– 331.
Xu, B., Jian, P., Zheng, H., Zou, H., Zhang, L. & Liu, D. 2005. U–Pb
zircon geochronology and geochemistry of Neoproterozoic volcanic
rocks in the Tarim Block of northwest China: implications for the
breakup of Rodinia supercontinent and Neoproterozoic glaciations.
Precambrian Research, 136, 107– 123.
Xu, B., Xiao, S. et al. 2009. SHRIMP zircon U– Pb age constraints on
Neoproterozoic Quruqtagh diamictites in NW China. Precambrian
Research, 168, 247– 258.
Zhan, S., Chen, Y., Xu, B., Wang, B. & Faure, M. 2007. Late Neopro-
terozoic paleomagnetic results from the Sugetbrak Formation of the
Aksu area, Tarim basin (NW China) and their implications to paleo-
geographic reconstructions and the Snowball Earth hypothesis. Pre-
cambrian Research, 154, 143–158.
Zhang, Q.-R. & Piper, J. D. A. 1997. Palaeomagnetic study of Neoproter-
ozoic glacial rocks of the Yangzi Block: palaeolatitude and configur-
ation of South China in the late Proterozoic Supercontinent.
Precambrian Research, 85, 173– 199.
Zhang, Q.-R., Li, X.-H., Feng, L.-J., Huang, J. & Song, B. 2008. A
new age constraint on the onset of the Neoproterozoic glaciations
in the Yangtze Platform, South China. Journal of Geology, 116,
423– 429.
Zhang, Q.-R., Chu, X. L. & Feng, L. J. 2009. Discussion on the Neopro-
terozoic glaciations in the South China Block and their related paleo-
latitudes. Chinese Science Bulletin, 54, 1797– 1800.
Zhang, S., Evans, D. A. D. et al. 2006. New paleomagnetic results from
the Nantuo Formation in south China and their paleogeographic
implications. EOS, Transactions of the American Geophysical
Union, abstract GP34A-04.
Zhang, S., Jiang, G. & Han, Y. 2008. The age of the Nantuo
Formation and Nantuo glaciation in South China. Terra Nova, 20,
289– 294.
Zhou, C., Tucker, R., Xiao, S., Peng, Z., Yuan, X. & Chen, Z. 2004.
New constraints on the ages of Neoproterozoic glaciations in south
China. Geology, 32, 437– 440.
 Chapter 8

Biostratigraphy and stratigraphic subdivision of Cryogenian successions

of Australia in a global context

KATHLEEN GREY1*, ANDREW C. HILL2 & CLIVE CALVER3

1
Geological Survey of Western Australia, Department of Mines and Petroleum, 100 Plain Street, Western Australia, 6076
2
Centro de Astrobiologı́a (INTA-CSIC), Instituto Nacional de Técnica Aeroespacial, Ctra de Ajalvir, km 4,
28850 Torrejón de Ardoz, Madrid, Spain
3
Mineral Resources Tasmania, PO Box 56, Rosny Park, Tasmania 7018, Australia
*Corresponding author (e-mail: kath.grey@dmp.wa.gov.au)

Abstract: Cryogenian correlation in Australia is based on an extensive data set from the Centralian Superbasin and Adelaide Rift
Complex and integrates biostratigraphy and isotope chemostratigraphy to provide a three-dimensional interpretation based on outcrop
and drill holes. Studies are ongoing, but newer data are consistent with the distributions discussed here. From the chemostratigraphic
and biostratigraphic viewpoint, the first appearance of the acritarch Cerebrosphaera buickii, coupled with a large negative isotope excur-
sion at c. 800 Ma, supported by the first appearance of the stromatolite Baicalia burra, seems to have potential for boundary placement. It
is widely recognized across Australia and seems to have potential globally.

Australia has extensive Cryogenian successions that can be readily
correlated using a variety of techniques. Cryogenian correlation is
based on both field sections and continuous core from drill holes.
In Western Australia, this includes drill holes Empress 1 and 1A,
Lancer 1, Vines 1, other drill holes, and outcrops in the western
Officer and Amadeus Basins and the Kimberley area. Results
from Western Australia are consistent with data from other drill
holes and field sections examined from across Australia, including
those from the eastern Officer, Amadeus, Georgina, and Ngalia
Basins, the Adelaide Rift Complex and Tasmania (Figs 8.1–
8.4). In general, biostratigraphic correlations (Figs 8.1 –8.3) corre-
spond to those obtained using lithostratigraphy, well-log and other
geophysical data and chemostratigraphy (d13C and 87Sr/86Sr;
Walter et al. 1995, 2000; Morton & Drexel 1997; Hill & Walter
2000; Hill et al. 2000a, b; Preiss 2000; Haines et al. 2004, 2008;
Grey et al. 2005; Mory & Haines 2005; Gorter et al. 2007; Hill,
unpublished data).
The Cryogenian was originally defined as a Neoproterozoic
chronometric period extending from 850 Ma to 650 Ma. Since
the name Cryogenian is preoccupied, it is questionable whether
it should be used for any proposed chronostratigraphic units
below the Ediacaran. However, for practical purposes, the term
Cryogenian is here used for the interval from the chronometrically
defined base of the Cryogenian System and Period at 850 Ma to the
chronometrically defined top of the Cryogenian at 650 Ma
(Gradstein et al. 2004), but also incorporates the interval that lies
between the 650 Ma boundary and the chronostratigraphically
defined base of the succeeding Ediacaran System and Period with
its Global Boundary Stratotype Section and Point (GSSP) at the
base of the Nuccaleena Formation, Flinders Ranges, South Australia
(Knoll et al. 2006). This pragmatic approach facilitates discussion
pending decisions of the International Subcommission on Neopro-
terozoic (Cryogenian and Ediacaran) Stratigraphy. Only Australian
Cryogenian successions are discussed here in detail because it is
still difficult to correlate between Australia and successions else-
where with any certainty. However, consistencies observed
across Australia auger well for eventual global correlation.
There are no unequivocal Tonian-age (.850 Ma) sedimentary
rocks in Australia, although several possible sections require
further investigation. The base of the recognized Cryogenian
succession is probably no older than c. 830 Ma (Fig. 8.5), based
on U –Pb ages for the Gairdner Dyke Swarm, which either
underlies or just intrudes the base of the succession in South
Australia (see below), and global chemostratigraphic correlation
(Macdonald et al. 2010). Early Cryogenian successions are wide-
spread and have been identified in the Adelaide Rift Complex
(Callanna and Burra Groups), the Amadeus Basin (Bitter Springs
Formation), the Georgina Basin (Yackah beds), the Ngalia Basin
(Vaughan Springs Quartzite and Albinia Formation), the Officer
Basin (Buldya Group and lateral equivalents in both Western
and South Australia, the Kimberley area (Ruby Plains Group),
parts of the Birrindudu Basin (Redcliff Pound Group), and
Tasmania (lower Togari Group and its equivalents). This part of
the succession underlies the Sturt glaciation (Preiss et al. 2011),
named after the Sturt Tillite in the Adelaide Rift Complex), and
which includes the Areyonga Formation in the Amadeus Basin
and lateral equivalents in central Australia and Tasmania, but is
represented by a hiatus in Western Australia, except in Vines 1,
where evidence of two glaciations, represented by the older
Pirrilyungka Formation and the younger Wahlgu Formation, may
be present (Haines et al. 2008; Hill et al. 2011).
The interval between the Sturt Tillite and Elatina Formation and
their equivalents is not well represented across Australia. There is
a near-continuous succession in the Adelaide Rift Complex (the
Umberatana Group) and a partial succession in the Amadeus
Basin (Aralka Formation). In Western Australia, this interval is
marked by a heavily karstified surface and disconformity in both
Empress 1A and Lancer 1. The succeeding Elatina glaciation
(Williams et al. 2008) and its equivalents (previously referred
to as the Marinoan glaciation) appear to be represented in all
Australian Neoproterozoic basins.
In Western Australia, the continuously cored Empress 1A and
Lancer 1 successions can be tied to each other and to other
Officer Basin drill holes using well-log data, lithostratigraphy,
seismic interpretation and chemostratigraphy (Grey et al. 2005).
Acritarch and stromatolite ranges are consistent across the basin
and proposed biostratigraphic correlations correspond closely to
non-biostratigraphic correlations. There is evidence to suggest
that Australian successions can be similarly correlated to succes-
sions elsewhere in the world.

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 113– 134. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.8
114 K. GREY ET AL.

120° 130° 140° 150°

10° 10°

KIMBERLEY
REGION

Fig. 8.1. Locality map showing distribution

GEORGINA of Neoproterozoic basins in Australia and
20° NGALIA BASIN 20° localities sampled for microfossil and
AMADEUS BASIN
BASIN AS1 palynology studies in the Australian
Arunta Province
LD1
D1 O1
Cryogenian (adapted from Walter et al.
MW1
T6
H1 W1
Bb1 1995; Preiss 2000; Grey 2005b).
T9 MC1
Western
L1 Y2
Y3 Distribution data from Grey & Cotter
Musgrave Province
Officer Basin Lu1
K1A PEAKE & DENISON (1996); Cotter 1997, 1999; Grey 1999a,
Probable extent of V1 RANGES
Centralian Superbasin E1A
OFFICER B1 Curnamona 2005a, b; Grey et al. 2005b. Abbreviations:
BASIN Eastern G1 Province
Officer AS1, BMR Alice Springs 1; Bb1, Bluebush
Basin Gawler
Craton B4 1; B1, Birksgate 1; B4, Western Mining
STUART
SHELF
BLD 04; D1, Eagle Dragoon 1; E1A,
30° 30°
GSWA Empress 1/1A; G1, Comalco Giles
P12 ADELAIDE
RIFT COMPLEX 1; H1, Eagle Corporation Hussar 1; K1A,
Shell Kanpa 1A; L1, GSWA Lancer 1; LD1,
Lake Disappointment 1 (LDDH 1); Lu1,
Shell Lungkarta 1; MC1, Transoil Mount
Drillhole with Bitter Springs Formation Assemblage
Charlotte 1; MW1, Mount Winter 1; O1,
Drillhole with Cerebrosphaera buickii Assemblage Magellan Ooraminna 1; PP12; North
Drillhole with both assemblages
Outcrop with Bitter Springs Formation Assemblage KING ISLAND
Broken Hill PP 12; T6, GSWA TWB 6
(Trainor Water Bore 6); T9, GSWA TWB 9
1000 km
40°
TASMANIA
40° (Trainor Water Bore 9); V1, GSWA Vines
1; W1, Wallara 1; Y2, Shell Yowalga 2; Y3,
110° 120° 130° 140° 150° 160°
Shell Yowalga 3.

120° 130° 140° 150°

10° 10°

KIMBERLEY
REGION
R

GEORGINA

20° NGALIA BASIN 20°

AMADEUS BASIN
T BASIN BS
T Arunta Province BS P

T BS
S LB Y3 BS W1
LB
Western L1
LB Musgrave Province M5
Officer Basin LB PEAKE & DENISON
Probable extent of LB BG BG RANGES
E1A
Centralian Superbasin OFFICER Curnamona
BASIN Eastern BG Province Fig. 8.2. Locality map showing distribution of
Officer
Basin Gawler the Acaciella australica stromatolite
Craton
STUART assemblage in Australia. Only localities with
SHELF 30° identified specimens are shown (data from
30° BG
ADELAIDE Walter 1972; Preiss 1973b, 1976, 1987; Walter
RIFT COMPLEX
et al. 1979; Grey 1995, 1999b, 2005a,
unpublished data; Stevens & Grey 1997; Grey
& Blake 1999; Hill et al. 2000a; Grey et al.
W1
Drillhole Outcrop
2005). Abbreviations: T, Tarcunyah Group;
Stratigraphy
T Tarcunyah Group BS Bitter Springs Formation LB, Lower Buldya Group; R, Ruby Plains
LB Lower Buldya Group P Plenty Group
R Ruby Plains Group BG Burra Group Group; S, Sunbeam Group; BS, Bitter Springs
S Sunbeam Group KING ISLAND
Formation; P, Plenty Group; BG, Burra Group;
1000 km
40° 40° E1A, GSWA Empress 1A; L1, GSWA Lancer
TASMANIA
1; M5, Manya 5; W1, Wallara 1; Y3, Shell
110° 120° 130° 140° 150° 160°
Yowalga 3.
 AUSTRALIAN CRYOGENIAN CORRELATIONS 115

120° 130° 140° 150°

10° 10°

KIMBERLEY
REGION

GEORGINA

20° NGALIA BASIN 20°

AMADEUS BASIN
BASIN
Arunta Province

UB L1 J
H1 W1
UB
Western UB Musgrave Province
Officer Basin UB UB PEAKE & DENISON
Probable extent of E1A RANGES
UB BG
Centralian Superbasin OFFICER Curnamona
N1 BASIN Eastern Province
Officer
Basin Gawler BG
Craton
BG
STUART
SHELF
Fig. 8.3. Locality map showing distribution
30° BG 30°
of the Baicalia burra stromatolite
BG BG ADELAIDE
RIFT COMPLEX assemblage in Australia. Only localities with
BG
identified specimens are shown (data from
Preiss 1972, 1974, 1976, 1985, 1987; Griffin
& Preiss 1976; Stevens & Grey 1997; Grey
Drillhole Outcrop
1999b, 2005a, unpublished data; Hill et al.
W1
Stratigraphy 2000a). Abbreviations: UB, Upper Buldyah
UB Upper Buldya Group BG Burra Group
J Johnnys Creek Beds Tg Togari Group KING ISLAND
Group; J, Johnnys Creek beds; BG, Burra
Group; Tg, Togari Group; E1A, GSWA
1000 km Tg
40°
TASMANIA
40° Empress 1A; H1, Eagle Corporation Hussar
1; L1, GSWA Lancer 1; N1, NJD 1; W1,
110° 120° 130° 140° 150° 160°
Wallara 1.

120° 130° 140° 150°

10° 10°

Fig. 8.4. Locality map showing

distribution of Neoproterozoic basins in
KIMBERLEY Australia and localities sampled for
REGION
chemostratigraphic studies in the
Australian Cryogenian (adapted from
GEORGINA Walter et al. 1995; Hill & Walter 2000;
20° 20° Preiss 2000; McKirdy et al. 2001; Grey
NGALIA BASIN
AMADEUS BASIN
BASIN
2005b). Drill hole abbreviations for
Arunta Province A1 chemostratigraphy: A1, AS 27; A2, AS
I LS
W1
D A2 A3 28; A3, AS 3; B1, BTD 1; B2, Blinman
L B
MC 2; E, Empress 1A; ET, ETM 5A-1; F,
Western
Officer Basin
E Musgrave Province
PEAKE & DENISON Forest 1; I, Illogwa Creek 6; L, Lancer
Probable extent of M5 C
RANGES 1; LN, LNM 10-1; M5, Manya 5; MC,
Centralian Superbasin OFFICER Curnamona
BASIN Eastern
UBPD
Province Mount Charlotte 1; S1, SAS 1; S3, SAR
Officer SR
Basin Gawler
DW
3; SC, SCYW 1a; SL, SLT 104; SR, SR
Craton B2
STUART
SC
S3 M
O LN 6; W1, Wallara 1, W2, Wokurna 2; W5,
ET
30°
SHELF SL EC
B1 30° Wokurna 5. Field section abbreviations
S1 DC
SD ADELAIDE for chemostratigraphy: A, Arthur
W2 UBMF RIFT COMPLEX River; B, Bluebush; C, Coominaree
W5
Dolomite type section; D, Dump; DC,
Depot Creek; DW, Doodney’s Well
Hills; EC, Enorama Creek; LS, Limbla
Syncline; M, Mallee Water;
Drillhole Outcrop
O, Oraparinna Diapir; SD, Skillogalee
KING ISLAND Dolomite in Port Germein Gorge;
F
1000 km A UBMF, Upper Burra Group,
40°
TASMANIA
40° mid-Flinders Ranges; UBPD, Upper
Burra Group, Peake and Denison
110° 120° 130° 140° 150° 160°
Ranges.
116 K. GREY ET AL.

ADELAIDE NW
KING KIMBERLEY
PER- AGE AMADEUS NGALIA GEORGINA RIFT TASMANIA REGION
ISLAND
IOD (Ma) SS NW Central Eastern Northeastern BASIN BASIN COMPLEX
C Cambrian Cambrian Cambrian Cambrian Cambrian Cambrian Cambrian
543
550 McFadden Lungkarta ‘lower’ Arumbera ‘lower’ Yuendumu
4 Fm Fm Sandstone Sandstone 12

LOUISA
Julie
Formation
EDIACARAN

volcanics
575 volcanics
CHD
MOPUNGA rhyodacite 11
UNGOOLYA GROUP WILPENA
Keppel

GRASSY GROUP
GROUP GROUP
Creek
600
3 Pertatataka
Newhaven Fm
Shale
Formation
Member

GROUP WDM W
625 Boondawari
Y

Fm W ? ? Pioneer/Olympic MDDM Black/Oorabra

UGE 9 10
650 2
6 7 8
Pirrilyungka 5
Fm Y illite S illite Member
675

700

T
Steptoe
Fm
‘CRYOGENIAN’

725 4
Kanpa
Fm BURRA
’ GROUP
750
BULDYA GROUP

1
?

775 Hussar
Fm

3 ? ?
800
2
CALANNA GROUP

Member Curdimurka
LM
Subgroup
Albinia
Yackah
WM CVE Fm
825 beds
Loves

SUNBEAM Units ROCKY CAPE

CDE GROUP
GROUP Arkaroola
Alinya Gillen Subgroup
Fm YCE Member
TQ V
850
1 1 BASEMENT 1

Key: Neoproterozoic glacial or Sequence boundary

partially glacial units
1 Geochronological age
Cap dolomite Volcanic
intrusion
Neoproterozoic sediments
Volcanics
Unconformity/disconformity

Fig. 8.5. Australian Neoproterozoic stratigraphy and correlations assuming a c. 635 Ma age (Hoffmann et al. 2004; Condon et al. 2005) for the Elatina Formation and
equivalents; however, see Williams et al. (2008) for an extended discussion of age constraints (adapted from Preiss 1987, 2000; Walter et al. 1995; Morton 1997; Grey
et al. 2005; Dunster et al. 2007). The length of time breaks at major unconformities and sequence boundaries are estimates only. Amadeus Basin successions were
described in Wells et al. (1970), Kennard et al. (1986), Korsch & Kennard (1991) and Lindsay (1993). Stratigraphic subdivision of the Adelaide Rift Complex and Stuart
Shelf follow Preiss (1987, 2000 and references therein), Haines (1988, 1990) and Reid & Preiss (1999). Officer Basin successions and correlations were detailed in
Jackson & van de Graaff (1981), Phillips et al. (1985), Townson (1985), Williams (1992), Preiss (1993), Walter & Gorter (1994), Walter et al. (1994), Bagas et al. (1995,
1999), Lindsay (1995), Lindsay & Reine (1995), Morton (1997), Hill et al. (2000a), Grey et al. (2005) and Haines et al. (2008). Georgina Basin successions were
described by Smith (1972), Walter (1980) and Dunster et al. (2007), the Ngalia Basin successions in Wells & Moss (1983), the Kimberley region in Dow & Gemuts
(1969), Blake et al. (1998), Corkeron & George (2001) and Corkeron (2007), and in Tasmania by Calver (1998), Calver & Walter (2000) and Calver et al. (2004).
Australia-wide correlations were discussed in Preiss et al. (1978), Coats & Preiss (1980), Walter et al. (1995, 2000), Calver & Walter (2000), Hill & Walter (2000), Hill
et al. (2000a, b), Calver et al. (2004) and Grey et al. (2005). SS, Supersequence number, Centralian Superbasin (Walter et al. 1995); CDE, Coominaree Dolomite
equivalent; CHD, Croles Hill Diamictite; CVE, Cadlareena Volcanics equivalent; LM, Lancer Member; MDDM, Mount Davenport Diamictite Member; TQ, Townsend
 AUSTRALIAN CRYOGENIAN CORRELATIONS 117

Geochronological constraints
Tonian (1000 –850 Ma) successions are probably not widespread
in Australia, although many units are inadequately dated. Units
such as the Pindyin Sandstone and Alinya Formation in the
eastern Officer Basin, and their presumed lateral equivalents, the
Townsend Quartzite and Lefroy Formation, in the western
Officer Basin, the Kulail Sandstone (Close et al. 2005) and poss-
ibly the Dean Quartzite in the Bloods Range region of the
Amadeus Basin, the Heavitree Quartzite and lateral equivalents
in the Georgina and Amadeus Basin, and parts of the Callanna
Group in the Adelaide Rift Complex could be either Tonian or
Cryogenian, depending on whether there is a stratigraphic break
between them and the overlying Bitter Springs Formation and its
lateral equivalents and whether they pre-date or were intruded
by the Gairdner Dyke Swarm. The Alinya Formation in Giles 1
(eastern Officer Basin), which seems to have an anomalous acri-
tarch assemblage (Zang 1995; see below), and the Lamil and
Throssell Range Groups along the northeastern Pilbara margins
could be either younger than (Bagas et al. 2002) or older than
the Cryogenian Tarcunyah Group. In Tasmania, the Rocky Cape
Group (c. 1000–750 Ma) is probably at least in part of Tonian
age (Black et al. 2004).
The mafic Gairdner Dyke Swarm, which intrudes the Musgrave
Province, Gawler Craton and Stuart Shelf (Parker et al. 1987;
Preiss 1987; Cowley & Flint 1993) and cross-cuts underlying
Mesoproterozoic successions, and its northwesterly extension,
the Amata Suite, probably relate to extension during the early tec-
tonic evolution of the Adelaide Rift Complex and the Centralian
Superbasin. It has not been determined whether the dykes underlie
or intrude basal successions in these areas, but it is probable that
they slightly post-date the onset of deposition. The dykes are prob-
able feeders for mafic lavas of the poorly dated Wooltana Volca-
nics near the top of the Callanna Group and other volcanic rocks,
such as the Beda Volcanics and those in the Bitter Springs
Formation (Zhao et al. 1994); however, more precise dating of
these formations is required before concluding they are all the
same age. A Stuart Shelf dyke gave a SHRIMP U – Pb age of
827 + 6 Ma on baddeleyite, and the Little Broken Hill Gabbro in
the Willyama Inlier gave an age of 827 + 9 Ma (Wingate et al.
1998). Sm –Nd isochrons of 867 + 47 and 802 + 35 Ma (Zhao
& McCulloch 1993; Zhao et al. 1994) were obtained from a dyke
intruding the Stuart Shelf (Preiss 2000). Another of the dykes,
which cross-cuts Mesoproterozoic units in the Western Musgrave
Complex, gave a zircon 207Pb/206Pb age of 824 + 4 Ma
(Glikson et al. 1996). These ages constrain the maximum age of
deposition of Australian Cryogenian successions to c. 830 Ma.
Few Australian Cryogenian dates can be considered reliable
(Preiss 2000; Fanning & Link 2008). A thermal-ionization mass
spectrometry (TIMS) U –Pb age of 802 + 10 Ma was reported
from the Rook Tuff, in the lower part of the Curdimurka Subgroup
of the Callanna Group in the Adelaide Rift Complex (Fanning et al.
1986). Recent dating of a bedded tuff in the Fifteenmile Group
of the Coal Creek Inlier, Ogilvie Mountains, NW Canada at
c. 811.5 Ma (Macdonald et al. 2010) is consistent with the Rook
Tuff age, and apparently lies at the same stratigraphic and
chemostratigraphic level.
Reid (2009) recently obtained an age of 795 + 5 Ma using
laser ablation – inductively coupled plasma mass spectrometry
(LA –ICPMS) on a porphyry that intrudes the Skillogalee Dolo-
mite and which is now thought to be ‘penecontemporaneous
with Skillogalee-age deposition’ (Drexel 2009). This conflicts
with an age on an underlying rhyolite from the Boucaut Volcanics
at the base of the Burra Group cited as giving a U –Pb SHRIMP
zircon age of 777 + 7 Ma by Preiss (2000); however, the
Boucaut Volcanics date was recently considered unreliable
by Fanning & Link (2008). A U –Pb zircon date on a thin volcani-
clastic layer in the lower Wilyerpa Formation in the Adelaide
Rift Complex (deposited during the waning phase of the Sturt
glaciation; Preiss et al. 2011), and previously referred to as
being within the Merinjina Tillite, gave an age of 659 + 6 Ma
(Fanning & Link 2008), and indicates an age of c. 660 Ma for
the Sturt glaciation. This is consistent with Re –Os ages from
black shales that closely succeed the Sturt glacials and their
equivalents: 657.2 + 5.4 Ma in the Aralka Formation in the
Amadeus Basin, 643 + 2.4 Ma on the Tindelpina Shale Member
in the basal Tapley Hill Formation in the Adelaide Rift Complex
(Kendall et al. 2006) and 641 + 5 Ma on a black shale above the
Julius River Member in Tasmania (Kendall et al. 2007). A detrital
zircon U –Pb SHRIMP age of 725 + 11 Ma from the middle
Kanpa Formation, .200 m below the Steptoe –Wahlgu Formation
disconformity (Nelson 2002), indicates that the Western Austra-
lian hiatus equivalent to the Sturt glaciation is also younger than
725 Ma. There is no firm evidence in Australia for the Sturt glacia-
tion extending back much beyond 660 Ma. This is in contrast with
recent dating from NW Canada, where ages of 716.5 Ma were
obtained for glacial units believed to be equivalents of the Sturt
glaciation (Kendall et al. 2009; Macdonald et al. 2010).
The age of the younger (Elatina) glaciation (Williams et al.
2008), and hence the base of the Ediacaran Period, remains uncer-
tain. The Ediacaran boundary lies at the base of the Nuccaleena
Formation, so it would be sensible for the underlying Elatina dia-
mictite and lateral equivalents to be included with units of Cryo-
genian age, even if this part of the succession is younger than
the previous upper chronometric boundary of 650 Ma (Gradstein
et al. 2004).
The Elatina Formation is currently correlated with the Ghaub
Formation of Namibia, dated at 635.5 + 1.2 Ma (Hoffmann
et al. 2004) and the Nantuo Tillite of China, dated by an ash bed
within the cap dolostone above the tillite at 635.2 + 0.6 Ma
(Condon et al. 2005) because of the presence of similar cap carbon-
ates. Lithologically, a succession on King Island (the Cottons
Breccia and overlying Cumberland Creek Dolostone) closely
resembles the Elatina and Nuccaleena Formations, and correlation
between the two successions appears robust (Calver & Walter
2000; Preiss 2000; Hoffman et al. 2009). The King Island succes-
sion is intruded by a 575 + 4 Ma subvolcanic sill, and overlain by
mafic volcanics with a Sm –Nd isochron age of 579 + 16 Ma
(Meffre et al. 2004), although the timing of the intrusion relative
to deposition is unclear. In northwestern Tasmania, the Croles
Hill Diamictite, upsection of the Sturt-equivalent Julius River
Member, is locally underlain by rhyodacite with a SHRIMP
U –Pb age of 582 + 4 Ma (Calver et al. 2004), indicating that
this could be a Gaskiers-age glaciation, that is, c. 580 Ma. If the
Cottons Breccia and Elatina Formation are correlatives of the
Croles Hill Diamictite, then the base of the Ediacaran (as defined
by the base of the Nuccaleena Formation) could be c. 580 Ma,
rather than 635 Ma. Better age constraints are required on the

Fig. 8.5. (Continued) Quartzite; UGE, Umberatana Group (of the Adelaide Rift Complex) equivalents; WDM, Wanapi Dolomite Member; WM, Woolnough Member;
WV, Wantapella Volcanics; YCE, Younghusband Conglomerate equivalent. Geochronology: (1) 827 + 6 Ma (Wingate et al. 1998) and 824 + 4 Ma (Glikson et al.
1996), Gairdner Dyke Swarm; (2) 802 + 10 Ma, Rook Tuff (Fanning et al. 1986); (3) 797 + 5 Ma, Skillogalee Dolomite (Drexel 2009); (4) 725 + 11 Ma detrital age,
uppermost Kanpa Formation (Nelson 2002); (5) 659 + 6 Ma, Wilyerpa Formation (Sturt Tillite correlative) (Fanning & Link 2008); (6) 657.2 + 5.4 Ma, Aralka
Formation (Kendall et al. 2006); (7) 647.2 + 10 Ma and 645.1 + 4.8 Ma, Tindelpina Shale Member (Kendall et al. 2006); (8) 640.7 + 4.7 Ma (Kendall et al. 2007),
Black River Dolomite; (9) 657 + 17 Ma detrital age, Marino Arkose Member (Ireland et al. 1998); (10) 575 + 3 Ma (Calver et al. 2004), intrusive in Yarra
Creek Shale; (11) 582 + 4 Ma, rhyodacite flow beneath Croles Hill Diamictite (Calver et al. 2004); (12) 556 + 24 Ma detrital age, Bonney Sandstone (Ireland
et al. 1998).
118 K. GREY ET AL.
Elatina Formation and Nuccaleena Dolomite to confirm their cor-
relation with the Chinese and Namibian glaciations and determine
whether only the Croles Hill Diamictite is a Gaskiers glaciation
equivalent, or whether other Australian Elatina glacial deposits
are also of this age.
Chronometric constraints on Ediacaran successions are poor
and restricted to a Rb –Sr whole-rock isochron of 588 + 35 Ma
on the postglacial Bunyeroo Formation, based on samples above
and below the level of the Acraman impact ejecta layer in drill
hole SCYW 1a (Hill et al. 2008) and a single detrital zircon
grain, dated at 556 + 24 Ma, from the lower part of the Rawnsley
Quartzite (Preiss 2000). The ejecta layer, a significant stratigraphic
tie-line, has not been dated but was estimated to be c. 580 Ma
(Williams & Gostin 2005) based on a c. 635 Ma age for the
Elatina glaciation.
The undated glaciogenic Egan Formation of the Kimberley area
appears to correlate with carbonate units (the Julie Formation of
the Amadeus Basin, the Elkera Formation of the Georgina Basin,
the basal Bonney Sandstone of the Adelaide Rift Complex and
the Wilari Dolomite Member, and the Tanana Formation of the
Officer Basin) that lie well above the cap carbonate and just
below the first record of bilaterians in other basins (Grey &
Corkeron 1998).
Biostratigraphy
Biostratigraphic correlation is increasingly significant for
Australian Neoproterozoic correlation, and is based on chert
microfossils, palynology and stromatolite biostratigraphy. Paly-
nology provides good biostratigraphic control as well as indicating
palaeoenvironment and thermal maturity. Microfossil assemblages
consist predominantly of leiospheres, filaments and mat fragments,
mostly of conservative, long-ranging species with simple mor-
phologies, but containing some more complex and time-diagnostic
species (Schopf 1968; Schopf & Barghoorn 1969; Fairchild 1976;
Oehler 1976; Jackson & Muir 1981; Jackson & van de Graaff
1981; Zang & Walter 1992; Zang 1995; Grey & Cotter 1996;
Cotter 1997, 1999; Grey & Stevens 1997; Grey 1999a, 2005a;
Hill et al. 2000a, b). Data have been obtained from at least 1000
samples from about 30 drill holes across Australia, several consist-
ing of continuous core (Figs 8.1 & 8.6). Stromatolite biostratigra-
phy is based on .200 samples collected from .80 localities,
together with .70 horizons examined in drill core (Figs 8.2, 8.3
and 8.7, Table 8.1). Only localities with identified stromatolites
are recorded here; there are numerous samples that have yet to
be identified. Core data can be tied to field outcrops using both stro-
matolite and acritarch biostratigraphy, and taxonomic ranges are
consistent across the basins and correspond to the correlations
determined by independent methods such as lithostratigraphy,
seismic and well-log data and chemostratigraphy.
Biostratigraphic correlation is more difficult globally because
so few successions have so far been adequately documented.
Although there are probably at least two hundred assemblages
recorded in the literature (Knoll 1996), ages are often poorly con-
strained, so it is difficult to know precisely when in the Neoproter-
ozoic (or in some cases, in the Proterozoic) some of the successions
occur. Several reviews have been published, most notably a com-
pilation of c. 3500 occurrences of Proterozoic and Early Cambrian
microfossils and microfossil-like objects, based on 470 strati-
graphic units and 316 published papers by Mendelson & Schopf
(1992). From the taxonomic point of view, the catalogue of acri-
tarchs published by Fensome et al. (1990) is invaluable, but did
not include microfossils other than those considered to be acri-
tarchs. Many significant publications were overlooked in this
catalogue and it is difficult to extract taxa that fall within the
age constraints of interest.
Given the numerous revisions to stratigraphy of recent years,
updating the records is a major undertaking and beyond the
scope of this review. Moreover, very few of the records contain
adequate range charts, so it is very difficult to determine the rela-
tive stratigraphic positions and ranges of each of the species.
Palynology
In Australia, identical palynological successions are present in
Empress 1A and Lancer 1 and are known from several other
western Officer Basin drill holes, particularly LDDH 1, Kanpa
1A and Yowalga 3 (Grey & Cotter 1996; Cotter 1997, 1999;
Grey & Stevens 1997; Grey 1999a, 2005a; Hill et al. 2000a)
(Fig. 8.1). Elements of the various assemblages have been
recorded in other drill holes, but distributions and ranges have
not yet been thoroughly documented (Grey, unpublished data).
The oldest described Cryogenian assemblage is probably that
from the Alinya Formation in Giles 1, eastern Officer Basin
(Zang 1995). This assemblage is puzzling because it contains Tra-
chyhystrichosphaera, Cymatiosphaeroides kullingii and Valeria
lophostriata (normally found in Tonian or Mesoproterozoic suc-
cessions), vase-shaped microfossils, assigned to Melanocyrillium
sp. (also usually found in early Cryogenian assemblages), Cryo-
genian species similar to those in the Bitter Springs and Browne
Formations, and spiny forms resembling acanthomorphs recorded
from the Ediacaran (although none of the latter seems identical to
other Australian Ediacaran species). This strange combination of
taxa may indicate reworking of older rocks and possibly the pres-
ence of two disconformable successions. This assemblage requires
reassessment.
Vase-shaped microfossils are locally abundant in chert in the
Black River Dolomite below the diamictitic Julius River
Member in Tasmania (Saito et al. 1988), and are consistent with
an age pre-dating the Sturt glaciation (Porter & Knoll 2000).
Elsewhere in Australia, acritarch and microfossil assemblages
show consistent distribution patterns. In the western Officer
Basin, they are present in the Browne Formation and its lateral
equivalents (Jackson & van de Graaff 1981; Grey & Cotter 1996;
Cotter 1997, 1999; Grey & Stevens 1997; Grey 1999a, 2005a).
Well-preserved microfossils from chert in the Browne Formation
at the Madley Diapirs (Jackson & van de Graaff 1981, p. 34;
Cotter 1997) show low species diversity and mainly represent
permineralized cyanobacterial mat. Palynomorphs from the
Browne Formation in Browne 1 and 2, Kanpa 1A, Dragoon 1
and Yowalga 2 and 3 include Myxococcoides sp., Siphonophycus
sp., Leiosphaeridia spp. and Synsphaeridium spp., many of
which are long ranging and have been recorded from younger
and older successions (Cotter 1997, 1999). Some species of
Simia and Satka colonialica may have more restricted stratigraphic
distributions. Similar taxa were subsequently identified in other
Officer Basin drill holes, including LDDH 1, Empress 1A and
Lancer 1 (Grey & Cotter 1996; Grey 1999a, 2005a). This lower
Buldya Group (Browne Formation) assemblage has many
species in common with assemblages from the Gillen Member of
the Bitter Springs Formation (Zang & Walter 1992).
Palynomorphs from the upper Buldya Group (Hussar, Kanpa and
Steptoe Formations) in Hussar 1, Lungkarta 1, Yowalga 2 and 3,
Empress 1A and Lancer 1 (Grey & Cotter 1996; Cotter 1999;
Grey 1999a, 2005a; Hill et al. 2000a; Grey et al. 2005) include
Cerebrosphaera buickii, C. ananguae, Chuaria sp. cf. circularis,
Coneosphaera sp., Eoentophysalis croxfordii, Eomicrocystis sp.
cf. elegans, Leiosphaeridia crassa, L. sp. cf. exsculpta, L. jacutica,
L. minutissima, L. tenuissima, L. ternata, Myxococcoides canta-
brigiensis, Ostiana microcystis, Pterospermopsimorpha insolita,
Satka colonialica, Simia annulare, Stictosphaeridium sinapticuli-
ferum, ?Symplassosphaeridium sp., Synsphaeridium spp., Tasma-
nites sp., as well as filamentous microfossils Calyptothrix sp. cf.
alternata, Clavitrichoides rugosus, Oscillatoriopsis amadeus,
Oscillatoriopsis sp., Siphonophycus kestron, S. robustum, S. septa-
tum, S. solidum and S. typicum. Although taxa from the previous
 AUSTRALIAN CRYOGENIAN CORRELATIONS 119

Fig. 8.6. (a– f) Cerebrosphaera buickii from Spitsbergen, Australia and the Grand Canyon. (a, b) Svanbergfjellet Formation, Spitsbergen: HUPC 62713B, P-2945-47M
(S-33-4), holotype (a); HUPC 62763, 86-G-33-2S, SEM (b). Photos in a and b are courtesy of N. Butterfield (specimens described in Butterfield et al. 1994). (c) Hussar
Formation, Officer Basin, Empress 1, 808.2 m/2, L-38-4. (d) Hussar Formation, Officer Basin, Lancer 1, 605.14 m/2, W-44-3, Officer Basin. (e) Pirilyungka Formation,
Vines 1, Officer Basin, 1685.33 m/2, L-57-2. (f ) BLD 4, Skillogalee Dolomite, Stuart Shelf, BLD 4, 1008.2 m, E-56-0. (g) PP12, Anama Siltstone, Stuart Shelf, PP12,
156.9 m, N-58-2. (h) ‘Finke beds’, Amadeus Basin, 1502.0 m, Q-33-0. (i) Lower Chuar Group (Galeros Fm), Grand Canyon, USA, SP14-63-14 (Nagy 2008, MSc thesis,
photo courtesy of S. Porter).

assemblage persist, the more elaborate forms tend to be fewer in
number, so assemblages are dominated by leiospheres, Siphono-
phycus spp., Synsphaeridium spp., Chuaria sp. cf. circularis and
Cerebrosphaera buickii.
In this assemblage, the distinctive acritarch Cerebrosphaera
buickii consistently first appears about the middle of the Hussar
Formation in the western Officer Basin (Grey & Cotter 1996;
Cotter 1999; Hill et al. 2000a). In the Adelaide Rift Complex,
C. buickii is present in the Anama Siltstone Member of the
Rhynie Sandstone, near the base of the Burra Group in drill hole
PP 12 and the Skillogalee Dolomite (in drill hole BLD 4) in the
middle Burra Group (Hill et al. 2000a). The same species was
recently identified in the ‘Finke beds’ between the Johnnys
Creek Member of the Bitter Springs Formation and the Areyonga
Formation in Wallara 1 in the Amadeus Basin (Grey, unpublished
observations).
C. buickii is a key species because it has been recognized
at about the same stratigraphic level across Australia (Hill
120 K. GREY ET AL.

Cambrian
c. 695

c. 600
c. 800

c. 635
c. 660
Approximate age (Ma)

11
14
15

12
13

10
Stratigraphic package

8
9

6
7

1
5

2
3
PQ/RBb Parmites f. indet.
GM Tungussia erecta
LCM2/BrF Minjaria pontifera
LCM2/ERD Linella avis

Acaciella australica
LCM2/BrF/?CD equiv. Kulparia alicia

assemblage
LCM2/BrF Jurusania nisvensis
LCM2/BrF/WWF Inzeria intia

Sturt glaciation
CD Gymnosolen f.indet.
LCM2/BrF Boxonia pertaknurra
LCM2/BrF/SHF Basisphaera irregularis
LCM2/BrF/SHF/KF/WWF/Yb/CD Acaciella australica
BuG/UBG/UTG/JCb/WHF/JRM (clasts) Baicalia burra

Baicalia burra
assemblage
UBG Tungussia n. form
UBG Stomatolite n. form (pseudocolumnar)
Elatina glaciation
UBG/UTG Conophyton n. form
UBG/BuG Tungussia wilkatanna
THF Omachtenia n. form
BgL Acaciella augusta

assemblage
multiplex
Inzeria
BgL Inzeria conjuncta
BgL Inzeria multiplex

BgL Boxonia melrosa

AnF Katavia costata

AnF Tungussia etina

AnF Linella f. indet.

EF Inzeria f. indet.

assemblage
munyallina
Linella
EF Linella munyallina
EF Jurusania burrensis
EF Kulparia kulparensis
?AF Clast supported stromatolite
PF Divergent branching stromatolite
TF unnamed stromatolite
PiS & OF/WF (in cap carbonate) Kotuikania juvensis = Elleria minuta
upper carbonate BdF Tesca stewartii

upper carbonate BwF Stromatolite form 1 Walter et al.

upper carbonate BwF Eleonora boondawarica

upper carbonate BwF Acaciella savoryensis

Basal BnS Linella f. indet.

JF/Basal BnS/WDM/EkF/EgF Tungussia julia

EgF Stromatolite form Grey & Corkeron

EkF Georgina howchini

Fig. 8.7. Range chart of Neoproterozoic stromatolite distributions in Australia (see Table 8.1 for guide to abbreviations and additional stratigraphic information). Details
of the stratigraphic units mentioned here can be found by searching the Geoscience Australia Stratigraphic Names database on http://dbforms.ga.gov.au/pls/www/
geodx.strat_units.int.
 AUSTRALIAN CRYOGENIAN CORRELATIONS 121

Table 8.1. Key to abbreviations in stromatolite range chart (Fig. 8.7)

Symbol Package Stratigraphic unit

AnF 8 Angepena Formation, Upalinna Subgroup, Umberatanna Group, Adelaide Rift Complex
AF 6 Aralka Formation (Mount Conner, Fenn Gap & Pioneer Ssst clast), Amadeus Basin
BdF 1 Boord Formation, western Amadeus Basin
BgL 9 Brighton Limestone, Upalinna Subgroup, Umberatanna Group, Adelaide Rift Complex
BnS 1 Bonney Sandstone, Pound Subgroup, Wilpena Group, Adelaide Rift Complex
BrF 13 Browne Formation, lower Buldya Group, western Officer Basin
BuG 12 Burra Group (Skillogalee Dolomiteand other units), Adelaide Rift Complex
BwF 1 Boondawari Formation, western Officer Basin
CD 13 Coominaree Dolomite, Callanna Group, Adelaide Rift Complex & eastern Officer Basin
EF 7 Etina Formation, Upalinna Subgroup, Umberatanna Group, Adelaide Rift Complex
EgF 1 Egan Formation, Louisa Downs Group, central Kimberley region
EkF 1 Elkera Formation, Mopunga Group, Georgina Basin
ERD 13 Eliot Range Dolomite, Ruby Plains Group, eastern Kimberley area
GM 14 Gillen Member, Bitter Springs Formation, Amadeus Basin
JCb 12 Johnnys Creek beds’, Amadeus Basin
JF 1 Julie Formation, Amadeus Basin
JRM 12 Julius River Member, Black River Dolomite, Togari Group, Smithton Basin
KF 13 Karara Formation, Tarcunyah Group, NE Pilbara
LCM2 13 Loves Creek Member unit 2, Bitter Springs Formation, Amadeus Basin
OF 2 Olympic Formation, Amadeus Basin
PF 5 Pertatataka Formation, Amadeus Basin
PiS 2 Pioneer Sandstone, Amadeus Basin
PQ 15 Paralana Quartzite, Arkaroola Subgroup, Callanna Group, Adelaide Rift Complex
RBb 15 River Broughton beds, Curdimurka Subgroup, Callanna Group, Adelaide Rift Complex
SHF 13 Skates Hills Formation, Sunbeam Group, western Officer Basin
TF 4 Trezona Formationm, Upalinna Subgroup, Umberatanna Group, Adelaide Rift Complex
THF 10 Tapley Hill Formation, Nepouie Subgroup, Umberatanna Group, Adelaide Rift Complex
UBG 12 Upper Buldya Group (upper Hussar, Kanpa & Steptoe Formations), western Officer Basin
UTG 12 Upper Tarcunyah Group Undifferentiated, Officer Basin, south east Pilbara
WDM 1 Wilari Dolomite Member, Tanana Formation, Ungoolya Group, eastern Officer Basin
WF 2 Wahlgu Formation, western Officer Basin
WHF 12 Wright Hill Formation, eastern Officer Basin
WWF 13 Waltha Woora Formation, Tarcunyah Group, NE Pilbara
Yb 13 Yackah beds, Plenty Group, Georgina Basin

Details of the stratigraphic units mentioned here can be found by searching the Geoscience Australia Stratigraphic Names
database: http://dbforms.ga.gov.au/pls/www/geodx.strat_units.int.

et al. 2000a) and could be an important index species for global
correlation (Fig. 8.6). It was previously recorded from the Svan-
bergfjellet Formation and overlying Draken Formation, Akademi-
kerbreen Group, Spitsbergen (Knoll et al. 1991; Butterfield et al.
1994) and the Tanner, Jupiter and Carbon Canyon Members of
the Galeros Formation, lower and middle Chuar Group, Grand
Canyon, USA (Nagy 2008). Considering that there is good geo-
chronological constraint on the first appearance (younger than
802 + 10 Ma for the Rook Tuff and older than 795 + 5 Ma for
the Skillogalee Dolomite), and the coincidence or near coincidence
with three other key markers (d13C, 87Sr/86Sr and the first appear-
ance of the stromatolite Baicalia burra), the first appearance of
C. buickii is a strong candidate for a boundary that could be corre-
lated globally (see section ‘Chemostratigraphy’ below).
Sturt glacial and post-glacial assemblages are poorly known
throughout Australia. In the Adelaide Rift Complex only a few
simple, poorly preserved leiospheres have been recorded
(Timofeev 1966; Grey, unpublished data). Elsewhere, successions
appear barren. In Western Australia, a hiatus with a karstified
surface characterizes the interval. In Vines 1, there may be a tran-
sition into the Pirrilyungka Formation, a newly defined diamictite
unit that is probably a Sturt Tillite equivalent, and is directly
overlain by the Wahlgu Formation, a diamictite correlated with
the Elatina Formation (Haines et al. 2008; Hill et al. 2011).
C. buickii is present, possibly in situ, in the turbiditic Unit 1 at
the base of the Pirrilyungka Formation (Haines et al. 2008), but
is only present as sparse, reworked fragments in the diamictite.
Sparse, reworked fragments are present in the Chambers Bluff
Tillite in Nicholson 2, eastern Officer Basin (Eyles et al. 2007),
another probable Sturt glacial equivalent. Elsewhere diamictite
and interglacial successions appear barren.
The overlying Ediacaran acritarch succession is very well con-
strained lithostratigraphically across Australia, and closely
matches correlations based on chemostratigraphy, seismic inter-
pretation and stromatolite biostratigraphy, even though absolute
dating remains poor (Grey 2005b; Willman et al. 2006; Willman
& Moczydłowska 2008). Acritarch distributions are consistent
and well constrained from above the Elatina glaciation up to the
change of facies associated with the appearance of the bilaterian
Ediacara fauna.
Although biostratigraphic distributions are consistent across
Australia, they seem to be inconsistent with some of the distri-
butions reported from NW Canada (Macdonald et al. 2010).
There does not seem to be anything equivalent to the Wynniatt
Group assemblage in Australia. Vase-shaped microfossils appear
sparse in the Australian successions, and there are no reports
of C. buickii at the appropriate stratigraphic levels in north
America, with the exception of the Chuar Group.
Stromatolites
Stromatolites have been recorded at various levels throughout the
Neoproterozoic of Australia (Walter 1972; Preiss 1972, 1973a, b,
122 K. GREY ET AL.

Fig. 8.8. (a– h) Taxa found in the Acaciella australica Stromatolite Assemblage: from Jay Creek, Amadeus Basin, Northern Territory (b, f, g); from Lancer 1, Officer
Basin, Western Australia (c, d, e). (a) Bioherms of A. australica in the Skates Hills Formation, Skates Hills, Officer Basin, Western Australia. (b) A. australica, polished
slab showing branching columns. (c) Basisphaera irregularis, drill core split face, 349 m. (d) Boxonia pertaknurra, drill core surface, 332.5 m. (e) Inzeria intia, drill core
split face 333 m. (f ) Kulparia alicia, GSWA 109264C, polished slab showing branching columns. (g) Linella avis, thin section, GSWA 109267. (h) Minjaria pontifera,
cut core face, Wallara 1, 1979 m, Amadeus Basin, Northern Territory.
 AUSTRALIAN CRYOGENIAN CORRELATIONS 123

Fig. 8.9. Taxa found in the Baicalia burra Stromatolite Assemblage. All specimens from upper Buldya Group, Officer Basin, Western Australia; all core samples from
Empress 1A. (b– f, h, i) Drill core split faces. (a) B. burra, GSWA 84708, upper Buldya Group, near Lake Throssell. (b) B. burra, core face, c. 775 m, Kanpa Formation.
(c) B. burra, small columns, 510.8 m, Steptoe Formation. (d, e) Tungussia wilkatanna, Steptoe Formation: (d) 656.9 m (e), 513.5 m. (f ) Tungussia new form,
Empress 1A, 1077.4 m, Hussar Formation. (g, h) Conophyton new form, (g) GSWA 139573, upper Buldya Group, near Constance Headland; (h); 779.0 m, Kanpa
Formation. (i) Unnamed pseudocolumnar stromatolite, 520.7 m, Steptoe Formation.
 124
120o 130
o
140o 150o ADELAIDE RIFT COMPLEX ADELAIDE RIFT COMPLEX
Peake and Denison Ranges Mid-North (Southern Flinders Ranges)
13
Thickness C (% ) 13
C (% )
Calthorinna (m) Thickness
-2 0 2 4 6 8 Belair (m)
KIMBERLEY Tillite
1528 Subgroup -2 0 2 4 6 8
REGION 1007

Dolomite
Member
Auburn
GEORGINA
20o 20
o
1400 900
AMADEUS BASIN
BASIN
Arunta Provin
ce 1300 800
Western
Officer W1 D A3
J
Basin
L B 1200 700
PEAKE & DENISON

Member
E
M5 C RANGES 1100 600
OFFICER

Kalachalp
Eastern
BASIN Officer
UBPD
o Basin 30o
30
Gawler
1000 500

W
Craton
STUART ADELAIDE 900 400
SHELF
SD
RIFT COMPLEX I
UBMF
800 300 J

Member
700 200

40
o
Drillhole
Field section KING ISLAND 40o
600 100 I
A
TASMANIA
500 0
Undalya
120o 130o 140o 150
o
Quartzite
400

s
ite
H

at y
om og
ol
Assemblage Str nol
300 Undalya

ly
2 4 6 8

Pa
Quartzite
578
200
WESTERN OFFICER BASIN 500 NW TASMANIA
100
Lancer 1 Empress 1A 400 Arthur River section
0 G
Unit A (mostly 300
s

13
ite

C (% )

s
m gy

ite
13 quartz sandstone)

m gy
ol

s
C (% )
ro lo

13

ol

ite
C (% ) Thickness
at

ro lo

m gy
St no

at
Depth 200 -2 0 2 4 6 8

ol
St no
Depth

ro lo
Member (m)
ly

at
(m) -2 0 2 4 6 8

ly

St no
Pa

-2 0 2 4 6 8
I (m)

Pa

ly
-2 0 2 4 6 8 ?F

Pa
act 100 500
500 H Steptoe J
1260
Keene 1200 0
Assemblage

Fm Assemblage Bungaree 400

Basalt 600 I Quartzite

K. GREY ET AL.
Kanp

600 1100
300
G
C.b.

Kanpa
700 H 1000 AMADEUS BASIN 200 ?E
700

s
Fm 900

ite
m gy
100

Assemblage
13

ol
C (% )

T
800

ro lo
at
St yno
800 800 Areyonga
Composite
-2 0 2 4 6 8 0

l
Pa
G 700
Formation
1049 ROCKY CAPE
900
G GROUP

C.b.
‘Finke
900 600 beds’ 1000 G
Hussar F

B.burra
Fm 1000
500
1000 Hussar
Fm
1100
F 400 900

1100 E 300 E
800
1200
E 200 ADELAIDE RIFT COMPLEX
1200
100 MTS
Peake and Denison Ranges
Lancer
Member Aeolian deposition 1300

s
700

ite
0
1300

W
act

ol
Ass.
D D

at
Depth
A.a.

om
Woolnough
A.a.

Woolnough
1400
Member
C Member
C Cadlareena (m)

r
D -2 0

St
600 Volcanics
1400
B 104
100

CALLANNA
B

Assemb.
EASTERN OFFICER BASIN B

Subgroup
Arkaroola
1500
A

GROUP

A.a.
A

CD
Basement Basement TD 1625 Manya-5 518
500
C or

Bluebush
50

outcrop
468 D
s
ite

YC
B
ol

Key
at

Depth 400
om

0
Cadlareena (m) -2 0 2 4 6 A weathered
r
St

Sandstone Stromatolite Volcanics

1139 372 outcrop
1150
Mudstone–siltstone Calcareous siltstone B 300
Assemblage

or
CDE

1200
Evaporite Dolomitic siltstone
D 200

Dolomite Missing section 1250

A
Limestone Basalt or 112
YCE

outcrop
100
Dump

1300
C
Unconformity Break in section
TD 1333 0
Heavitree Quartzite

Fig. 8.10.
 AUSTRALIAN CRYOGENIAN CORRELATIONS 125

1974, 1976, 1985, 1987; Griffin & Preiss 1976; Walter et al. 1979,
1994; Grey 1995, 1999b, 2005a; Grey & Stevens 1997; Hill et al.
2000a; Grey et al. 2005, unpublished data; Williams et al. 2007;
Planavsky & Grey 2008), but discussion here is restricted mainly
to those of demonstrated stratigraphic significance (Figs 8.2,
8.3 and 8.7–8.9, Table 8.1). Throughout Australia, pre-Sturt-
glaciation Cryogenian successions are characterized by two wide-
spread, apparently time-restricted stromatolite assemblages: the
lower Acaciella australica Stromatolite Assemblage (Figs 8.2,
8.7 and 8.8) and the upper B. burra Stromatolite Assemblage
(Stevens & Grey 1997; Hill et al. 2000a; Figs 8.3, 8.7 and 8.9).
The value of these two assemblages for biostratigraphy is clearly
established by the presence of both assemblages in the same rela-
tive stratigraphic relationships in three drill holes across two
basins. Two other assemblages, the Inzeria multiplex Stromatolite
Assemblage and the Linella munyallina Stromatolite Assemblage
(Fig. 8.7), are present in the interglacial succession, but so far are
known only from the Adelaide Rift Complex. Stromatolites are
less common in the Ediacaran succession and seem to be of
more limited biostratigraphic application than the Cryogenian
ones (Fig. 8.7, Table 8.1).
The Acaciella australica Stromatolite Assemblage (Figs 8.2, 8.7
and 8.8, Table 8.1), epitomized by Unit 2 of the Loves Creek
Member, Bitter Springs Formation, Amadeus Basin, is character-
ized by Acaciella australica and 12 associated forms (Stevens &
Grey 1997; Hill et al. 2000a). In Western Australia, this assem-
blage is known from the Browne Formation (including in continu-
ous core in Empress 1A and Lancer 1), the Skates Hills Formation
of the western Officer Basin (Hill et al. 2000a; Grey 1995, 2005a;
Grey et al. 2005) and unassigned dolomite in the Tarcunyah Group
(Grey, unpublished data).
The Acaciella australica Stromatolite Assemblage is wide-
spread in the Amadeus Basin, including the Wallara 1 drillcore
(Walter 1972; Walter et al. 1979; Grey, unpublished data). It is
also present in the Coominaree Dolomite of the Adelaide Rift
Complex, equivalent successions in the eastern Officer Basin,
the Yackah beds of the Georgina Basin, and in the Eliot Dolomite
in the eastern Kimberley (Preiss 1973b, 1987; Grey 1995; Grey &
Blake 1999; Grey et al. 2005; Planavsky & Grey 2008). In South
Australia, the A. australica Stromatolite Assemblage in the Coo-
minaree Dolomite is older than the 802 + 10 Ma Rook Tuff. Its
stratigraphic position relative to other biostratigraphic markers,
d13C and 87Sr/86Sr excursions and the Sturt glaciation is well-
documented in Australia (Hill & Walter 2000; Hill et al. 2000a,
b; Hill 2005; unpublished data). One form, Linella avis, was first
described from the Maly Karatau Ridge, Chichkan Formation,
Malokaroy Group and the Talass Range in Kazakhstan (Krylov
1967, 1975; Sergeev 1989; Grey & Blake 1999; Raaben et al.
2001), which is possibly equivalent in age to the Bitter Springs
Formation (Sergeev 1989; pers. comm. 1996). Poorly documented
forms from the Mackenzie Mountains of Canada appear to belong
to this assemblage and require further investigation (Grey, unpub-
lished data).
In Western Australia, the overlying B. burra Stromatolite
Assemblage (Figs 8.3, 8.7 and 8.9, Table 8.1) ranges from about
the middle Hussar Formation, through the Kanpa Formation and
into the overlying Steptoe Formation. The assemblage is
dominated by B. burra, which is geographically widely distributed.
In the Kanpa Formation, B. burra is accompanied by Conophyton
new form, and in both the Officer Basin and Adelaide Rift
Complex by T. wilkatanna. The assemblage has been recorded in
Empress 1A, Lancer 1, Kanpa 1A and Wallara 1, as well as
being known from many Officer Basin field localities (Grey
et al. 2005). B. burra and Conophyton form indet. have also
been collected from an unassigned unit near the top of the
Tarcunyah Group near Constance Headland in the northwestern
Officer Basin (Grey, unpublished data). A previously unknown
form, tentatively identified as Tungussia? form indet., is present
in both Empress 1A and Lancer 1.
Both B. burra and T. wilkatanna are present in the Burra Group
of the Adelaide Rift Complex (Preiss 1972, 1974, 1985, 1987),
supporting correlation between the upper Buldya Group and
at least part of the Burra Group. B. burra has also been recorded
in Tasmania (Griffin & Preiss 1976), as erratics in Sturt glacial
deposits (Julius River Member) of the Black River Dolomite
(Calver 1998; Grey, unpublished data). Unpublished data
suggest that this assemblage is present in western USA and
Canada in the Mackenzie Mountains, but none of the stromato-
lites in this area or in other areas of NW Canada has been
formally identified. Testing whether or not the apparently abun-
dant stromatolites show similar distribution patterns to those
of Australia could provide a significant step forward in
global correlation.
Chemostratigraphy
In the last decade, advances in Neoproterozoic chemostratigraphic
(d13C and 87Sr/86Sr) correlations throughout Australia have pro-
vided ties between the Australian Cryogenian and successions in
Canada, Svalbard, Namibia and Mongolia (Calver & Lindsay
1998; Calver 2000; Calver & Walter 2000; Hill & Walter 2000;
Hill et al. 2000b; Walter et al. 2000; McKirdy et al. 2001; Hill
2005, unpublished data). Figures 8.10 and 8.11 show a compilation
of pre-Sturt glaciation, and Sturt glaciation and post-Sturt glacia-
tion d13C stratigraphy of Australia, respectively. There are few
reliable Australian Cryogenian 87Sr/86Sr ratios, because dolomite
predominates over limestone. Despite minimal utility for
Australia-wide correlations, they are extremely important globally
(see section ‘Discussion’).
There is no official terminology in d13C or 87Sr/86Sr chemostra-
tigraphy. In the Neoproterozoic most authors focus on correlating
negative parts of d13C curves, because Neoproterozoic rocks pre-
dominantly have positive d13C values and any negative d13C
values occur only for relatively brief periods of time. This has
led to the use of ‘excursion’ and ‘anomaly’, often preceded by
‘negative’, the connotation being that these are short-lived
events. Even when not preceded by ‘negative’ it could confuse
the reader into that interpretation. There are significant ‘positive’
anomalies or excursions, particularly in Cryogenian rocks, that
are important for global correlation. The term ‘stage’ has been
used to describe the long interval of negative d13C values in the
Loves Creek Member of the Bitter Springs Formation (Halverson
et al. 2005), but could equally be applied to long intervals of

Fig. 8.10. (Continued) Pre-Sturt glaciation d13C stratigraphy of Australia. Location of drill hole and outcrop sections shown in inset (see Fig. 8.4 for explanation of
abbreviations). Horizontal lines in palynology and stromatolites columns indicate first appearances: (A.a) Acaciella australica; (B) Baicalia burra Stromatolite
Assemblage; (C.b) Cerebrosphaera buickii. See text for an explanation of d13C correlation intervals A–J. Western Officer Basin: lithostratigraphy (Stevens & Apak
1999; Haines et al. 2004), d13C (Hill & Walter 2000; Hill 2005) and biostratigraphy (Grey 1999a, b, 2005a); eastern Officer Basin: lithostratigraphy (Morton 1997), d13C
(Hill & Walter 2000) and biostratigraphy (Preiss 1987; Grey 1995); Peake and Denison Ranges: lithostratigraphy (Fairchild 1976; Ambrose et al. 1981), d13C (Hill &
Walter 2000) and biostratigraphy (Preiss 1987; Hill et al. 2000a); southern Flinders Ranges: lithostratigraphy (Preiss 1987), d13C (Hill & Walter 2000) and
biostratigraphy (Hill et al. 2000a); Amadeus Basin: lithostratigraphy (Hill et al. 2000b, and references therein), d13C (Hill et al. 2000b) and biostratigraphy (unpublished
data); Tasmania: lithostratigraphy and d13C (Calver 1998). Abbreviations: CD, Coominaree Dolomite; CDE, Coominaree Dolomite equivalent; MTS, molar tooth
structure; TD, total depth of drill hole; WM, Woolnough Member; YC, Younghusband Conglomerate; YCE, Younghusband Conglomerate equivalent.
126 K. GREY ET AL.
positive values. The problem is that ‘stage’ has lithostratigraphic
and biostratigraphic connotations. In this manuscript we prefer
the use of ‘d13C interval’ or just ‘interval’ to describe those
parts of d13C curves that are significant for Australian and global
correlation and that may include positive or negative d13C excur-
sions, or both, or parts of d13C curves that do not have anomalously
low or high values. This terminology is preferable for two reasons.
The first is that it does not exclude those parts of d13C curves that
are not anomalous; a long interval of moderately positive d13C
values can be equally important for correlation. Second, in some
cases it minimizes focusing on every peak or trough in d13C
curves that may be a result of local or regional factors, and
instead concentrates on the broader-scale d13C trends or patterns.
Ten distinct d13C intervals (A–J) have been identified below the
base of the Sturt glaciation in Australia based on correlations
between at least two separate drill hole or outcrop sections (see
also Hill & Walter 2000; Walter et al. 2000) (Fig. 8.10).
Beneath interval A there is only a d13C stratigraphy in the
Amadeus Basin (Gillen Member, Bitter Springs Formation) so
no Australia-wide correlations are possible; however, this interval
is potentially important for global correlation. Intervals A –D are
older than c. 800 Ma and coincide with the A. australica Stroma-
tolite Assemblage described above. The palynological and micro-
fossiliferous records at this stratigraphic level across Australia are
also consistent. The predominant chemostratigraphic feature at
this stratigraphic level is the presence of 13C-depleted d13C
values ( –4 to 0‰) (intervals B and D; best represented by Units
1 & 2 of the Loves Creek Member, Bitter Springs Formation,
Amadeus Basin), which are separated by a comparatively brief
interval of positive d13C values (c. þ6‰) (interval C). There are
no reliable 87Sr/86Sr ratios at this stratigraphic level, but there
are in older sediments, in the Gillen Member of the Bitter
Springs Formation.
Intervals E –J are younger than c. 800 Ma and older than
c. 660 Ma, are characterized by predominantly positive d13C
values (mostly þ2 to þ6‰), and coincide with the first appearance
of C. buickii and the B. burra Stromatolite Assemblage. First
records of these taxa occur in interval E in the Amadeus Basin
and interval F in the western Officer Basin (Grey, unpublished
data), and in the lowermost Burra Group in the Adelaide Rift
Complex (this chapter), which probably corresponds with interval
E (however, further d13C stratigraphy needs to be completed at this
level in the Adelaide Rift Complex in order to verify this). d13C
values of between –3 and 0‰ in the middle Hussar Formation
may correlate with the Black River Dolomite in Tasmania (interval
F); however, this correlation is tentative because the Arthur River
section is not as thick as in Lancer 1 and Empress 1A in the western
Officer Basin. It is probable that this part of the succession (interval
F) is missing from the central (Wallara 1 locality) and northeastern
(AS 27 and AS 28 localities) Amadeus Basin. Interval F has also
not yet been identified in the Adelaide Rift Complex; further
work needs to be completed to see whether it does exist, because
it could potentially be important in strengthening Australian corre-
lations. Within the ranges of C. buickii and B. burra, d13C values
peak at about þ8‰ (interval H), immediately preceding an inter-
val of 13C-depleted d13C values (interval I). d13C values then return
to between þ4 and þ6‰ (interval J) just prior to the Sturt glacial
deposits (best represented in the Kalachalpa Formation of the
Peake and Denison Ranges.
The positive shift in d13C values from between – 4 and 0‰
(interval D) to between þ2 to þ6‰ (interval E) coincides with
a rise in 87Sr/86Sr from 0.7057 to 0.7063, as recorded in the
Bitter Springs Formation (Fanning 1986, reproduced in Hill
et al. 2000b, unpublished data). The positive shifts in d13C and
87
Sr/86Sr are followed by the first records of C. buickii and the
B. burra Stromatolite Assemblage (interval E). Therefore, there
is an approximate coincidence between the first appearance of
C. buickii and positive shifts in d13C and 87Sr/86Sr at c. 800 Ma,
which is consistent with data from Svalbard and Canada (see
section ‘Discussion and Global Correlation’). The next youngest
well-preserved 87Sr/86Sr ratios in Australia are from within the
Sturt glacial deposits in northwestern Tasmania, where the
lowest ratio is 0.7063 (Calver 1998).
In Sturt glacial deposits and post-Sturt –pre-Elatina glaciation
(interglacial) deposits there are lithological, geochemical (d13C
intervals K – P) and geochronological similarities between the
Amadeus Basin, Stuart Shelf, Adelaide Rift Complex and north-
western Tasmania. The remainder of the interglacial succession,
excluding Elatina glaciation deposits, is only known in the Ade-
laide Rift Complex, so no interbasinal comparison is possible.
However, the Adelaide Rift Complex d13C record compares
closely with global records (see section ‘Discussion and Global
Correlation’). A rise in 87Sr/86Sr ratios from 0.7063 within the
Sturt glacial deposits (northwestern Tasmania; Calver 1998) to
0.7071 in the interglacial succession (Adelaide Rift Complex;
McKirdy et al. 2001) also compares closely with global records.
For Elatina glaciation units there are data from dolomite beds –
originally primary precipitates – within the Olympic Formation
(d13C, –1 to 3.5‰; Kennedy et al. 2001; Skotnicki et al. 2008)
and Pioneer Sandstone (d13C, –2 to 2‰; Calver 1995) that
define an upward increasing trend. These d13C data are comparable
to those within the Ghaub Formation in Namibia (Kennedy et al.
2001) (see section ‘Discussion and Global Correlation’).
Discussion and global correlation
From correlations shown in Figures 8.10 and 8.11, a composite
Australian Cryogenian chemostratigraphy (d13C and 87Sr/86Sr)
is compared with other global successions including the first
records of C. buickii and the stromatolite B. burra (Fig. 8.12). In
Spitsbergen and the USA, only Baicalia has been identified. To
avoid ambiguous d13C correlations between global successions,
the tie-lines in Figure 8.12 (dark green dashed lines) are chosen
to coincide with the most extreme values, which avoids the more
nuanced fluctuations that may be a result of more localized
(regional) heterogeneity. In this way it is possible to show simi-
larities in d13C and 87Sr/86Sr that extend globally.
Three ‘broad’ intervals of the global d13C curves are possible,
which are consistent with radiometric dates, 87Sr/86Sr and
palaeontological data.
Australia – Canada– Svalbard: .800 Ma
This correlation is based largely on an extended stratigraphic inter-
val where d13C values fall between –4 and 0‰ (intervals B & D in
Australia; Fig. 8.10). The c. 830–800 Ma age for this anomaly is
based on radiometric dates from Australia (see section ‘Geochro-
nology’) and is consistent with a c. 780 Ma age for the Little Dal
Basalt at the top of the Mackenzie Mountains Supergroup in
Canada (Jefferson & Parrish 1989; Harlan et al. 2003) and a
c. 811.5 Ma age for a bedded tuff in the Fifteenmile Group of
the Coal Creek Inlier, Ogilvie Mountains, northwestern Canada
(Macdonald et al. 2010), which lies just beneath the extended
interval of negative d13C values. It also corresponds with a
rise in seawater 87Sr/86Sr from 0.7055–0.7057 to 0.7062–
0.7063. A third independent line of evidence for the uniqueness
of this part of the Cryogenian record is the first record of C.
buickii. In Spitsbergen, C. buickii first occurs in the upper parts
of the Lower Dolomite Member of the Svanbergfjellet Formation
(Butterfield et al. 1994), which is in the uppermost part of the
extended interval of negative d13C values between c. 830 and
800 Ma (intervals B and D in Australia; Fig. 8.10). In Australia,
the first record of C. buickii is in slightly younger rocks but there
are no known suitable siltstone facies in Australia at this time
within which C. buickii is typically found. In the Grand Canyon,
USA, the first record of C. buickii is in the lowermost Chuar
 o
120o 130o 140o 150

KIMBERLEY
REGION

GEORGINA
ADELAIDE RIFT COMPLEX
o
20
o
20
AMADEUS BASIN
BASIN e
Western
Arunta Provinc
LS
Composite field and drillhole sections
Officer W1
Basin 13
Musgrave Province PEAKE & DENISON C (‰)
RANGES (m)
OFFICER
BASIN Eastern
Thickness
-10 -8 -6 -4 -2 0 2 4 6 8 10
Officer
30o Basin SC 30
o

Gawler
B2 El
M O
Craton 500
STUART
EC
ADELAIDE Y
DC
SHELF RIFT COMPLEX

Trezona
Fm.
1000 km
P
Drillhole 0
40
o
Field section KING ISLAND 40
o
Thickness -10 -8 -6 -4 -2 0 2 4 6 8 10
F
390

Enorama
TASMANIA

Shale
o
120o 130o 140o 150

NE O
0
AMADEUS BASIN Thickness -10 -8 -6 -4 -2 0 2 4 6 8 10 Key

AUSTRALIAN CRYOGENIAN CORRELATIONS

880
Limbla Syncline field section
Sandstone Glacial diamictite

Etina Formation
13
Thickness C (‰)
(m)
-6 -4 -2 0 2 4 500 Mudstone–siltstone Diamictite
562
N
Ringwood Member

Dolomite Ooids

UMBERATANA GROUP
500 ?
Limestone Missing section
0
Thickness -10 -8 -6 -4 -2 0 2 4 6 8 10
130

Brighton
Unconformity

Limest.
100
400
13
0 Cdol (‰ VPDB)
Depth -10 -8 -6 -4 -2 0 2 4 6 8 10
Aralka Formation

0 13
STUART SHELF WD
Ccarb (‰ VPDB)
13
estimated from + Corg
300 SCYW 1a drillhole
13
M Depth C (‰)
Whyalla

Tapley Hill Formation

(m) -2 0 2 4 6
Unamed member

Sandstone 500
1140

MCGM
BL
200 M
CENTRAL 1200
Appila Tapley Hill Formation

AMADEUS BASIN 1000

Wallara-1 drillhole M NW TASMANIA
100 Forest-1 drillhole
13
C (‰) 1300 13
Pioneer Depth C (‰)
Depth
L Sandstone (m) -4 -2 0 2 4 Kanunnah
(m) -6 -4 -2 0 2 4
Tillite TSM

1281 1500 Subgroup

790
Ar L L TSM L

TOGARI
0 1300 L 800

BRD
A

Formation

GP

JRM
K
Wilyerpa

900 K
Formation
Areyonga

1400 K 975

K 1450
2031
T.D. 1075m
1400
1425
Johnnys Creek
beds

Fig. 8.11. Sturt glaciation and post-Sturt glaciation d13C stratigraphy of the Australian Cryogenian. Location of drill hole and outcrop sections shown in inset (see Fig. 8.4 for explanation of abbreviations). See text
for an explanation of d13C correlation intervals K–P. Stratigraphy and d13C data from the Amadeus Basin and Stuart Shelf (Walter et al. 2000), the Adelaide Rift Complex (Walter et al. 2000; McKirdy et al. 2001), and
northwestern Tasmania (Calver 1995, 1998). In the Adelaide Rift Complex the following sections were used: drill hole Blinman 2 for the Wilyerpa (Sturt Tillite correlative) and Tapley Hill Formations; Depot Flat field section
for the Brighton Limestone; field section east of the Oraparinna Diapir for the Etina Formation; Mallee Water field section for the Enorama Shale; Doodney’s Well Hills field section for the Trezona Formation. Abbreviations: A,
Areyonga Formation; Ar, Aralka Formation; BL, Brighton Limestone; BRD, Black River Dolomite; El, Elatina Formation; JRM, Julius River Member; MCGM, Mount Caernarvon Greywacke Member; TSM, Tindelpina

127
Shale Member; WD, Wockerawirra Dolomite; Y, Yudnamutana Subgroup.
 AUSTRALIA CANADA SVALBARD NAMIBIA MONGOLIA
87
87
Sr/86Sr (0.70..) 87
Sr/86Sr (0.70..)
87
Sr/86Sr (0.70..) 87
Sr/86Sr (0.70..) Sr/86Sr (0.70..)
60 70 80

128
13 13 60 70 80 13 13 13
AGE 60 70 80 60 70 80 60 70 80
(Ma) -8 -6 -4 -2 0 2 4 6 8 -8 -6 -4 -2 0 2 4 6 8 -8 -6 -4 -2 0 2 4 6 8 -8 -6 -4 -2 0 2 4 6 8 -8 -6 -4 -2 0 2 4 6 8
635
T W

KEELE
0.7071

ELBO
0.7072 0.7071
650 0.7071
0.7071

TW
6 7 8 0.7068

T
5 0.7063 0.7068
S Rapit

,
675

CLG
0.7064
, 700

DRAKEN BACKLUNDTOPPEN ELBO

0.7067
USA
Wilyerp
13
725
illite 4
-8 -6 -4 -2 0 2 4 6 8
S

Kalachalp 0.7069
750

Kanp
775
* -8 -6 -4 -2 0 2 4 6 8
’

SVANBERGFJ

K. GREY ET AL.
0.7062 -8 -6 -4 -2 0 2 4 6 8
((

3
800
Johnnys
,
2
* * 60 70 80

0.7063

0.7062
13
Ccarb (‰)
GP

825 Siliciclastic
1 13
Carbonate/siliciclastic Corg (‰)
C. buickii 1st record
0.7057 Carbonate
Glacial * B. burra 1st record

((
850
-8 -6 -4 -2 0 2 4 6 8
-8 -6 -4 -2 0 2 4 6 8
60 70 80
60 70 80 0.7055

0.7055

-8 -6 -4 -2 0 2 4 6 8
60 70 80

Fig. 8.12. Global d13C, 87Sr/86Sr and biostratigraphic correlation of the Australian Cryogenian. The circled numbers on the time scale correspond to geochronological ages in Figure 8.5. Australia: 87Sr/86Sr ratios from Fanning
(1986) and reproduced in Hill et al. (2000b) (Gillen Member, unpublished data) (Unit 3, Loves Creek Member), Calver (1998) (Julius River Member, Black River Dolomite), and McKirdy et al. (2001) (Brighton Limestone
and Trezona Formation). Canada: d13C data from Halverson (2006) (Little Dal Group and Coates Lake Group), and Hoffman & Schrag (2002) and James et al. (2001) (Windemere Supergroup). 87Sr/86Sr ratios from Narbonne
et al. (1994), Kaufman et al. (1997) and Halverson et al. (2007b). c. 780 Ma age for the Little Dal Basalt (Jefferson & Parrish 1989; Harlan et al. 2003). Abbreviations: CLG, Coates Lake Group; Tw, Twitya Formation; Keele,
Keele Formation. Svalbard: d13C data from Halverson et al. (2005, 2007a). 87Sr/86Sr ratios from Halverson et al. (2005, 2007b). First record of Cerebrosphaera buickii, upper Lower Dolomite Member, Svanbergfjellet
Formation (Butterfield et al. 1994). Abbreviations: Elbo, Elbobreen Formation. Namibia: d13C data from Halverson et al. (2005). 87Sr/86Sr ratios from Yoshioka et al. (2003) and Halverson et al. (2007b). 746 Ma age from
Hoffman et al. (1996) and 760 Ma age from Halverson et al. (2005). Mongolia: d13C and 87Sr/86Sr data from Shields et al. (2002). USA: d13C data from Dehler et al. (2005). First record of Cerebrosphaera buickii, Tanner
Member, Galeros Formation, Chuar Group (S. Porter and Nagy, unpublished data). 770 Ma age from (Williams et al. 2003) and 742 Ma age from Karlstrom et al. (2000).
 AUSTRALIAN CRYOGENIAN CORRELATIONS
Group, which has a maximum age of c. 770 Ma (Dehler et al. 2005,
and references therein) and is therefore consistent with the
approximate ages of the first records in Australia and Spitsbergen.
Only limited palynological studies have been carried out on the
Mackenzie Mountains successions, and have been mainly on Edia-
caran to Cambrian samples (Baudet et al. 1989), but the Wynniatt
Formation of the Shaler Supergroup in the Minto Inlier has 13 acri-
tarch species (Butterfield & Rainbird 1998). A synthesis of lithos-
tratigraphic studies of arctic Canada reveals close similarities
between the Mackenzie Mountains and Shaler Supergroups, and
more specifically between the Rusty Shale and Wynniatt For-
mation (Rainbird et al. 1996). The absence of C. buickii in the
Wynniatt Formation, whose probable correlative, the Rusty
Shale, is just beneath the extended interval of negative d13C
values between c. 830 and 800 Ma, is consistent with the strati-
graphic level of C. buickii in Spitsbergen. Although further paly-
nology in Canada above this stratigraphic level is necessary to
determine exactly where C. buickii first appears, the first record
of this acritarch is globally consistent. A fourth, and potentially
more controversial line of evidence is the first record of the B.
burra Assemblage, which across Australia consistently first
appears at about the same stratigraphic level as C. buickii. The stro-
matolite Baicalia has been reported in Svalbard (Halverson et al.
2007b) and the USA (Dehler et al. 2005), just above the first
appearance of C. buickii, but has not been identified to form level.
Together, d13C, 87Sr/86Sr and the first records of C. buickii and
B. burra have the potential to improve global correlation and poss-
ibly inform a chronostratigraphic definition of the Cryogenian
Period, and that c. 800 Ma could be a potential lower age limit
for the co-occurrence of significant geochemical and palaeontolo-
gical anomalies. The authors recognize, however, that more
research, in particular on the palaeontological record, needs to
be completed to strengthen these arguments.
Australia – Svalbard – Namibia – USA: 800– 700 Ma
d13C values in this period are mostly moderately 13C-enriched with
several short-lived intervals to between –2 and 0‰. The dis-
tinguishing feature of the d13C record is the occurrence of values
of about þ8 to þ10‰ in Australia, Svalbard, Namibia, and poss-
ibly the USA. (The Chuar Group record is problematic because it is
predominantly based on d13Corg. None of the d13Ccarb values
exceed þ4‰ and many of the ‘calculated’ (d13Ccarb ¼ 1TOC –
d13Corg) d13Ccarb values exceed þ10‰, the maximum known
values in carbonate successions from Svalbard and Namibia.) A
radiometric age of 760 Ma in Namibia places a maximum age con-
straint on the d13C peak. In Svalbard and Namibia there are two
stratigraphic levels of þ8 to þ10‰ values, which makes precise
correlation with Australia more ambiguous than lower in the suc-
cession. In the Coates Lake Group of Canada there are extremely
13
C-depleted d13C values that may correlate with the lower Polar-
isbreen Group of Svalbard, but it is difficult to be certain because in
the Coates Lake Group d13C values rapidly shift to þ8‰ and there
is no record of this in Svalbard or elsewhere. An equivalent mag-
nitude d13C shift occurs between the two large Cryogenian glacia-
tions on all continents; however, 87Sr/86Sr values of 0.7064 in the
Coates Lake Group strongly suggest it is a pre-Sturt glacial unit.
The 87Sr/86Sr record in Svalbard suggests a rise in values from
0.7062 to 0.7069 concomitant with the rise in d13C values to
their peak of þ8 to þ10‰. In Svalbard, the 87Sr/86Sr values
then decrease to 0.7067. Ratios of 0.7064 in the Coates Lake
Group and 0.7063 in the Sturt glaciation in Tasmania suggest a
further decline in seawater 87Sr/86Sr. A similar rise then decline
in seawater 87Sr/86Sr is known in Greenland (Fairchild et al.
2000) and, more recently, Scotland (Sawaki et al. 2010), so a
stratigraphic horizon at c. 760 Ma could be a strong candidate
for chronostratigraphic definition of the Cryogenian Period
because of the coincidence of maximum pre-Sturt glaciation
129
peaks in both d13C and 87Sr/86Sr within the stratigraphic range
of C. buickii and B. burra.
Australia – Canada – Svalbard – Namibia – Mongolia:
660– 635 Ma
The ‘interglacial’ part of the Cryogenian succession has a remark-
ably consistent d13C and 87Sr/86Sr record. In immediately post-
Sturt glaciation sediments, d13C values mostly fall within
the range of –4 to –2‰, with the exceptions of Australia ( –6 to
–4‰) and Svalbard ( –1 to 0‰). With the exception of Svalbard
(d13C between þ6 and þ7), d13C values then rise rapidly in all
countries to between þ8 and þ10‰, and even higher in Mongolia.
87
Sr/86Sr values rise concomitantly from 0.7068 to 0.7071. In
Australia, Canada and Namibia, d13C values then rapidly decline
to between –6 and –10‰ (the Trezona ‘anomaly’; Halverson
et al. 2005). These interglacial peaks and troughs are short-lived
compared to the Ediacaran Shuram anomaly, and are more likely
to represent a global ocean signal (Bristow & Kennedy 2008,
and references therein). Unless evidence is presented to the con-
trary, the stratigraphic level at which the maximum interglacial
peaks in d13C and 87Sr/86Sr occur is a candidate for further subdi-
vision of the Cryogenian Period, between c. 650 and 635 Ma. A
stratigraphic level within or close to the lithological boundaries
of the Sturt glaciation is not appropriate because of the apparent
range in radiometric ages for the Sturt glaciation globally
(Fanning & Link 2004, 2008).
Conclusions
The thick and well-preserved Cryogenian successions in several
Australian basins can be correlated using a variety of means, and
these techniques are most effectively used in combination.
However, poor preservation or lack of appropriate facies often
limits the applicability of biostratigraphy and chemostratigraphy,
and the latter is often ambiguous owing to the non-uniqueness of
patterns of variation. Chronometric constraints are particularly
important, especially in testing global correlations, but such con-
straints are regrettably few. Recent dating shows that the Sturt
glaciation ended later (c. 660 Ma: Kendall et al. 2007; Fanning &
Link 2008) than presumed correlatives on other continents. That
the same may be true of the Elatina glaciation is suggested (but
not proved) by dating from Tasmania (Calver et al. 2004). More
dates are sorely needed, including a focus on the age of the Elatina
glacial, the close of which marks the end of the Cryogenian Period.
As was the case with the base of the Ediacaran, it is probably not
advisable to place the lower Cryogenian boundary at a horizon
associated with glaciation (in this case, the Sturt glaciation),
which is marked by major disconformities and unconformities.
Additionally, although the Sturt glaciation appears to be consistent
in age across Australia, it may not be synchronous with glacial
units elsewhere that are normally correlated with it.
The A. australica and B. burra Stromatolite Assemblages are
Australia-wide markers, and unpublished data suggest they may
be present on other continents. However, the global distributions
of these readily correlatable Australian taxa have yet to be demon-
strated. The first appearance of the acritarch C. buickii, coupled
with the first appearance of the stromatolite B. burra, appears to
be a significant marker, possibly globally, and should be con-
sidered for determining the positioning of a boundary.
When global correlations are attempted with successions in
Canada, Svalbard, Namibia, the USA and Mongolia there are con-
sistencies between d13C and 87Sr/86Sr excursions beneath the Sturt
glaciation (c. 660 Ma in Australia) and between the Sturt and
Elatina glaciations, and the approximate ages at which these
anomalies occur in other global successions are consistent with
130 K. GREY ET AL.
the Australian time scale. It thus appears that Cryogenian d13C and
87
Sr/86Sr excursions are of global extent and probably synchro-
nous and, together with the traditional tool of period definition,
first and last appearances of fossils, provides a way forward for
further subdivision of the Neoproterozoic, at the base of the Cryo-
genian Period.
There are two stratigraphic levels where it may be possible to
make a chronostratigraphic definition of the base of the Cryogen-
ian Period. The first is in rocks of c. 800 Ma (based on radiometric
ages in Australia and Canada), where there is an interval of
negative d13C values ( –4 to 0‰) that coincides with a rise in
87
Sr/86Sr ratios from 0.7055 to 0.7062 and the first appearance
of the distinctive acritarch C. buickii. Three independent lines of
evidence make this stratigraphic level a strong candidate for
basal definition of the Cryogenian Period. The second is at
c. 760 Ma (a radiometric age from Namibia), where the pre-Sturt
glaciation d13C record peaks at about þ8‰ in Australia, Svalbard,
Namibia and the USA. 87Sr/86Sr ratios from Svalbard show a
concomitant increase from 0.7062 to 0.7069, and then a later
decrease to 0.7064 (Coates Lake Group, Canada). There is no
equivalent record of the 87Sr/86Sr rise from 0.7062 to 0.7069 in
other countries due to incompatible lithologies, which weakens
the case for boundary placement at this stratigraphic level. These
age rocks coincide with the stratigraphic ranges of C. buickii and
B. burra. The consistency of the data in numerous sections in Aus-
tralia, and limited reports of their occurrence elsewhere, indicate
that these two levels have potential for global boundary placement.
The synchroneity of glaciations in Australia and elsewhere in
the world remains unresolved, with new dates from NW Canada
conflicting with the limited data from Australia (Kendall et al.
2009). Biostratigraphic data through this interval is also sparse,
but between the Sturt and Elatina glaciations (c. 660–635 Ma)
there is the possibility for further subdivision of the Cryogenian
Period at a stratigraphic level that coincides with a peak in
d13C of between þ8 and þ10‰ and 87Sr/86Sr of between
0.7071 and 0.7072.
Appendix 1 Taxonomic citations
The names of authors of scientific names have been omitted in the
text and text figures and instead are presented here. The names are
the names of authors of taxa, not references, and therefore they are
not necessarily cited in the reference list.
Spheroidal microfossils and acritarchs
Cerebrosphaera ananguae Cotter 1999
Cerebrosphaera buckii Butterfield 1994 in Butterfield et al. 1994
Chuaria sp. cf. Ch. circularis Walcott 1899; emend. Vidal and
Ford 1985
Coneosphaera sp. Luo 1991
Cymatiosphaeroides kullingii Knoll 1984, emend. Knoll et al.
1991
Eoentophysalis croxfordii (Muir 1976) Butterfield 1994
Eomicrocystis sp. cf. E. elegans Golovenok and Belova 1984
Leiosphaeridia crassa (Naumova 1949) Jankauskas 1989
Leiosphaeridia jacutica (Timofeev 1966) emend. Mikhailova and
Jankauskas in Jankauskas et al. 1989
Leiosphaeridia minutissima (Naumova 1949) emend. Jankauskas
in Jankauskas et al. 1989
Leiosphaeridia sp. cf. L. exsculpta (Timofeev 1969) emend.
Mikhailova in Jankauskas et al. 1989
Leiosphaeridia tenuissima Eisenack 1958
Leiosphaeridia ternata (Timofeev 1966) emend. Mikhailova in
Jankauskas et al. (1989)
Melanocyrillium Bloeser 1979 ex Bloeser 1985
Myxococcoides cantabrigiensis Knoll 1982
Ostiana microcystis Hermann 1976
Pterospermopsimorpha insolita Timofeev 1969
Satka colonialica Jankauskas 1979
Simia annulare Timofeev 1969 emend. Mikhailova in Jankauskas
et al. (1989)
Stictosphaeridium sinapticuliferum
Symplassosphaeridium Timofeev 1959 ex Timofeev 1969
Synsphaeridium Eisenack 1965
Tasmanites Newton 1875
Trachyhystrichosphaera Timofeev & German 1976 emend.
Jankauskas et al. 1989
Valeria lophostriata (Jankauskas), Jankauskas 1982
Filamentous microfossils
Calyptothrix sp. cf. Ca. alternata Jankauskas 1980
Clavitrichoides rugosus Mikhailova in Jankauskas et al. 1989
Oscillatoriopsis amadeus (Schopf & Blacic 1971) Butterfield 1994
Siphonophycus kestron Schopf 1968
Siphonophycus robustum (Schopf 1968) Knoll et al. 1991
Siphonophycus septatum (Schopf 1968) Knoll et al. 1991
Siphonophycus solidum (Golub 1979) emend. Butterfield 1994
Siphonophycus typicum (Hermann 1974) Butterfield et al. 1994
Stromatolites
Acaciella australica (Howchin 1914) Walter 1972
Baicalia burra Preiss 1972
Basisphaera irregularis Walter 1972
Boxonia pertaknurra Walter 1972
Conophyton Maslov 1937
Elleria minuta Walter and Krylov in Walter et al. 1979
Inzeria intia Walter 1992
Inzeria multiplex Preiss 1973
Jurusania nisvensis Raaben 1964
Kulparia alicia (Cloud and Semikhatov 1969) Walter 1972
Linella avis Krylov 1967
Linella munyallina Preiss 1974
Tungussia Semikhatov 1962
Tungussia wilkatanna Preiss 1974
The ideas and figures presented in this manuscript have benefited from discus-
sions with numerous researchers, but in particular J. Gehling, P. Haines, R.
Hocking, W. Preiss and G. Shields. Field samples and data have been collected
by many geologists for more than 100 years, especially M. Walter, W. Preiss, I.
Williams, N. Planavsky and M. Stevens. K.G. publishes with the permission of
the Executive Director of the Geological Survey of Western Australia and is
an Honorary Associate at Monash University and the University of Western
Australia. This paper is a contribution to International Geological Correlation
Program 512: Neoproterozoic Ice Ages.
References
Ambrose, G. J., Flint, R. B. & Webb, A. W. 1981. Precambrian and
Paleozoic geology of the Peake and Denison Ranges. Geological
Survey of South Australia Bulletin 50.
Bagas, L., Grey, K. & Williams, I. R. 1995. Reappraisal of the Paterson
Orogen and Savory Basin. Western Australia Geological Survey
Annual Review 1994–95, 55 – 63.
Bagas, L., Grey, K., Hocking, R. M. & Williams, I. R. 1999. Neoproter-
ozoic successions of the northwestern Officer Basin: a reappraisal.
Western Australia Geological Survey Annual Review 1998– 99,
39 – 44.
Bagas, L., Camacho, A. & Nelson, D. R. 2002. Are the Neoproterozoic
Lamil and Throssell Groups of the Paterson Orogen allochthonous?
Geological Survey of Western Australia Annual Review, 2000/01,
45 – 52.
Baudet, D., Aitken, J. D. & Vanguestaine, M. 1989. Palynology of
uppermost Proterozoic and lowermost Cambrian Formations,
 AUSTRALIAN CRYOGENIAN CORRELATIONS
central Mackenzie Mountains, northwestern Canada. Canadian
Journal of Earth Sciences, 26, 129–148.
Black, L. P., Calver, C. R., Seymour, D. B. & Reed, A. 2004. SHRIMP
U –Pb detrital zircon ages from Proterozoic and Early Palaeozoic
sandstones and their bearing on the early geological evolution of
Tasmania. Australian Journal of Earth Sciences, 51, 885–900.
Blake, D. H., Tyler, I. M. & Sheppard, S. 1998. Geology of the Ruby
Plains 1:100 000 sheet area (4460), Western Australia. Australian
Geological Survey Organisation, Canberra.
Bristow, T. F. & Kennedy, M. J. 2008. Carbon isotope excursions and
the oxidant budget of the Ediacaran atmosphere and ocean.
Geology, 36, 863–866.
Butterfield, N. J. & Rainbird, R. H. 1998. Diverse organic-walled
fossils, including ‘possible dinoflagellates’, from the early Neoproter-
ozoic of arctic Canada. Geology, 26, 963– 966.
Butterfield, N. J., Knoll, A. H. & Swett, K. 1994. Paleobiology of
the Neoproterozoic Svanbergfjellet Formation, Spitsbergen. Fossils
and Strata, 34, 84.
Calver, C. R. 1995. Ediacarian Isotope Stratigraphy of Australia. PhD
thesis, Macquarie University, Sydney, Australia.
Calver, C. R. 1998. Isotope stratigraphy of the Neoproterozoic Togari
Group, Tasmania. Australian Journal of Earth Sciences, 45, 865– 874.
Calver, C. R. 2000. Isotope stratigraphy of the Ediacarian (Neoprotero-
zoic III) of the Adelaide Rift Complex, South Australia, and the
overprint of water column stratification. Precambrian Research,
100, 121– 150.
Calver, C. R. & Lindsay, J. F. 1998. Ediacarian sequence and isotope
stratigraphy of the Officer Basin, South Australia. Australian
Journal of Earth Sciences, 45, 513–532.
Calver, C. R. & Walter, M. R. 2000. The late Neoproterozoic Grassy
Group of King Island, Tasmania: correlation and palaeogeographic
significance. Precambrian Research, 100, 299– 312.
Calver, C. R., Black, L. P., Everard, J. L. & Seymour, D. B. 2004.
U –Pb zircon age constraints on late Neoproterozoic glaciation in
Tasmania. Geology, 32, 893– 896.
Close, D. F., Edgoose, C. J. & Scrimgeour, I. R. 2005. The Tjauwata
Group: a late Mesoproterozoic rift succession underlying the south-
western Amadeus Basin. Central Australian Basins Symposium:
Petroleum and Minerals Potential, 16 –18 August 2005, Northern
Territory Geological Survey, 39 – 40.
Coats, R. P. & Preiss, W. V. 1980. Stratigraphic and geochronological
reinterpretation of Late Proterozoic glaciogenic sequences in the
Kimberley Region, Western Australia. Precambrian Research, 13,
181– 208.
Condon, D., Zhu, M., Bowring, S., Wang, W., Yang, A. & Jin, Y. 2005.
U –Pb ages from the Doushantuo Formation, China. Science, 308,
95 – 98.
Corkeron, M. 2007. ‘Cap carbonates’ and Neoproterozoic glacigenic
successions from the Kimberley region, north-west Australia. Sedi-
mentology, 54, 871– 903.
Corkeron, M. L. & George, A. D. 2001. Glacial incursion on a Neopro-
terozoic carbonate platform in the Kimberley. Geological Society of
America Bulletin, 113, 1121– 1132.
Cotter, K. L. 1997. Neoproterozoic microfossils from the Officer Basin,
Western Australia. Alcheringa, 21, 247–270.
Cotter, K. L. 1999. Microfossils from Neoproterozoic Supersequence 1
of the Officer Basin, Western Australia. Alcheringa, 23, 63 – 86.
Cowley, W. M. & Flint, R. B. 1993. Epicratonic igneous rocks and
sediments. In: Drexel, J. F., Preiss, W. V. & Parker, A. J. (eds)
The Geology of South Australia, Volume 1, The Precambrian.
South Australian Geological Survey Bulletin, 54, 142– 147.
Dehler, C. M., Elrick, M., Bloch, J. D., Crossey, L. J., Karlstrom, K.
E. & Des Marais, D. J. 2005. High-resolution d13C stratigraphy of
the Chuar Group (ca. 770– 742 Ma), Grand Canyon: Implications
for mid-Neoproterozoic climate change. Geological Society of
America Bulletin, 117, 32 –45.
Dow, D. B. & Gemuts, I. 1969. Geology of the Kimberley Region,
Western Australia: The East Kimberley. Geological Survey of
Western Australia Bulletin, 120, 135.
Drexel, J. F. 2009. Review of the Burra Mine Project, 1980– 2008 – a
progress report. South Australia. Department of Primary Industries
and Resources Report Book, 2008/16, 76.
131
Dunster, J. N., Kruse, P. D., Duffett, M. L. & Ambrose, G. J. 2007.
Geology and resource potential of the southern Georgina Basin.
Northern Territory Geological Survey, Digital Information
Package, DIP007.
Eyles, C. H., Eyles, N. & Grey, K. 2007. Palaeoclimate implications
from deep drilling of Neoproterozoic strata in the Officer Basin and
Adelaide Rift Complex of Australia: a marine record of wet-based
glaciers. Palaeogeography, Palaeoclimatology, Palaeoecology,
248, 291–312.
Fairchild, I. J., Spiro, B., Herrington, P. M. & Song, T. 2000. Controls
on Sr and C isotope compositions of Neoproterozoic Sr-rich lime-
stones of East Greenland and North China. In: Grotzinger, J. P.
& James, N. P. (eds) Carbonate Sedimentation and Diagenesis in
the Evolving Precambrian World. SEPM Special Publication, 67,
297– 313.
Fairchild, T. R. 1976. The geological setting and palaeobiology of a late
Precambrian stromatolitic microflora from South Australia. PhD
thesis, University of California, USA.
Fanning, C. M. 1986. 87Sr/86Sr of gypsum/anhydrite and carbonate
samples. Australian Mineral Development Laboratories Report G
6696/86 (unpublished).
Fanning, C. M. & Link, P. K. 2004. U– Pb SHRIMP ages of the Neopro-
terozoic (Sturtian) glaciogenic Pocatello Formation, southeastern
Idaho. Geology, 32, 881–884.
Fanning, C. M. & Link, P. K. 2008. Age constraints for the Sturtian
Glaciation; data from the Adelaide Geosyncline, South Australia
and Pocatello Formation, Idaho, USA. In: Gallagher, S. J. &
Wallace, M. W. (eds) Neoproterozoic Extreme Climates and the
Origin of Early Metazoan Life. Selwyn Symposium of the GSA
Victoria Division, September 2008, Geological Society of Australia
Extended Abstracts, 91, 57– 62.
Fanning, C. M., Ludwig, K. R., Forbes, B. G. & Preiss, W. V. 1986.
Single and multiple grain U– Pb zircon analyses for the early Adelai-
dean Rook Tuff, Willouran Ranges, South Australia. Geological
Society of Australia Abstracts, 15, 255– 304.
Fensome, R. A., Williams, G. L., Barss, M. S., Freeman, J. M. & Hill,
J. M. 1990. Acritarchs and fossil prasinophytes: an index to genera,
species and infraspecific taxa. American Association of Stratigraphic
Palynologists Foundation, Contributions Series 25, 771.
GEOSCIENCE AUSTRALIA, Stratigraphic Names database. World
Wide Web address: http://dbforms.ga.gov.au/pls/www/geodx.
strat_units.int
Glikson, A. Y., Stewart, A. J. et al. 1996. Geology of the western
Musgrave Block, central Australia, with particular reference to the
mafic– ultramafic Giles Complex. Australian Geological Survey
Organisation Bulletin, 239, 206.
Gorter, J., Grey, K. & Hocking, R. 2007. The petroleum exploration
potential of the Australian Infracambrian (Ediacaran) of the
Amadeus and Officer basins. Australian Petroleum Production and
Exploration Association Journal, 47, 391–392 (extended abstract).
Gradstein, F. M., Ogg, J. G., Smith, A. G., Bleeker, W. & Lourens,
A. J. 2004. A new geologic time scale, with special reference to
Precambrian and Neogene. Episodes, 27, 83 – 100.
Grey, K. 1995. Neoproterozoic stromatolites from the Skates Hills For-
mation, Savory Basin, Western Australia, and a review of the distri-
bution of Acaciella australica. Australian Journal of Earth Sciences,
42, 123– 132.
Grey, K. 1999a. Appendix 7: Proterozoic palynology of samples from
Empress 1A. In: Stevens, M. K. & Apak, S. N. GSWA Empress 1
and 1A well completion report, Yowalga Sub-basin, Officer Basin,
Western Australia. Western Australia Geological Survey Record,
1999/4, 68 –69.
Grey, K. 1999b. Appendix 8: Proterozoic stromatolite biostratigraphy of
Empress 1A. In: Stevens, M. K. & Apak, S. N. GSWA Empress 1 and
1A Well Completion Report, Yowalga Sub-basin, Officer Basin,
Western Australia. Western Australia Geological Survey Record,
1999/4, 70 –72.
Grey, K. 2005a. Preliminary report on the Proterozoic biostratigraphy,
Lancer 1, Officer Basin, Western Australia. In: Mory, A. J. &
Haines, P. W. (eds) Lancer 1 Well Completion Report (interpretive
papers), Officer and Gunbarrel Basins, Western Australia. Western
Australia Geological Survey Record, 2005/4, 81.
132 K. GREY ET AL.
Grey, K. 2005b. Ediacaran palynology of Australia. Memoir of the
Association of Australasian Palaeontologists, 31, 439.
Grey, K. & Blake, D. 1999. Neoproterozoic (Cryogenian) stromatolites
from the Wolfe Basin, east Kimberley, Western Australia: correlation
with the Centralian Superbasin. Australian Journal of Earth Sciences,
46, 329– 341.
Grey, K. & Corkeron, M. 1998. Late Neoproterozoic stromatolites in
glacigenic successions of the Kimberley region, Western Australia:
evidence for a younger Marinoan glaciation. Precambrian Research,
92, 65 –87.
Grey, K. & Cotter, K. L. 1996. Palynology in the search for Proterozoic
hydrocarbons. Western Australia Geological Survey Annual Review,
1995– 96, 70– 80.
Grey, K. & Stevens, M. K. 1997. Neoproterozoic palynomorphs of
the Savory Sub-basin, Western Australia, and their relevance to
petroleum exploration. Western Australia Geological Survey
Annual Review, 1996– 97, 49– 54.
Grey, K., Hocking, R. M. et al. 2005. Lithostratigraphic nomenclature
of the Officer Basin and correlative parts of the Paterson Orogen,
Western Australia. Western Australia Geological Survey Report,
93, 89.
Griffin, B. J. & Preiss, W. V. 1976. The significance and provenance of
stromatolitic clasts in a probable Late Precambrian diamictite
in northwestern Tasmania. Proceedings of the Royal Society of
Tasmania Papers, 110, 111–127.
Haines, P. W. 1988. Storm-dominated mixed carbonate/siliciclastic shelf
sequence displaying cycles of hummocky cross-stratification, Late
Proterozoic Wonoka Formation, South Australia. Sedimentary
Geology, 58, 237–254.
Haines, P. W. 1990. A late Proterozoic storm-dominated carbonate shelf
sequence: the Wonoka Formation in the central and southern Flinders
Ranges, South Australia. In: Jago, J. B. & Moore, P. S. (eds) The
Evolution of a Late Precambrian –Early Palaeozoic Rift Complex:
Adelaide Geosyncline. Geological Society of Australia Special
Publication, 16, 177– 198.
Haines, P. W., Mory, A. J., Stevens, M. K. & Ghori, K. A. R. 2004.
GSWA Lancer 1 well completion report (basic data), Officer and
Gunbarrel Basins, Western Australia. Geological Survey of
Western Australia Record, 2004/10, 39.
Haines, P. W., Hocking, R. M., Grey, K. & Stevens, M. K. 2008. Vines
1 revisited: are older Neoproterozoic glacial deposits preserved
in Western Australia? Australian Journal of Earth Sciences, 55,
397– 406.
Halverson, G. 2006. A Neoproterozoic chronology. In: Xiao, S. &
Kaufman, A. (eds) Neoproterozoic Geobiology and Paleobiology.
Topics in Geobiology 27. Springer, New York, 231–271.
Halverson, G. P., Hoffman, P., Schrag, D. P., Maloof, A. C. &
Rice, A. 2005. Towards a Neoproterozoic composite carbon
isotope record. Geological Society of America Bulletin, 117,
1181– 1207.
Halverson, G. P., Dudás, F. Ö., Maloof, A. C. & Bowring, S. A.
2007a. Evolution of the 87Sr/86Sr composition of Neoproterozoic
seawater. Palaeogeography, Palaeoclimatology, Palaeoecology,
256, 103– 129.
Halverson, G. P., Maloof, A. C., Schrag, D. P., Dudás, F. Ö. &
Hurtgen, M. 2007b. Stratigraphy and geochemistry of a ca. 800
Ma negative carbon isotope interval in northeastern Svalbard.
Chemical Geology, 237, 23 –45.
Harlan, S., Heaman, L., LeCheminant, A. & Premo, W. 2003. Gunbar-
rel mafic magmatic event: a key 780 Ma time marker for Rodinia
plate reconstructions. Geology, 31, 1053–1056.
Hill, A. C. 2005. Stable isotope stratigraphy, GSWA Lancer 1, Officer
Basin, Western Australia. In: Mory, A. J. & Haines, P. W. (eds)
GSWA Lancer 1 Well Completion Report (Interpretive Papers),
Officer and Gunbarrel Basins, Western Australia. Geological
Survey of Western Australia Record, 2005/4, 1– 11.
Hill, A. C. & Walter, M. R. 2000. Mid-Neoproterozoic (830–750 Ma)
isotope stratigraphy of Australia and global correlation. Precambrian
Research, 100, 181– 211.
Hill, A. C., Cotter, K. L. & Grey, K. 2000a. Mid-Neoproterozoic
biostratigraphy and isotope stratigraphy in Australia. Precambrian
Research, 100, 281– 298.
Hill, A. C., Arouri, K., Gorjan, P. & Walter, M. R. 2000b. Geochem-
istry of marine and non-marine environments of a Neoproterozoic
cratonic carbonate/evaporite: the Bitter Springs Formation, Central
Australia. In: Grotzinger, J. P. & James, N. P. (eds) Carbonate
Sedimentation and Diagenesis in the Evolving Precambrian. SEPM
Special Publication, 67, 327–344.
Hill, A. C., Haines, P. W., Grey, K. & Willman, S. 2008. New records
of Ediacaran Acraman ejecta in drill holes from the Stuart Shelf and
Officer Basin, South Australia. Meteoritics and Planetary Science,
42, 1883– 1891.
Hill, A. C., Haines, P. W. & Grey, K. 2011. Neoproterozoic glacial
deposits of central Australia. In: Arnaud, E., Halverson, G. P. &
Shields-Zhou, G. (eds) The Geological Record of Neopro-
terozoic Glaciations. Geological Society, London, Memoirs, 36,
677– 691.
Hoffmann, K.-H., Condon, D. J., Bowring, S. A. & Crowley, J. L.
2004. U– Pb zircon date from the Neoproterozoic Ghaub
Formation, Namibia: constraints on Marinoan glaciation. Geology,
32, 817– 820.
Hoffman, P. & Schrag, D. 2002. The snowball Earth hypothesis: testing
the limits of global change. Terra Nova, 14, 129–155.
Hoffman, P., Hawkins, D., Isachsen, C. & Bowring, S. 1996.
Precise U–Pb zircon ages for early Damaran magmatism in the
Summas Mountains and Welwitschia Inlier, northern Damara belt,
Namibia. Communications of the Geological Survey of Namibia,
11, 47 – 52.
Hoffman, P. F., Calver, C. R. & Halverson, G. P. 2009. Cottons
Breccia of King Island, Tasmania: glacial or non-glacial, Cryogenian
or Ediacaran? Precambrian Research, 172, 311– 322.
Ireland, T. R., Flöttmann, T., Fanning, C. M., Gibson, G. M. &
Preiss, W. V. 1998. Development of the Early Paleozoic Pacific
Margin of Gondwana from detrital zircon ages across the Delamerian
Orogen. Geology, 26, 243– 246.
Jackson, M. J. & Muir, M. D. 1981. The Babbagoola Beds, Officer
Basin, Western Australia: correlations, micropalaeontology, and
implications for petroleum prospectivity. BMR Journal of Australian
Geology and Geophysics, 6, 81 – 93.
Jackson, M. J. & van de Graaff, W. J. E. 1981. Geology of the Officer
Basin. Bureau of Mineral Resources Bulletin, 206, 1 – 102.
James, N., Narbonne, G. & Kyser, T. 2001. Late Neoproterozoic cap
carbonates: Mackenzie Mountains, northwestern Canada: precipi-
tation and global glaciation. Canadian Journal of Earth Science,
38, 1229– 1262.
Jefferson, C. & Parrish, R. 1989. Late Proterozoic stratigraphy, U/Pb
zircon ages and rift tectonics, Mackenzie Mountains, northwestern
Canada. Canadian Journal of Earth Sciences, 26, 1784– 1801.
Karlstrom, K. E., Bowring, S. A. et al. 2000. Chuar Group of the
Grand Canyon: Record of breakup of Rodinia, associated change in
the global carbon cycle, and ecosystem expansion by 740 Ma.
Geology, 28, 619–622.
Kaufman, A., Knoll, A. & Narbonne, G. 1997. Isotopes, ice ages and
terminal Proterozoic Earth history. Proceedings of the National
Academy of Sciences, 95, 6600–6605.
Kendall, B., Creaser, R. A. & Selby, D. 2006. Re– Os geochronology
of postglacial black shales in Australia: consequences for timing of
the Sturtian glaciation. Geology, 34, 729–732.
Kendall, B., Creaser, R. A., Calver, C. R., Raub, T. D. & Evans,
D. A. D. 2007. Neoproterozoic paleogeography, Rodinia breakup,
and Sturtian glaciation: constraints from Re– Os black shale ages
from southern Australia and northwestern Tasmania. Geological
Society of America Abstracts with Programs, 39, 335.
Kendall, B. S., Creaser, R. A., Calver, C. R., Raub, T. D. & Evans,
D. A. D. 2009. Correlation of Sturtian diamictite successions in
southern Australia and northwestern Tasmania by Re– Os black
shale geochronology and the ambiguity of ‘Sturtian’-type diamictite-
cap carbonate pairs as chronostratigraphic marker horizons. Precam-
brian Research, 172, 301–310.
Kennard, J. M., Nicoll, R. S. & Owen, M. 1986. Late Proterozoic and
Early Palaeozoic Depositional Facies of the Northern Amadeus
Basin, Central Australia. 12th International Sedimentological
Congress, Field Excursion 25B. Bureau of Mineral Resources,
Canberra, 125.
 AUSTRALIAN CRYOGENIAN CORRELATIONS
Kennedy, M. J., Christie-Blick, N. & Prave, A. R. 2001. Carbon isoto-
pic composition of Neoproterozoic glacial carbonates as a test of
paleoceanographic models for snowball Earth phenomena.
Geology, 29, 1135– 1138.
Knoll, A. H. 1996. Chapter 4: Archaean and Proterozoic palaeontology.
In: Jansonius, J. & McGregor, D. C. (eds) Palynology: Principles
and Applications. American Association of Stratigraphic Palynolo-
gists Foundation 1, Publishers Press, Salt Lake City, 51– 80.
Knoll, A. H., Swett, K. & Mark, J. 1991. Paleobiology of a Neopro-
terozoic tidal flat/lagoonal complex: the Draken Conglomerate
Formation, Spitsbergen. Journal of Paleontology, 65, 531–570.
Knoll, A. H., Walter, M. R., Narbonne, G. M. & Christie-Blick, N.
2006. The Ediacaran Period: a new addition to the geologic time
scale. Lethaia, 39, 13 – 30.
Korsch, R. J. & Kennard, J. M. 1991. Geological and geophysical
studies in the Amadeus Basin, central Australia. BMR Bulletin, 236,
511– 524.
Krylov, I. N. 1967. Rifeyskie i nizhne-kembriyskie stromatolity Tyan’-
Shanya i Karatau [Riphean and Lower Cambrian stromatolites of
Tien-Shan and Karatau]. Akademiya Nauk, SSSR, Trudy, Geologi-
cheskii Institut Leningrad, 171, 76 (in Russian).
Krylov, I. N. 1975. Stromatolity Rifeya i Fanerozoya SSSR. [Riphean and
Phanerozoic stromatolites in the USSR]. Akademiya Nauk SSSR,
Trudy, Geologicheskii Institut, 274, 243 (in Russian).
Lindsay, J. F. 1993. Geological Atlas of the Amadeus Basin. Australian
Geological Survey Organisation, 25 plates.
Lindsay, J. F. 1995. Geological Atlas of the Officer Basin South Australia.
Australian Geological Survey Organisation and Mines and Energy
Department, South Australia, 30 plates.
Lindsay, J. F. & Reine, K. 1995. Well-log data, Officer Basin, South
Australia. Australian Geological Survey Organisation Canberra
Record 1995/2, 13.
Macdonald, F. A., Schmitz, M. D. et al. 2010. Calibrating the Cryo-
genian. Science, 327, 1241– 1243.
McKirdy, D. M., Burgess, J. M. et al. 2001. A chemostratigraphic
overview of the late Cryogenian interglacial sequence in the Adelaide
Fold-Thrust Belt, South Australia. Precambrian Research, 106,
149– 186.
Meffre, S., Direen, N. G., Crawford, A. J. & Kamenetsky, V. S.
2004. Mafic volcanic rocks on King Island, Tasmania: evidence for
579 Ma break-up in east Gondwana. Precambrian Research, 135,
177– 191.
Mendelson, C. V. & Schopf, J. W. 1992. Proterozoic and selected Early
Cambrian microfossils and microfossil-like objects. In: Schopf, J. W.
& Klein, C. (eds) Evolution of the Proterozoic Biosphere – A Multi-
disciplinary Study. Cambridge University Press, New York,
865– 895.
Morton, J. G. G. 1997. Lithostratigraphy and environments of depo-
sition. In: Morton, J. G. G. & Drexel, J. F. (eds) Petroleum
Geology of South Australia, Volume 3: Officer Basin. South
Australian Department of Mines and Energy Resources Report
Book, 97/19, 47– 86.
Morton, J. G. G. & Drexel, J. F. (eds) 1997. Petroleum Geology of South
Australia Volume 3: Officer Basin. South Australia, Department of
Mines and Energy Resources Report Book, 97/19, 173.
Mory, A. J. & Haines, P. W. 2005. GSWA Lancer 1 well completion
report (interpretive papers), Officer and Gunbarrel Basins, Western
Australia. Geological Survey of Western Australia Record, 2005/4,
90.
Nagy, R. M. 2008. Microfossils from the Neoproterozoic Chuar Group,
Grand Canyon, Arizona: taxonomy, paleoecological analysis and
implications for life during the onset of Neoproterozoic glaciation.
PhD thesis, University of California, Santa Barbara.
Narbonne, G., Kaufman, A. & Knoll, A. 1994. Integrated chemostra-
tigraphy and biostratigraphy of the Windermere Supergroup, north-
west Canada: implications for Neoproterozoic correlations and the
early evolution of animals. Geological Society of America Bulletin,
106, 1281–1292.
Nelson, D. R. 2002. Compilation of geochronology data, 2001. Western
Australia Geological Survey Record, 2002/2, 282.
Oehler, D. Z. 1976. Transmission electron microscopy of organic micro-
fossils from the Late Precambrian Bitter Springs Formation of
133
Australia: techniques and survey of preserved ultrastructure. Journal
of Paleontology, 50, 90 –106.
Parker, A. J., Rickwood, P. C. et al. 1987. Mafic dyke swarms of
Australia. In: Halls, H. C. & Fahrig, W. F. (eds) Mafic Dyke
Swarms. Geological Association of Canada, Special Papers, 34,
401– 417.
Phillips, B. J., James, A. W. & Philip, G. M. 1985. The geology and
hydrocarbon potential of the northwestern Officer Basin. APEA
Journal, 25, 52– 61.
Planavsky, N. & Grey, K. 2008. Stromatolite branching in the Neopro-
terozoic of the Centralian Superbasin, Australia: an example of shift-
ing sedimentary and microbial control of stromatolite morphology.
Geobiology, 6, 33– 45.
Porter, S. M. & Knoll, A. H. 2000. Testate amoebae in the Neoproter-
ozoic Era: evidence from vase-shaped microfossils in the Chuar
Group, Grand Canyon. Paleobiology, 26, 360–385.
Preiss, W. V. 1972. The systematics of South Australian Precambrian and
Cambrian stromatolites, part I. Transactions of the Royal Society of
South Australia, 96, 67– 100.
Preiss, W. V. 1973a. The Systematics of South Australian Precambrian
and Cambrian Stromatolites, Part II. Transactions of the Royal
Society of South Australia, 97, 91 –125.
Preiss, W. V. 1973b. Early Willouran stromatolites from the Peake and
Denison Ranges and their stratigraphic significance. South Australian
Department of Mines and Energy, Report Book, 73/208, 27.
Preiss, W. V. 1974. The Systematics of South Australian Precambrian and
Cambrian Stromatolites, Part III. Transactions of the Royal Society of
South Australia, 98, 185–208.
Preiss, W. V. 1976. Proterozoic stromatolites from the Nabberu and
Officer Basins, Western Australia, and their stratigraphic signifi-
cance. South Australia Geological Survey, Report of Investigations,
47, 51.
Preiss, W. V. 1985. Stratigraphy and tectonics of the Worumba Anticline
and associated intrusive breccias. South Australia Geological Survey
Bulletin, 52, 85.
Preiss, W. V. 1987. The Adelaide Geosyncline –late Proterozoic strati-
graphy, sedimentation, paleontology and tectonics. South Australian
Geological Survey Bulletin, 53, 438.
Preiss, W. V. 1993. Neoproterozoic. In: Drexel, J. F., Preiss, W. V. &
Parker, A. J. (eds) The Geology of South Australia, Volume 1, The
Precambrian. South Australian Geological Survey Bulletin, 54,
171– 203.
Preiss, W. V. 2000. The Adelaide geosyncline of South Australia and its
significance in Neoproterozoic continental reconstruction. Precam-
brian Research, 100, 21 –63.
Preiss, W. V., Walter, M. R., Coats, R. P. & Wells, A. T. 1978. Litho-
logical correlations of Adelaidean glaciogenic rocks in parts of the
Amadeus, Ngalia and Georgina Basins. Bureau of Mineral Resources
Geology & Geophysics Australia Journal, 3, 43 –53.
Preiss, W. V., Gostin, V. A., McKirdy, D. M., Ashley, P. M.,
Williams, G. E. & Schmidt, P. W. 2011. The glacial succession of
Sturtian age in South Australia – the Yudnamutana Subgroup. In:
Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The
Geological Record of Neoproterozoic Glaciations. Geological
Society, London, Memoirs, 36, 701– 712.
Raaben, M. E., Sinha, A. K. & Sharma, M. 2001. Precambrian Stroma-
tolites of India and Russia: A Catalogue of Type-Form-Genera.
Birbal Sahni Institute of Palaeobotany, Lucknow, 125.
Rainbird, R. H., Jefferson, C. W. & Young, G. M. 1996. The early Neo-
proterozoic sedimentary Succession B of northwestern Laurentia:
correlations and palaeogeographic significance. Geological Society
of America Bulletin, 108, 454–470.
Reid, A. 2009. Appendix G: U– Pb zircon dating of the porphyry.
Burra Mine. In: Drexel, J. F. (ed.) Review of the Burra Mine
Project, 1980– 2008 – A Progress Report. South Australia. Depart-
ment of Primary Industries and Resources Report Book, 2008/16,
68– 75.
Reid, P. W. & Preiss, W. V. 1999. Parachilna map sheet (second edition).
South Australia Geological Survey, Geological Atlas 1:250 000
Series, Sheet SH54-13.
Saito, Y., Tiba, T. & Matsubara, S. 1988. Precambrian and Cambrian
cherts in northwestern Tasmania. Bulletin of the National Science
134 K. GREY ET AL.
Museum, Series C: Geology and Paleontology, National Science
Museum, Tokyo, 14, 59– 70.
Sawaki, Y., Kawai, T. et al. 2010. 87Sr/86Sr chemostratigraphy of Neo-
proterozoic Dalradian carbonates below the Port Askaig Glaciogenic
Formation, Scotland. Precambrian Research, 179, 150– 164.
Schopf, J. W. 1968. Microflora of the Bitter Springs Formation, Late
Precambrian, Central Australia. Journal of Paleontology, 42,
651– 688.
Schopf, J. W. & Barghoorn, E. S. 1969. Microorganisms from the Late
Precambrian of South Australia. Journal of Paleontology, 43,
111– 118.
Sergeev, V. N. 1989. Microfossils from transitional Precambrian –
Phanerozoic strata of central Asia. Himalayan Geology, 13, 269– 278.
Shields, G. A., Brasier, M. D., Stille, P. & Dorjnamjaa, D. 2002.
Factors contributing to high d13C values in Cryogenian limestones
of Western Mongolia. Earth & Planetary Science Letters, 196,
99 –111.
Skotnicki, S. J., Hill, A. C., Walter, M. R. & Jenkins, R. 2008. Strati-
graphic relationships of Cryogenian strata disconformably overlying
the Bitter Springs Formation, northeastern Amadeus Basin, Central
Australia. Precambrian Research, 165, 243–259.
Smith, K. G. 1972. Stratigraphy of the Georgina Basin. Bureau of Mineral
Resources Australia Bulletin, 111, 156.
Stevens, M. K. & Apak, S. N. 1999. Empress 1 and 1A well completion
report, Yowalga Sub-basin, Officer Basin, Western Australia.
Western Australia Geological Survey, Record, 1999/4, 110.
Stevens, M. K. & Grey, K. 1997. Skates Hills Formation and Tarcunyah
Group, Officer Basin, carbonate cycles, stratigraphic position and
hydrocarbon prospectivity. Western Australia Geological Survey
Annual Review for 1996–97, 55 – 60.
Timofeev, B. V. 1966. Mikropaleofitologicheskoe issledovanie drevnikh
svit [Microphytological investigations of ancient formations]. Akade-
miya Nauk SSSR, Isdatelskvo Nauka [Science Publishing House],
Moscow, 1 – 147. (Russian, published in English translation dated
1974 by British Library – Lending Division, Yorkshire, England,
214.)
Townson, W. G. 1985. The subsurface geology of the western Officer
Basin – results of Shell’s 1980– 1984 petroleum exploration cam-
paign. APEA Journal, 25, 34 – 51.
Walter, M. R. 1972. Stromatolites and the biostratigraphy of the Austra-
lian Precambrian and Cambrian. Palaeontological Association
London, Special Papers in Palaeontology, 11, 190.
Walter, M. R. 1980. Adelaidean and Early Cambrian stratigraphy of the
southwestern Georgina Basin: correlation chart and explanatory
notes. Bureau of Mineral Resources Australia Report, 214, 21.
Walter, M. R. & Gorter, J. D. 1994. The Neoproterozoic Centralian
Superbasin in Western Australia: the Savory and Officer Basins. In:
Purcell, P. G. & Purcell, R. R. (eds) The Sedimentary Basins of
Western Australia. Petroleum Exploration Society of Australia,
West Australian Basins Symposium, Perth, WA, 1994, Proceedings,
851– 864.
Walter, M. R., Krylov, I. N. & Preiss, W. V. 1979. Stromatolites from
Adelaidean (Late Proterozoic) sequences in central and South
Australia. Alcheringa, 3, 287– 305.
Walter, M. R., Grey, K., Williams, I. R. & Calver, C. R. 1994. Strati-
graphy of the Neoproterozoic to early Palaeozoic Savory Basin,
Western Australia, and correlation with the Amadeus and Officer
Basins. Australian Journal of Earth Sciences, 41, 533– 546.
Walter, M. R., Veevers, J. J., Calver, C. R. & Grey, K. 1995. Neo-
proterozoic stratigraphy of the Centralian Superbasin, Australia.
Precambrian Research, 73, 173–195.
Walter, M. R., Veevers, J. J., Calver, C. R., Gorjan, P. & Hill, A. C.
2000. Dating the 840– 544 Ma Neoproterozoic interval by isotopes of
strontium, carbon and sulfur in seawater, and some interpretative
models. Precambrian Research, 100, 371–433.
Wells, A. T. & Moss, J. F. 1983. The Ngalia Basin, Northern Territory:
Stratigraphy and structure. Australian Bureau of Mineral Resources
Bulletin, 212, 88.
Wells, A. T., Forman, D. J., Ranford, L. C. & Cook, P. J. 1970. Geology
of the Amadeus Basin, central Australia. Bureau of Mineral
Resources Australia Report, 100.
Williams, G. E. & Gostin, V. A. 2005. Acraman– Bunyeroo impact event
(Ediacaran), South Australia, and environmental consequences:
twenty-five years on. Australian Journal of Earth Sciences, 52,
607– 620.
Williams, G. E., Jenkins, R. J. F. & Walter, M. R. 2007. No heliotrop-
ism in Neoproterozoic columnar stromatolite growth, Amadeus
Basin, central Australia: geophysical implications. Palaeogeography,
Palaeoclimatology, Palaeoecology, 249, 80 – 89.
Williams, G. E., Gostin, V. A., McKirdy, D. M. & Preiss, W. V. 2008.
The Elatina glaciation, late Cryogenian (Marinoan Epoch), South
Australia: sedimentary facies and palaeoenvironments. Precambrian
Research, 163, 307– 331.
Williams, I. R. 1992. Geology of the Savory Basin, Western Australia.
Western Australia Geological Survey Bulletin, 141, 115.
Williams, M., Crossey, L. J. et al. 2003. Dating sedimentary sequences:
in situ U/Th-Pb microprobe dating of early diagenetic monazite and
Ar – Ar dating of marcasite nodules; case study from Neoproterozoic
black shales in the southwestern. U.S. Geological Society of America
Abstracts with Programs, 35, 595.
Willman, S. & Moczydłowska, M. 2008. Ediacaran acritarch biota
from the Giles 1 drill hole, Officer Basin, Australia, and its potential
for biostratigraphic correlation. Precambrian Research, 162,
498– 530.
Willman, S., Moczydłowska, M. & Grey, K. 2006. Neoproterozoic
(Ediacaran) diversification of acritarchs – a new record from the
Murnaroo 1 drillcore, eastern Officer Basin, Australia. Review of
Palaeobotany and Palynology, 139, 17 –39.
Wingate, M. T. D., Campbell, I. H., Compston, W. & Gibson, G. M.
1998. Ion microprobe U–Pb ages for Neoproterozoic basaltic mag-
matism in south-central Australia and implications for the breakup
of Rodinia. Precambrian Research, 87, 135–159.
Yoshioka, H., Asahara, Y., Tojo, B. & Kawakami, S. 2003.
Systematic variations in C, O and Sr isotopes and elemental con-
centrations in Neoproterozoic carbonates in Namibia: implications
for glacial to interglacial transition. Precambrian Research, 124,
69 – 85.
Zang, W. 1995. Early Neoproterozoic sequence stratigraphy and acritarch
biostratigraphy, eastern Officer Basin, South Australia. Precambrian
Research, 74, 119– 175.
Zang, W. & Walter, M. R. 1992. Late Proterozoic and Early Cambrian
microfossils and biostratigraphy, Amadeus Basin, central Australia.
Memoirs of the Association of Australasian Palaeontologists 12,
1 –132.
Zhao, J.-X. & McCulloch, M. T. 1993. Sm– Nd mineral isochron ages of
Late Proterozoic dyke swarms in Australia: evidence for two distinc-
tive events of mafic magmatism and crustal extension. Chemical
Geology (Isotope Geoscience Section), 109, 341–354
Zhao, J.-X., McCulloch, M. T. & Korsch, R. J. 1994. Characterization
of a plume-related 800 Ma magmatic event and its implications for
basin formation in central-southern Australia. Earth & Planetary
Science Letters, 121, 349– 367.
 Chapter 9

A user’s guide to Neoproterozoic geochronology

DANIEL J. CONDON1* & SAMUEL A. BOWRING2

1
NERC Isotope Geoscience Laboratories, British Geological Survey, Keyworth NG12 5GS, UK
2
Department of Earth, Atmospheric and Planetary Sciences, Massachusetts Institute of Technology, Cambridge, MA 02139, USA
*Corresponding author (e-mail: dcondon@bgs.ac.uk)

Abstract: Geochronology is essential for understanding Neoproterozoic Earth history. Here we review the types of rocks and minerals
that are used to date geologic events and the analytical protocols for the different radio-isotopic decay systems employed. We discuss the
limitations and potential of these methodologies for dating Neoproterozoic stratigraphy, highlighting the major sources and magnitudes
of uncertainties and the assumptions that often underpin them.

Crucial to understanding the nature and causes of glaciations, large
perturbations of biogeochemical cycles and key evolutionary inno-
vations that took place during the Neoproterozoic is our ability
to precisely correlate and sequence disparate stratigraphic
sections. Relative ages of events can be established within single
sections or by regional correlation using litho-, chemo- and/or
bio-stratigraphy. However, relative chronologies do not allow for
testing of event synchroneity, robustness of correlations or calcu-
lation of rates and duration of change. At present, a major limit-
ation to our understanding of Neoproterozoic history is the
dearth of precise and accurate radio-isotopic dates. However, the
dramatic increase in geochronological constraints over the past
five years is cause for optimism.
This chapter reviews the radio-isotopic techniques used for
dating of Neoproterozoic stratigraphy. It is aimed at the many geol-
ogists, climate scientists, palaeobiologists and geophysicists who
use Neoproterozoic geochronology, especially those less familiar
with the process of obtaining a date from a rock or mineral. We
provide an outline of the strengths, weaknesses and limitations
of the different techniques, as well as critical evaluation of the
assumptions that underpin the accuracy and precision of the calcu-
lated dates. We discuss specific examples where some of these
assumptions are not valid, resulting in inaccurate dates assigned
to certain stratigraphic levels. Overall, we emphasize the U –Pb
(zircon) geochronology of volcanic ash beds. However, we also
review Re – Os, Lu –Hf and Pb/Pb geochronology and their appli-
cation to dating sedimentary rocks. Although the K – Ar and
40
Ar/39Ar methods are mainstays of Mesozoic and younger time
scale calibration, there are few examples where they have been
successfully applied to dating eruption ages of Neoproterozoic
volcanic rocks. This is most likely due to the susceptibility of
K-bearing minerals to alteration and open system behaviour
related to metamorphism.
What can be dated?
The absolute age of Neoproterozoic sedimentary successions can
be determined via radio-isotopic dating of minerals separated
from volcanic rocks. These minerals, such as zircon, are highly
enriched in the parent isotope (U in the case of zircon, see
below) and can be dated via radio-isotopic decay schemes (such
as U –Pb, Re – Os and K –Ar/40Ar/39Ar techniques). As these min-
erals crystallize (or become ‘closed’ to the loss of daughter
isotope) at the approximate time of the magmatic eruption, their
age is assumed to approximate the depositional age of the volcanic
rock. It is also possible in some cases to apply radio-isotopic dating
directly to sedimentary chemical precipitates and organic-rich
components enriched in parent isotopes (U, Re, Lu) (see section
‘Whole-rock geochronometers (Re – Os, Lu –Hf, Pb/Pb)’).
Dating accessory minerals from volcanic rocks
Zircon (ZrSiO4) is a common accessory mineral in silicic volcanic
rocks ranging from lavas to air-fall tuffs to volcaniclastic sedimen-
tary rocks and is a nearly ubiquitous component of most clastic
sedimentary rocks. The refractory and durable nature of zircon
over a wide range of geological conditions means that it is likely
to retain its primary crystallization age even through subsequent
metamorphic events. Silicic air-fall tuffs are the most common vol-
canic rocks in fossil-bearing sequences and are found in layers that
range in thickness from a millimetres to many metres. They are
commonly preserved in marine settings. In most of these rocks
the primary volcanic ash has been altered, probably soon after
deposition, to clay minerals in a process that does not affect zircon.
What makes zircon ideal for U –Pb dating is the fact that,
because U has a similar charge and ionic radius to Zr, it substitutes
readily into the zircon crystal structure (in modest amounts, typi-
cally in the tens to hundreds of parts per million (ppm) range),
whereas Pb has a different charge and larger ionic radius,
leading to its effective exclusion from the crystal lattice. There-
fore, at the time of crystallization (t0), there is effectively no Pb
present in a crystal (although mineral and fluid inclusions may
contain Pb) and the present-day Pb is the direct product of in situ
U decay since t0 (see section ‘Uranium – lead’ for further
details). An additional factor that makes zircon a robust chron-
ometer is its high closure temperature (.900 8C) for Pb diffusion
(Cherniak & Watson 2003), or the temperature below which U and
Pb do not undergo significant thermally activated volume diffu-
sion. This means that zircons tend to preserve their primary ages,
even in volcanic rocks metamorphosed to amphibolites-facies
conditions.
The refractory and durable nature of zircon means that it is often
recycled through crustal processes of erosion, metamorphism and
magmatism, which, combined with its high closure temperature,
means that it is possible to ‘inherit’ older zircon in newly
formed igneous rocks. Silicic volcanic rocks are often the
product of melting or assimilation of older zircon-bearing rocks
and often contain inherited zircon. This phenomenon is most com-
monly observed as older cores surrounded by younger, magmatic
zircon overgrowths. Such grains pose an analytical challenge, as
the different domains should be analysed separately. This is best
achieved using either microbeam techniques (see section ‘U –Pb
microbeam techniques’), which employ the high spatial resolution

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 135– 149. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.9
136 D. J. CONDON & S. A. BOWRING
of focused ion beams or lasers to analyse different domains within
a single crystal, or mechanical micro-sampling of single grains fol-
lowed by conventional analysis. In addition, silicic volcanic rocks
often contain inherited whole grains that range in age from only
slightly older than the eruption to tens to hundreds of millions of
years older. These zircons are the refractory residue from assimila-
tion of older rocks in the magma or are incorporated during an
explosive eruption. A further complication arises when one con-
siders that the crystallization history of a magma is not necessarily
an instantaneous event and, in some circumstances (such as large-
volume silicic eruptions), zircons (and other minerals) may crys-
tallize tens to hundreds of thousands of years prior to an eruption
and/or over a protracted interval. Thus the zircons separated
from an air-fall tuff may record a range of dates from the eruption
age to millions of years prior. In such situations, high-precision
single-grain analyses are required to isolate different populations
and assign an eruption age to the sample.
It is also possible to date other U-bearing accessory phases from
volcanic ash beds such as monazite and titanite; however, these are
much less common in their occurrence than zircon. Monazite
[(LREE)PO4] is typical of peraluminous magmas and meta-
morphic rocks, but it is also sometimes found in volcanic ash
beds (e.g. Roden et al. 1990). Monazite has a similarly high
closure temperature (.900 8C) to zircon and can incorporate
ThO2 at the percent level in addition to U into its crystal
lattice, making it ideally suited to U – Th –Pb dating. Titanite
(CaTiSiO5) is a common accessory phase in plutonic and meta-
morphic rocks, but somewhat less common in volcanic rocks.
A potential advantage of titanite is that it has a lower closure temp-
erature than zircon and monazite (c. 650 8C) and thus does not
retain Pb until it cools below that temperature. Thus, in volcanic
rocks, pre-eruptive residence is not a limiting factor for its use in
dating volcanic eruptions. However, titanite also incorporates
some initial Pb into its crystal structure that limits the precision
of titanite U –Pb dates.
Dating chemical precipitates and organic residues
Chemically precipitated rocks and organic residues from sedimen-
tary rocks can be exploited as chronometers using the isochron
approach, and calculated dates are interpreted as the time of pre-
cipitation and/or early diagenesis. The most commonly dated
Neoproterozoic rocks using this approach include carbonates,
phosphates and organic-rich shales, as they contain high concen-
trations of the different parent nuclides (e.g. Re, U and Lu).
U– Pb (Pb/Pb) dating of carbonates. At the time of formation, car-
bonates incorporate U into their crystal lattice, typically with a
concentration of several ppm (although in some cases this can
be many tens of ppm), as well as (initial) Pb with concentrations
typically in the parts per billion (ppb) range, and therefore have
the potential for U –Pb dating. Carbonates form in a variety of ter-
restrial and marine environments, but most Neoproterozoic succes-
sions targeted for U –Pb dating are marine. Fluid-mediated
recrystallization of carbonates is a common process during burial
and diagenesis, and very few (if any) Neoproterozoic carbonates
have escaped this process. The variable mobility of Pb and U
means that it is common for the two to become decoupled,
especially during recent exposure and weathering, in which case
the U – Pb systematics are often unreliable. If the decoupling of
the U and Pb occurred recently, then the Pb/Pb systematics are
often relied upon for age information. Several studies have gener-
ated Pb/Pb isochrons from Neoproterozoic successions with vari-
able success (Babinski et al. 1999, 2007; Folling et al. 2000;
Barfod et al. 2002; Chen et al. 2004), and most demonstrate evi-
dence for disturbance of the Pb/Pb systematics during burial/
metamorphic events (Babinski et al. 1999; Folling et al. 2000).
A major issue related to Pb/Pb isochron dates is how to assess
closed-system behaviour, and it is common to use either a statisti-
cal measure of coherence (see section ‘Sources and types of uncer-
tainty’) or pre-existing age constraints to aid in the interpretation.
However, although closed-system behaviour results in statistical
coherence, the inverse is not always true (see section ‘Uncertain-
ties as a result of geologic complexity’). Recent studies have
employed combined textural and Sr isotope analyses of the carbon-
ates as an independent proxy for disturbance during fluid flow
events (Babinski et al. 2007). For a recent review of U –Pb
dating of carbonates see Rasbury & Cole (2009).
Re– Os dating of organic-rich sediments. Both Re and Os are redox-
sensitive metals with multiple oxidation states. Re decays to Os
with a half-life of c. 41 Ga (Table 9.1). Both metals can become
concentrated within anoxic sediments via redox reactions near
the sediment –water interface and also have been shown to be pre-
ferentially incorporated into the organic matter of shales (Creaser
et al. 2002). Following deposition, both metals are often relatively
immobile and can behave as a closed system through burial, dia-
genesis and low-grade metamorphism, allowing exploitation as a
chronometer. In the past five years, the Re –Os geochronometer
has been applied to several organic-rich Neoproterozoic strati-
graphic intervals (Schaefer & Burgess 2003; Kendall et al. 2004,
2006), as well as several Phanerozoic intervals (Cohen et al.
1999; Creaser et al. 2002; Selby & Creaser 2005), and has given
new hope for dating thick sedimentary sequences devoid of volca-
nic rocks. For a recent review of Re – Os dating of organic-rich
sediments, see Kendall et al. (2009).
U–Pb (Pb/Pb) and Lu–Hf dating of phosphates. Phosphate minerals
are enriched (relative to the fluid from which they precipitate) in
rare earth elements (REE), Th and U, and therefore have potential
for geochronology using the Lu – Hf, Th –Pb and U –Pb decay
schemes. Few studies have successfully exploited the Lu –Hf
and U – Pb system for the precise dating of sedimentary phosphates
(Barfod et al. 2002).
In addition to dating ‘bulk’ phosphates, both monazite and xeno-
time (YPO4) are known to form during early diagenesis (Evans
et al. 2002; Rasmussen 2005). Both minerals have highly favour-
able U(Th)/Pb concentrations, but their occurrence as very small
grains or overgrowths makes them difficult to isolate and
analyse. Xenotime occurs as syntaxial overgrowths on zircons,
but due to their small size (a few tens of micrometres) and textural
complexity, an in situ isotopic technique with a spatial resolution
of ,10 mm is required to successfully date xenotime; this has so
far only been achieved by ion microprobe. An alternative approach
is the Th – U –total Pb electron microprobe technique (Williams
et al. 2007). This technique has the disadvantage that it requires
high concentrations of Th, U and Pb for detection, limiting it to
phases such as monazite; however, it does have high spatial resol-
ution (beam width down to ,5 mm) and therefore the ability to
analyse very small grains in their petrographic context, but with
uncertainties of several percent.
Table 9.1. Radiometric decay systems used in geochronology (see text for
further details)
Radioactive Radiogenic Half-life (years)
parent nuclide daughter nuclide
238
U 206
Pb 4.468  109
235
U 207
Pb 7.038  108
187
Re 187
Os 4.16  1010
176
Lu 176
Hf 3.71  1010
87
Rb 86
Sr 4.944  1010
40
K 40
Ar, 40Ca 1.25  109
 USER’S GUIDE TO NEOPROTEROZOIC GEOCHRONOLOGY
Precipitation of xenotime also occurs during fluid and thermal
events, which introduces the potential for complexity through mul-
tiple episodes of growth (Rasmussen 2005). In contrast to xeno-
time, diagenetic monazite tends to occur as impure nodules (up
to 2 mm in diameter) in shales. Studies of Palaeozoic diagenetic
monazite nodules have demonstrated the utility of the Pb/Pb and
Th –Pb systems, although the U –Pb systematics have been per-
turbed, indicating recent U remobilization (Evans et al. 2002).
Neither (diagenetic) monazite or xenotime U – Pb geochronology
have been successfully applied to the dating of a Neoproterozoic
sedimentary succession, although recent Th –U –total Pb investi-
gation of monazite from the Adelaide Rift complex (Mahan
et al. 2010) does show potential.
In some cases the ‘whole-rock geochronometers’ may be useful
for obtaining temporal constraints on sedimentary successions
devoid of volcanic material. In general, best-case uncertainties
associated with calculated isochron dates are in the 0.5 –1%
level, comparable to those of U –Pb (zircon) microbeam dates.
Despite the relatively large uncertainties intrinsic to the dating of
these types of rocks, the Pb/Pb isochron method has been
applied and interpreted as an estimate of depositional ages (see
above). Problems associated with isochron geochronometers are
centred around the lack of an independent check on open-system
behaviour.
Maximum and minimum age constraints
Not all sedimentary successions are amenable to direct dating via
radio-isotopic methods. In the absence of zircon-bearing volcanic
rocks, or chemical sediments for isochron dating, another approach
is to obtain maximum age constraints by dating detrital zircons.
Detrital zircons in sedimentary rocks can range from being con-
siderably older than the estimated depositional age they are con-
tained within (often many hundreds of millions of years older) to
close to the age of deposition, and in this latter case can provide
useful constraints (Bingen et al. 2005). The youngest age deter-
mined in a detrital population gives a maximum depositional age
for the unit. Minimum age constraints can be provided by over-
lying strata, which may contain age-diagnostic fossils, and/or
cross-cutting igneous intrusions.
Radio-isotopic geochronometers
A relatively small number of radioactive decay schemes are suit-
able for dating Neoproterozoic rocks and are listed in Table 9.1.
Each of these is based upon the radioactive decay of a parent
nuclide to a stable daughter nuclide. Obtaining an accurate date
using these decay systems requires that (i) the decay constant of
the parent nuclide is accurately and precisely determined; (ii)
closed-system behaviour, which can be simply stated to mean
that the parent/daughter ratio has only changed by radioactive
decay; and (iii) the initial daughter nuclide, if present, can be
precisely and accurately accounted for. In this section, we
outline the basic principles of the radio-isotopic geochronometers
in Table 9.1, separating the U –Pb system applied to U-bearing
accessory minerals from those that use an isochron approach
(Re – Os, Lu – Hf, Pb/Pb, etc.) applied to chemical precipitates
and organic residues.
Uranium – lead
U –Pb geochronology is often regarded as the gold standard of
geochronology because, unlike all other chronometers, it exploits
two independent decay schemes, 235U to 207Pb and 238U
to 206Pb, and both the 238U and 235U decay constants are relatively
precise and accurate (Jaffey et al. 1971). Two separate dates for a
137
ln (1+ 206Pb*/238U )
U
206Pb/238U date =
238
λ238
950 Ma
Pb*/
206
ln (1+ 207Pb*/235U )
850 Ma
0.14
λ235
750 Ma
0.12
650 Ma
0.10 λ235t 235U
207Pb/206Pb date = e –1
λ238t 238U
e –1 today
207 235
Pb*/ U
0.9 1.1 1.3 1.5
Fig. 9.1. U–Pb (Wetherill) concordia diagram for the age range 542– 1000 Ma.
The grey band represents the concordia curve plotted to reflect the uncertainties
in the 235U and 238U decay constants.
zircon based on each individual decay scheme may be calculated
and plotted on a concordia diagram (Fig. 9.1). On a conventional
(Wetherill) concordia diagram the X and Y axes are the ratio of
radiogenic 207Pb to 235U and radiogenic 206Pb to 238U ratios,
respectively, and the concordia curve represents the simultaneous
solution of the decay equations for a given age. A third date, the
207
Pb/206Pb date, can be determined from only Pb isotopic
measurements by knowing both the 235U and 238U decay constants
and the present-day 235U/238U ratio, which is assumed to be
137.88 (Steiger & Jager 1977). Calculation of the U –Pb dates
requires determination of the Pb* –U ratio (Pb* denotes radiogenic
Pb). See Figure 9.1 for age equations.
The different half-lives of 238U and 235U (c. 4.5 and c. 0.7 Ga,
respectively) mean that by the Neoproterozoic time, much
smaller amounts of 235U (relative to 238U) remained due to
the higher decay rate, and smaller amounts 207Pb were produced
per increment of time relative to 206Pb. The advantage of two inde-
pendent chronometers in the same mineral is that it is possible to
detect small amounts of open-system behaviour such as Pb loss
or the inheritance of older Pb. This is a major factor in our
ability to make reliable, high-precision age determinations, as
we can evaluate whether a number of analyses represents a
single time of mineral growth. It has been known for several
decades that zircons often show evidence of post-crystallization
Pb loss. This has the effect of lowering the U –Pb ratios and the
derived dates. The analyses plot below the concordia curve and
are termed normally ‘discordant’, with 207Pb/206Pb dates . 207
Pb– 235U dates . 206Pb – 238U dates (Fig. 9.2). Silver & Duetsch
(1963) demonstrated that Pb loss was correlated with radiation
damage. Since that time it has become widely appreciated that
Pb is not lost by thermally activated volume diffusion but rather
by fast-pathway diffusion from damaged parts of the crystal
lattice. In order to minimize/eliminate the effects of post-
crystallisation Pb loss, it is possible to pre-treat zircons to prefer-
entially remove domains that have lost Pb, leaving zircon that
has remained a closed system. One major approach involves phys-
ically abrading away the exterior portions of the zircons (Krogh
1982a), based on the observation that the outer portions were
richest in U and thus susceptible to radiation damage and Pb
loss. Krogh (Krogh 1982b) also demonstrated that the careful
selection of the least magnetic zircons (diamagnetic) often corre-
sponded to those with lowest U contents, radiation damage, and
least affected by Pb loss. These approaches were widely applied
138 D. J. CONDON & S. A. BOWRING
U Upper intercept or 207Pb/206Pb date: 720
238
713.65 ±0.70 (±4.5) Ma
Pb/
206
206Pb/238Udate:
0.1165 711.65 ±0.58 (±1.4) Ma
0.1155 705
207Pb/235Udate:
712.35 ±0.83 (±1.9) Ma
700
0.1145
ss
-lo
695
Pb
0.1135
690 207 235
0.975 0.99 1.00 1.01 1.02 Pb/ U
Fig. 9.2. Schematic U–Pb concordia diagram illustrating the 206Pb– 238U,
207
Pb– 235U and 207Pb/206Pb dates that can be calculated. The data presented are
a subset of analyses from sample WM54 (Bowring et al. 2007). Uncertainties
are the 2s internal uncertainties, whereas those in parentheses are the 2s
internal plus the systematic decay constant uncertainties: 0.11% for 238U and
0.14% for 235U (Jaffey et al. 1971). The grey band is the concordia line plotted
to reflect the uncertainties in the U decay constants. Error ellipses (white) are
plotted with 2s internal uncertainties, and black error bars represent 2s internal
plus systematic decay constant uncertainties.
until the development of a new technique described as ‘chemical
abrasion’ (CA-TIMS) (Mattinson 2005). This technique involves
annealing zircon grains at 800–900 8C followed by partial dissol-
ution in hydrofluoric acid. This method effectively ‘mines out’ or
preferentially dissolves the higher U parts of the zircon that have
been damaged by radiation and are thus susceptible to
fast-pathway diffusion of Pb from the zircon crystal. In many
cases these domains are irregular in shape and occur in the grain
interiors. This method seems to offer the promise of effective elim-
ination of open-system behaviour in most zircon. Microbeam tech-
niques (see section ‘U– Pb microbeam techniques’) have not
typically used pre-treatment techniques, as they assume that Pb
loss is restricted to the exterior portions of grains, which they
attempt to avoid during the in situ analyses. In addition, grains
treated using this method are often extremely fragile and not amen-
able to mounting and polishing.
Whole-rock geochronometers (Re – Os, Lu – Hf, Pb/Pb)
When rocks or minerals do not have extreme enrichment in the
parent/daughter ratio but instead have a range, the isochron
approach is often useful. The isochron method involves analyses
of multiple cogenetic samples (minerals or subsamples of a rock
from an identical stratigraphic level) and is used for systems
where initial daughter atoms are present and there is the possibility
of a range in the parent/daughter ratio. The Re –Os, Lu –Hf and
U –Pb chronometers can be used to date carbonates, phosphates
and organic-rich shales that incorporate both parent and daughter
isotopes at the time of their formation. Ideally, the isochron
approach allows determination of both an age and an initial isoto-
pic composition of the daughter element.
For multiple cogenetic samples to preserve the time of system
closure, samples must begin with (i) a homogeneous initial daugh-
ter isotopic composition and (ii) a spread in the parent/daughter
ratio such that, over time, different samples with different
parent/daughter ratios will evolve and will define a straight line
slope = e λt -1
D*/Dref
t2
t1
to
Y-axis intercept = D0/Dref P/Dref
Fig. 9.3. Schematic isochron diagram illustrating (i) the situation at t0 where
different samples from the same stratigraphic interval record a spread in
parent(P)/daughter(Dref) ratios but a constant initial daughter isotopic
composition. D0 is the initial amount of daughter isotope and D* is the
radiogenic daughter isotope (from decay and initial) such that the daughter
produced solely from decay (since t0) ¼ D* –D0.
with a slope equal to e lt –1 (Fig. 9.3). A typical isochron is
plotted P/Dref on the X axis and D*/Dref on the Y axis, where P
is the number of parent atoms in the sample, Dref is the number
of atoms of a stable reference isotope of the daughter element
and D* the total number of radiogenic daughter atoms plus
initial atoms (D0) of the same isotope (amount of daughter atoms
due to decay ¼ D* –D0) (Fig. 9.3). At the time of sample for-
mation (t0), all samples would define a horizontal line. However,
as the parent decays over time (t1, t2, etc.), each sample will
evolve along a slope of –1, and samples with higher initial
parent/daughter ratio (P/Dref) will be displaced most (Fig. 9.3),
such that the isochron rotates to a positive slope. Assuming closed-
system behaviour since the formation of the sample, a linear
regression through the points allows calculation of the slope and
Y-intercept from which the age and the initial isotopic composition
of the daughter can be determined (Fig. 9.3).
Analytical methodologies
The majority of age constraints for Neoproterozoic strata have
been obtained by U –Pb zircon dating, so most of this section
is concerned with U –Pb analytical methods. The analytical
methods used for other chronometers have some similarities with
the ID-TIMS U –Pb method. The differences are discussed below.
U– Pb methodologies
There are two main approaches to U – Pb zircon geochronology: in
situ ‘microbeam’ techniques and isotope dilution thermal ioniz-
ation mass spectrometry (ID-TIMS). The major difference is that
in ID-TIMS geochronology, zircon is dissolved and the U and
Pb separated from the other elements before analysis, but in
microbeam techniques the zircon is analysed using either a laser
that ablates the sample prior to injection into a plasma or a
focused ion beam that sputters the sample and generates secondary
ions from a mineral mounted in epoxy and polished or a thin
section.
U –Pb ID-TIMS. ID-TIMS analyses of zircon (either as multi-grain
fractions, single grains or grain fragments) involve dissolution of
the zircon in the presence of tracer isotopes and are called
isotope dilution. Because the tracer is added before dissolution,
the tracer/sample isotope ratios stay constant, despite incomplete
recovery during chemistry and low ionization efficiency during
 USER’S GUIDE TO NEOPROTEROZOIC GEOCHRONOLOGY
analysis. For U –Pb ID-TIMS analyses, the most common tracers
are 205Pb and 235U. 205Pb is an artificial isotope that does
not occur in nature, whereas 235U occurs naturally, although an
enriched form is used as a tracer. Natural U in the zircon is
assumed to have 238U/235U ¼ 137.88 (Steiger & Jager 1977),
so the tracer can be used to determine the number of moles of
238
U and 235U in a sample. Following dissolution, the sample
undergoes chemical purification using anion exchange chemistry
that allows separation of Zr and REEs from Pb and U, and Pb
and U from one another.
Following purification, Pb and U are analysed separately by
thermal ionization mass spectrometry, where the ratios of
sample isotopes (204Pb, 206Pb, 238U, etc.) to the tracer isotopes
(205Pb, 235U) can be measured. As the amount of tracer isotope
added to the sample is known, the number of atoms of each
naturally occurring isotope in the sample can be determined
based upon the measured ratio of sample isotope/tracer isotope
ratio. After corrections for mass-dependent isotope fractionation,
and the minor contribution of common Pb and U from the
reagents, the tracer and labware, the sample 206Pb/207Pb,
206
Pb/238U and 207Pb/235U ratios can be determined and
206
Pb/207Pb, 206Pb/238U and 207Pb/235U dates calculated. See
section ‘Calibrating tracers for isotope-dilution’ for further discus-
sion of the isotope dilution technique.
Optimization of this technique means that it is now possible
to date zircons with ,10 pg radiogenic Pb with a precision of
,0.1% on the U –Pb ratio for single-grain analyses. However,
it is a very labour-intensive technique, as each single U –Pb
analysis takes several hours of mass spectrometry and chemical
purification in an ultraclean laboratory environment, making it
time-consuming to develop high-n datasets.
U– Pb microbeam techniques
U –Pb geochronology by microbeam techniques has revolutio-
nized geochronology over the past two decades. The two major
techniques are secondary ion mass spectrometry (SIMS), typified
by the SHRIMP (sensitive high resolution ion microprobe) and
laser ablation inductively coupled plasma mass spectrometry
(LA-ICP-MS). Both techniques (collectively termed ‘microbeam’
techniques) offer high-spatial-resolution analyses using either a
focused ion beam to sputter a volume of zircon (SIMS) or a laser
that is used to vaporize a volume of zircon (LA-ICP-MS).
Microbeam techniques allow in situ analysis of very small
volumes and thus high spatial resolution. A typical volume of
zircon analysed by an ion probe is cylindrical, 20 –30 mm in diam-
eter and several micrometres deep, with somewhat larger volumes
for LA-ICP-MS (Kosler & Sylvester 2003). In addition, the ana-
lyses can be done relatively rapidly (many tens of analyses per
day for LA-ICP-MS and tens for SIMS) and are amenable to
automation. Furthermore, these techniques also allow analysis of
other isotopes/elements of interest (Hf, O, REEs) and can be
made on the same zircon grains in close proximity to the volume
analysed for geochronology.
Fundamental to the microbeam U – Pb zircon methods is use of a
primary standard against which the U – Pb ratio of the unknown
zircon is calibrated. For SIMS techniques this calibration involves
analyses of standard zircon to develop a calibration curve for a
known U –Pb ratio (which is determined via ID-TIMS analyses)
and against which analyses of unknown zircons can be compared.
This is achieved through analytical sessions where a standard
zircon is repeatedly analysed interspersed with analyses of
unknown zircons (this is termed sample-standard bracketing).
During the course of a single analytical session, the measured
ratio/date of a standard can drift by several percent. For LA-
ICP-MS the approach is somewhat similar in that sample-standard
bracketing is applied to determine the inter-elemental fraction-
ation, which is then applied to the unknown zircons. In both
139
SIMS and LA-ICP-MS techniques, the 207Pb/206Pb ratio is a
direct measurement; for SIMS mass-dependent fractionation
appears to be minimal and the measured ratio is commonly used,
whereas in LA-ICP-MS analyses mass-dependent fractionation is
quantifiable and is corrected for, either using sample-standard
bracketing and/or using a solution with known 205Tl/203Tl ratio
to correct for mass bias on Pb isotopic ratios. For further details
of microbeam techniques see Ireland & Williams (2003) for a
review of SIMS U – Pb geochronology and Kosler & Sylvester
(2003) for a review of LA-ICP-MS geochronology.
There is a tradeoff between the benefit of the high spatial
resolution provided by microbeam techniques and the precision
of individual spot analyses. LA-ICP-MS and SIMS analyses are
approximately an order of magnitude less precise than ID-TIMS
(Ireland & Williams 2003; Kosler & Sylvester 2003). In situ tech-
niques are without question essential tools for characterizing
complex (zoned) zircons from volcanic and metamorphic rocks
and for characterizing detrital populations, which in some cases
can provide robust estimates of the maximum age of a sequence.
Isochron techniques
Isochron techniques involve analysis of multiple samples assumed
to be the same age, which have a spread in parent/daughter ratio
and have remained closed systems. To maximize the probability
that the samples are the same age and have the same initial isotopic
composition it is preferable to design an appropriate sampling
strategy. For example, Re – Os geochronology is often attempted
on laminated organic-rich shales, so it is essential to have multiple
samples from the same restricted stratigraphic interval (e.g.
Kendall et al. 2004) to avoid samples that span an amount of
time much larger than uncertainties. In some cases, such as
working with core samples, this is not always possible (Schaefer
& Burgess 2003; Kendall et al. 2006). Samples that integrate a sig-
nificant amount of time could introduce scatter if there is temporal
variation in the initial isotopic composition of the daughter
element, and/or the sample represents an amount of time that far
exceeds measurement uncertainty, especially in condensed sec-
tions. For U – Pb in carbonates, different cement domains may
have very different U –Pb ratios, but is difficult to know a priori
whether the cements are all exactly the same age and have the
same initial ratios. In both cases, deviations from the assumption
of the same age and initial ratio lead to increased scatter and uncer-
tainties in calculated dates.
Sample dissolution and purification techniques are similar to the
procedures for U –Pb ID-TIMS. Before isotope ratio mass spec-
trometry, samples undergo dissolution and chemical purification.
For multi-element systems (such as Re –Os, U –Pb and Lu –Hf ),
isotopic tracers are added before dissolution for the isotope
dilution (see above), whereas for systems where only daughter
isotopes are measured (i.e. Pb/Pb), direct measurements of the
isotope ratios are made. The isotopic composition is determined
via thermal ionization mass spectrometry, although it is also poss-
ible to use solution-mode ICPMS for most elements.
The accuracy and precision of isochron techniques is largely
controlled by having a sufficient spread in initial parent/daughter
ratio, closed-system behaviour and an identical initial isotopic
composition for all samples. For precipitates such as carbonates
and phosphates there is often no significant detrital input;
however, this is not the case for organic-rich shales targeted for
Re –Os where there is potential for significant concentrations of
initial Os from multiple sources. This has been demonstrated in
several studies (Creaser et al. 2002; Kendall et al. 2004);
however, it is possible to limit the detrital Os contribution by
selective dissolution of the organic component using a CrO3 –
H2SO4 dissolution approach. Kendall et al. (2004) compared two
dissolution methods (aqua regia v. CrO3 –H2SO4 dissolution) on
greenschist facies organic-rich shale from the Old Fort Point
140 D. J. CONDON & S. A. BOWRING
Formation in Western Canada. Both dissolution techniques were
used on the same powders. However, the aqua regia method
yielded scattered data and a resulting ‘isochron’ regression with
a mean square weighted deviation (MSWD) of 65 and a large
age uncertainty (9%) in comparison to the CrO3 –H2SO4 dissol-
ution method, which yielded an isochron with much less scatter
(MSWD ¼ 1.2) and a relatively low uncertainty (0.8% 2s)
(Kendall et al. 2004).
Sources and types of uncertainty
Without an accurate estimation of total uncertainty, the radio-
isotopic age of a given rock or mineral is of limited value. For
example, suppose a date of 618 Ma is reported for a detrital
zircon from a unit that underlies a Marinoan-type glacial deposit.
If the date is relatively precise (and accurate), +2 Ma, then it
could be inferred that the onset of glacial sediment accumulation
post-dates deposition of the detrital component at 618 + 2 Ma.
On the other hand, if the 618 Ma date has an uncertainty of
100 Ma then the detrital zircon could be as old as 718 Ma or
young as 518 Ma, making it of limited use. Consider the case
where the 618 Ma is the 206Pb – 238U date with an uncertainty
of 10 Ma, but the 207Pb/206Pb date is 650 + 400 Ma. Although
the two dates overlap (and are therefore technically concordant),
the lack of precision on the 207Pb/206Pb date renders it impossible
to assess open-system behaviour (such as Pb loss or inheritance)
within the limits imposed by the precision of the 207Pb/206Pb
date, so the accuracy of the 206Pb– 238U date is not known; if
the zircon has lost Pb, the 206Pb– 238U date is a minimum date.
This latter example may seem extreme, but data of this type
have been published and are often uncritically cited. For
example, Ireland et al. (1998) published an extensive dataset of
SHRIMP U –Pb zircon dates on detrital zircons from the Kanman-
too Group in Australia (it should be noted that it was not the inten-
tion of this study to constrain the timing of sediment
accumulation). The units sampled included the Marino Arkose,
where 50 detrital zircons were analysed, the majority of which
were .1 Ga (n ¼ 48). Two grains yielded 206Pb– 238U dates of
649 + 17 and 655 + 17 Ma, which many researchers use to indi-
cate that the Marino Arkose is c. 650 Ma (or younger) (Zhou et al.
2004; Halverson et al. 2005; Peterson et al. 2005). The 207Pb/206Pb
dates associated with these two analyses are 470 + 440 Ma
and 666 + 307 Ma, respectively. Although it is possible that the
two zircons are indeed c. 650 Ma, it is not legitimate to assume
the 206Pb – 238U date is an accurate estimate of the age of the
zircons without considering both the 207Pb– 235U and 206Pb/207Pb
uncertainties. In practical terms these two detrital zircon dates
provide no significant constraint on the timing of sedimentation.
It is advisable for consumers of geochronological data to under-
stand the various sources of error and when one must consider the
total uncertainty of a given date as opposed to its constituent parts.
Although the uncertainty of each date contains an internal/random
component in the total uncertainty, there are also components that
are systematic, such as errors from the decay constants. When
comparing ages determined by the same isotopic system these
can be ignored, offering a potential increase in resolving power.
In this section we review the different sources of uncertainties
and the assumptions that underlie the often quoted (or not)
errors. For more detailed treatment of uncertainties in geochronol-
ogy the following articles are recommended: Ireland & Williams
(2003), Stern & Amelin (2003), Schmitz & Schoene (2007), and
various papers by Ludwig (1980, 1991, 1998, 2003).
Random/internal uncertainties
Random/internal uncertainties are those relating to the measure-
ment of isotopic ratios of the sample, standards and blanks, and
are used in the derivation of errors of the radiogenic ratios.
These uncertainties are intrinsic to each analysis and represent
the minimum uncertainty that must be considered. Most of these
sources of random uncertainty relate to the mass spectrometry
measurements and our ability to measure and reproduce isotopic
ratios, such as corrections made for laboratory blank. Factors
such as the electronic noise of detectors place a theoretical limit
on the precision that can be achieved by detecting a certain
number of ions over a finite period of time (counting statistics).
However, for almost all geochronological applications, other
factors such as correction for mass-dependent fractionation that
occurs during sample ionization, and correction for laboratory
blank and/or initial parent and daughter nuclide, dominate the
analytical uncertainty budget, especially for small samples.
The analytical uncertainties associated with U – Pb ID-TIMS
dates have decreased substantially over the past decade. This is
due in large part to a reduction in the common Pb levels introduced
in the laboratory (laboratory blank) and correction for mass frac-
tionation though use of a ‘double-spike’, where two tracer isotopes
of the same element (202Pb/205Pb or 233U/235U, for example) are
used for real-time mass fractionation correction, helping reduce
the uncertainty in the mass-dependent fractionation. See Schmitz
& Schoene (2007) and Bowring et al. (2006) for a more complete
discussion of sources of uncertainty in U – Pb ID-TIMS analyses.
In SIMS and LA-ICP-MS (in ‘dry’ mode) analyses, common Pb
levels are intrinsically low and the uncertainty related to any cor-
rection for common Pb is usually insignificant compared to the
uncertainty in the U –Pb normalization (see below). In ‘wet’
mode, as is typical for LA-ICP-MS U –Pb analyses, the correction
for Pb blank and isobaric interferences on the Pb and U peaks can
be a significant source of uncertainty, requiring a correction to be
made (Horstwood et al. 2003).
Microbeam U –Pb standardization. Microbeam U –Pb data are
acquired in analytical sessions where the analysis of unknown
zircons are alternated with standards. The raw measured U –Pb
ratio of the standard varies or ‘drifts’ during an analytical
session due to slight changes in instrument parameters. Therefore
there is an uncertainty associated with the Pb–U standardization
that must be considered. This uncertainty is on the order of 1%
(Stern & Amelin 2003), although the attributed magnitude is
dependent on the frequency with which the standard is analysed.
There are differences of opinion on how the uncertainty related
to the U –Pb standardization is factored into the total uncertainty of
a date. Some groups consider that the ‘standardization’ value is
constant for a given session, so the uncertainty is systematic and
need only be considered when comparing data collected in differ-
ent analytical sessions. In this approach the session Pb– U uncer-
tainty can be simply added to the weighted mean uncertainty in
a manner analogous to the ID-TIMS tracer calibration uncertainty.
Conversely, other groups consider that the reproducibility of the
standard is a reflection of the true external reproducibility of all
analyses and that it should be incorporated into individual analyses
of unknown zircon (Ireland & Williams 2003; Stern & Amelin
2003). Typically, larger uncertainties for each analysis result in a
reduction of the MSWD for weighted mean dates from multiple
analyses, but has obvious implications for the identification of out-
liers (see discussion in Ireland & Williams 2003). Information
regarding the approach taken to the standardization uncertainty
is often recorded in the footnotes to the data table or in the data
repository, but can be crucial when trying to precisely sequence
rocks or calculate durations of events.
Typical internal (2s) uncertainties for ID-TIMS 207Pb/206Pb
and 206Pb – 238U zircon dates are c. 0.5 –0.2% and c. 0.1– 0.05%,
respectively, and for microbeam techniques internal (2s) uncer-
tainties on 207Pb/206Pb and 206Pb– 238U dates are c. 3 –5% and
c. 1 –2%, respectively. The difference is about an order of magni-
tude, and for a 206Pb– 238U date of 600 Ma amounts to the differ-
ence in uncertainty on a single date between 0.3 and 0.6 Ma for
ID-TIMS and 6 and 12 Ma for microbeam U –Pb dates.
 USER’S GUIDE TO NEOPROTEROZOIC GEOCHRONOLOGY
Systematic/external uncertainties
Systematic uncertainties are those related to a parameter that has a
fixed value and is effectively constant for a suite of determinations,
and is usually determined by experiments. An example of such a
parameter used in U –Pb geochronology with systematic errors
would include the 238U decay constant (Jaffey et al. 1971),
which will affect all 238U – 206Pb dates and thus can be ignored
when comparing any 238U derived dates. The level at which
these systematic errors have to be considered depends upon their
nature; if they are specific to a single laboratory (i.e. tracer cali-
bration) then the systematic uncertainty has to be considered
when comparing dates determined using a different tracer. Simi-
larly, decay constant uncertainties affect all dates derived from
that decay constant equally, so they only need to be considered
when comparing dates derived from different decay constants
(i.e. 238U – 206Pb and Re –Os dates). They can otherwise be
ignored when calculating differences between dates based upon
the same decay scheme. This distinction between random and
systematic (or internal and external) types of uncertainty is impor-
tant, as in some circumstances where the same parameter (decay
constant, lab tracer) is used for calculating a suite of dates the
effective uncertainty can be reduced. This results in increased tem-
poral resolution at which synchronicity, rates and durations of
proxy record evolution can be assessed.
Decay constants. One source of systematic uncertainty that affects
all radio-isotopic dates is that related to uncertainty in the decay
constants (Table 9.1). Three approaches have been taken to deter-
mine the decay constants (the probability that a given atom will
decay per unit of time) of the long-lived radionuclide: (i) direct
counting; (ii) in-growth; and (iii) geological comparison. Direct
counting involves the detection of alpha, beta or gamma activity
relative to the total number of radioactive atoms. In-growth
relies upon the quantification of a decay product that is accumu-
lated from a quantity of high-purity parent nuclide over a well-
defined period of time. Geological comparison involves analyses
of cogenetic rocks or minerals with multiple chronometers, assum-
ing that each chronometer should yield the same date. This
approach has the potential for relative intercalibration of the
decay constants, but accurate intercalibration requires that at
least one decay constant is accurate and known with some pre-
cision (Mattinson 2000; Schoene et al. 2006). This is usually
assumed to be the 238U and 235U due to the precision with which
the decay constants have been determined (Jaffey et al. 1971)
and the internal check provided by closed-system zircon analyses
(Mattinson 2000; Schoene et al. 2006).
The counting experiments of Jaffey et al. (1971) determined the
238
U and 235U decay constants with uncertainties of 0.11% and
0.14%, respectively. These values have been adopted for use
in geochronology (Steiger & Jager 1977). The 187Re and 176Lu
decay constants have been determined by both direct counting
experiment and through geological comparison with the U –Pb
system, and uncertainties are estimated at c. 0.4–0.5% (Scherer
et al. 2001; Selby et al. 2007).
The incorporation of decay constant uncertainties is becoming
increasingly important as the internal precision of dates is
reduced and multiple geochronometers are being used to investi-
gate the same time intervals. The decay constant uncertainties
for isochron dates are typically ,20% of the total uncertainty
budget; in contrast, the uncertainties in the U decay constants
are often .50% of the total uncertainty budget of U –Pb
ID-TIMS dates (Fig. 9.2). The situation for the ID-TIMS U –Pb
community is that 206Pb– 238U and 207Pb/206Pb dates often do
not overlap within analytical precision, and the U decay constant
uncertainties must be considered (Ludwig 2000; Begemann et al.
2001; Schoene et al. 2006). As the ‘consumer’ often uses these
dates interchangeably, we are now seeing 206Pb– 238U and
207
Pb/206Pb age uncertainties presented as +X/Y/Z and +X/Z
141
respectively, where X is the analytical/internal uncertainty, Y is
the analytical uncertainty plus the systematic tracer calibration
uncertainty, and Z is the total uncertainty including X, Y and the
decay constant uncertainties. This permits use of the data with
the level of uncertainty that is appropriate to the problem being
addressed.
Age of primary standards for microbeam U –Pb dating. As discussed
above, U –Pb microbeam techniques rely upon measurement of
the U – Pb ratio relative to standard minerals of known age. The
U –Pb dates of these minerals are determined via ID-TIMS ana-
lyses with typical total uncertainties of 0.1 –0.3% that should be
propagated into the total uncertainty of the final U/Pb microbeam
date. Because the systematic uncertainty related to the age of
the primary standard is about an order of magnitude less than
the random errors related to the U – Pb determination of the
unknown mineral, it is often not considered significant.
Intra- and intercrystal homogeneity is a fundamental require-
ment of a zircon standard for microbeam U – Pb geochronology
as the U – Pb ratio of the standard is considered invariant. Isotopic
homogeneity is assessed by multiple ID-TIMS analyses on single
crystals and/or crystal fragments (Black et al. 2003; Schmitz et al.
2003) with variability assessed on a microgram scale. In contrast,
microbeam techniques require standards that are homogeneous on
the submicrometre scale. At present, zircon standards are either
chips of megacrysts (e.g. SL13 and 91500) or multicrystal
mineral separates from plutonic rocks (e.g. Temora, R33). In
general the zircon standards are relatively homogeneous at the
level that can be detected by either microgram ID-TIMS analyses
or nanogram SIMS analyses. However there have been issues with
at least one of the megacryst standards (SL13), which is hetero-
geneous at the micrometre-scale (Ireland & Williams 2003). The
fact that all zircon standards are natural means they are not
perfect, as they are likely to be affected by zonation and/or Pb
loss and/or other (matrix-related) differences. For most well-
characterized zircon standards this variation occurs well below
the level of quantification/detection and is not significant.
Calibrating tracers for isotope-dilution. For isotopic analyses that
use the addition of isotopic tracers (isotope dilution), the accuracy
of the tracer calibration (isotopic composition and concentrations)
has major control over the accuracy of the derived dates for a
mineral or rock. Calibration of tracers is performed through admix-
ing the tracer with another solution of known isotopic composition
and, importantly, known purity. High-purity metals or salts (see
Selby et al. 2007 for details of a Re –Os tracer, Wasserburg
et al. 1981 for Sm– Nd, Schoene et al. 2006 for details of a
U –Pb tracer calibration) are used as the basis of the gravimetric
reference solutions against which the concentration of the tracer
isotope can be determined. Therefore, the purity of the metal or
salt, and the accuracy of the weighing prior to dissolution, controls
the precision and accuracy of the calibration. This total uncertainty
is typically estimated at c. 0.1%. For multi-element tracers, the
elemental (i.e. U –Pb) ratio is fixed, so the uncertainty in the
tracer calibration is systematic and can be ignored for the practical
purposes of age determinations generated using the same tracer.
This is particularly useful when attempting to determine the
relative time difference between samples such as determining sedi-
ment accumulation rates (Bowring et al. 2007), or assessing syn-
chroneity of events (Condon et al. 2005). At present, it is typical
that each isotope laboratory has their own tracer. This means
that the tracer uncertainty has to be considered when comparing
dates with other laboratories and other techniques. Recently, the
U –Pb ID-TIMS community has made an effort to eliminate this
inter-laboratory uncertainty through the development and cali-
bration of a large amount of 205Pb– 233U – 235U tracer for commu-
nity use under the auspices of the EARTHTIME Initiative (Condon
et al. 2007). As analytical uncertainties are reduced it is becoming
apparent that inter-laboratory (or inter-session) variations are
142 D. J. CONDON & S. A. BOWRING
comparable to, or greater than, internal precision. Although much
of this can be eliminated by a common tracer solution for U –Pb
ID-TIMS, as discussed above, it is also desirable that along with
analyses of unknown zircons, secondary standards (minerals or
synthetic solutions) are analysed in order to provide an accurate
assessment of long-term reproducibility and inter-laboratory
agreement.
Calculating an age from multiple dates
A significant proportion of temporal constraints for Neoprotero-
zoic strata are derived from U –Pb dates on zircons from volcanic
rocks. The final reported date and associated uncertainty are often
weighted mean dates derived from a number (n) of individual dates
on different zircons (or zircon sub-domains). This is the case for
data acquired using both ID-TIMS and microbeam techniques.
The weighted mean weights each individual analysis (such as a
single SIMS spot or single-grain ID-TIMS analyses) according
to its precision, so analyses with a low uncertainty contribute
more to the weighted mean than those with high uncertainty.
Importantly, the use of a weighted mean algorithm (or other aver-
aging) is underpinned by the expectation of a single population
with normally distributed errors (i.e. there is no correlation
between precision of analyses and age). If the errors on the individ-
ual analyses are approximately equal (as is typical for microbeam
U –Pb
pffiffiffi data), then the weighted mean uncertainty is proportional to
1/ n. In this case, high-n datasets can be used to reduce the
overall age uncertainty for data collected on a single population
with normally distributed errors. If the uncertainties on the individ-
ual analyses are variable then the weighted mean uncertainty is
controlled by the most precise analyses making the weighted
mean date less proportional to n. High-n datasets are critical for
assessing analytical v. geological scatter; however, the real limit
on the precision is the analytical uncertainty of single (spot or
grain) analyses, as this controls our ability to resolve real variation
within a series of analyses due to ‘open-system’ behaviour.
A common measure of the ‘coherence’ of a dataset is a statistical
parameter called the MSWD (mean square of the weighted devi-
ates; York 1966, 1967). A value of c. 1 indicates that the scatter
in the data can be explained by analytical uncertainties alone.
Values much less than 1 indicate that analytical uncertainties
have been overestimated. Values greater than 1 can indicate
either that the uncertainties have been underestimated or that
another source of scatter, often called ‘geological’ scatter is
present. Furthermore, the actual MSWD value for which the
scatter of the data can be considered due to analytical factors
alone is not restricted to a value of 1, but in fact varies according
to the number of data points in the calculation (Wendt & Carl
1991). So, to be 95% confident that the scatter of the data is due
to the analysis when n ¼ 5, an acceptable MSWD range would
be 0.2 –2.2, but for n ¼ 25 this would be 0.6– 1.5 (Wendt & Carl
1991). Although not often explicitly stated, an MSWD of 1 does
not necessarily mean there is a single (age) population. Rather, it
indicates that if real (age) variation is present, it cannot be resolved
within the precision of the individual analyses. It is not uncommon
to see both isochron and weighted mean dates reported with
MSWDs much greater than 1. These dates likely have little geo-
logical significance and should not be used.
Uncertainties as a result of geologic complexity
Uncertainty as a result of geological complexity is the most diffi-
cult to quantify. The most common cause of excess scatter is open-
system behaviour resulting from either inheritance of older zircon
and/or Pb loss. For U – Pb zircon analyses, reduced errors on
single analyses often exposes fine-scale variability that may
reflect protracted or punctuated crystallization of zircon crystals
in a magma chamber or the effects of very subtle open-system be-
haviour. Thus, high-precision analyses do not always transform
into reduced uncertainties in calculated weighted mean dates.
Complex U –Pb zircon systematic. In the past decade, errors associ-
ated with ID-TIMS analyses have been reduced by almost an order
of magnitude. These reduced errors offer unprecedented precision,
but also expose geological complexity at the ,0.1% level, some-
times resulting in scatter that exceeds analytical uncertainties. It is
now common for a geochronologist to be faced with a population
of zircon analyses that do not form a coherent cluster, and the
crucial question is how to interpret the data to arrive at an eruption
and/or depositional age. The advent of CA-TIMS pre-treatment
for the elimination of Pb loss has been extremely important, as it
gives one confidence that in many cases Pb loss need not be con-
sidered as a cause of excess scatter. Furthermore, for Neoprotero-
zoic rocks, the concordia curve has a shallow enough slope, and the
207
Pb – 235U dates measured precisely enough to be able to evaluate
discordance at the per mil level. However, this is not the case for
microbeam U – Pb dates.
As outlined above, microbeam U – Pb dates on volcanic rocks
rely upon the averaging of a relatively high-n dataset (10 –20) of
relatively imprecise (c. 2– 4%) U –Pb determinations to get a
weighted mean date with a precision of c. 1% or less. Underpinning
these lower uncertainties is the assumption of a single population
with normally distributed errors. However, it is the low precision
of each analysis combined with variability of the standard analyses
that bracket unknowns that often precludes the detection of subtle
amounts of Pb loss or inheritance. Stated another way, if the
amount of Pb loss or inheritance is less than the precision of a
single spot analysis, then it cannot be detected via normal statisti-
cal proxies (such as the MSWD), so the assumption of a normal
distribution may be invalid (Fig. 9.4). If Pb loss is the main
source of open-system behaviour, this will have the effect of
lowering the 206Pb – 238U date on some analyses, as well as the
weighted mean 206Pb– 238U date. If unrecognized, the younger
dates could be misconstrued as the true crystallization age.
To highlight the problem of assigning dates to samples from
multiple U –Pb analyses, we compare two published studies on
zircons from the same ash bed from the lower Doushantuo For-
mation of China. Zhang et al. (2005) obtained a weighted-mean
206
Pb – 238U age of 621 + 7 Ma (n ¼ 13, MSWD ¼ 1.13, does
not include the 0.48% error in U – Pb calibration) from concordant
U –Pb zircon analyses on their sample 04SC20A, which occurs c.
2 m above the cap carbonate in the Yangtze Gorge region. Condon
et al. (2005) sampled the same ash bed (their sample YG04-2) and
analysed the zircons by the ID-TIMS method. The ID-TIMS
dataset showed a spread in U –Pb dates that formed a linear
array below Concordia, indicating zero-age Pb loss. Three of the
ten zircons analysed were ‘concordant’ and yield a weighted
mean 206Pb– 238U date of 632.3 + 0.5 Ma (MSWD ¼ 1.15, does
not include the 0.1% error in U –Pb tracer calibration). The two
dates are statistically different and do not overlap, even when esti-
mates of random and systematic error are considered; however,
when compared at the precision of a single analysis, all the
SHRIMP U –Pb dates overlap with the ID-TIMS dates (Fig. 9.4).
A likely interpretation of this dataset is that a number of SIMS ana-
lyses were located on domains that had lost a small amount of Pb,
resulting in a lower mean 206Pb– 238U date when all the analyses
are included. The higher precision of the ID-TIMS analyses
allows small amounts of Pb loss to be recognized and those data
points not included in calculating the final 206Pb– 238U date. A
third U –Pb date has been published for this same ash bed: Yin
et al. (2005) obtained a weighted mean 206Pb – 238U age of
628.3 + 5.8 Ma (MSWD ¼ 0.86). This publication does not
include any details of the U –Pb data such as error in U –Pb nor-
malization, but it is presumed this weighted mean uncertainty
does not include any systematic sources of uncertainty. This date
 USER’S GUIDE TO NEOPROTEROZOIC GEOCHRONOLOGY 143

is indistinguishable from either the other SHRIMP date (Zhang

(a)
680 et al. 2005) or the ID-TIMS date (Condon et al. 2005).
As a second example of problems related to determining a
U
238

sample age from a population of U –Pb (zircon) dates we present

Pb/

new ID-TIMS data on zircon from another sample that was pre-
206

viously analysed by SHRIMP to derive a U – Pb date (Fig. 9.5).

Fanning and Link (Fanning & Link 2004) dated samples from the
640
0.104 Neoproterozoic Pocatello Formation, Idaho, using the SHRIMP
U –Pb method on zircon from a series of volcanic layers and
620 clasts within diamictite. We obtained new ID-TIMS U – Pb
0.100 zircon data on one of these samples, 06PL00, that had yielded a
weighted mean SHRIMP 206Pb/238U date of 708 + 8 Ma
600 (n ¼ 18, MSWD ¼ 1.7, includes additional 0.35% error in U/Pb
0.096 normalization). For a population of this size an MSWD of 1.7 is
at the upper limit of acceptability (Wendt & Carl 1991), possibly
indicating scatter due to non-analytical causes, that is real age vari-
0.092 ation in the population.
560
Six single zircon crystals have been analysed by ID-TIMS
207 235 (Table 9.2 & Fig. 9.5) at the Massachusetts Institute of Technology
0.5 0.6 0.8 0.9 Pb/ U
following the analytical procedure outlined in Bowring et al.
(b) (2007). Three zircons are from the same mineral separate used
for the SHRIMP study (06PL00) and three are from an attempted
U

recollection (36PL05) at the same stratigraphic level as sample

238

06PL00 in the Scout Mountain Member of the Pocatello For-

Pb/

635
mation, with the authors of the original study. One zircon
206

(00PL06 z2) is concordant at c. 705 Ma, whereas the remaining

five zircons form a population at c. 690 Ma. Within this population,
0.1030
four zircons give equivalent 206Pb– 238U dates with a weighted
mean of 687.4 + 0.6/1.3 Ma, and the fifth zircon (34PL05 z4)
0.1025 gives a similar 207Pb– 206Pb date but a slightly younger
206Pb– 208U date, which we interpret to reflect Pb loss.

This new ID-TIMS U – Pb date of 687.4 + 1.3 Ma (includes 1‰

625
tracer calibration error) is distinctly younger than the weighted
mean SHRIMP U –Pb date of 705 + 8 Ma (includes 0.35%
U –Pb calibration error) for the same sample and stratigraphic
level (Fanning & Link 2004). However, when we look at the
data in terms of single analysis data points, we can see that there
0.1010 are some similarities between the two datasets (Fig. 9.5): they
207 235
0.848 0.852 0.856 0.860 0.864 0.868
Pb/ U both contain zircon 206Pb– 238U dates of c. 705 Ma and c. 690 Ma.
The proportions of these different age populations is different,
with the ID-TIMS dataset being dominated by c. 690 Ma grains,
(c)
whereas this age population is subordinate in the SHRIMP
dataset, most likely reflecting the low-n nature of the ID-TIMS
660 dataset and perhaps differences in pre-selection of zircons prior
to analysis based upon morphology and so on. Importantly, the
relatively low precision of the SHRIMP U – Pb dates means that
640 it is not possible to discern different populations at the ,3%
level. The assumption of a single (age) population and resultant
weighted mean calculation results in a U –Pb date that is
620
c. 20 Ma older than the concordant ID-TIMS population, which
is considered the most robust estimate of the age of the rock.
These two case studies from the lower Doushantuo Formation,
China, and the Scout Mountain Member, Idaho, serve to illustrate
600
the problems related to the recognition of real scatter, be it due to
Pb loss and/or a mixed age population, at a resolution that is less
than the precision of single analysis. In such cases, exploiting
580 high-n datasets to derive weighted mean dates with lower uncer-
SHRIMP analyses (Zhang et al. 2005)
ID-TIMS analyses tainties can result in inaccurate age assignments for a given
(Condon et al. 2005) sample. This section is not meant to suggest all SHRIMP U –Pb
data are inaccurate but to demonstrate that beyond the level of a
Fig. 9.4. U –Pb zircon data on samples from the lower Doushantuo Formation
single analysis uncertainty (i.e. in the calculation of weighted
(2 m above the cap carbonate), Yangtze Gorge region, China. (a) U– Pb mean dates) assumptions are being made about the nature of the
concordia plot of SHRIMP U –Pb (zircon) dataset obtained on zircon from population. If these assumptions are valid then the weighted
(Zhang et al. 2005). (b) U– Pb concordia plot of ID-TIMS U–Pb (zircon) mean date and associated error will be accurate; however, if
dataset for the same ash bed (Condon et al. 2005). (c) Plot of single analysis these assumptions are not valid (i.e. it is not a single age population
206
Pb – 238U dates of both the SHRIMP and ID-TIMS datasets. The grey bar with normally distributed errors), then the calculated weighted
reflects the weighted mean SHRIMP U–Pb date (all points included) and mean date and associated error may be an inaccurate reflection
ID-TIMS date (three concordant analyses included). The error bar incorporates of the sample age. In situ techniques are without question essential
error in U–Pb calibration (SHRIMP and TIMS). tools for characterizing complex zircons from volcanic and
144 D. J. CONDON & S. A. BOWRING
Fig. 9.5. U– Pb zircon data on samples from the Scout Mountain Member,
Pocatello Formation, southern Idaho. (a) U– Pb concordia plot of SHRIMP
U–Pb (zircon) dataset obtained on sample 00PL00 (Fanning & Link 2004).
(b) New U –Pb concordia plot of ID-TIMS U–Pb (zircon) dataset on zircons
from the 00PL00 mineral separate and a recollection from the same
stratigraphic level (see Table 9.2). (c) Plot of single analysis 206Pb– 238U dates
of both the SHRIMP and ID-TIMS datasets. The grey bar reflects the weighted
mean SHRIMP U–Pb date (all points included) and ID-TIMS date (00PL06 z2a
and 34PL05 z4 excluded) and error incorporates error in U–Pb calibration.
metamorphic rocks and for rapid analysis of detrital populations.
Ideally, microbeam techniques can be used to rapidly characterize
a population of zircons by analysing a small volume of many
zircons, which could then be followed by conventional high-
precision geochronology of selected grains.
Linear arrays v. isochrons. As outlined above, a limitation of the
isochron approach is the lack of an independent check for open-
system behaviour, unlike the dual decay scheme of the U –Pb
system. Most studies use a combination of bracketing age con-
straints and/or a statistical measure of coherence (MSWD or
uncertainty) to assess whether the system has been perturbed;
however, precision and amount of scatter cannot be used as a
proxy for closed-system behaviour. For example, organic-rich
sediments from the Aralka Formation, Australia, have been ana-
lysed for Re –Os geochronology using both the aqua regia and
CrO3 – H2SO4 dissolution methods. Schaffer & Burgess (2003)
used the aqua regia dissolution method and obtained a three-point
isochron (samples integrated over 1.6 m stratigraphic thickness) of
592 + 14 Ma (MSWD  1). An expanded dataset collected over
10 m stratigraphic thickness yielded a nine-point regression and an
age of 623 + 18 Ma (MSWD ¼ 5.2). Subsequent Re –Os analyses
(on samples from a 2 m interval within the 10 m interval sampled
by Schaefer & Burgess (2003) using the CrO3 – H2SO4 dissolution
method yielded a 10-point isochron with an age of 657.2 + 5.4 Ma
(2s internal uncertainties, MSWD ¼ 1.2) (Kendall et al. 2006).
This difference is attributed to either a sampling and/or analytical
artefact(s) related to sample digestion and/or sample-spike equili-
bration (Kendall et al. 2006, 2009).
It is clear from the Aralka case study that one cannot use
coherence of a dataset as a means to assess accuracy, especially
with isochrons based upon low-n datasets (or sub-sets). The
bottom line is that a suite of samples with the same initial ratio
and a range of parent/daughter ratios that evolve in a closed
system can yield an isochron; however, a suite of samples that
define a linear array alone cannot be inferred to represent closed-
system behaviour, as simple mixing of two reservoirs can yield
linear arrays.
Dating the Neoproterozoic
Geochronology plays an important role in the independent inte-
gration of geographically disparate stratigraphic sections and
therefore plays a key role in addressing questions pertaining to
the correlation of glaciogenic intervals and their synthesis in
terms of number of glacial events, their duration and geographic
extent. Several recent studies have assessed the published age con-
straints for Neoproterozoic glaciations in order to address such
questions, and we briefly summarize the current state of play in
light of what we consider the key datasets that pertain to the
tempo of glaciation, the development of early animals and other
interesting events.
How many glacial intervals?
In the past ten years, the number and quality of geochronological
data have increased considerably and (at least) three distinct
glacial ‘intervals’ have been geochronologically delineated.
These temporally distinct glacial intervals are also broadly
distinct in terms of the physical characteristics of associated
strata (Kennedy et al. 1998; Hoffman & Schrag 2002; Halverson
et al. 2005).
I. Ediacaran glaciations. Ediacaran-age (c. 635 to 542 Ma) glacial
deposits are preserved on at least three palaeocontinents. The
Gaskiers Formation on the eastern Avalonian Peninsula has been
directly dated by U –Pb (zircon) ID-TIMS dates on ash beds that
 USER’S GUIDE TO NEOPROTEROZOIC GEOCHRONOLOGY 145

Table 9.2. ID-TIMS U-Pb data on zircons from 06PL00 and 34PL05

Concentrations Age (Ma)

206 208 207 207 206 206 207 207
U Pb Th Pb* Pb Pb Pb Pb corr. Pb Pb Pb
Fractions1 (ppm)2 (ppm)2 U3 Pb*/Pbc4 204
Pb5 206
Pb6 206
Pb6 % err7 235
U6 % err7 238
U6 % err7 coef. 238
U8 235
U8 206
Pb8

34PL05 z19 55 7 0.69 33.5 1935.4 0.2160 0.0625 0.13 0.96940 0.18 0.11254 0.12 0.696 687.49 688.15 690.35
34PL05 z29 33 4 0.50 35.8 2174.6 0.1559 0.0626 0.24 0.97056 0.31 0.11253 0.19 0.655 687.44 688.75 693.04
34PL05 z49 67 9 1.07 74.3 3915.9 0.3331 0.0625 0.09 0.96463 0.12 0.11203 0.08 0.690 684.51 685.69 689.56
06PL00 z1 13 2 0.49 10.9 672.6 0.1541 0.0625 0.35 0.96970 0.48 0.11252 0.30 0.676 687.37 688.31 691.35
06PL00 z2 42 5 0.78 22.5 1278.0 0.2433 0.0630 0.13 1.00320 0.21 0.11551 0.16 0.779 704.68 705.43 707.82
06PL00 z3 52 7 0.78 18.4 1052.3 0.2426 0.0625 0.15 0.96939 0.25 0.11248 0.19 0.795 687.13 688.15 691.53

1
z1, z2, etc. are fractions composed of single grains. Grains were annealed and leached prior to dissolution. See Bowring et al. (2007) for analytical procedures.
2
Volumes estimated from images, absolute uncertainty is c. 50%.
3
Model Th–U ratio calculated from radiogenic 208Pb/206Pb ratio and 207Pb/206Pb age.
4
Ratio of radiogenic Pb (including 208Pb) to common Pb.
5
Measured ratio corrected for spike and fractionation only. Mass fractionation corrections were based on analysis of NBS-981 and NBS-983. Corrections of 0.25 + 0.04%/
amu (atomic mass unit) and 0.07 + 0.04%/amu were applied to single-collector Daly analyses and dynamic Faraday– Daly analyses, respectively.
6
Corrected for fractionation, spike and blank. All common Pb was assumed to be procedural blank.
7
Errors are 2s, propagated using the algorithms of Ludwig (1980).
8
Calculations are based on the decay constants of Jaffey et al. (1971).
9
34PL05 was collected within a few hundred metres of the original 06PL00 sample locality and at the same stratigraphic level within the Scout Mountain Member. See
Bowring et al. (2007) for details of the analytical methods.

occur below, within and above the glacial Gaskiers Formation,
constraining its age to c. 584 to 582 Ma with a maximum duration
of 2.6 Ma (Bowring et al. 2003). Similar-age (,600–542 Ma)
glacial successions occur elsewhere on Avalonia (Thompson &
Bowring 2000), the NE margin of Laurentia (Condon & Prave
2000), Baltica (Bingen et al. 2005) and the Tarim Block, NW
China (Xu et al. 2009); however, the age of these units is less
well constrained, making it impossible to assess the synchronicity
of Ediacaran glaciation(s).
II. ‘Marinoan’ glaciations. Marinoan-type glacial deposits are con-
sidered those associated with cap carbonates with distinctive litho-
logical and isotopic similarities (Kennedy et al. 1998; Hoffman
et al. 2007) and they occur on nearly every palaeocontinent. A
maximum age constraint for the Marinoan glaciation comes
from China and Australia. In China, the top of pre-glacial
Datangpo Formation is constrained by a U – Pb (zircon) SHRIMP
date of 654.5 + 3.8 Ma (Shihong et al. 2008). In Australia, a
Re –Os isochron age of 643 + 2.4 Ma (Kendall et al. 2006) was
obtained from black shale of the Tindelpina Member at the base
of the pre-Marinoan Tapley Hill Formation. Synglacial age con-
straints come from 635.5 + 1.2 Ma U– Pb (zircon) ID-TIMS
date from an ash bed within the basinal facies of the Ghaub For-
mation (Hoffmann et al. 2004) and a similar 636.3 + 4.9 Ma
U –Pb (zircon) SHRIMP date from the base of the Nantuo For-
mation (Shihong et al. 2008). Termination of this glacial episode
is constrained by an ash bed within the cap carbonate of the
Nantuo Tillite (Yangtze Platform), which has been dated at
635.2 + 0.6 Ma (Condon et al. 2005). Combined, these data indi-
cate a duration of ,10 Ma and globally synchronous termination
of the Marinoan glaciation at c. 635 Ma. It should be noted that
there still remains a lack of consensus over the age of the actual
Marinoan glacial in its type section and inferentially the base of
the Ediacaran Period.
III. Sturtian glaciation(s). Sturtian-type glacial deposits are typi-
cally defined as being stratigraphically lower than Marinaon-type
deposits and are commonly associated with dark, organic-rich
cap carbonates and sedimentary ironstones. On the Congo
craton, the Chuos Formation overlies the Naauwpoort volcanics
dated at 746 + 2 Ma (Hoffman et al. 1996), providing a robust
maximum age constraint. In Oman, the glaciogenic Ghubrah
Formation is dated at c. 712 Ma (Bowring et al. 2007), although
this glacial deposit, where dated, is not associated with a cap car-
bonate. U –Pb (zircon) SHRIMP dates from the Pocatello For-
mation, Idaho (Fanning & Link 2004), are commonly cited as
age constraints for Sturtian glaciation. However, as we have out-
lined above, there are accuracy issues with some of these dates
and, perhaps more importantly, their geological context has been
questioned (Fanning & Link 2008). Similarly, commonly cited
age constraints from the Edwardsburg Formation preserved in
roof pendants of the Idaho Batholith (Lund et al. 2003) lack
robust sedimentological and stratigraphic constraint in order to
assess their relevance for the Sturtian glacial episode.
Relevant age constraints from post-Sturtian and pre-Marinoan
glacial deposits come from both Australia and China. In Australia,
a Re –Os date of 657.2 + 5.4 Ma (Kendall et al. 2006) for the basal
Aralka Formation provides a minimum constraint on the age of
the underlying Areyonga glacial deposits in the Amadeus Basin,
which is correlated with the type-Sturtian glacial deposits
(Walter et al. 2000). In China, a U –Pb (zircon) ID-TIMS date of
663 + 4 Ma from the Datangpo Formation (Zhou et al. 2004)
directly overlies the glacial Tiesiao Formation and its associated
Mn-rich cap carbonate, again providing a minimum age constraint
on a Sturtian-type glacial deposit. Thus, Sturtian-type glacial
deposits are constrained to the interval c. 746– 663 Ma.
However, the nature of the current dataset, with issues remaining
about the glacial affinity/stratigraphic context of certain succes-
sions as well as those relating to the radio-isotopic dates them-
selves, means that it is not possible to infer anything about
duration and/or synchronicity of the glacial deposits and, inferen-
tially, the nature of the glacial episode(s) they reflect.
IV. Pre-Sturtian? glaciations. Evidence for a pre-Sturtian glaciation
(c. 740 –760 Ma) comes from a small number of sections in
Namibia (Frimmel et al. 1996), Zambia (Key et al. 2001) and
NW China (Xu et al. 2009); however, in each case, either the
exact relationship of the dates to the glacial unit is inferred and a
complicated local stratigraphy lessens the confidence that the
dated unit directly constrains a glacial deposit (Frimmel et al.
1996; Key et al. 2001) and/or the glacial nature of the dated unit
remains equivocal (Frimmel et al. 1996; Xu et al. 2009). As
such, the community should remain circumspect about an older
Cryogenian glaciation until more equivocal evidence is presented.
146 D. J. CONDON & S. A. BOWRING
Dating the earliest fossils
There is growing consensus on the development of Ediacaran-age
organisms based upon the sequencing of key assemblages via
high-precision U –Pb dating. We briefly review the nature of,
and age constraints for, key assemblages/Fauna.
I. Avalon assemblage. The Avalon assemblage is preserved in
relatively deep-water marine settings within the Avalon zone of
Newfoundland and England, and is characterized by rangemorphs,
frond- and bush-shaped colonies, which lacked mobility. The ear-
liest Ediacaran (macroscopic) fossil (Charnia, C. masoni,
C. wardi), fronds some 2 m in length, occur in the Drook For-
mation (Narbonne & Gehling 2003) and are dated at c. 577 Ma
(Bowring et al. 2003), some 5 Ma after the end of the Gaskiers
glaciation. This assemblage extends in range to c. 560 Ma in
Newfoundland and England; however, this upper limit requires
more robust constraint.
II. White Sea assemblage. The White Sea assemblage is preserved in
a shallow marine setting in the Vendian sections of the White Sea
and the Ediacaran Member of Australia (Narbonne 2005). This
assemblage includes fronds similar to those of the Avalon assem-
blage, but also includes segmented and discoid fossils (such as
Kimberella) as well as horizontal traces indicating mobile bilater-
ians were present, marking the advent of muscularity recorded in
the fossil record. In the White Sea sections, this assemblage is con-
strained to c. 555 Ma (Martin et al. 2000), probably extending to
close to the base of the Cambrian.
III. Nama assemblage. Nama assemblage fossils, typified by those
of the Nama Group, Namibia, include rangemorphs (mainly multi-
frondate fronds) and horizontal traces similar to the White Sea
assemblage, with an age c. 550 Ma to the base of the Cambrian.
An important addition to the Nama assemblage are the weakly
calcified metazoans, Cloudina and Namacalathus, representing a
major evolutionary development (calcification) and providing a
record of predation (Hua et al. 2003). These weakly calcified
metazoans occur in many terminal Ediacaran successions
(Oman, Namibia, China and Canada), and their age range is con-
strained from c. 550 Ma to the base of the Cambrian (Grotzinger
et al. 1995; Amthor et al. 2003; Condon et al. 2005).
IV. Doushantuo fauna. The Doushantuo fauna comprise phospha-
tized eggs and embryos (Xiao et al. 1998, 2000), as well as micro-
scopic sponges (Li et al. 1998), and thus provides critical
constraint upon the earliest animals. The age of the Doushantuo
Formation is constrained to c. 635 –550 Ma (Condon et al.
2005). These fossils occur in the middle of the formation and are
therefore likely c. 590 to 570 Ma, although this remains to be con-
strained by direct dating. The Miaohe biota of the upper Doushan-
tuo Formation contains diverse macroalgal assemblages (Xiao
et al. 2002). Although not part of the White Sea assemblage, the
Miaohe biota is similar in age, being slightly older than the
550.1 + 0.6 Ma ash from the top of the Doushantuo Formation
(Condon et al. 2005), perhaps indicating a temporal coincidence
between the development of macroalgae and bilaterians (White
Sea assemblage, see above).
Recently Love et al. (2009) documented an abundance of a
Demospongiae-specific biomarker from Cryogenian (c. 710–
635 Ma) sedimentary rocks in Oman. These data indicate a
pre-Ediacaran age for the advent of multicellularity.
Finally, the disappearance of these Ediacaran fossil assemblages
(Rangemorphs, weakly calcified metazoans) is coincident with
the base of the Cambrian and a negative d13C isotope excursion
(Grotzinger et al. 1995; Amthor et al. 2003). The base of the Cam-
brian is dated at c. 541 Ma (Amthor et al. 2003; Bowring et al.
2007). The Early Cambrian fossil record documents a period of
enhanced biomineralization and increased muscular development.
Other Neoproterozoic events
In addition to the climatic and biological events/developments
that took place during the Neoproterozoic, there were also a
series of other events that took place, including impact events
and major isotope excursions. Age constraints for most of these
events are based upon correlation-based inference and as such
we do not discuss them here. One event, the Shuram/Wonoka
carbon isotope excursion, is directly constrained by radio-isotopic
dating and is discussed further below.
I. Shuram/Wonoka carbon isotope excursion. The Shuram/Wonoka
carbon isotope excursion (CIE) is a pronounced large-magnitude
(c. 15‰) d13C excursion that is recorded in Oman, Australia,
China, Siberia, Namibia and several other locations (Burns &
Matter 1993; Calver 2000; Condon et al. 2005; Melezhik
et al. 2005, 2008). Termination of this event has been dated at
c. 550 Ma in the Doushantuo Formation, where d13C cross
from negative to positive (Condon et al. 2005). The onset (and
inferentially the duration) has yet to be satisfactorily constrained,
with detrital zircons from the base of the CIE providing
maximum age constraints dated at c. 620 Ma (Bowring et al.
2007, 2009). The duration of this unique CIE is a matter of some
debate, with estimates varying from c. 50 Ma (Le Guerroue et al.
2006) to c. 5– 11 Ma (Bowring et al. 2007). Constraining
the timing of onset of the Shuram/Wonoka CIE is critically
important to assessing hypotheses that implicate this event in
oxygenation of oceans and early animal evolution (Condon et al.
2005; Fike et al. 2006).
Future directions
Clearly, research on the Neoproterozoic Earth System would
greatly benefit from more and, perhaps more importantly, better
dates. The number of possible dates is limited by the identification
of suitable sections/materials for analyses: ash bed bearing succes-
sions for U– Pb (zircon) dating, organic-rich sediments for Re – Os
analyses, and so on. There is undoubtedly a wealth of material to be
analysed from intervals of interest. Timely identification of targets
among the background of similar-looking material, which will not
yield meaningful age constraints, requires a change in approach,
with the development of better screening protocols, in order to
maximize efficiency
From the U –Pb zircon perspective we propose there is great
utility in breaking down the microbeam v. ID-TIMS barrier. We
have attempted to highlight in this chapter that these two analytical
methods have complementary attributes: the microbeam U –Pb
techniques are capable of high sample throughput, a factor of
c. 10 –20 greater than U –Pb ID-TIMS, whereas the ID-TIMS
U –Pb analyses are c. 10– 20 times more precise. Sample through-
put for ID-TIMS analyses is an issue, and what may not be appreci-
ated by the non-practitioner is that for every dated sample that
makes it into the literature, there will have been many more that
have been analysed and yielded no meaningful chronological
information, most likely due to a predominance of inherited/xeno-
crystic material. An optimized approach for an enhanced U –Pb
(zircon) based chronology for the Neoproterozoic would involve
a first-phase microbeam U – Pb dating, where samples are effec-
tively ‘screened’, prior to a second phase involving ID-TIMS
U –Pb dating of samples that yield microbeam U –Pb dates of inter-
est. Where samples contain mixed-age zircon populations it would
be possible to either remove zircons from a resin mount of
ID-TIMS analyses or, preferably, to simply mount the zircons on
double-sided sticky tape, making removal for ID-TIMS analysis
straightforward. This approach is not suitable for SIMS U – Pb ana-
lyses, but is easily adaptable for LA-ICP-MS analyses. Combining
the efficiency of the microbeam techniques with the precision of
ID-TIMS would be the most effective plan of attack for improving
 USER’S GUIDE TO NEOPROTEROZOIC GEOCHRONOLOGY
the chronology of Neoproterozoic strata, and this would require a
collaborative approach between the various groups involved.
Conclusions
There are a number of radio-isotopic dating techniques
that can be used to constrain the age of Neoproterozoic
sedimentary successions
These include U –Pb dating of minerals (predominantly zircon)
from interlayered volcanic rocks, and Re – Os or Pb/Pb isochron
dating of organic-rich sediments or carbonates. Most of the Neo-
proterozoic time scale is calibrated by U –Pb zircon dates from
air-fall tuffs, but this dataset is augmented by a growing number
of Re –Os and Pb/Pb isochron dates.
Each of the radio-isotopic systems has different strengths and
weaknesses. U –Pb (zircon) dating is considered the premier geo-
chronometer; however, it is of limited use in successions devoid of
zircon-bearing volcanics. Realizing the goal of a robust and highly
resolved temporal framework for the Neoproterozoic will require
exploitation of all these different methodologies and refined
searches for volcanic air fall deposits.
All dates involve some subjective interpretation
Dates interpreted as depositional ages are usually based on mul-
tiple analyses from the same stratigraphic horizon. In the case of
U –Pb (zircon) dates from volcanic rocks, the reported age is
usually based on a weighted mean 206Pb – 238U date in which a vari-
able number of single 206Pb– 238U dates (either single-grain or
single-spot analyses) are weighted (based upon their associated
uncertainty) and a mean calculated. Underpinning this mean date,
and its lower uncertainty, is the assumption of a single population
with normally distributed errors, an assumption that is not always
valid. Isochron dates are similarly based upon the linear regression
of a number of data points that are assumed to be cogenetic, have a
common initial daughter isotopic composition, and the samples
analysed have acted as a ‘closed system’ since their formation.
Although most attempt to produce age constraints with the
lowest possible uncertainty, the best measure of high-precision
age constraints is the precision of a single analysis. It is this pre-
cision that controls the evaluation of ‘open-system’ behaviour
and thus the accuracy of the final date. Although the coherence
of a dataset is often used as a proxy for closed-system behaviour,
the coherence is limited by the precision of the individual analyses.
The uncertainty of a date is as important as the date itself
(Ludwig 2003)
The total uncertainty of a radio-isotopic date comprises random (or
internal) and systematic (or external) components. The random/
internal uncertainties are related to the measurement of the isotopic
ratios and the corrections applied. Systematic uncertainties are
those related to the uncertainty in absolute value of various con-
stant parameters used in the calculation of either an isotopic
ratio or in the calculation of the date itself. Analytical uncertainties
should reflect the ability to reproduce a given isotopic ratio and
represent the minimum uncertainty that may be considered. For
microbeam U –Pb dates, the standard calibration is best considered
as a non-systematic uncertainty and should be incorporated into
each individual spot U –Pb date uncertainty (Ireland & Williams
2003). If comparing dates generated using different techniques
or using different calibration materials (such as mineral standards
for microbeam dates or isotopic tracer for ID-TIMS), then the
systematic uncertainties related to these calibrations must be
147
considered. An additional systematic uncertainty arises from
decay constant uncertainties. These uncertainties have been deter-
mined experimentally or assessed via geological comparison with
another decay scheme.
There is considerable variation in the published literature
regarding the treatment of the constituent parts of the total uncer-
tainty. In some cases, such as when using ID-TIMS 206Pb– 238U
dates generated using a single isotopic tracer solution, certain
components can be ignored (in this case tracer calibration uncer-
tainty and 238U decay constant uncertainty). This can be useful
when attempting to determine sediment accumulation rates
(Bowring et al. 2007) or assess the synchronicity of events
(Condon et al. 2005).
The constituent of the total uncertainty that is most difficult to
quantify is that related to the geology. This can be considered at
the scale of relating a given mineral to a stratigraphic level (i.e.
do the magmatic zircons date the exact age of a given air-fall
tuff?). However, a first-order uncertainty is often the relationship
of a dated unit to the section of interest. Regardless of how
robust the radio-isotopic data are, if the geological context of a
dated sample is poorly constrained, the date is of little use.
A robust, highly resolved chronology for the Neoproterozoic
is yet to be fully realized
Although there has been a marked increase in the number of radio-
isotopic age constraints for Neoproterozoic sedimentary succes-
sions in the past decade, there is still much we do not know, limit-
ing our ability to test correlations and rate dependent hypotheses.
This is in part due to the lack of suitable material to date in key sec-
tions, although it is also due to the reproducibility of some of the
radio-isotopic techniques and systems employed. Accelerated pro-
gress could be made through increased collaboration within the
community, so the most appropriate and efficient methodologies
are applied to the intervals of interest.
We thank E. Arnaud, M. Babinbski, J. Etienne, M. Horstwood, S. Kamo, B.
Kendall, A. Maloof and T. Prave for constructive reviews. This chapter is an
outgrowth of a continuing research programme supported by the NSF, the
NASA Astrobiology Program and NERC. This represents a contribution of the
IUGS- and UNESCO-funded IGCP (International Geoscience Programme)
Project #512.
References
Amthor, J. E., Grotzinger, J. P., Schroder, S., Bowring, S. A.,
Ramezani, J., Martin, M. W. & Matter, A. 2003. Extinction of
Cloudina and Namacalathus at the Precambrian – Cambrian boundary
in Oman. Geology, 31, 431– 434.
Babinski, M., Van Schmus, W. R. & Chemale, F. 1999. Pb/Pb dating
and Pb isotope geochemistry of Neoproterozoic carbonate rocks
from the Sao Francisco basin, Brazil: implications for the mobility
of Pb isotopes during tectonism and metamorphism. Chemical
Geology, 160, 175– 199.
Babinski, M., Vieira, L. C. & Trindade, R. I. F. 2007. Direct dating of
the Sete Lagoas cap carbonate (Bambui Group, Brazil) and impli-
cations for the Neoproterozoic glacial events. Terra Nova, 19,
401– 406.
Barfod, G. H., Albarede, F., Knoll, A. H., Xiao, S. H., Telouk, P.,
Frei, R. & Baker, J. 2002. New Lu– Hf and Pb/Pb age constraints
on the earliest animal fossils. Earth and Planetary Science Letters.
201, 203–212.
Begemann, F., Ludwig, K. R. et al. 2001, Call for an improved set of
decay constants for geochronological use. Geochimica et Cosmochi-
mica Acta, 65, 111– 121.
Bingen, B., Griffin, W. L., Torsvik, T. H. & Saeed, A. 2005. Timing of
Late Neoproterozoic glaciation on Baltica constrained by detrital
zircon geochronology in the Hedmark Group, south-east Norway.
Terra Nova, 17, 250–258.
148 D. J. CONDON & S. A. BOWRING
Black, L. P., Kamo, S. L., Allen, C. M., Aleinikoff, J. N., Davis,
D. W., Korsch, R. J. & Foudoulis, C. 2003. TEMORA 1: a new
zircon standard for Phanerozoic U– Pb geochronology. Chemical
Geology, 200, 155–170.
Bowring, S., Myrow, P., Landing, E. & Ramezani, J. 2003. Geochro-
nological constraints on terminal Neoproterozoic events and the rise
of metazoans. Geophysical Research Abstracts, 5, 219.
Bowring, S. A., Schoene, B., Crowley, J. L., Ramezani, J. & Condon,
D. J. 2006. High-precision U –Pb zircon geochronology and the
stratigraphic record: progress and promise. The Paleontological
Society Papers, 12, 25– 46.
Bowring, S. A., Grotzinger, J. P., Condon, D. J., Ramezani, J.,
Newall, M. & Allen, P. A. 2007. Geochronologic constraints of
the chronostratigraphic framework of the Neoproterozoic Huqf
Supergroup, Sultanate of Oman. American Journal of Science, 307,
1097– 1145.
Bowring, S. A., Grotzinger, J. P., Condon, D. J., Ramezani, J. &
Newall, M. J. 2009. Reply to Comment: Oman Chronostratigraphy.
American Journal of Science, 309, 91– 96.
Burns, S. J. & Matter, A. 1993. Carbon isotopic record of the latest Pro-
terozoic from Oman. Eclogae Geologicae Helvetiae, 86, 595–607.
Calver, C. R. 2000. Isotope stratigraphy of the Ediacaran (Neoproterozoic
III) of the Adelaide Rift Complex, Australia, and the overprint of water
column stratification. Precambrian Research, 100, 121–150.
Chen, D. F., Dong, W. Q., Zhu, B. Q. & Chen, X. P. 2004. Pb/Pb ages of
Neoproterozoic Doushantuo phosphorites in South China: constraints
on early metazoan evolution and glaciation events. Precambrian
Research, 132, 123– 132.
Cherniak, D. J. & Watson, E. B. 2003. Diffusion in Zircon. Reviews in
Mineralogy and Geochemistry, 53, 113– 143.
Cohen, A. S., Coe, A. L., Bartlett, J. M. & Hawkesworth, C. J. 1999.
Precise Re–Os ages of organic-rich mudrocks and the Os isotope
composition of Jurassic seawater. Earth and Planetary Science
Letters, 167, 159–173.
Condon, D. J. & Prave, A. R. 2000. Two from Donegal: Neoproterozoic
glacial episodes on the northeast margin of Laurentia. Geology, 28,
951– 954.
Condon, D., Zhu, M. Y., Bowring, S., Wang, W., Yang, A. H. & Jin,
Y. G. 2005. U– Pb ages from the Neoproterozoic Doushantuo
Formation, China. Science, 308, 95– 98.
Condon, D., Schoene, B., Bowring, S., Parrish, R., McLean, N.,
Noble, S. & Crowley, Q. 2007. EARTHTIME; isotopic tracers
and optimized solutions for high-precision U–Pb ID-TIMS geochro-
nology. Eos, Transactions, American Geophysical Union, 88, Fall
Meeting Supplement, Abstract V41E-06.
Creaser, R. A., Sannigrahi, P., Chacko, T. & Selby, D. 2002. Further
evaluation of the Re– Os geochronometer in organic-rich sedimentary
rocks: a test of hydrocarbon maturation effects in the Exshaw
Formation, Western Canada Sedimentary Basin. Geochimica et
Cosmochimica Acta, 66, 3441– 3452.
Evans, J. A., Zalasiewicz, J. A., Fletcher, I., Rasmussen, B. &
Pearce, N. J. G. 2002. Dating diagenetic monazite in mudrocks:
constraining the oil window? Journal of the Geological Society,
159, 619– 622.
Fanning, C. M. & Link, P. K. 2004. U –Pb SHRIMP ages of Neoproter-
ozoic (Sturtian) glaciogenic Pocatello Formation, southeastern Idaho.
Geology, 32, 881–884.
Fanning, C. M. & Link, P. K. 2008. Age constraints for the Sturtian gla-
ciation: data from the Adelaide Geosyncline, South Australia and
Pocatello Formation, Idaho, USA, Selwyn Symposium Vol. 91. Geo-
logical Society of Australia Abstracts, Melbourne, Geological
Society of Australia, 57 – 62.
Fike, D. A., Grotzinger, J. P., Pratt, L. M. & Summons, R. E. 2006.
Oxidation of the Ediacaran Ocean. Nature, 444, 744–747.
Folling, P. G., Zartman, R. E. & Frimmel, H. E. 2000. A novel
approach to double-spike Pb/Pb dating of carbonate rocks: examples
from Neoproterozoic sequences in southern Africa. Chemical
Geology, 171, 97 –122.
Frimmel, H. E., Klotzli, U. S. & Siegfried, P. R. 1996. New Pb/Pb
single zircon age constraints on the timing of Neoproterozoic glacia-
tion and continental break-up in Namibia. Journal of Geology, 104,
459– 469.
Grotzinger, J. P., Bowring, S. A., Saylor, B. Z. & Kaufman, A. J.
1995. Biostratigraphic and geochronological constraints on early
animal evolution. Science, 270, 598–604.
Halverson, G. P., Hoffman, P. F., Schrag, D. P., Maloof, A. C. &
Rice, A. H. N. 2005. Toward a Neoproterozoic composite carbon-
isotope record. Geological Society of America Bulletin, 117,
1181– 1207.
Hoffman, P. F. & Schrag, D. P. 2002. The snowball Earth hypothesis:
testing the limits of global change. Terra Nova, 14, 129–155.
Hoffman, P. F., Hawkins, D. P., Isachsen, C. E. & Bowring, S. A. 1996.
Precise U–Pb zircon ages for early Damaran magmatism in the
Summas Mountains and Welwitschia Inlier, northern Damara Belt.
Communications of the Geological Survey of Namibia, 11, 47– 53.
Hoffman, P. F., Halverson, G. P., Domack, E. W., Husson, J. M.,
Higgins, J. A. & Schrag, D. P. 2007. Are basal Ediacaran (635
Ma) post-glacial ‘cap dolostones’ diachronous? Earth and Planetary
Science Letters, 258, 114– 131.
Hoffmann, K. H., Condon, D. J., Bowring, S. A. & Crowley, J. L.
2004. U– Pb zircon date from the Neoproterozoic Ghaub Formation,
Namibia: constraints on Marinoan glaciation. Geology, 32, 817– 820.
Horstwood, M. S. A., Foster, G. L., Parrish, R. R., Noble, S. R. &
Nowell, G. M. 2003. Common-Pb corrected in situ U– Pb accessory
mineral geochronology by LA-MC-ICP-MS. Journal of Analytical
Atomic Spectrometry, 18, 837– 846.
Hua, H., Pratt, B. R. & Zhang, L.-Y. 2003. Borings in Cloudina shells:
complex predator– prey dynamics in the terminal Neoproterozoic.
Palaios, 18, 454– 459.
Ireland, T. R. & Williams, I. S. 2003. Considerations in zircon geochro-
nology by SIMS. Reviews in Mineralogy and Geochemistry, 53,
215– 241.
Ireland, T. R., Flottmann, T., Fanning, C. M., Gibson, G. M. &
Preiss, W. V. 1998. Development of the early Paleozoic Pacific
margin of Gondwana from detrital-zircon ages across the Delamerian
orogen. Geology, 26, 243–246.
Jaffey, A. H., Flynn, K. F., Glendenin, L. E., Bentley, W. C. &
Essling, A. M. 1971. Precision measurement of half-lives and
specific activities of 235U and 238U. Physics Reviews, C4, 1889– 1906.
Kendall, B. S., Creaser, R. A., Ross, G. M. & Selby, D. 2004. Con-
straints on the timing of Marinoan ‘Snowball Earth’ glaciation by
187
Re/187Os dating of a Neoproterozoic, post-glacial black shale in
Western Canada. Earth and Planetary Science Letters, 222,
729– 740.
Kendall, B., Creaser, R. A. & Selby, D. 2006. Re– Os geochronology
of postglacial black shales in Australia: constraints on the timing of
‘Sturtian’ glaciation. Geology, 34, 729–732.
Kendall, B., Creaser, R. A. & Selby, D. 2009. 187Re– 187Os Geochro-
nology of Precambrian Organic-Rich Sedimentary Rocks. Geological
Society, London, Special Publications, 326, 85 –107.
Kennedy, M. J., Runnegar, B., Prave, A. R., Hoffmann, K. H. &
Arthur, M. A. 1998. Two or four Neoproterozoic glaciations?
Geology, 26, 1059–1063.
Key, R. M., Liyungu, A. K., Njamu, F. M., Somwe, V., Banda, J.,
Mosley, P. N. & Armstrong, R. A. 2001. The western arm of the
Lufilian Arc in NW Zambia and its potential for copper mineraliz-
ation. Journal of African Earth Sciences, 33, 503.
Kosler, J. & Sylvester, P. J. 2003. Present trends and the future of zircon
in geochronology: laser ablation ICPMS. Zircon, 53, 243– 275.
Krogh, T. E. 1982a. Improved accuracy of U– Pb zircon ages by the
creation of more concordant zircon systems using an air abrasion
technique. Geochimica et Cosmochimica Acta, 46, 637– 649.
Krogh, T. E. 1982b. Improved accuracy of U –Pb zircon dating by
selection of more concordant fractions using a high-gradient mag-
netic separation technique. Geochimica et Cosmochimica Acta, 46,
631– 635.
Le Guerroue, E., Allen, P. A., Cozzi, A., Etienne, J. L. & Fanning,
M. 2006. 50 Myr recovery from the largest negative d13C excursion in
the Ediacaran ocean. Terra Nova, 18, 147– 153.
Li, C. W., Chen, J. Y. & Hua, T. E. 1998. Precambrian sponges with
cellular structures. Science, 279, 879– 882.
Love, G. D., Grosjean, E. et al. 2009. Fossil steroids record the appear-
ance of Demospongiae during the Cryogenian period. Nature, 457,
718– 721.
 USER’S GUIDE TO NEOPROTEROZOIC GEOCHRONOLOGY
Ludwig, K. R. 1980. Calculation of uncertainties of U– Pb isotope data.
Earth and Planetary Science Letters, 46, 212–220.
Ludwig, K. R. 1991. Isoplot – a plotting and regression program for
radiogenic isotope data. USGS Open File Report, 91– 445.
Ludwig, K. R. 1998. On the treatment of concordant uranium –lead ages.
Geochimica et Cosmochimica Acta, 62, 665– 676.
Ludwig, K. R. 2000. Decay constant errors in U– Pb concordia-intercept
ages. Chemical Geology, 166, 315–318.
Ludwig, K. R. 2003. Mathematical-statistical treatment of data and errors
for 230Th/U geochronology. Reviews in Mineralogy and Geochemis-
try, 52, 631– 656.
Lund, K., Aleinikoff, J. N., Evans, K. V. & Fanning, C. M. 2003.
SHRIMP U–Pb geochronology of Neoproterozoic Windermere
Supergroup, central Idaho: implications for rifting of western Lauren-
tia and synchroneity of Sturtian glacial deposits. Geological Society
of America Bulletin, 115, 349–372.
Mahan, K. H., Wernicke, B. P. & Jercinovic, M. J. 2010. Th–U –total
Pb geochronology of authigenic monazite in the Adelaide rift
complex, South Australia, and implications for the age of the type
Sturtian and Marinoan glacial deposits. Earth and Planetary
Science Letters, 289, 76– 86.
Martin, M. W., Grazhdankin, D. V., Bowring, S. A., Evans, D. A. D.,
Fedonkin, M. A. & Kirschvink, J. L. 2000. Age of Neoproterozoic
bilatarian body and trace fossils, White Sea, Russia: implications for
metazoan evolution. Science, 288, 841– 845.
Mattinson, J. M. 2000. Revising the ‘gold standard’ – the uranium decay
constants of Jaffey et al. 1971. EOS, AGU Fall Meeting Supplement
Abstact V61A-02.
Mattinson, J. M. 2005. Zircon U –Pb chemical abrasion (‘CA-TIMS’)
method: combined annealing and multi-step partial dissolution analy-
sis for improved precision and accuracy of zircon ages. Chemical
Geology, 220, 47 –66.
Melezhik, V. A., Fallick, A. E. & Pokrovsky, B. G. 2005. Enigmatic
nature of thick sedimentary carbonates depleted in 13C beyond the
canonical mantle value: the challenges to our understanding of the
terrestrial carbon cycle. Precambrian Research, 137, 131–165.
Melezhik, V. A., Roberts, D., Fallick, A. E. & Gorokhov, I. M. 2008.
The Shuram– Wonoka event recorded in a high-grade metamorphic
terrane: insight from the Scandinavian Caledonides. Geological
Magazine, 145, 161– 172.
Narbonne, G. M. 2005. The Ediacara biota: Neoproterozoic origin of
animals and their ecosystems. Annual Review of Earth and Planetary
Sciences, 33, 421– 442.
Narbonne, G. M. & Gehling, J. G. 2003. Life after snowball: the oldest
complex Ediacaran fossils. Geology, 31, 27– 30.
Peterson, K. J., McPeek, M. A. & Evans, D. A. D. 2005. Tempo and
mode of early animal evolution: inferences from rocks, Hox, and
molecular clocks. Paleobiology, 31, 36– 55.
Rasbury, E. T. & Cole, J. 2009. Directly dating geologic events: U–Pb
dating of carbonates. Reviews of Geophysics, 47, RG3001, doi:
10.1029/2007RG000246.
Rasmussen, B. 2005. Radiometric dating of sedimentary rocks: the appli-
cation of diagenetic xenotime geochronology. Earth-Science
Reviews, 68, 197.
Roden, M. K., Parrish, R. R. & Miller, D. S. 1990. The absolute age of
the Eifelian Tioga Ash Bed, Pennsylvania. Journal of Geology, 98,
282– 285.
Schaefer, B. F. & Burgess, J. M. 2003. Re– Os isotopic age constraints
on deposition in the Neoproterozoic Amadeus Basin: implications for
the ‘Snowball Earth’. Journal of the Geological Society, 160,
825– 828.
Scherer, E., Munker, C. & Mezger, K. 2001. Calibration of the lute-
tium – hafnium clock. Science, 293, 683– 687.
Schmitz, M. D. & Schoene, B. 2007. Derivation of isotope ratios,
errors, and error correlations for U–Pb geochronology using
205
Pb-235U-(233U)-spiked isotope dilution thermal ionization
mass spectrometric data. Geochemistry Geophysics Geosystems, 8,
Q08006, doi: 10.1029/2006GC001492.
Schmitz, M. D., Bowring, S. A. & Ireland, T. R. 2003. Evaluation of
Duluth Complex anorthositic series (AS3) zircon as a U– Pb geochro-
nological standard: new high-precision isotope dilution thermal ion-
ization mass spectrometry results. Geochimica et Cosmochimica
Acta, 67, 3665–3672.
149
Schoene, B., Crowley, J. L., Condon, D. J., Schmitz, M. D. &
Bowring, S. A. 2006. Reassessing the uranium decay constants for
geochronology using ID-TIMS U– Pb data. Geochimica et Cosmochi-
mica Acta, 70, 426– 445.
Selby, D. & Creaser, R. A. 2005. Direct radiometric dating of the
Devonian-Mississippian time-scale boundary using the Re –Os
black shale geochronometer. Geology, 33, 545– 548.
Selby, D., Creaser, R. A., Stein, H. J., Markey, R. J. & Hannah, J. L.
2007. Assessment of the Re-187 decay constant by cross
calibration of Re –Os molybdenite and U– Pb zircon chronometers
in magmatic ore systems. Geochimica et Cosmochimica Acta, 71,
1999–2013.
Shihong, Z., Ganqing, J. & Yigui, H. 2008. The age of the Nantuo
Formation and Nantuo glaciation in South China. Terra Nova, 20,
289– 294.
Silver, L. T. & Duetsch, S. 1963. Uranium – lead isotopic variations in
zircon: a case study. Journal of Geology, 71, 721– 758.
Steiger, R. H. & Jager, E. 1977. Subcommission on geochronology –
convention on use of decay constants in geochronology and
cosmochronology. Earth and Planetary Science Letters, 36,
359– 362.
Stern, R. A. & Amelin, Y. 2003. Assessment of errors in SIMS zircon
U–Pb geochronology using a natural zircon standard and NIST
SRM 610 glass. Chemical Geology, 197, 111–142.
Thompson, M. D. & Bowring, S. A. 2000. Age of the Squantum ‘tillite’,
Boston Basin, Massachusetts: U –Pb zircon constraints on terminal
Neoproterozoic glaciation. American Journal of Science, 300,
630– 655.
Walter, M. R., Veevers, J. J., Calver, C. R., Gorjan, P. & Hill, A. C.
2000. Dating the 840– 544 Ma Neoproterozoic interval by isotopes of
strontium, carbon, and sulfur in seawater, and some interpretative
models. Precambrian Research, 100, 371–433.
Wasserburg, G. J., Jacobsen, S. B., Depaolo, D. J., Mcculloch, M. T.
& Wen, T. 1981. Precise determination of Sm/Nd ratios, Sm and
Nd isotopic abundances in standard solutions. Geochimica et Cosmo-
chimica Acta, 45, 2311– 2323.
Wendt, I. & Carl, C. 1991. The statistical distribution of the mean
squared weighted deviation. Chemical Geology: Isotope Geoscience,
86, 275– 285.
Williams, M. L., Jercinovic, M. J. & Hetherington, C. J. 2007.
Microprobe monazite geochronology: understanding geologic pro-
cesses by integrating composition and chronology. Annual Review
of Earth and Planetary Sciences, 35, 137–175.
Xiao, S. H., Zhang, Y. & Knoll, A. H. 1998. Three-dimensional preser-
vation of algae and animal embryos in a Neoproterozoic phosphorite.
Nature, 391, 553– 558.
Xiao, S. H., Yuan, X. L. & Knoll, A. H. 2000. Eumetazoan
fossils in terminal Proterozoic phosphorites? Proceedings of the
National Academy of Sciences of the United States of America, 97,
13 684– 13 689.
Xiao, S. H., Yuan, X. L., Steiner, M. & Knoll, A. H. 2002. Macroscopic
carbonaceous compressions in a terminal Proterozoic shale: a sys-
tematic reassessment of the Miaohe biota, south China. Journal of
Paleontology, 76, 347– 376.
Xu, B., Xiao, S. et al. 2009. SHRIMP zircon U– Pb age constraints on
Neoproterozoic Quruqtagh diamictites in NW China. Precambrian
Research, 168, 247–258.
Yin, C. Y., Tang, F. et al. 2005. U–Pb zircon age from the base of
the Ediacaran Doushantuo Formation in the Yangtze Gorges,
South China: constraint on the age of Marinoan glaciation. Episodes,
28, 48– 49.
York, D. 1966. Least squares fitting of a straight line. Canadian Journal
of Physics, 44, 1079– 1086.
York, D. 1967. The best isochron. Earth and Planetary Science Letters, 2,
479– 482.
Zhang, S. H., Jiang, G. Q., Zhang, J. M., Song, B., Kennedy, M. J. &
Christie-Blick, N. 2005. U –Pb sensitive high-resolution ion
microprobe ages from the Doushantuo Formation in south China:
constraints on late Neoproterozoic glaciations. Geology, 33,
473– 476.
Zhou, C., Tucker, R., Xiao, S., Peng, Z., Yuan, X. & Chen, Z. 2004.
New constraints on the ages of Neoproterozoic glaciations in south
China. Geology, 32, 437– 440.
 Chapter 10

Modelling the Snowball Earth

YVES GODDÉRIS1*, GUILLAUME LE HIR2 & YANNICK DONNADIEU3

1
LMTG, CNRS-Observatoire Midi-Pyrénées, 14 avenue Edouard Belin, 31400 Toulouse, France
2
IPGP, 4 place Jussieu, 75252 Paris, France
3
LSCE, CEA-CNRS, Orme des Merisiers, 91191Gif-sur-Yvette, France
*Corresponding author (e-mail: godderis@lmtg.obs-mip.fr)

Abstract: We review most of the modelling studies performed to date to understand the initiation and melting of a Snowball Earth, as
well as to describe the glacial environment during the glaciation itself. All the described scenarios explaining the onset of glaciation rely
on a sufficient decrease in the concentrations of atmospheric greenhouse gases (GHGs), typically resulting from the equatorial palaeo-
geography of the late Proterozoic. It is still heavily debated whether or not the oceanic ice cover was thick during the glaciation itself.
However, a consensus has arisen that the most climatically stable scenarios imply the existence of a globally frozen ocean, with a thick ice
cover caused by the flowing of high-latitude sea-ice glaciers towards the equator. Depending on the characteristics of the ice, a thin ice
layer may have persisted along the equator, but this numerical solution is rather fragile. During the snowball event itself, model results
suggest the existence of wet-based continental glaciers. Some parts of the continents may have remained ice-free. From the modelling
perspective, the most significant problem in the snowball hypothesis, particularly in its ‘hard snowball’ version (the most stable numeri-
cally), is the melting phase. With improved modelling, the CO2 threshold required to melt the snowball is much higher than initially
thought, significantly above 0.29 bar. Indeed, because of the very cold conditions prevailing at the surface of the Earth during the
glacial event, the atmosphere becomes vertically isothermal, strongly limiting the efficiency of the greenhouse effect. This melting
problem is further highlighted by geochemical modelling studies that show that weathering of the oceanic crust might be an active
sink of CO2 during the glacial event, limiting the rise in atmospheric CO2. The solution might be found by considering the input
of dark dust from catastrophic volcanic eruptions that would efficiently decrease the albedo of the ice. Finally, modelling studies
also explore the aftermath of the glaciation. The world might have been drier than initially anticipated, resulting in the persistence of
the supergreenhouse effect for at least one million years after the melting phase.

The Snowball Earth hypothesis is an exciting research field
for climate modellers. First, it requires the use of diverse models
to explore the salient aspects of the theory, from global geochem-
ical models to ice sheet and climate models (Table 10.1). Second,
coupling of these models is often required, which results in the
construction of new numerical tools that can be applied to other
palaeoclimate problems. Finally, the modelling of such an
extreme environment leads scientists to push their numerical
models to their physical limits, which is always instructive in
terms of the numerical and physical behaviour of these models.
The first modelling studies on the ice-albedo instability that
underpins the Snowball Earth hypothesis were performed inde-
pendently by William Sellers and Mikhail Budyko (1969), with
a global Energy Balance Model (EBM, or 0D model). With their
respective models, they demonstrated that reducing the current
solar constant by even a few percent drives a rapid transition of
the Earth climate into an ice-covered state, characterized by a
high albedo (0.85) and global mean temperature of –100 8C
(Sellers 1969). Modelling of the nonlinear response of the ice
line as a function of the incoming absorbed energy demonstrates
the existence of three stables states of the Earth climate system:
(1) ice-free (e.g. the warm Cretaceous), (2) partially ice-free
(e.g. the Phanerozoic glaciations and the present-day state of the
Earth system) and (3) globally ice-covered (e.g. the Neoprotero-
zoic Snowball Earth). Historically, the first two of these states
have been the focus of most climatic modelling. However, since
the publication of Hoffman et al. (1998), significant efforts have
been made to understand the third stable state with respect to the
Neoproterozoic glaciations.
Below we present a summary of the results and the main
modelling efforts carried out to date to understand the causes of
Neoproterozoic Snowball glaciations, the environment during gla-
ciation, and the mechanisms able to trigger deglaciation. We will
focus on hypotheses that have been quantified using complex
numerical models – that is, models not restricted to first-order
mass balance calculations, but rather accounting for the physics
of a snowball event. For this reason, we do not attempt to
discuss all the hypotheses proposed to explain multiple equatorial
glaciations.
How to initiate a snowball glaciation?
Thus far, three scenarios have been proposed and numerically
tested to explain the onset of a snowball glaciation. They all rely
on large drops in the partial pressure of greenhouse gases (GHGs).
Scenario 1: It has been suggested that the partial pressure of atmos-
pheric methane might have been quite high during the Proterozoic
(Pavlov et al. 2003), reaching up to 100– 300 ppmv because of
intense activity of methanogenic bacteria in an anoxic ocean.
The progressive oxidation of the atmosphere might have led to
the collapse of the atmospheric methane stock in the Neopro-
terozoic, leaving the atmosphere with a low CO2 content, thus
triggering global glaciations before pCO2 could self-adjust via
the silicate-weathering feedback (Walker et al. 1981). This mech-
anism has not been fully tested to date and requires specific but
unvalidated assumptions: the rise in O2 and subsequent collapse
in CH4 must have occurred during the Neoproterozoic, and the
global glaciation must have followed this event within no more
than a few million years, based on the inferred residence time of
carbon in the ocean – atmosphere system (François & Goddéris
1998). Should this time span be longer, the imbalance between
solid Earth degassing and continental silicate weathering would
force CO2 to rise and compensate for the methane collapse.
The two following scenarios rely on the palaeogeographic
setting of the Neoproterozoic. This is still a matter of intense
debate, but there is now a general consensus about the dominantly

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 151– 161. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.10
Table 10.1. A summary of all modelling studies of the Neoproterozoic ice ages (with type of model employed and reference)

Tools Authors Note

Snowball Earth inception

EBM Rose & Marshall (2009) Ice-albedo instability
EMIC Micheels & Montenari (2008) Slushball conditions
GEO Goddéris & Donnadieu (2008) Comment on Peltier (2007)
AGCM Ishiwatari et al. (2007) Ice-albedo instability
EMIC Lewis et al. (2007) Sea-ice dynamics
AGCM-GEO Goddéris et al. (2007) Continental drift and CO2 decrease
AGCM-GEO Peltier et al. (2007) CO2 scenario for a slushball
AGCM Romanova et al. (2006) Test of the climatic factors
AGCM Pavlov et al. (2005) Impact of interstellar dust
EMIC Donnadieu et al. (2004b) Sea-ice dynamics
EMIC-GEO Donnadieu et al. (2004a) Continental drift and CO2 decrease
EMIC Lewis et al. (2004) Test of the climatic factors
OAGCM Poulsen & Jacob (2004) Sea-ice dynamics
AGCM-EMIC Ramstein et al. (2004) Review
EMIC-GEO Donnadieu et al. (2004c) Continental drift and CO2 decrease
AGCM-ISM Pollard & Kasting (2004) Glacial deposits
OAGCM Peltier et al. (2004) Ice-albedo instability
AGCM Jenkins (2004) Ice-albedo instability
EMIC Stone & Yao (2004) Ice-albedo instability
EBM-GEO Goddéris et al. (2003) Igneous provinces and CO2 decrease
EMIC Lewis et al. (2003) Ocean dynamics
OAGCM Poulsen (2003) Climatic feedbacks
AGCM-ISM Baum & Crowley (2003) Climatic feedbacks
EMIC Bendtsen (2002) Climatic feedbacks
EMIC Bendtsen & Bjerrum (2002) Climatic feedbacks
AGCM Donnadieu et al. (2002) Testing the high obliquity
OAGCM Poulsen et al. (2002) Climatic feedbacks
GEO Schrag et al. (2002) Methane and CO2 decrease
OAGCM Poulsen et al. (2001) Climatic feedbacks
AGCM Baum & Crowley (2001) Climatic feedbacks
EBM-ISM/AGCM Hyde et al. (2000) Slushball solution
AGCM Chandler & Sohl (2000) Climatic feedbacks
AGCM Jenkins (2000) Climatic feedbacks
AGCM Jenkins & Smith (1999) Climatic feedbacks
AGCM Jenkins & Frakes (1998) Climatic feedbacks
EBM Crowley & Baum (1993) Climatic feedbacks
EBM-GEO Marshall et al. (1988) Ice-albedo instability
EBM Budyko (1969) Ice-albedo instability
EBM Sellers (1969) Ice-albedo instability
During the Snowball Earth
GEO Le Hir et al. (2008b) CO2 evolution, carbon cycling
GEO Le Hir et al. (2008a) Seawater composition
EBM þ ice-shelf flow Warren & Brandt (2006) Sea-ice thickness
EBM þ ice-shelf flow Pollard & Kasting (2006) Sea-ice thickness
EBM þ ice-shelf flow Goodman (2006) Comment
EBM þ ice-shelf flow Pollard & Kasting (2005) Sea-ice thickness
GCM-ISM Donnadieu et al. (2003) Continental ice dynamics
Complex 1D ice model Goodman & Pierrehumbert (2003) Sea-ice thickness
Spectral model Warren et al. (2002) Sea-ice thickness
Sea ice model McKay (2000) Sea-ice thickness
Melting the Snowball Earth
AGCM-dust model Le Hir et al. (2010) Explicit effect of volcanic ash on snow/ice albedo
AGCM Abbot & Pierrehumbert (2010) Explicit effect of volcanic ash on snow/ice albedo
AGCM Le Hir et al. (2007) Inversion of the atmospheric vertical thermal gradient
EMIC Lewis et al. (2006) Sensitivity to snow and ice albedo values
AGCM Pierrehumbert (2005) Inversion of the atmospheric vertical thermal gradient
AGCM Pierrehumbert (2004) Inversion of the atmospheric vertical thermal gradient
EBM-ISM Crowley et al. (2001) Near-Snowball Earth
EBM with CO2 ice clouds Caldeira & Kasting (1992) CO2 ice cloud effect
Tools Year/ Authors / Journal Note

Aftermath of the Snowball Earth

GCM-GEO/weathering model Le Hir et al. (2009) Weathering rates, climate, CO2 restoring
GEO Higgins & Schrag (2003) Carbon isotopic signal

Energy balance models (EBMs) only compute the radiative energy budget of the atmosphere. The transport is implicitly included as well as the water
cycle. Radiative convective models (RCMs) include a description of the radiative energy budget of the atmosphere, and of the upward transport of
energy by convection. General circulation models (GCMs) resolve explicitly the transport of energy and matter within the atmosphere (AGCM) and
the ocean (OGCM) on a three-dimensional grid and physically compute the water cycle. Earth system Models of Intermediate Complexity (EMIC)
bridge the gap between the simple RCM, the EBM and the highly complex GCMs. They couple simplified atmospheric and oceanic models and are
generally designed for long-term simulations (104 to 105 years). GEO stands for global biogeochemical cycle models and ISM for ice-sheet model.
 MODELLING SNOWBALL EARTH 153

low-latitude location of most of the continental blocks from

800 Ma to 600 Ma (Evans 2000; Meert & Powell 2001; Torsvik
et al. 2001; Trindade et al. 2003; Macouin et al. 2004; Meert &
Torsvik 2004; Hoffman & Li 2009).

Scenario 2: This scenario (Schrag et al. 2002) was developed to

explain the unusual relationship between the onset of glaciation
and a large negative d13C (carbonate) anomaly that preceded it
(Halverson et al. 2002). High pre-glacial d13C values (.þ5‰)
were assumed to result from a combination of high phosphorous
flux to the oceans and efficient burial of organic carbon in the
deltas of large equatorial watersheds from continents clustered
in the tropics. Analogous present-day deltas include those of the
Amazon and southeastern Asian rivers such as the Irrawady,
Ganges and Salween. However, it is worth noting that the
organic matter being buried in the Amazon and Bengal fans is
today partly of terrestrial origin, which would not have been the
case in the Neoproterozoic. Furthermore, the high sedimentation
rates and resulting efficient burial of organic matter in southeastern
Asia (up to 100% of the amount of organic carbon reaching the
sediment) is very high, not because of the low-latitude location
of these catchments, but because of their source in the Himalayas
(Galy et al. 2007). Thus, it is not clear that Neoproterozoic palaeo-
geography alone could account for the unusually efficient organic
carbon burial.
The subsequent decrease in d13C prior to the onset of snowball
glaciation is interpreted to be the result of a protracted release of
CH4 derived from methanogenesis within the previously buried
organic carbon pool. In the Schrag et al. (2002) model, after avail-
able reservoirs for methane clathrate on continental margins were
filled, CH4 began to leak to the anoxic water column and then to the
atmosphere, where it drove an increase in CH4 partial pressure.
The gradual substitution of CO2 by CH4 then generated a severe
vulnerability in global climate because CH4 has much shorter resi-
dence times than CO2 and is not similarly regulated through sili-
cate weathering. Thus, if this situation persisted for several
hundreds of thousands of years, then analogous to the previous
scenario, a sharp decrease in the CH4 flux would precipitate a
sudden cooling event.
This model for triggering a Snowball Earth has the advantage of
being the only one that can account for the major negative d13C
anomalies preceding Cryogenian glaciations (Halverson et al.
2005; Prave et al. 2009). However, this scenario has not been
thoroughly modelled, and some aspects remain unresolved, such
as what triggered the collapse in the CH4 flux and whether O2
partial pressure was sufficiently low to allow the accumulation
of high concentrations of atmospheric methane.
Scenario 3: The prominent tectonic event of the Neoproterozoic
was the break-up of the Rodinia supercontinent. The breakup
started c. 800 Ma ago, when continental masses were distributed
in the tropics (Li et al. 2003). Donnadieu et al. (2004a) calculated
the climatic and geochemical impact of this break-up, through the
development of a numerical model coupling a 2.5D climate model
(Petoukhov et al. 2000) and a global biogeochemical cycle model
(Goddéris & Joachimski 2004). The use of an explicit climate
model allows the subsequent increase in rainfall and runoff
above the continents to be captured as the breakup proceeds. As
a result, CO2 consumption through continental weathering is
stimulated, driving atmospheric CO2 levels towards the threshold
required to initiate a snowball glaciation (Fig. 10.1). This hypoth-
esis therefore allows atmospheric CO2 to decrease to the threshold
level without invoking a change in CO2 degassing. Donnadieu
et al. (2004a) computed a decrease in atmospheric CO2 from
1830 ppmv down to 510 ppmv during the breakup of Rodinia
and a concomitant drop in global mean annual temperature from
10.8 8C to 2 8C, assuming a solar constant reduced by 6% relative
to the present. Importantly, the drastic cooling produced in this
model is the result of dominantly longitudinal break-up in the
Fig. 10.1. CO2 consumption by continental weathering rate for a
supercontinental (Rodinia) and a dispersed configuration in 104 mol of Ca2þ or
Mg2þ/km2/a, for a fixed 6.5 PAL of CO2. Source: Donnadieu et al. (2004a).
low latitudes, in contrast to the latitudinal break-up of Pangaea,
which did not trigger widespread glaciation.
However, Rodinia break-up alone is not sufficient to initiate a
snowball. Indeed atmospheric CO2 stabilizes above the required
threshold (a value dependent on the model used, see next
section). However, the possible solution lies in the geodynamics
of supercontinental break-up, which is heralded by the onset of
large magmatic provinces as a result of the accumulation of heat
below large continental assemblages (Courtillot et al. 1999) and
subsequent release through mafic eruptions. Specifically, a
mantle superplume, starting at 830 Ma and lasting some 85
million years, is associated with the early stages of Rodinia
break up (Li et al. 1999; Li et al. 2003). Tholeiitic magmatism
has been identified between 825 and 755 Ma, especially in
Australia (Wingate et al. 1998; Wingate & Giddings 2000) and
northwestern Laurentia (Park et al. 1995), but also in south
China (Li et al. 1999) and on the Congo craton (Key et al.
2001). Collectively, these magmatic events appear to define a
plume time-cluster based on the database of Ernst and Buchan
(2002). The most widespread flood basalt event probably occurred
on Laurentia, with a minimum size of 3  106 km2 and possibly
equivalent in area to the original Siberian Traps.
The onset of continental flood basalts results first in a massive
but short-lived release of CO2 into the atmosphere, leading to an
enhanced greenhouse effect. However, a few million years after
the end of the eruptive phase, this CO2 excess is rapidly consumed
by intensified silicate weathering. In about 4 million years, the CO2
level stabilizes at a value lower than the pre-eruption level
because highly weatherable basalt on the surface consumes CO2
(Dessert et al. 2001). The efficiency of this long-term cooling
effect depends on the surface area of the large igneous province
as well as on the precipitation and temperature where the traps
occur (Goddéris et al. 2003). Donnadieu et al. (2004a) have
demonstrated that the weathering of freshly erupted basaltic
surface as a disaggregating Rodinia drifts towards the humid equa-
torial regions is sufficient to trigger a snowball glaciation. Strictly
154 Y. GODDÉRIS ET AL.
speaking, this scenario can only apply to the first snowball glacia-
tion (i.e. Sturtian episode) and does not explicitly account for the
negative d13C anomaly that precedes this glaciation (Prave et al.
2009). Nevertheless, this study indicates that a globally cool
climate is a predictable consequence of the low-latitude break-up
of Rodinia.
The ice and climate of the Snowball Earth
The onset of the snowball: ice-albedo instability
Decrease in atmospheric GHGs will always cool global climate.
However, even with very low GHG levels, is the Snowball Earth
a plausible climatic end-member? A first set of numerical studies
focused on the possibility of sea ice reaching the equator. Simple
energy balance models all display a climatic instability once the
ice line reaches about 308 latitude (Budyko 1969). This runaway
ice-albedo feedback occurs once the surface radiation entering
the open ocean regions cannot balance the heat loss by reflection
of sunlight from low-latitude ice-covered regions, resulting in
rapid ice growth and global cooling. However, initial coupled
ocean – atmosphere general circulation models did not follow
this behaviour prescribed by EBMs. Using the coupled Fast
Ocean Atmosphere Model (FOAM), which accounts for ocean
dynamics, Poulsen (2003) performed a simulation with a unique
atmospheric pCO2 fixed at 140 ppm, a solar constant reduced by
7%, an idealized equatorial supercontinent and a sea-ice thickness
of 1000 m. He forced the ice line to be located at 108 latitude. Once
the prescribed condition for the location of the ice line was released,
it did not continue to the equator but rather retreated to higher lati-
tudes, .408N and S. This result was used to argue that a full snow-
ball climate is climatically untenable (Poulsen et al. 2001; Poulsen
2003). However, Lewis et al. (2004) demonstrated that this instan-
taneous sea-ice retreat was caused by the absence of sea-ice
dynamics and was an artefact of the way the energy required to
melt the sea ice was treated. Using a similar model, Lewis et al.
(2007) then obtained the surprising result that the ocean completely
froze over when pCO2 fell below 3000 ppmv.
The potential of the onset of a full snowball glaciation in a
coupled dynamic model has been confirmed by a study performed
with an Earth Model of Intermediate Complexity (EMIC) that took
into account sea-ice thermodynamics and ocean dynamics, where
two realistic continental configurations were tested (superconti-
nent Rodinia and a disaggregated Rodinia along the equator;
Donnadieu et al. 2004b). Probably the most important conclusion
arising from this study was that climatic instability exists for both
continental configurations, meaning that ocean dynamics cannot
prevent the onset of a Snowball Earth. However, they also found
that the CO2 threshold required to trigger a snowball is (i) highly
dependent on the continental configuration, being higher for the
dispersed continental configuration, and (ii) below the threshold
values calculated with an atmospheric model, confirming the
important role of the ocean dynamics on the onset of a Snowball
Earth, but also emphasizing the role of atmospheric meridional
heat transport. Indeed, the Hadley cells strengthen as the ice
moves equatorward in response to decreasing atmospheric CO2
pressure, bringing more and more heat from the equator to the
ice front, thus slowing its advance. But this strong negative feed-
back collapses once the ice line reaches 308 latitude. Indeed, as
the ice line is now located on the descending branch of the
Hadley cell, the cold air in contact with the ice is now efficiently
transported back towards the equator along the lower meridional
branch of the cell (Bendsten 2002). Air temperatures at the
equator then rapidly decrease to zero, shutting down the climatic
negative feedback, allowing the onset of snowball glaciation.
Heat transport in the ocean is more complex and largely depends
on the continental configuration. Nevertheless, Donnadieu et al.’s
(2004b) key finding is that heat transport during the advance of
the ice and the final collapse into a snowball state is mainly
driven by the atmosphere, and the ocean plays only a minor role.
The snowball CO2 thresholds were computed to be 90 ppmv for
a supercontinental configuration and 150 ppmv for the dispersed
configuration along the equator. These levels are much lower
than initially expected from atmospheric model simulations
(Jenkins & Smith 1999). Similar conclusions regarding the critical
role played by the Hadley transport once the ice line reaches 308
latitude were obtained recently with a general circulation model
(GCM) coupled to a thermodynamic sea-ice model, assuming an
idealized supercontinent located at the equator (Poulsen & Jacob
2004).
During the Snowball Earth: the slushball theory as a solution
to explain the survival of photosynthetic life?
How can climate models that predict a totally ice-capped ocean be
reconciled with palaeontological evidence that establishes the per-
sistence of photosynthetic activity throughout the snowball event?
Hyde et al. (2000) proposed a scenario where the equatorial ocean
remains ice-free during snowball events. This ‘tropical oasis’ sol-
ution is colloquially referred to as the ‘slushball’ hypothesis and
has gained traction amongst geologists who see evidence for
open ocean conditions during the presumed snowball glaciations
(e.g. Leather et al. 2002; Allen & Etienne 2008). This result was
obtained by coupling a simple energy balance model and an ice-
sheet model under a solar constant reduced by 6% and assuming
a very low CO2 level of 140 ppmv. Under such conditions,
massive icecaps develop on continents, as well as thick ice
sheets on the global ocean. Then, the calculated icecaps were
specified as boundary conditions of a GCM, in which the CO2
partial pressure was prescribed at 700 ppmv. The GCM output
showed that the ocean remains free of ice below 258 latitude.
However, this climatic solution is not self-consistent: if the gla-
ciers were allowed to respond to the 700 ppm pCO2, they would
retreat. And if CO2 was reduced to 140 ppmv, the equatorial
ocean would freeze. This slushball simulation is therefore not
robust, because the threshold for the onset of a Snowball Earth
simulated with a GCM is generally far below 700 ppmv (Poulsen
2003; Donnadieu et al. 2004a) and the stability range of the
slushball solution.
An additional argument against the slushball model comes from
sea-ice dynamics. As demonstrated by Lewis et al. (2007), a
sea-ice margin reaching low latitudes without driving a total
collapse of the ocean is only obtained in simulations omitting
the sea-ice dynamic process. In oceanic models including sea-ice
dynamics, the ice-albedo instability feedback does not allow a
stable slushball solution.
During the Snowball Earth: a clear equatorial thin-ice
as a solution for life?
Based on observations of lakes in the Antarctic Dry Valley,
McKay (2000) formulated an alternative scenario to explain the
survival of photosynthesis. Using a simple method to reconstruct
the energy balance of the ice cover, he showed that if the mean
annual temperature at the equator was around –30 8C, then the
equilibrium thickness of the bare sea ice in this area might have
been as little as a few metres, assuming an ice albedo of ,0.75
(McKay 2000). This result was obtained assuming low absorption
efficiency of the ice within the visible part of the solar spectrum, so
that heat can be transferred to the water below the ice, thus limiting
the growth of the ice sheet through seawater freezing. Such a low
absorption coefficient for the ice requires a slow rate of freezing,
which prevents the incorporation of air bubbles and keeps the ice
relatively clear. However, such clear ice is inconsistent with
high albedo, which requires a high bubble density. The calculated
 MODELLING SNOWBALL EARTH
thickness is critically dependent on the ice albedo, exceeding
several tens of metres if the ice albedo is .0.8.
The thin-ice model has been challenged by Warren et al. (2002),
who described the absorption of solar energy as a function of wave-
length and depth, accounting for the scattering of radiation by air
bubbles inside the ice. They found that the absorption of incident
energy within the first centimetres of the ice sheet was much
higher than previously expected (McKay 2000), particularly in
the near-infrared wavelength band (0.7 –3 mm). Furthermore, the
visible part of the spectrum is largely scattered by air bubbles,
brine inclusions and cracks. Heat is thus not efficiently transferred
to the water below the ice. As a consequence, their results sug-
gest that the ocean was everywhere capped by at least several
hundred metres of ice. In Warren et al.’s (2002) model, very low
albedos (,0.4) of the ice would be required to allow a thin-ice
solution, implying a very low air bubble density. Such albedo
values are not realistic under the tropical sun, because it would
warm the ice above the melting point. Another argument supports
the thick-ice hypothesis. The latitudinal air-temperature gradient
in a snowball environment would produce thicker sea ice at
higher latitudes than around the equator. This is not dynamically
stable, so ice would flow from high to low latitudes as sea glaciers
reducing the thickness gradient, but without forming a uniform
thickness everywhere due to slow flow rates (Goodman &
Pierrehumbert 2003).
The question of thick v. thin ice remains open. A recent model-
ling study coupling an EBM with an ice-shelf model demonstrates
the possibility of the existence of areas of thin ice (,3 m thick)
below a latitude of 208 (Pollard & Kasting 2005) (Fig. 10.2).
Climate simulations of the Snowball Earth conditions predict net
positive surface accumulation (precipitation –evaporation) of a
few mm/a in mid to high latitudes. Snow thus accumulates on
the ice, strongly increasing the albedo and allowing the growth
of thick sea ice. But in low latitudes, the net surface accumulation
is negative, the sea ice remains free of snow, and the thin-ice
solution is valid.
Pollard & Kasting (2005) calculated that the snow limit is
located around a latitude of 128. Below this latitude, the sea ice
is a marine ice accumulated through basal freezing, much less
bubbly and clearer than the snow-derived glacial ice. The high-
latitude thick ice (150 m thick) flows from high latitude at a rate
of 800 m/a and eventually crosses the 128 latitude limit.
Because the precipitation –evaporation (P– E) budget becomes
negative, the snow accumulated on this ice starts to sublimate or
melt. As a result, up to 5 W m22 is absorbed by the ice, and internal
melting occurs between latitudes of 128 and 108. Thin ice may thus
be maintained at low latitude, where the albedo can be as low as
0.45, despite the flowing of high-latitude sea-ice glaciers.
Fig. 10.2. Sea-ice thickness during a snowball in the Pollard & Kasting (2005)
model, as a function of latitude. The dashed line indicates bubbly ice, and the
solid line clear ice. Flowing of sea glaciers is accounted for.
155
Furthermore, Pollard & Kasting (2005) argue that, in any case,
low-latitude large lakes and confined seas like the modern Mediter-
ranean, protected from planetary-scale sea-glacier flow by sur-
rounding continents, would maintain areas with thin ice cover.
This model has been challenged by Goodman (2006) and
Warren & Brandt (2006). Goodman (2006) noted that the atmos-
pheric and oceanic water cycle must be internally balanced,
because the glaciation lasts for millions of years, to avoid any con-
centration of the water mass into the atmosphere or the ocean.
Because Pollard & Kasting (2005) calculated that the ice sheet
was formed by freezing of seawater and that net sublimation trans-
fers water from the ice sheet to the atmosphere in the tropics, either
some meteoric ice would have to return water to the ocean (which
did not happen in the model) or else marine ice could not reach the
surface, and all of the ice-sheet surface would consist of meteoric
ice. However, meteoric ice is bubbly and opaque, with a high
albedo, thus counteracting the internal melting process proposed
by Pollard & Kasting (2005). In conclusion, whether the thin-ice
solution can be maintained or not is still a matter of debate. We
can only conclude that very clear thin ice is a fragile solution,
and can potentially be formed only where meteoric ice is absent.
To date, model studies have reconstructed the Snowball Earth as
a hostile environment for photosynthetic marine algae and associ-
ated food chains, with no ice-free equatorial ocean, nor tropical
thin-ice. However, the first description of the snowball theory
suggested the existence of cracks in the sea ice, whether it was
thin or not (Hoffman et al. 1998). In a cold environment, the
equatorial sea ice would form cracks if its speed exceeds
50 m/a. Newly formed sea ice in cracks will contain brine chan-
nels, which today host a variety of organisms including photo-
trophs. Moreover, if the ice was thin, winds and tidal forces
were probably strong enough to generate leads (Pollard &
Kasting 2005). Also, active hydrothermal areas in shallow water
should maintain open waters in their direct vicinity. Hence the
existence of several types of refugia required to explain the per-
sistence of photosynthesis cannot be ruled out from numerical
simulations alone. Indeed, large populations of microscopic
organisms can populate a small volume in which light remains
available. They can preserve a relative diversity even if the remain-
ing oasis is small (Knoll 2003).
During the Snowball Earth: glacial deposits and continental
ice behaviour
The behaviour of the terrestrial glacial regime during the Neopro-
terozoic glaciations is still a matter of debate. Some Snowball Earth
detractors claim that the glacial sequences cannot be explained
with the Snowball Earth scenario. Indeed, the near shutdown of
the hydrological cycle simulated by climatic models, once the
Earth is entirely glaciated, stands in contrast to the requirements
for active, wet-based continental ice sheets to produce the
observed thick glacial deposits. The extent and dynamics of the
continental ice has been estimated through the forcing of an ice-
sheet model with the output of a GCM in snowball conditions
(Donnadieu et al. 2003). The GCM (LMDz) was run assuming
1 PAL (preindustrial atmospheric level, or 280 ppmv) of CO2, a
solar constant reduced by 6%, and a realistic palaeogeographic
configuration including an estimate of the continental relief
(Greenvillian-aged mountains, with a prescribed elevation of c.
2000 m, relics of the last major orogenic cycle). With those for-
cings, the ice line rapidly reaches 308 latitude, reaching the
instability threshold and resulting in the onset of a snowball
state. The ice-sheet model (Ritz et al. 2001) is then run using the
climate fields generated by the GCM. The three main conclusions
of this study were as follows. First, thick continental ice-sheets
build up within a few hundred thousand years, with a mean of
2500 m and a maximum thickness of 5000 m above Laurentia
and Antarctica. The ice sheets nucleate on high relief and in less
156 Y. GODDÉRIS ET AL.
than 50 ka spread across the continental interior. Second, the
model predicts a final relative fall in sea level of 200 m.
However, the drop in sea level is highly dependent on several com-
peting phenomena. Glacioeustasy is the dominant control; for
example, with an average thickness of ice on all continents of
2000 m, the fall will be c. 500 m, assuming an emerged continental
surface of one-quarter the area of the oceans. However,
glacio-isostatic rebound, hydro-isostatic rise of the seafloor (due
to a reduction in water load) and the tectonic subsidence balance
the initial glacio-eustatic fall, reducing the relatively low sealevel
fall despite a very significant land ice volume. Performing similar
GCM – ice sheet model simulations, but without prescribing any
relief on the continents, Pollard & Kasting (2004) also show that
thick ice sheets build up on continents in the tropics, but the
location of the foci of the ice sheets is determined by the dynami-
cally controlled regions of positive P– E.
Third a large area of the calculated continental ice sheet is wet-
based, so basal sliding is expected to occur. It is particularly active
in the equatorial area where the calculated sliding speed can reach
several tens of m/a. If the sea ice in the equatorial area was thin
(see above discussion), such sliding of the ice sheets may
explain the main features of glacial deposits accumulated during
the Snowball Earth event (Donnadieu et al. 2003). Some parts of
the continents remain ice-free, even after the ice sheet reaches a
steady-state volume after 400 ka in the ice-sheet model. Such
ice-free areas may act as dust sources, darkening the snow over
large surfaces through aeolian deposition, helping to maintain
areas of thin ice on the ocean (Pollard & Kasting 2005).
Melting the Snowball Earth
The commonly accepted scenario for the melting of the Snowball
Earth is from Kirschvink (1992) and Hoffman et al. (1998).
However, the solution for escape from a snowball was first pro-
posed by Walker et al. (1981) in their seminal paper on silicate
weathering and the long-term stabilization of global temperatures.
Before the publication of the Snowball Earth hypothesis, they had
already proposed a solution to the melting problem:
‘If global glaciations were to occur, the rate of silicate weathering should fall
very nearly to zero, and carbon dioxide should accumulate in the atmosphere
at whatever rate it is released from volcanoes. Even the present rate of release
would yield 1 bar of carbon dioxide in only 20 million years. The resultant
large greenhouse effect should melt the ice cover in a geological short
period of time.’
This general prescription for the recovery from a snowball
remains essentially valid. Indeed, the snowball hypothesis funda-
mentally relies on the strong reduction of the carbon sink during
the snowball. Organic carbon burial was certainly maintained at
a very low level during the event, because most of the ocean was
covered by thick ice. Also, silicate weathering was strongly inhib-
ited by a drastic decrease in the continental water cycle and the low
temperatures. As a result, in the absence of rains to scrub the atmos-
phere of CO2, the carbon originating from the degassing of the solid
Earth should accumulate in the atmosphere to very high levels, until
a threshold for deglaciation is reached. Caldeira & Kasting (1992)
calculated the oft-cited figure of 450 times the present-day (i.e.
1992) CO2 level (or 0.126 bar) as the threshold where CO2 levels
counteract the strong ice-albedo feedback and initiate melting. At
the present-day degassing rate of 6.8  1012 mol/a (a value
chosen to balance the present-day continental silicate weathering;
Gaillardet et al. 1999), it would take only 3.5 million years to
reach this threshold. However, as in all facets of modelling of the
Snowball Earth, the situation is more complicated.
The melting problem: the climatic modelling argument
The threshold value of 0.126 bar CO2 for initiating snowball
melting was calculated with an EBM assuming the present-day
insolation. This threshold rises to 0.29 bar with the same model
if a realistic decrease of 6% in the solar constant is used. It would
then take 8 million years to melt the hard snowball, still assuming
the present-day degassing rate, without accounting for possible
dissolution of carbon into the ocean (Higgins & Schrag 2003),
which would increase the time required to accumulate threshold
CO2 levels. Furthermore, an energy balance model is probably
not the best tool with which to explore high CO2 climate, because
too many parameterizations used in EBMs (such as cloud
forcing) are calibrated on the present climate. Working with an
atmospheric GCM (the FOAM GCM) and increasing CO2 partial
pressure in a fully glaciated world, Pierrehumbert (2004) showed
that 0.29 bar is strictly a minimum value for the threshold, and
most likely it is much higher. Through an extrapolation of the
CO2 radiative forcing function, he showed that 2 bar of CO2
might have been required to melt the snowball, corresponding to
an unrealistic glacial duration of .50 million years.
These simulations were carried out assuming an idealized
rectangular supercontinent and present-day orbital parameters.
The greenhouse efficiency G can be defined as the difference
between the amount of energy emitted by the Earth surface and
the amount of energy lost as long-wave radiation at the top of
the atmosphere, the outgoing long-wave radiation (OLR):
G ¼ sTS4  OLR (1)
where Ts is the average surface temperature. OLR is a function
of the effective planetary temperature at the top of the atmosphere,
Te, and is equal to sTe4 (for the modern Earth, Te ¼ 255 K). Today,
the atmosphere is not isothermal, and the effective temperature is
lower than the temperature at the surface. Consequently G is posi-
tive. During the snowball event, in the winter hemisphere, the
atmosphere is isothermal with height, because of inefficient solar
heating of the surface (high albedo) and convection due to
absence of a warm ocean. The air-temperature profile as a function
of height even displays an inversion at ground level, due to the
extremely cold conditions at the Earth surface. As a result, Te is
almost equal to Ts, and the greenhouse efficiency approaches
0. Adding CO2 to the system will thus not efficiently warm the
Earth, as long as the temperature profile remains isothermal. The
outgoing long-wave radiation might even exceed the black-body
radiation at the Earth surface, and the net greenhouse effect is
slightly negative for the winter hemisphere. In the summer hemi-
sphere, the low tropopause limits the vertical temperature gradient,
reducing the efficiency of the greenhouse effect, which is com-
pounded by the virtual lack of water vapour feedback at such
cold temperatures. Furthermore, the cloud greenhouse effect is
reduced, because the high-altitude clouds (contributing the most
to the greenhouse effect) are thin or absent because of the very
low water vapour concentration in the cold upper troposphere in
the summer hemisphere, and at any level in the winter hemisphere
(Pierrehumbert 2004).
A solution to this problem may be found in reduction of the
albedo. Decreasing the albedo would increase the amount of
solar flux absorbed at the surface and promote the onset of a
normal vertical temperature profile of the atmosphere in the
winter hemisphere and increase the efficiency of the greenhouse
effect. Causes of the reduction in albedo are not yet defined, but
the thin-ice solution or the presence of dirty snow generated by
volcanic eruptions or windblown dust from deserts are reasonable
mechanisms (Pierrehumbert 2004).
Expanding on the work of Pierrehumbert (2004, 2005), Le Hir
et al. (2010) explored the impact of a large volcanic eruption in
the final stage of the snowball glacial event when atmospheric
CO2 had already reach 0.2 bar. Huge eruptions, similar to the
Toba eruption at 74 ka, occur once or twice per million years
and release up to 2  1016 g of dust into the atmosphere. Given
the expected duration of the glacial event (close to 10 million
years), it seems plausible that a major volcanic eruption occurred
 MODELLING SNOWBALL EARTH
during the final glacial stage. A GCM (LMD) was used that included
an improved calculation of the albedo of the snow, including
the effect of aging of the snow and dust input. Le Hir et al. (2010)
showed that the sudden release of 2  1016 g of dust into the atmos-
phere decreases the albedo of the surface along the equator by about
0.2 units, but actually increases the albedo by about 0.1 units
between 208 and 408 latitude. The decrease in albedo along
the equator is due to dust accumulation in an area where net ablation
of the snow cover is predicted. Temperature rises along the equator,
activating Hadley cell circulation and leading to an increase in
fresh snow precipitation above 208 latitude. Dust is buried
under the fresh snow and albedo increases. Overall, the mean
annual temperature between latitudes of –10 and þ108 rises by
about 6 8C, reaching values as high as –2 8C. Although a volcanic
eruption does not seem to be sufficient to generate the dramatic
melting of the Snowball Earth, dust input does seem to be a plausible
mechanism to reduce the required CO2 threshold.
The melting problem: the geochemical
modelling argument
Problems also arise from geochemical modelling. Using a model
describing the global carbon and alkalinity cycles during the
Snowball Earth event, Le Hir et al. (2008b) calculated that an
open water area of 3000 km2 during glaciation is large enough to
allow efficient diffusion of atmospheric carbon into the ocean
such that the atmosphere and the ocean are in equilibrium with
respect to carbon at the million-year timescale. Present-day
emerged geothermal environments occupy a surface with an area
of c. 1.5  106 km2 (Dessert et al. 2003). Hence, during global gla-
ciations, some small open-water areas, easily in excess of
3000 km2, may be maintained by heat production and lava flows.
For open water to be maintained, two conditions must be fulfilled.
First, the hydrothermal zone must not be too deep, otherwise their
heat would be diffused over an area so large that it cannot melt the
overlying sea ice. Second, the hydrothermal zone must not be too
shallow on the continental shelf where the heat flux would only
melt the sea ice in the immediate vicinity, resulting in a lake
above the thick sea ice which may be disconnected from the
global ocean under the thick sea ice (Goodman 2006).
If only 2% of geothermal environments had fulfilled these con-
ditions, then massive dissolution of CO2 should have occurred,
acidifying the ocean. Even assuming the existence of a buffering
mechanism through seafloor carbonate dissolution, the pH of the
water drops to about 6 within 3 to 4 million years after the onset
of the Snowball glaciation, with atmospheric CO2 levels reaching
more than 0.1 bar (Le Hir et al. 2008b). Such acidification of the
waters enhances the dissolution rate of the oceanic basaltic crust
by a factor of more than 4, even assuming a glacial thermohaline
circulation fixed at 1% of its present-day value. Seafloor weather-
ing consumes exospheric carbon, trapping it as carbonates in the
veins of the oceanic crust (Alt & Teagle 1999). This means that
all carbon sinks are not drastically reduced during the glacial
event, thus partly counteracting the rise in atmospheric CO2
thought to be required to melt the snowball. Le Hir et al.
(2008b) calculated that within 30 million years (thought to be
the maximum duration of a snowball event), atmospheric CO2
cannot rise above 0.24 bar, a level below the predicted melting
threshold determined by Pierrehumbert (2004). Thus, these
results further exacerbate the problem of whether a snowball
could actually be melted (Fig. 10.3).
Aftermath of the Snowball Earth
Few studies have dealt with the global environment and climate
during the melting of a snowball. The extreme greenhouse atmos-
phere expected to be typical of the melting phase is not well
157
Fig. 10.3. Snowball duration as a function of the CO2 threshold needed to
initiate melting. The stars correspond to estimated levels with various models
and boundary conditions (stars), assuming accumulation of CO2 in the
atmosphere, and no diffusion in the ocean. S is the solar constant normalized to
the present-day value. The triangle indicates a simulation where atmospheric
CO2 is allowed to dissolve into the ocean through cracks, assuming the 0.29 bar
threshold (EBM simulation with a solar constant reduced by 6%). The square
further assumes consumption of CO2 by seafloor weathering, buffering the rise
in CO2 and hence delaying the melting.
understood, and no investigations have been performed with
complex numerical models to date. Three studies explore the geo-
chemical global cycles in the aftermath of the snowball. Two of
them are extremely simple (box models) (Higgins & Schrag
2003; Le Hir et al. 2008a). In both cases, the principle difficulty
in simulating the post-glacial environment arises from the quanti-
fication of the weathering rates of the continental surfaces under a
very warm climate. Weathering during the melting phase is crucial
for at least two reasons (Higgins & Schrag 2003): (i) cap-carbonate
accumulations are thought to be the consequence of the transfer of
alkalinity to the ocean from the extremely fast dissolution of con-
tinental carbonates exposed to acid rain under several hundred
times the present-day CO2 atmospheric pressure, and (ii) the rate
at which continental silicate rocks dissolve controls the time
span needed to restore normal climatic conditions, because silicate
weathering consumes atmospheric CO2 and stores it as carbonate
sediments on the seafloor. However, weathering is a function of
air temperature and continental runoff. It can be safely assumed
that temperatures are high under 0.3 bar of atmospheric CO2
once the melting is completed, but global precipitation patterns
in such a climate are not easy to predict. Lee Kump (personal com-
munication, cited by Higgins & Schrag 2003) suggested a
maximum sevenfold increase in continental runoff compared to
the present-day global value. However, a recent extensive numeri-
cal study in which a GCM was coupled to a transport-reactive
model simulating weathering rates demonstrated that such an
increase is not realistic (Le Hir et al. 2009). Indeed, the total
amount of energy that can be used to evaporate waters from the
ocean and to produce continental rainfall in a supergreenhouse
environment (post-snowball) is limited by the total incoming
solar energy. Le Hir et al. (2009) showed that above 0.11 bar of
CO2, all incoming solar radiation is used as latent heat. Accounting
for a realistic geographical distribution of the continents and an
explicit modelling of the water cycle, the model predicts that the
continental runoff cannot exceed 1.2 times the present-day
runoff under more than 0.11 bar of CO2.
Nevertheless, assuming a sevenfold increase in the continental
runoff, Higgins & Schrag (2003) suggested that continental weath-
ering (carbonate þ silicate rocks) may have increased by a factor
of 50 directly after melting, relative to present-day values. As a
158 Y. GODDÉRIS ET AL.
result, normal values of atmospheric CO2 (around several hundred
ppmv) would be restored in 200 ka, starting from 0.126 bar. In
their study, Le Hir et al. (2009) found that CO2 consumption
through continental weathering cannot exceed 10 times the
present-day value, even accounting for the high surface area of
rock flour covering the continents, because weathering rates are
inhibited by the moderate increase in precipitation. Consequently,
it takes at least one million years for atmospheric CO2 to return to
pre-glacial values in response to silicate weathering in the
post-snowball environment.
One of the most striking features of the post-snowball environ-
ment is the accumulation of up 18.5 m (on average) of cap dolo-
stones on continental shelves, characterized by very low d13C
(c. –4‰, Pee Dee Belemnite Standard (PDB)). Heretofore, mod-
elling this peculiar phase of the snowball sequence has proven
challenging, because modelling the melting of huge icecaps
under very high CO2 levels is not straightforward. For that
reason, no explicit modelling has been performed and significant
simplification is required. Regarding the d13C signature of cap car-
bonates, Higgins & Schrag (2003) calculated that the nadir of the
negative d13C excursion is observed 10 ka after the resumption of
continental weathering. This decrease in d13C directly after
melting initiation is linked to the decrease in the carbon isotopic
fractionation between CO2 and carbonates, itself forced by a pre-
scribed rise in sea surface temperature of 30 8C. In their model,
with a very high prescribed weathering flux, cap-dolostone
accumulation occurs in a few ka. In a similar study, coupling a
radiative –convective climate model to a global biogeochemical
cycle model, Le Hir et al. (2008a) found that the most negative
d13C value was attained 100 ka after the resumption of continental
weathering, suggesting a longer time span for cap-dolostone
accumulation. This longer time span was also an outcome of the
more extensive modelling in Le Hir et al. (2009). Accounting for
the total amount of water released by the melting of the huge
icecap (200  103 km3; Donnadieu et al. 2003), and estimating
the total flux of Mg2þ and Ca2þ released by weathering during
the transgressive phase with a reactive-transport model, Le Hir
et al. (2009) calculated that only 50 cm of pure dolostone would
accumulate on the shelves in the first 10 ka, far below the estimated
worldwide average thickness of 18.5 m (Hoffman et al. 2007).
This slow rate of accumulation suggests that cap-dolostone depo-
sition may have spanned longer than 10 ka, collaborating argu-
ments for a longer timescale for cap-dolostone deposition based
on the existence of multiple magnetic reversal events measured
in various cap dolostones (Trindade et al. 2003; Raub & Evans
2006). However, a corollary of this conclusion is that, because
the cap dolostone was strictly deposited during the post-glacial
melting phase (Hoffman et al. 2007), it requires that deglaciation
was also protracted, and it is not clear if this is physically feasible.
Therefore, the timescale of melting and cap-dolostone precipi-
tation remains an unresolved question, awaiting careful modelling
of the melting phase.
Discussion and conclusions
Do the climate models teach us something about the causes and
consequences of the Neoproterozoic climatic events? The answer
is surely yes if the models have a physical basis. They allow us
to explore the space of parameters of the snowball glacial
events, the variability of these parameters being fixed by field
studies or by the known physical behaviour of the climate. The
answer is no if they are mere mathematical manipulations per-
formed without heed to realistic physical parameters and geologi-
cal constraints. The most important results of the modelling studies
can be summarized as follows:
(1) Onset of snowball glaciations. At least two Cryogenian gla-
ciations were global in extent (Macdonald et al. 2010) and
preceded by negative d13C anomalies (Prave et al. 2009),
supporting the idea that snowball events are based on
major perturbations to the carbon cycle, most likely very
low pCO2. The general tectonic context of the Neoprotero-
zoic is probably the main precondition for the onset of
global glaciation, facilitating atmospheric CO2 pumping by
continental weathering. The break-up of the Rodinia super-
continent triggered an increase in continental weathering
through enhanced humidity above continental surfaces that
led to a generally cool climate during the Neoproterozoic.
As long as the continents were dispersed along the equator,
climate remained cool. Within this cool context, additional
triggers, such as the weathering of magmatic provinces
erupted during the initial phase of the break-up, or atmos-
pheric methane collapse, were required to initiate snowball
glaciations.
(2) During the snowball. First, sea ice is allowed to reach the
equator once the CO2 threshold for initiating a runaway
ice-albedo feedback is reached. Whether this sea ice was
thick or thin is still debatable, but the thin-ice solution
cannot be excluded, thus potentially allowing a vast refuge
for primary producers. Like modern ice sheets, snowball
ice sheets on the continents grew to several kilometres in
thickness and were wet-based, consistent with the thick
accumulation of glacial deposits during the glaciations.
Finally, massive dissolution of atmospheric CO2 into the
ocean occurred if even a small surface area of the ocean
(3000 km2) remained open. In this scenario the ocean may
have been largely acidified during the snowball events,
with a pH as low 6, promoting the increase in the consump-
tion of carbon by the oceanic crust by a factor of 4. Hence,
with such an efficient carbon sink, the CO2 accumulation
would not be linear, but asymptotic, approaching around
0.25 bar of atmospheric CO2.
(3) Melting the snowball. Many unresolved questions remain
concerning snowball modelling, but the most important of
them concerns the initiation of melting. A significant con-
clusion of the modelling studies is that a very high CO2
level is required to melt the snowball, probably .0.29 bar.
This level might never be reached, because carbon is con-
sumed by the enhanced dissolution of the basaltic oceanic
crust. Efforts should focus on this problem, exploring pro-
cesses that might reduce the albedo of the Earth and enable
melting at lower CO2 levels.
(4) In the aftermath of the snowball glaciations. The alkalinity
required to accumulate the cap dolostone results from the
weathering of continental carbonate during the supergreen-
house following the snowball event. This post-snowball
environment needs to be explored in more detail. It seems
that the most prominent and rapid environmental pertur-
bations occurred during the onset and subsequent relaxation
of the supergreenhouse in the direct aftermath of the snow-
ball. Modelling studies contradict earlier assertions of a
highly vigorous hydrological cycle during the supergreen-
house event, with runoff only as much as 20% higher com-
pared to its present-day value. Consequently, weathering
rates were also lower, and rather than a few hundred thousand
years (Hoffman et al. 1998; Higgins & Schrag 2003), it most
likely took .1  106 years for atmospheric CO2 to return
to pre-glacial values. These results also suggest that cap-
dolostone accumulation endured on the order of 100 ka.
If the conditions for initiation of the Snowball Earth seem to be
reasonably well understood (Table 10.1), much work remains to
resolve the melting and climate throughout the deglaciation. Cor-
rectly simulating the duration of deglaciation with fully coupled
climate and ice-sheet models could provide important constraints
on global cap-dolostone precipitation. However, nature is more
complex than models, and it is unrealistic to expect that models
 MODELLING SNOWBALL EARTH
will ever capture the full complexity of the Neoproterozoic ice
ages. Nevertheless, models provide important clues and some
quantitative constraints on the controversial question of whether
these catastrophic events really did occur.
This represents a contribution of the IUGS- and UNESCO-funded IGCP (Inter-
national Geoscience Programme) Project #512.
References
Abbott, D. S. & Pierrehumbert, R. T. 2010. Mudball: surface dust and
snowball Earth deglaciation. Journal of Geophysical Research, 115,
doi: 10.1029/2009JD012007.
Allen, P. A. & Etienne, J. L. 2008. Sedimentary challenge to Snowball
Earth. Nature Geoscience, 1, 817– 825.
Alt, J. C. & Teagle, D. A. H. 1999. The uptake of carbon during the
alteration of ocean crust. Geochimica Cosmochimica Acta, 63,
1527– 1535.
Baum, S. K. & Crowley, T. J. 2001. GCM response to late Precambrian
(590 Ma) ice-covered continents. Geophysical Research Letters,
28, doi: 10.1029/2000GL011557.
Baum, S. K. & Crowley, T. J. 2003. The snow/ice instability as a mech-
anism for rapid climate change: a Neoproterozoic snowball Earth
model example. Geophysical Research Letters, 30, doi: 10.1029/
2003GL017333.
Bendtsen, J. 2002. Climate sensitivity to changes in the solar insolation
in a simple coupled climate model. Climate Dynamics, 18, 595– 609.
Bendtsen, J. & Bjerrum, C. J. 2002. Vulnerability of climate on Earth
to sudden changes in insolation. Geophysical Research Letters, 29,
doi: 10.1029/2002GL014829.
Budyko, M. I. 1969. The effect of solar radiation variations on the climate
of the Earth. Tellus, 21, 611–619.
Caldeira, K. & Kasting, J. F. 1992. Susceptibility of the early Earth to
irreversible glaciation caused by carbon dioxide clouds. Nature, 359,
226– 228.
Chandler, M. A. & Sohl, L. E. 2000. Climate forcings and the initiation
of low-latitude ice sheets during the Neoproterozoic Varanger glacial
interval. Journal of Geophysical Research – Atmospheres, 105,
20 737– 20 756.
Condon, D., Zhu, M., Bowring, S., Wang, W., Yang, A. & Jin, Y. 2005.
U –Pb ages from the Neoproterozoic Doushantuo Formation, China.
Science, 308, 95 – 98.
Courtillot, V., Jaupart, C., Manighetti, I., Tapponnier, P. &
Besse, J. 1999. On causal links between flood basalts and continental
breakup. Earth and Planetary Science Letters, 166, 177–195.
Crowley, T. J. & Baum, S. K. 1993. Effect of decreased solar luminosity
on Late Precambrian ice extent. Journal of Geophysical Research,
98, 16 723– 16 732.
Crowley, T. J., Hyde, W. T. & Peltier, W. R. 2001. CO2 levels required
for deglaciation of a ‘Near-Snowball’ Earth. Geophysical Research
Letters, 28, 283– 286.
Dessert, C., Dupré, B., François, L. M., Schott, J., Gaillardet, J.,
Chakrapani, G. J. & Bajpai, S. 2001. Erosion of Deccan Traps
determined by river geochemistry: impact on the global climate and
the 87Sr/86Sr ratio of seawater. Earth and Planetary Science
Letters, 188, 459– 474.
Dessert, C., Dupré, B., Gaillardet, J., François, L. M. & Allègre,
C. J. 2003. Basalt weathering laws and the impact of basalt weather-
ing on the global carbon cycle. Chemical Geology, 202, 257– 273.
Donnadieu, Y., Ramstein, G., Fluteau, F., Besse, J. & Meert, J. 2002.
Is high obliquity a plausible cause for Neoproterozoic glaciations?
Geophysical Research Letters, 29, doi: 10.1029/2002GL015902.
Donnadieu, Y., Fluteau, F., Ramstein, G., Ritz, C. & Besse, J. 2003.
Is there a conflict between the Neoproterozoic glacial deposits and
the snowball Earth interpretation: an improved understanding with
numerical modeling. Earth and Planetary Science Letters, 208,
101– 112.
Donnadieu, Y., Goddéris, Y., Ramstein, G., Nédelec, A. & Meert,
J. G. 2004a. Snowball Earth triggered by continental break-up
through changes in runoff. Nature, 428, 303– 306.
159
Donnadieu, Y., Ramstein, G., Fluteau, F., Roche, D. & Ganopolski,
A. 2004b. The impact of atmospheric and oceanic heat transport on
the sea ice-albedo instability during the Neoproterozoic. Climate
Dynamics, 22, 293– 306.
Donnadieu, Y., Ramstein, G., Goddéris, Y. & Fluteau, F. 2004c.
Global tectonic setting and climate of the Late Neoproterozoic: a
climate–geochemical coupled study. In: Jenkins, G. S., McMena-
min, M. A. S., McKay, C. P. & Sohl, L. (eds) The Extreme Protero-
zoic: Geology, Geochemistry, and Climate. Geophysical Monograph
146. American Geophysical Union, Washington, DC, 79 – 89.
Ernst, R. E. & Buchan, K. L. 2002. Maximum size and distribution
in time and space of mantle plumes: evidence from large igneous
provinces. Journal of Geodynamics, 34, 309–342.
Evans, D. A. D. 2000. Stratigraphic, geochronological, and paleomagnetic
constraints upon the Neoproterozoic climatic paradox. American
Journal of Science, 300, 347–433.
François, L. M. & Goddéris, Y. 1998. Isotopic constraints on
the Cenozoic evolution of the carbon cycle. Chemical Geology,
145, 177–212.
Gaillardet, J., Dupré, B., Louvat, P. & Allègre, C. J. 1999. Global
silicate weathering and CO2 consumption rates deduced from the
chemistry of the large rivers. Chemical Geology, 159, 3– 30.
Galy, V., France-Lanord, C., Beyssac, O., Faure, P., Kudrass, H. &
Palhol, F. 2007. Efficient organic carbon burial in the Bengal fan
sustained by the Himalayan erosional system. Nature, 450, 407– 410.
Goddéris, Y. & Joachimski, M. M. 2004. Global change in the late
Devonian: modelling the Frasnian–Famennian short-term carbon
isotope isotope excursions. Palaeogeography, Palaeoclimatology,
Palaeoecology, 202, 309– 329.
Goddéris, Y. & Donnadieu, Y. 2008. Carbon cycling and snowball
Earth. Nature, 456, doi: 10.1038/nature07653.
Goddéris, Y., Donnadieu, Y. et al. 2003. The Sturtian glaciation: fire
and ice. Earth and Planetary Science Letters, 211, 1 –12.
Goddéris, Y., Donnadieu, Y. et al. 2007. Coupled modeling of global
carbon cycle and climate in the Neoproterozoic: links between
Rodinia breakup and major glaciation. Comptes Rendus Geoscience,
339, 212–222.
Goodman, J. C. 2006. Through thick and thin: marine and meteoric ice
in a ‘Snowball Earth’ climate. Geophysical Research Letters, 33,
doi: 10.1029/2006GL026840.
Goodman, J. C. & Pierrehumbert, R. T. 2003. Glacial flow of floating
marine ice in ‘Snowball Earth’. Journal of Geophysical Research,
108, doi: 10.1029/2002JC001471.
Halverson, G. P., Hoffman, P. F., Schrag, D. P. & Kaufman, A. J.
2002. A major perturbation of the carbon cycle before the Ghaub
glaciation (Neoproterozoic) in Namibia: prelude to snowball
Earth? Geochemistry, Geophysics, Geosystems, 3, doi: 10.1029/
2001GC000244.
Halverson, G. P., Hoffman, P. F., Schrag, D. P., Maloof, A. C. &
Rice, A. H. N. 2005. Towards a Neoproterozoic composite carbon
isotope record. Geological Society of America Bulletin, 117,
1181–1207.
Higgins, J. A. & Schrag, D. P. 2003. Aftermath of a snowball Earth. Geo-
chemistry Geophysics Geosystems, 4, doi: 10.1029/2002GC000403.
Hoffman, P. F. & Li, Z.-X. 2009. A palaeogeographic context for Neopro-
terozoic glaciation. Palaeogeography, Palaeoclimatology, Palaeo-
ecology, 277, 158– 172.
Hoffman, P. F., Kaufman, A. J., Halverson, G. P. & Schrag, D. P.
1998. A Neoproterozoic Snowball Earth. Science, 281, 1342– 1346.
Hoffman, P. F., Halverson, G. P., Domack, E. W., Husson, J. M.,
Higgins, J. A. & Schrag, D. P. 2007. Are basal Ediacaran (635
Ma) post-glacial ‘cap dolostones’ diachronous? Earth and Planetary
Science Letters, 258, 114–131.
Hyde, W. T., Crowley, T. J., Baum, S. K. & Peltier, W. R. 2000.
Neoproterozoic ‘snowball Earth’ simulations with a coupled
climate/ice sheet model. Nature, 405, 425– 429.
Ishiwatari, M., Nakajima, K., Takehiro, S. & Hayashi, Y.-Y. 2007.
Dependence of climate states of gray solar atmosphere on solar con-
stant: from the runaway greenhouse to the snowball states. Journal of
Geophysical Research, 112, doi: 1029/2006JD007368.
Jenkins, G. S. 2000. Global climate model high-obliquity solutions to
the ancient climate puzzles of the Faint-Young Sun Paradox and
160 Y. GODDÉRIS ET AL.
low-altitude Proterozoic Glaciation. Journal of Geophysical
Research, 105, 7357– 7370.
Jenkins, G. S. 2004. High obliquity as an alternative hypothesis to early
and late Proterozoic extreme climate conditions. In: Jenkins, G. S.,
McMenamin, M. A. S., McKay, C. P. & Sohl, L. (eds) The
Extreme Proterozoic: Geology, Geochemistry, and Climate. Geophy-
sical Monograph 146. American Geophysical Union, Washington,
DC, 73– 78.
Jenkins, G. S. & Frakes, L. A. 1998. GCM sensitivity test using increased
rotation rate, reduced solar forcing and orography to examine low
latitude glaciation in the Neoproterozoic. Geophysical Research
Letters, 25, 3525–3528.
Jenkins, G. S. & Smith, S. R. 1999. GCM simulations of snowball Earth
conditions during the Late Proterozoic. Geophysical Research
Letters, 26, 2263–2266.
Key, R. M., Liyungu, A. K., Njamu, F. M., Somwe, V., Banda, J.,
Mosley, P. N. & Armstrong, R. A. 2001. The western arm of the
Lufilian Arc in NW Zambia and its potential for copper mineraliz-
ation. Journal of African Earth Science, 33, 503–528.
Kirschvink, J. L. 1992. Late Proterozoic low-latitude global glaciation:
the snowball earth. In: Schopf, J. W. & Klein, C. (eds) The Proter-
ozoic Biosphere. Cambridge University Press, Cambridge, 51 –52.
Knoll, A. H. 2003. Life on a Young Planet. Princeton University Press,
New Jersey.
Leather, J., Allen, P. A., Brasier, M. D. & Cozzi, A. 2002. Neoproter-
ozoic snowball Earth under scrutiny: evidence from the Fiq glaciation
of Oman. Geology, 30, 891–894.
Le Hir, G., Ramstein, G., Donnadieu, Y. & Pierrehumbert, R. T.
2007. Investigating plausible mechanisms to trigger a deglaciation
from a hard snowball Earth. Comptes Rendus Geoscience, 339,
274– 287.
Le Hir, G., Goddéris, Y., Donnadieu, Y. & Ramstein, G. 2008a. A geo-
chemical modelling study of the evolution of the chemical compo-
sition of seawater linked to a snowball glaciation. Biogeosciences,
5, 253–267.
Le Hir, G., Goddéris, Y., Ramstein, G. & Donnadieu, Y. 2008b.
A scenario for the evolution of the atmospheric pCO2 during a
snowball Earth. Geology, 36, 47 –50.
Le Hir, G., Donnadieu, Y. et al. 2009. The snowball Earth aftermath:
exploring the limits of continental weathering processes. Earth and
Planetary Science Letters, 277, 453–463.
Le Hir, G., Donnadieu, Y., Krinner, G. & Ramstein, G. 2010. Toward
the snowball Earth deglaciation. Climate Dynamics, 35, 285–297.
Lewis, J. P., Weaver, A. J., Johnston, S. T. & Eby, M. 2003. Neopro-
terozoic ‘snowball Earth’: dynamic sea ice over a quiescent ocean.
Paleoceanography, 18, doi: 10.1029/2003PA000926.
Lewis, J. P., Eby, M., Weaver, A. J., Johnston, S. T. & Jacob, R. L.
2004. Global glaciation in the Neoproterozoic: reconciling previous
modelling results. Geophysical Research Letters, 31, doi: 10.1029/
2004GL019725.
Lewis, J. P., Weaver, A. J. & Eby, M. 2006. Deglaciating the snowball
Earth: sensitivity to surface albedo. Geophysical Research Letters,
33, doi: 10.1029/2006GL027774.
Lewis, J. P., Weaver, A. J. & Eby, M. 2007. Snowball versus slushball
Earth: dynamic versus nondynamic sea ice? Journal of Geophysical
Research, 112, doi: 10.1029/2006JC004037.
Li, Z. X., Li, X. H., Kinny, P. D. & Wang, J. 1999. The break up of
Rodinia: did it start with a mantle plume beneath South China?
Earth and Planetary Science Letters, 173, 171–181.
Li, Z. X., Li, X. H., Kinny, P. D., Wang, J., Zhang, S. & Zhou, H. 2003.
Geochronology of Neoproterozoic syn-rift magmatism in the Yangtze
Craton, South China and correlations with other continents: evidence
for a mantle superplume that broke up Rodinia. Precambrian
Research, 122, 85 – 109.
Macdonald, F. A., Schmitz, M. D. et al. 2010. Calibrating the Cryo-
genian. Science, 327, 1241–1243.
Macouin, M., Besse, J., Ader, M., Gilder, S., Yang, Z., Sun, Z. &
Agrinier, P. 2004. Combined paleomagnetic and isotopic data
from the Doushantuo carbonates, South China: implications for the
‘snowball Earth’ hypothesis. Earth and Planetary Science Letters,
224, 387– 398.
Marshall, H. G., Walker, J. C. G. & Kuhn, W. R. 1988. Long-term
climate change and the geochemical cycle of carbon. Journal of
Geophysical Research, 93, 791– 801.
McKay, C. P. 2000. Thickness of tropical ice and photosynthesis on a
snowball Earth. Geophysical Research Letters, 27, 2153–2156.
Meert, J. G. & Powell, C. M. 2001. Assembly and break-up of Rodinia:
introduction to the special volume. Precambrian Research, 110, 1– 8.
Meert, J. G. & Torsvik, T. H. 2004. Paleomagnetic constraints on
Neoproterozoic ‘Snowball Earth’ continental reconstructions. In:
Jenkins, G. S., McMenamin, M. A. S., McKay, C. P. & Sohl, L.
(eds) The Extreme Proterozoic: Geology, Geochemistry, and
Climate. Geophysical Monograph 146. American Geophysical
Union, Washington, DC, 5 –11.
Micheels, A. & Montenari, M. 2008. A snowball Earth versus a slush-
ball Earth : results from Neoproterozoic climate modeling sensitivity
experiments. Geosphere, 4, 401–410.
Park, J. K., Buchan, K. L. & Harlan, S. S. 1995. A proposed giant
dyke swarm fragmented by the separation of Laurentia and
Australia based on paleomagnetism of ca.780 Ma mafic intrusions
in western North America. Earth and Planetary Science Letters,
132, 129– 139.
Pavlov, A. A., Hurtgen, M. T., Kasting, J. F. & Arthur, M. A. 2003.
Methane-rich Proterozoic atmosphere. Geology, 31, 87 –90.
Pavlov, A. A., Toon, O. B., Pavlov, A. K., Bally, J. & Pollard, D.
2005. Passing through a giant molecular cloud: ‘snowball’ glaciations
produced by interstellar dust. Geophysical Research Letters, 32, doi:
10.1029/2004GL021890.
Peltier, W. R., Tarasov, L., Vettoretti, G. & Solheim, L. P. 2004.
Climate dynamics in deep time: modeling the ‘snowball bifurcation’
and assessing the plausibility of its occurrence. In: Jenkins, G. S.,
McMenamin, M. A. S., McKay, C. P. & Sohl, L. (eds) The
Extreme Proterozoic: Geology, Geochemistry, and Climate. Geophy-
sical Monograph 146. American Geophysical Union, Washington,
DC, 107–124.
Peltier, W. R., Liu, Y. & Crowley, J. W. 2007. Snowball Earth pre-
vention by dissolved organic carbon remineralization. Nature, 450,
813– 818.
Petoukhov, V., Ganopolski, A., Brovkin, V., Claussen, M., Eliseev,
A., Kubatzki, C. & Rahmstorf, S. 2000. CLIMBER-2: a
climate system model of intermediate complexity. Part I: model
description and performance for present climate. Climate Dynamics,
16, 1 – 17.
Pierrehumbert, R. T. 2004. High levels of atmospheric carbon dioxide
necessary for the termination of global glaciation. Nature, 429,
646– 649.
Pierrehumbert, R. T. 2005. Climate dynamics of a hard Snowball Earth.
Journal of Geophysical Research, 110, doi: 10.1029/2004JD005162.
Pollard, D. & Kasting, J. F. 2004. Climate-ice sheet simulations of
Neoproterozoic glaciation before and after the collapse to snowball
Earth. In: Jenkins, G. S., McMenamin, M. A. S., McKay, C. P. &
Sohl, L. (eds) The Extreme Proterozoic: Geology, Geochemistry,
and Climate. Geophysical Monograph 146. American Geophysical
Union, Washington, DC, 91 –105.
Pollard, D. & Kasting, J. F. 2005. Snowball Earth: a thin-ice solution
with flowing sea glaciers. Journal of Geophysical Research, 111,
doi: 10.1029/2004JC002525.
Pollard, D. & Kasting, J. F. 2006. Reply to comment by Stephen
G. Warren and Richard E. Brandt on ‘Snowball Earth: A thin-ice sol-
ution with flowing sea glaciers’. Journal of Geophysical Research
111, C09017, doi: 10.1029/2006JC003488.
Poulsen, C. J. 2003. Absence of a runaway ice-albedo feedback in the
Neoproterozoic. Geology, 31, 115– 118.
Poulsen, C. J. & Jacob, R. L. 2004. Factors that inhibit snowball Earth
simulation. Paleoceanography, 19, doi: 10.1029/2004PA001056.
Poulsen, C. J., Pierrehumbert, R. T. & Jacob, R. L. 2001. Impact of
ocean dynamics on the simulation of the Neoproterozoic ‘snowball
Earth’. Geophysical Research Letters, 28, 1575–1578.
Poulsen, C. J., Jabob, R. L., Pierrehumbert, R. T. & Huynh, T. T.
2002. Testing paleogeographic controls on a Neoproterozoic snow-
ball Earth. Geophysical Research Letters, 29, doi: 10.1029/
2001GL014352.
 MODELLING SNOWBALL EARTH
Prave, A. R., Fallick, A. E., Thomas, C. W. & Graham, C. M. 2009. A
composite C-isotope profile for the Neoproterozoic Dalradian Super-
group of Scotland and Ireland. Journal of the Geological Society,
London, 166, 1 –13.
Ramstein, G., Donnadieu, Y. & Goddéris, Y. 2004. Proterozoic glacia-
tions. Comptes Rendus Geoscience, 336, 639–646.
Raub, T. D. & Evans, D. A. D. 2006. Magnetic reversals in basal Edia-
caran cap carbonates: a critical review. Eos, Transactions American
Geophysical Union, 87, abstract GP41-02.
Ritz, C., Rommelaere, V. & Dumas, C. 2001. Modeling the evolution of
Antarctic ice sheet over the last 420 000 years: implications for alti-
tude changes in the Vostok region. Journal of Geophysical Research,
106, 31 943– 31 964.
Romanova, V., Lohmann, G. & Grosfeld, K. 2006. Effect of land
albedo, CO2, orography, and oceanic heat transport on extreme cli-
mates. Climate of the Past, 2, 31 –42.
Rose, B. E. J. & Marshall, J. 2009. Ocean heat transport, sea ice, and
multiple climate states: insights from energy balance models.
Journal of Atmospheric Sciences, 66, 2828– 2843.
Schrag, D. P., Berner, R. A., Hoffman, P. F. & Halverson, G. P. 2002.
On the initiation of a Snowball Earth. Geochemistry, Geophysics,
Geosystems, 3, doi: 10.1029/2001GC000219.
Sellers, W. D. 1969. A global climatic model based on the energy
balance of the Earthatmosphere system. Journal of Applied Meteorol-
ogy, 8, 392– 400.
Stone, P. H. & Yao, M. S. 2004. The ice-covered Earth instability in a
model of intermediate complexity. Climate Dynamics, 22, 815– 822.
161
Torsvik, T. H., Carter, L. M., Ashwal, L. D., Bhushan, S. K., Pandit,
M. K. & Jamtveit, B. 2001. Rodinia refined or obscured: paleomag-
netism of the Malani igneous suite (NW India). Precambrian
Research, 108, 319–333.
Trindade, R. I. F., Font, E., D’Agrella-Filho, M. S., Nogueira,
A. C. R. & Riccomini, C. 2003. Low-latitude and multiple geomag-
netic reversals in the Neoproterozoic Puga cap carbonate, Amazon
craton. Terra Nova, 15, 441–446.
Walker, J. C. G., Hays, P. B. & Kasting, J. F. 1981. A negative
feedback mechanism for the long-term stabilization of Earth’s
surface temperature. Journal of Geophysical Research, 86,
9776–9782.
Warren, S. G. & Brandt, R. E. 2006. Comment on ‘Snowball Earth:
a thin-ice solution with flowing sea glaciers’ by D. Pollard and
J.F. Kasting. Journal of Geophysical Research, 111, C09016, doi:
10.1029/2005JC003411.
Warren, S. G., Brandt, R. E., Grenfell, T. C. & McKay, C. P.
2002. Snowball Earth: ice thickness on the tropical ocean. Journal
of Geophysical Research, 107, doi: 10.1029/2001JC001123.
Wingate, M. T. D., Campbell, I. H., Compston, W. & Gibson, G. G.
1998. Ion microprobe U– Pb ages for Neoproterozoic basaltic mag-
matism in south-central Australia and implications for the breakup
of Rodinia. Precambrian Research, 87, 135–159.
Wingate, M. T. D. & Giddings, J. W. 2000. Age and paleomagnetism
of the Mundine Well dyke swarm, Western Australia: implications
for an Australia– Laurentia connection at 755 Ma. Precambrian
Research, 100, 335–357.
 Chapter 11

The record of Neoproterozoic glaciation in the Taoudéni Basin, NW Africa

G. A. SHIELDS-ZHOU1*, MAX DEYNOUX2 & LAWRENCE OCH1

1
Department of Earth Sciences, University College London, Gower Street, London, WC1E 6BT, UK
2
Laurélie, 12270 Bor et Bar, France
*Corresponding author (e-mail: g.shields@ucl.ac.uk)

Abstract: The Taoudéni Basin covers over 1 000 000 km2 of the West African Craton, bounded by Pan-African orogenic belts. Four
supergroups separated by craton-scale unconformities are recognized, with Neoproterozoic glaciogenic deposits occurring at the base
of Supergroup 2. The Jbéliat Group occurs along a continuous, 1300-km-long, narrow belt from the Adrar region of Mauritania to
the eastern limit of the Hank in Algeria and comprises thin glacial drift capped widely by periglacial polygonal structures, with more
complex glacial sequences preserved in palaeo-depressions. A thicker, variously marine and continental glaciogenic succession can
be found in southern parts, while fully marine, glacially influenced successions are only known from the extreme SW of the basin.
The ‘triad’ sequence of diamictites overlain by barite-bearing ‘cap’ dolostones and then by green shales and/or bedded cherts (silexites)
is ubiquitous and has long been used to correlate the Supergroup 1/2 boundary across the basin and into the surrounding orogenic belts.
The bedded cherts commonly show a volcanic influence and are cemented by early marine calcite at their base at Adrar, Mauritania.
Although fossil-based age constraints are scarce and ambiguous, regional tectonic events indicate that ‘triad’ deposition occurred
between the Bassaride (665– 655 Ma) and Dahomeyide (610–580 Ma) orogens. Recent U –Pb zircon studies of ignimbrite tuffs
provide a minimum age for the glaciation of c. 600 Ma. Correlation of supergroup 2 glacial deposits with the c. 635 Ma end-Cryogenian
(‘Marinoan’) glaciation is likely and is supported by limited carbon and strontium isotope data. Barite is commonly found within the cap
carbonate and may relate to methane seepage and/or unusual oceanographic conditions after deglaciation. Several studies have attributed
sequence complexity within the post-glacial succession to isostatic reequilibration. The Taoudéni Basin represents a rare Neoproterozoic
example of terrestrial tillites and associated periglacial facies.

On the West African Craton, the Neoproterozoic and Palaeozoic
sedimentary cover can be subdivided into four supergroups (or
megasequences; e.g. Deynoux et al. 2006) bounded by craton-
scale unconformities (Trompette 1973). The lithostratigraphic
‘triad’ association (diamictite-cap dolostone-bedded chert), which
forms the base of Supergroup 2, has long been used as a marker
horizon (Zimmermann 1960; Leprun & Trompette 1969). The
first detailed studies of the glacial sedimentology of this region
were conducted during the 1970s, and what follows relies
heavily on over three decades of study by French researchers
who recorded the extraordinary preservation of glacial and perigla-
cial features across vast areas of arid North Africa (e.g. Deynoux
1982; Deynoux et al. 2006). This chapter mainly focuses on
some better known sections of the Taoudéni Basin that have
recently been the subject of isotopic studies (Alvaro et al. 2007;
Shields et al. 2007a, b). Owing to the large size of the Taoudéni
Basin, these descriptions are necessarily oversimplified and
the reader is referred to the published literature for more detailed
information on these and other (e.g. Bassaride, Rokelide, Anti-
Atlas, Hoggar-Iforas and Dahomeyide belts, and the Tindouf
Basin) Neoproterozoic glaciogenic successions of the West
African craton (Deynoux 1980, 1985; Deynoux & Trompette,
1981; Proust & Deynoux 1994; Deynoux et al. 2006). Glaciogenic
deposits of the neighbouring Volta Basin (Fig. 11.1) are presumed
to be correlative with those of the Taoudéni Basin Supergroup 2
and will also be considered here.
Structural framework and basin setting
The West African Craton has been a tectonic entity since at least
1600 Ma and is rimmed by mobile belts that became active
during late Proterozoic and Palaeozoic times and are related to
the Transaharan suture zone (Fig. 11.1). On the craton, the sedi-
mentary cover is preserved in widespread, variably connected
basins containing upper Proterozoic and Palaeozoic deposits of
which the largest is the Taoudéni Basin, forming a thin (3 km on
average), continuous tabular blanket over distances of 1000–
1500 km (Fig. 11.1). The Taoudéni Basin was possibly a less
rigid part of the platform, and is rimmed by the Mauritanide and
Hoggar-Iforas fold belts. No substantial sediment thickening
occurs in the central part of the basin, but extra subsidence has
caused some thickening in peripheral troughs and sub-basins,
such as the Gourma Basin to the east and the Madina-Kouta and
Bové basins to the west, which were involved in the Pan-African
and/or Hercynian fold belts.
With the exception of the area fringing the peripheral fold belts,
the Taoudéni Basin has been affected by regional-scale (epeiro-
genic) tilting, which has generated low angular unconformities,
and the reactivation of basement faults frequently intruded by
diabase sills and dykes. The sedimentary deposits are generally
devoid of tectonism, while metamorphism is limited to the
thermal effects of diabase intrusions in the proximal host material
that took place when the Atlantic Ocean began opening during
the Jurassic Period. In this regard, some sedimentary rocks may
have experienced locally high-grade, hydrothermal alteration. In
spite of their age, the sediments have undergone only modest
burial diagenesis because the sedimentary pile is relatively thin
and the platform has remained above sea level since the end of
the Carboniferous Period.
The Volta Basin thickens eastward, where it becomes gently
folded before disappearing beneath the Dahomeyide thrust belt.
According to sparse contextual information (Deynoux et al.
2006), the Volta Basin can be considered either to be a single fore-
land basin or as part of the Pan-African passive margin overlain by
a foreland basin made up of the Tamale Supergroup, lateral equiva-
lents of which can be found in intermontane, molasse-filled
grabens of the Dahomeyide belt (Affaton et al. 1991).
Stratigraphy
On the West-African platform, the Neoproterozoic and Palaeozoic
sedimentary cover can be subdivided into four supergroups

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 163– 171. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.11
164 G. A. SHIELDS-ZHOU ET AL.
(megasequences) separated by craton-scale unconformities
(Trompette 1973). Supergroup 1 rests with major unconformity
upon metamorphic and granitic basement across the Taoudéni
Basin and contains sedimentary strata of Mesoproterozoic age
(Clauer et al. 1982; Rooney et al. 2010). Supergroup 2 begins
with glacial deposits resting with an erosional and slightly
angular unconformity upon Supergroup 1 or directly upon the
basement. The glacial deposits are capped by a thin, but continuous
unit of calcareous dolomite, which contains barite in places.
Bedded cherts and green shales with a thin but continuous lime-
stone bed at their base generally overlie this cap-dolostone unit
and form the uppermost part of the so-called ‘triad’ of diamictite-
cap dolostone-bedded chert. Supergroup 3 consists of Late Ordo-
vician glacial deposits and post-glacial Silurian shales, whereas
Supergroup 4 corresponds to unconformably overlying Devonian
and Carboniferous strata.
In the Adrar region of Mauritania, glacial deposits of Super-
group 2 are unconformably underlain by early Neoproterozoic
Fig. 11.1. Simplified geological map of the
Taoudéni Basin and adjacent areas in NW
Africa (Source: Deynoux et al. 2006).
strata of the Char, Atar and Assabet el Hassiane Groups
(Fig. 11.2). The Atar Group and overlying siliciclastic rocks of
the Assabel-el-Hassiane Group in Mauritania experienced minor
tilting during Pan-African events before being peneplained by sub-
sequent uplift and erosion to form a largely flat, sub-glacial sub-
strate (Trompette 1973; Deynoux, 1980). Glacial deposits and
their ‘cap dolostone’ form the Jbéliat Group, which outcrops
nearly continuously for 1300 km along a narrow belt from the
Adrar region to the eastern limit of the Hank in Algeria (Fig. 1).
The Adrar cap carbonate package consists of one or two dolostone
units, with an intervening siliciclastic package of up to 40 metres
thickness, and a laterally extensive, thin limestone bed that discon-
formably overlies the uppermost dolostone forming the base of the
bedded cherts and shales of the Téniagouri Group.
In the Walidiala Valley, which straddles the Guinea-Senegal
border at the extreme southwestern margins of the Taoudéni
Basin, Supergroups 1 and 2 overlie basement rocks of the
Kenieba Inlier and are intruded and protected from erosion by
 NEOPROTEROZOIC GLACIATION IN THE TAOUDÉNI BASIN
plateau-forming diabase sills of Jurassic age. Supergroup 1 con-
sists of intertidal to supratidal, coarse-grained, red wackestones
and intervening siltstone forming the Ségou Group. The lower
part of Supergroup 2 is represented by the Mali Group, which
rests on the Mesoproterozoic Ségou Group with deep erosional
unconformity. The Mali Group (Fig. 11.3) consists of siliciclastic,
mostly siltstone units and comprises in its basal part the regionally
correlative triad of glaciogenic strata, dolostone and bedded chert.
The Mali Group has been subdivided into two formations (Culver
& Hunt 1991): the Hassanah Diallo Formation, 50 –120 m thick,
165
Fig. 11.2. Stratigraphy of the northern
part of the Taoudéni Basin in the Adrar area
and the Hank-Fersiga area (after Deynoux
et al. 2006). Ages are from Rooney et al.
2010 and Lahondère et al. 2005.
which represents the glacially related deposits, and the Nandou-
mari Formation, up to 130 m thick, which comprises at its base
quartz arenites overlain by dolostone, siltstone and bedded chert
(Shields et al. 2007a).
The southern region of the Taoudéni Basin stands out due to the
presence below the triad and associated craton-wide unconformity
of a 400–500-m-thick complex succession showing glacial influ-
ence (Fig. 11.4). Both triad and underlying glacially influenced
strata were initially assigned to the ‘Bakoye Group’ (Simon 1979);
however, the marked erosional and angular unconformity below
Fig. 11.3. Schematic geological map and
stratigraphic column of the Mali Group of
the southwestern Taoudéni Basin in the
Walidiala Valley (modified after Shields
et al. 2007a).
166 G. A. SHIELDS-ZHOU ET AL.
Fig. 11.4. Stratigraphy and main depositional environments of the glacially
related deposits in the southwestern part of the Taoudéni Basin in Mali
(Tambaoura plateaux) (modified from Deynoux et al. 2006).
the diamictites of the triad suggests that the underlying succession
records a distinct history. Deynoux et al. (1989, 2006) subsequently
subdivided the Bakoye Group into a Koniakari Group, correspond-
ing to the diamictites and post-glacial carbonate unit of the triad, and
a Wassangara Group, corresponding to the underlying, at least
partly glacially influenced succession. The Wassangara Group
forms most of the Tambaoura and Mandingue Plateaux in western
Mali and part of the Afollé Massif in southern Mauritania
(Fig. 11.1). It rests with an erosional and locally angular unconfor-
mity upon the Souroukoto Group.
The Volta Basin (Fig. 11.5) succession rests with major
unconformity on the Eburnean basement of the Leo Shield. The
lower (Bombouaka) Supergroup is up to 1000 m thick, comprising
largely siliciclastic rocks interspersed with thin calcareous beds in
its middle part, and is in general reminiscent of the lower part of
Supergroup 1 in the southern part of the Taoudéni Basin. The
middle (Pendjari or Oti) Supergroup is much thicker (2500 –
4000 m) and begins with glaciogenic deposits at its base, resting
unconformably on various parts of the Bombouaka Supergroup
or directly on basement. The triad of diamictite, dolostone and
chert passed upwards into green turbiditic siltstones and shales,
which contain phosphorite deposits. The upper (Tamale) Super-
group (500 m thick) is mainly exposed in Ghana and lies uncon-
formably on the Pendjari Supergroup. It comprises terrestrial
coarse-grained siliciclastic deposits with intercalated fine-grained
and calcareous horizons.
Glaciogenic deposits and associated strata
Palaeogeographic reconstructions of the triad-associated glacia-
tion of NW Africa imply an ice sheet centred towards the north
of the Reguibat Shield (Fig. 11.1) with inferred glacial movement
southward on the platform and laterally towards oceanic troughs or
basins located at the present position of the Pan-African belts.
Accordingly, continental glaciation is recorded in the northern
part of the Taoudéni Basin by a thin, irregular (0 –50 m thick)
veneer of terrestrial tillites with subordinate proglacial outwash
deposits preserved in limited shallow depressions. Towards the
south of the platform, the glacial drift thickens (150 –200 m),
showing marine influence in small intracratonic basins in western
Mali (Kayes area) and becoming wholly marine at the margins
of oceanic troughs bordering the Pan-African Bassaride Belt in
eastern Senegal and Guinea.
Along the northern margin of the Taoudéni Basin, glacial drift
comprises just a few metres of terrestrial tillite, with polygenic
and often striated pebbles to boulders overlying striated pave-
ments. Locally, a complex facies association is preserved in
smooth large-scale palaeo-depressions such as in the Jbéliat area
in Adrar, Mauritania (Fig. 11.1). In these palaeo-depressions, the
glacial deposits thicken up to 50 m and comprise distinct glacial
sequences suggesting at least three glacial advances and retreats,
with terrestrial tillites, outwash sandstones and various glacial fea-
tures such as striated clasts and pavements, roches moutonnées,
lacustrine varves with dropstones, and remarkable polygonal struc-
tures and sand wedges related to permafrost just below the cap
dolostone (Deynoux & Trompette 1976; Deynoux 1980, 1982,
1985). These deposits represent probably one of the only well-
preserved, purely continental records of a Pre-Pleistocene ice
sheet (Eyles & Januszczak 2004). Although true tillites are rarer
in the southern part of the Taoudéni Basin, diamictites of the
Koniakari Group and associated sandstones and shales form an
up to 200-m-thick marine to continental glacial succession well
exposed in the Kayes area of Mali (Rossi et al. 1984; Deynoux
et al. 1991). In areas of only marine sedimentation like the Wali-
diala Valley of Senegal, glacial influence can only be recognized
due to the occurrence of dropstones between debris flow deposits
of the Pelel Member and in shales of the overlying Diagoma
Member (Culver & Hunt, 1991; Shields et al. 2007a).
The post-glacial cap-carbonate succession in Adrar has been
described recently by Shields et al. (2007b) and Alvaro et al.
(2007), and only a brief summary is given here. The generally buff-
coloured cap dolostone is either well-bedded or brecciated with
thin and discontinuous intervening stringers of black chert.
Bedding surfaces are commonly erosional at all scales, that is,
scoured, rather than gradational or crinkled as with microbialites,
and show crossbedding only rarely. When passing to the brecciated
facies, beds become laterally and vertically folded and disrupted.
Fissures are common and are generally filled with ferroan
dolomite, black chert, white quartz or barite. A second generation
of fissures is filled with very coarse ferroan calcite spar. Micro-
scopically, the Jbéliat dolostone comprises alternating grain sizes
from dolomicrite to dolosparite with poorly developed or residual
cement made up of a mixture of clay, cryptocrystalline silica, side-
rite and pyrite.
The cap dolostone splits locally into two distinct stratigraphic
horizons providing space for 40 (Adrar) to 80 m (Hank) of onlap-
ping green to purple shales and siltstones including sandy dolos-
tone intercalations and passing upward in the Hank region into
sandy tidal deposits and eolian sandstone (Moussine-Pouchkine
& Bertrand-Sarfati 1997). Along the Atar cliff, the upper dolostone
horizon becomes progressively sandy until it forms a bank up to
7 m thick made up entirely of coarse-grained to granular, trough
crossbedded, dolomitic sandstone with isolated lenses of finer
dolostone. Barite is associated with the brecciated facies at
Jbéliat, Mauritania (mostly in the upper cap-dolostone unit) and
at Pont de Kabaté in Mali (Shields et al. 2007b), and appears as
palisadic crystals filling cracks or cavities, as well as monocrystal-
line geodes and laminated domes up to 30 cm across.
In the Jbéliat area and part of the Atar cliff area, over
c. 30– 40 km of the cap-dolostone exposure, a persistent
 NEOPROTEROZOIC GLACIATION IN THE TAOUDÉNI BASIN 167

2°W 0° 2°E Panafrican internal units

100 km
12° Suture zone

Kara struct. Unit

Panafrican
Atacora struct. Unit external units
Buem struct. Unit
10°

Tamale Megasequence
A
Flyschoid succession
Peniari or Oti
Tillite and cap dolostone Megasequence
B
8°
KFZ

Yemboure Gr.

Fosse aux lions Gr. Bambouaka

Megasequence
Dapaong Gr.

6° Eburnean basement
c Ocean
Atlanti
Thrust

A B

Fig. 11.5. Simplified geological map and a

synthetic cross-sections of the Volta Basin
50 km and Dahomeyide belt (modified after
Deynoux et al. 2006).

30– 50-cm-thick horizon of limestone is intercalated between
the cap dolostone and bedded cherts/siltstones of the overlying
Téniagouri Group. It is characterized by its purple to grey-green
colour and its brecciated texture defined by oriented, millimetre-
to centimetre-scale angular flakes of detrital, largely volcanogenic
material and authigenic barite (Hoffman & Schrag 2002; Shields
et al. 2007b), cemented by early marine calcite. At both Jbéliat
and the Atar Cliffs area, this limestone horizon may comprise
reverse-graded, calcite nodules or spherulites, which exhibit
cores of volcanogenic chert. The contact of the purple limestone
unit with the underlying cap dolostone is generally sharp and
marked by an erosional hardground surface; however, its contact
with the overlying cherts is transitional.
In the marine succession at Walidiala Valley, the Bowal
Member dolostone similarly consists of centimetre-to-decimetre
bedded, internally laminated dolostone and dolomitic siltstone
beds, overlying similar siltstone containing abundant lonestones
(dropstones). The dolostone unit shows widespread slumping
associated with the arrival of debris comprising volcanogenic
diamict that underwent pervasive dolomitization after redeposi-
tion. The bedded dolostone unit is characterized by various
forms of brittle deformation: fractures, tepee-like buckling and
discontinuous fitted brecciation with created space being filled
by early chert and/or by dolospar. In undeformed beds, this fine-
grained dolostone exhibits millimetre-scale layering defined by
mechanical laminations with no textural evidence for any
microbial influence. Microscopically, the Bowal Member dolo-
stone comprises equigranular grains of dolomite (generally
.90%) with accessory quartz and biotite grains of similar size.
Individual beds have sharp or erosional bases and fine upwards
from a surface of detrital dolomite/quartz lag of c. 100 mm size
to dolomicrite of ,10 mm in diameter (Shields et al. 2007a).
The Volta Basin tillite fills an erosional and slightly angular
unconformity of glacial origin with striated glacial pavements
along NW –SE trending palaeo-valleys. Its modest thickness
(c. 2 m) and geological context suggests that this deposit represents
a ‘true’ terrestrial tillite, which is overlain by a ,6 m cap dolos-
tone (Porter et al. 2004). At Bwipe (Ghana), the cap dolostone
unit is thicker, comprising up to 12 m of finely laminated,
‘microbial’ dolostone (Nedelec et al. 2007).
Boundary relations with overlying and underlying
non-glacial units
Tilting during Pan-African orogenic events and largely pre-glacial
peneplanation caused glacial and glacially influenced units of
Supergroup 2 to be deposited unconformably upon marine sedi-
mentary rocks of Supergroup 1 throughout the Taoudéni Basin,
and apparently across NW Africa, including in the Volta Basin.
In the Adrar region of Mauritania, the top of the glacial drift is
marked by highly indurated, black, coarse-grained sandstone,
which forms patterned ground with polygonal structures and
sand wedges typical of permafrost (Deynoux 1982). Laminated
dolostone drapes the immediately post-glacial topography or
rests directly upon the underlying and commonly striated perigla-
cial substrate in both Taoudéni and Volta basins.
168 G. A. SHIELDS-ZHOU ET AL.
In the Walidiala Valley section in Senegal, glaciomarine influ-
ence wanes up section as dropstones become both smaller and
rarer in mudstones of the Diagoma Member, the top 10 m of
which seem to be devoid of any dropstones. In places, channels
cut into these mudstones and are filled with coarse-grained, cross-
bedded sandstone and gravels of the Tanagué Member. A 2– 7 m
thick, silty cap dolostone, the Bowal Member, caps the glaciogenic
sequence and rests erosionally on the underlying siltstone due to
slumping or possibly conformably on coarse-grained siliciclastics
(Shields et al. 2007a).
Chemostratigraphy
Only limited isotope work has been carried out in Neoprotero-
zoic strata of the Taoudéni and Volta basins. Earlier studies
reported Sr- and C-isotope data from the type section of the uncon-
formably underlying Atar Group, in Mauritania (Veizer et al.
1983; Fairchild et al. 1990), but only recently have studies
focused on post-glacial strata across the basin (Alvaro et al.
2007; Shields et al. 2007a, b). Despite pervasive recrystallization
and multiple phases of cementation, which makes primary
carbon-isotope trends (e.g. the Jbéliat section, Mauritania, and
the Goumaré section, Mali) hard to recognize, published results
reveal a consistent picture of a post-glacial cap dolostone unit
marked by low d13CPDB (averaging – 4‰) trending up-section to
more negative values; this is most clearly seen in the Walidiala
Valley sections of Senegal and the Koniakari section, Mali.
Lowermost d13C values (– 6‰) are reported from a limestone
unit at the base of the Téniagouri Group, which overlies post-
glacial cap-dolostone strata in the Atar Cliffs area of Mauritania
(Alvaro et al. 2007). Shields et al. (2007b) trace a trend to more
positive values through this limestone unit with d13C as high as
þ3.7‰ in places. High-resolution isotopic data from four sections
of the Volta Basin (Burkina Faso: Arli, Kodjari, Koundjouari; and
Ghana: Bwipe) show a similarly decreasing upward trend (Porter
et al. 2004; Nedelec et al. 2007). Positive correlation between
d13C and d18O appears to be characteristic (Porter et al. 2004;
Shields et al. 2007a).
Barite samples from Mali and Mauritania have been analysed
for Sr and S isotopes (Shields et al. 2007b). 87Sr/86Sr ratios are
consistent between barite deposits (0.70773 –0.70814) with the
results of leaching experiments indicating that Sr in the barite
was derived from seawater with 87Sr/86Sr ¼ 0.7077 –8. In con-
trast, barite d34S compositions range widely, exhibiting values
between þ20‰ and þ46‰. This indicates that barite sulphate
derives from a marine source with d34S of about þ20‰, and
that it has undergone considerable isotopic fractionation due to sul-
phate reduction (Shields et al. 2007b). Sr isotope compositions
have also been reported for the cap dolostone at Bwipe (Volta
Basin), where they range from 0.7061 to 0.7073 (Nedelec et al.
2007). Such low values are atypical for basal Ediacaran dolo-
stones, which generally exhibit variably radiogenic ratios (e.g.
Yoshioka et al. 2003) that are unlikely to be representative of
ambient seawater.
Other characteristics
Economic deposits associated with the glaciogenic strata are
limited. Barite deposits are commonly associated with the cap
dolostone and are of high purity but are too localized to be of econ-
omic interest. The Atar and Assabet el Hassiane groups and their
equivalents below the glacial level in Mauritania and Mali (as
well as the Silurian shales above the Late Ordovician glacial depos-
its) have attracted interest as petroleum source rocks. Despite a gas
show in the 1970s, only four wells are known to have been drilled to
date (Craig et al. 2009). Banded haematite cherts (BIFs) are found
within shales of the Wassangara Group (Taoudeni Basin, Mali),
sandwiched between terrestrial strata showing fluvial, eolian and
glacial influence (Deynoux et al. 1991).
Simple discoidal impressions were reported from strata below
glaciogenic deposits by Bertrand-Sarfati et al. (1995) in the
Hank region of northern Mauritania; however, these are not
necessarily of Ediacaran age, as simple circular impressions are
also known from Cryogenian strata in Canada and older Protero-
zoic strata elsewhere (e.g. Cruse & Harris 1994). The discovery
of Cambrian-type small shelly fossils, including the proto-
gastropod Aldanella attleborensis in dolostone scree in Walidiala
Valley, Senegal (Culver et al. 1988) caused some to infer a Cam-
brian age for the Taoudéni Basin glaciation (Bertrand-Sarfati et al.
1995; Evans 2000). Although the block was believed to derive
from the post-glacial Bowal Member cap dolostone, subsequent
investigations have not been able to confirm the findings and
recent publications tend to play down their significance (Porter
et al. 2004; Deynoux et al. 2006; Alvaro et al. 2007; Shields
et al. 2007a, b).
Palaeolatitude and palaeogeography
There are no firm palaeomagnetic constraints on the position of
the West African craton during the Neoproterozoic. Existing data
are consistent with a position at mid to high latitudes; however,
the primary nature of all published results can be questioned
(Evans 2000).
Geochronological constraints
Early geochronological studies of the 1970s and 1980s applied
K –Ar and Rb –Sr isotope techniques to mudstone units on the
West African craton and demonstrated a late Neoproterozoic age
for the ‘Infracambrian’ glaciogenic units of Supergroup 2
(Clauer et al. 1982; Clauer & Deynoux 1987). Subsequent work
has largely confirmed this viewpoint, although some confusion
was injected by the discovery (Culver et al. 1988) of Cambrian-
type fossils in dolostone scree of the Walidiala Valley, Senegal
(see discussion above). The most robust age constraints derive
from U – Pb studies of zircons from bedded tuffs within the
lower Téniagouri Group, Mauritania (Lahondère et al. 2005).
These new age constraints (609.7 + 5.5 Ma, U –Pb TIMS zircon
and 606 + 6 Ma, U –Pb SHRIMP zircon) provide a minimum
age for the glacial Jbéliat Group and its cap dolostone of about
600 Ma, which is consistent with correlation with the comparably
well-dated c. 635 Ma glaciogenic successions of South China
(Condon et al. 2005) and Namibia (Hoffmann et al. 2004).
Deynoux et al. (2006) discuss existing age constraints in more
detail and conclude that triad deposition across the Taoudéni
Basin, and likely also in the Volta Basin, took place between the
Bassaride orogen (Pan-African I, 665– 655 Ma) and the Daho-
meyide orogen (610 – 580 Ma).
Isotope studies of post-glacial dolostone units across the basin
(Shields et al. 2007a, b; Alvaro et al. 2007) support this interpret-
ation (see Discussion below), and fossils are scarce and do not
provide additional age constraints. Although some glaciogenic
successions of the Taoudéni Basin may show multiple glacial
advances and retreats, for example, in Mali (Deynoux et al.
1991), there is at present no firm evidence for any older, mid-
Cryogenian (‘Sturtian’) glaciation. According to Villeneuve
(1988) and Villeneuve & Cornée (1994), the glacially influenced
Wassangara Group, which lies unconformably below the triad in
the Kayes area of Mali (Proust & Deynoux 1994), also postdates
the collision between the West African craton and a western Sene-
galese block at 660 Ma (Pan-African 1 orogeny). The above con-
straints permit regional correlation of Neoproterozoic to Cambrian
strata of the West African craton and surrounding Pan-African
belts (Deynoux et al. 2006, fig. 14).
 NEOPROTEROZOIC GLACIATION IN THE TAOUDÉNI BASIN
Discussion
The glacial formations on the rim of the northern Taoudéni Basin
consist of terrestrial tillites and proglacial outwash sandstones,
which are capped in the Adrar region of northern Mauritania by
a persistent horizon of large-scale polygonal structures and sand
wedges. These polygonal structures are evidence that a prolonged
periglacial period with temperatures low enough for the develop-
ment of permafrost (Deynoux 1982) followed the retreat of gla-
ciers in this part of the craton. Analogous polygonal structures
and sand wedges can also be found 500 km farther south in the
Afollé Massif, south of Kiffa, indicating that permafrost was not
restricted to the Adrar region but extended over a large part of
the Taoudéni Basin. Post-glacial cap dolostones of the northern
Taoudéni Basin mark an abrupt change in the regional depositional
environment from terrestrial to marine.
The cap dolostones and overlying shales (with calcitic cherts at
the base) of the Téniagouri Group are considered to relate to degla-
cial to post-glacial eustatic transgression, respectively. However,
the presence of intervening shales and sandstones, levelling
palaeoreliefs (see above) as observed in the Adrar and Hank
regions, highlights the complexity of this transgression. According
to Aı̈t-Kaci Ahmed & Moussine-Pouchkine (1994), and Bertrand-
Sarfati et al. (1997), in the Hank Algerian border region (Guetta-
tira-Grizim-Fersiga area), the cap carbonates and intervening
shales, which contain intercalations of phosphatic grainstones,
were deposited during the glacioeustatic transgression. Owing
to the balance between glacioeustasy and isostatic rebound, the
cap carbonates were locally subjected on topographic highs to sub-
aerial exposure with the development of a complex phosphatic
profile, including mini-stromatolites, while sedimentation contin-
ued in the depressions. Intervening green shales pass progressively
upward into sandy tidal deposits and eolian sandstones in the
Hank. These sandstones are overlain by a transgressive surface
covered by a few metres of conglomeratic and phosphatic sand-
stones that preceded the large-scale development on the whole
platform of shales equivalent to the Téniagouri Group in Maurita-
nia. Accordingly, the post-glacial cap-carbonate succession across
the Taoudéni Basin may represent a condensed deposit that relates
to the onset of the glacioeustatic transgression, with the transgres-
sive and highstand system tracts preserved only in palaeo-
depressions.
As in the Hank, the cap dolostone in the Adrar reveals a
complex, relative sea-level record, with the post-glacial ‘purple’
limestone representing a condensation horizon that marks the
beginning of a second, marine transgression, apparently coinciding
with regional volcanism. Mineral relationships confirm that
regional volcanism occurred after palaeo-topographic highs had
for a time become emergent, allowing karst dissolution to occur.
Carbonate saturation levels were high enough to grow calcite crys-
tals rapidly on and in caverns immediately beneath the seafloor
before glassy debris could disintegrate. The transition from dolo-
mite to calcite precipitation is similar to other such transitions
around the world (Hoffman & Schrag 2002), and could be correla-
tive based on the co-occurrence of barite and isotopic trends.
At the base of the Neoproterozoic-age Mali Group of the south-
western Taoudéni Basin, debris flows and turbidite-like, sandy
units of the Pelel Member pass upward into siltstone and shale
of the Diagoma Member. These two units represent the progressive
evolution from some portion of a fan delta fed by a nearby ice shelf
to a more distal environment disturbed only by the occasional
fallout from passing icebergs. The appearance of coarse-grained,
crossbedded sandstone beds and gravels of the overlying
Tanagué Member heralds a return to a shallower, fluvially influ-
enced environment before abrupt transgression caps the glacio-
genic succession. The succession in the Walidiala Valley could
represent therefore the stratigraphic expression of a glacial
retreat in a proximal glaciomarine environment affected by gla-
cioeustasy and isostatic rebound (Shields et al. 2007a).
169
The transgressive unit consists of a regionally extensive, 2 –
7-m-thick, silty dolostone, the Bowal Member, which is isotopi-
cally and petrographically indistinguishable from c. 635 Ma cap-
dolostone units elsewhere in NW Africa and worldwide (Shields
et al. 2007a). A large volcaniclastic debris flow has caused slump-
ing and soft-sediment deformation within the cap dolostone of the
Bowal Member. The widespread association of pyroclastic depos-
its with cap dolostone in geographically distant successions of the
Taoudéni Basin and neighbouring Hoggar-Iforas Belt (Deynoux
et al. 2006) implies that volcanism and deglaciation were roughly
contemporaneous across a large area. Early cementation is demon-
strated in this case by pervasive dolomitization of the debris flow
matrix and supports the notion of unusually high carbonate satur-
ation levels during the aftermath of latest Cryogenian glaciation
(Hoffman & Schrag 2002).
The origin of the barite mineralization in the post-glacial
carbonate succession was specifically addressed by Shields et al.
(2007b), who considered that barite formed during the second
marine transgression but before deposition of the overlying
Téniagouri Group shales and bedded cherts. Isotopic evidence
suggests that barite precipitated due to the mixing of sulphate-rich
seawater with a more reducing Ba-rich fluid. The consistent
association of barite occurrences throughout the basin with terres-
trial periglacial deposits (Deynoux 1980, 1982) and the sporadic
occurrence of the barite permit the suggestions that barite mineral-
ization was controlled by local seepage of permafrost methane
following deglaciation. Such a scenario implies a relatively short
time interval between deglaciation and barite genesis, which is
somewhat supported by the general agreement between barite
Sr- and S-isotope values and those of other basal Ediacaran
cap-carbonate successions (Shields et al. 2007b). The common
association of capdolostone to limestone transitions with barite
mineralization could, however, point to a global oceanographic
origin for the barite, whereby deeper Ba-rich seawater beneath a
persistent pycnocline mixed with sulphate-rich surface seawater
after deglaciation.
The negative d13C trend shown in Taoudéni Basin post-glacial
dolostones is matched by identical trends in correlative sections
of the neighbouring Volta Basin (Porter et al. 2004; Nedelec
et al. 2007) and worldwide (Kennedy 1996; Halverson et al.
2004), all from dolostone units immediately overlying glaciogenic
strata interpreted to belong to the latest Cryogenian (commonly
referred to as ‘Marinoan’) glaciation. The affiliation of the Taou-
déni Basin triad with other diamictite – cap dolostone associations
from around the world is consistent with the abovementioned geo-
chronological constraints and Sr- and S- isotope data, thus lending
further support to the notion of a global deglaciation at c. 635 Ma.
In this regard, the limestone unit reported from above post-glacial
cap dolostones in Adrar, Mauritania (Alvaro et al. 2007; Shields
et al. 2007b) may be equivalent to isotopically similar post-glacial
limestones around the world (cf. Hoffman & Schrag 2002). The
anomalously low 87Sr/86Sr ratios reported by Nedelec et al.
(2007) remain inexplicable; they could relate to diagenetic isotopic
exchange with juvenile volcanic minerals or to the fact that
contemporaneous seawater 87Sr/86Sr was significantly lower
than previously believed.
This represents a contribution of the IUGS- and UNESCO-funded IGCP (Inter-
national Geoscience Programme) Project #512.
References
Affaton, P., Rahaman, M. A., Trompette, R. & Sougy, J. 1991. The
Dahomeyide orogen: tectonothermal evolution and relationships
with the Volta Basin. In: Dallmeyer, R. D. & Lécorché, J. P.
(eds) The West African Orogens and Circum-Atlantic Correlatives.
Springer-Verlag, New York, 107–122.
Aı̈t-Kaci Ahmed, A. & Moussine-Pouchkine, A. 1994. Les formation
cambriennes de Fersiga (Sud-Ouest du Tanezrouft): nouvelle
170 G. A. SHIELDS-ZHOU ET AL.
interprétation de la sédimentation glaciaire et post-glaciaire sur le
Craton Ouest Africain. Bulletin du Service géologique d’Algérie,
5, 3 –21.
´ lvaro, J. J., Macouin, M., Bauluz, B., Clausen, S. & Ader, M.
A Fairchild, I. J., Marshall, J. D. & Bertrand-Sarfati, J. 1990. Strati-
2007. The Ediacaran sedimentary architecture and carbonate pro-
ductivity in the Atar Cliffs, Adrar, Mauritania: palaeoenvironments,
chemostratigraphy and diagenesis. Precambrian Research, 153,
236– 261.
Bertrand-Sarfati, J., Moussine-Pouchkine, A. Q., Amard, B. & Aı̈t
Kaci Ahmed, A. 1995. First Ediacaran fauna found in western
Africa and evidence for an Early Cambrian glaciation. Geology, 23,
133– 136.
Bertrand-Sarfati, J., Flicoteaux, R., Moussine-Pouchkine, A. &
Aı̈t Kaci Ahmed, A. 1997. Lower Cambrian apatitic stromatolites
and phospharenites related to the glacio-eustatic cratonic rebound
(Sahara, Algeria). Journal of Sedimentary Research, 67, 957– 974.
Clauer, N. & Deynoux, M. 1987. New information on the probable
isotopic age of the Late Proterozoic glaciation in West Africa. Pre-
cambrian Research, 37, 89– 94.
Clauer, N., Caby, R., Jeanette, D. & Trompette, R. 1982. Geochro-
nology of sedimentary and metasedimentary Precambrian rocks of
the West African Craton. Precambrian Research, 18, 53– 71.
Condon, D., Zhu, M., Bowring, S., Wang, W., Yang, A. & Jin, Y. 2005.
U –Pb ages from the Neoproterozoic Doushantuo Formation, China.
Science, 308, 95 –98.
Craig, J., Thurow, J., Thusu, B., Whitham, A. & Abuttaruma, Y.
2009. Global Neoproterozoic Petroleum Systems: The Emerging
Potential in North Africa. Geological Society, London, Special
Publications, 326.
Cruse, T. & Harris, L. B. 1994. Ediacaran fossils from the Stirling Range
Formation, Western Australia. Precambrian Research, 67, 1 –10.
Culver, S. J. & Hunt, D. 1991. Lithostratigraphy of the Precambrian –
Cambrian boundary sequence in the southwestern Taoudeni Basin,
West Africa. Journal of African Earth Sciences, 13, 407–413.
Culver, S. J., Pojeta, J. & Repetski, J. E. 1988. First record of Early
Cambrian shelly microfossils from West Africa, Geology, 16,
695– 599.
Deynoux, M. 1980. Les formations glaciaires du Précambrien terminal
et de la fin de l’Ordovicien en afrique de l’Ouest. Deux exemples
de glaciation d’inlandsis sur une plate-forme stable. Travaux du Lab-
oratoire des Sciences de la Terre St-Jerome, Marseille (B), 17, 554.
Deynoux, M. 1982. Periglacial polygonal structures and sand wedges in
the late Precambrian glacial formations of the Taoudeni Basin in
Adrar of Mauritania (West Africa). Palaeogeography, Palaeoclima-
tology, Palaeoecology, 39, 55– 70.
Deynoux, M. 1985. Terrestrial or waterlain glacial diamictites? Three
case studies from the Late Precambrian and Late Ordovician glacial
drifts in West Africa. Palaeogeography, Palaeoclimatology,
Palaeoecology, 51, 97– 141.
Deynoux, M. & Trompette, R. 1976. Late Precambrian mixtites:
glacial and/or nonglacial? A discussion dealing especially with the
mixtites of West Africa. American Journal of Science, 276,
1302– 1315.
Deynoux, M. & Trompette, R. 1981. Late Precambrian tillite of the
Taoudeni Basin, West Africa. In: Hambrey, M. J. & Harland,
W. B. (eds) Earth’s Pre-Pleistocene Glacial Record. Cambridge
University Press, New York, 123– 134.
Deynoux, M., Marchand, J. & Proust, J. N. 1989. Notice explicative de
la carte géologique du Mali occidental az 1/200.000. Feuilles Kan-
kossa, Kayes, Kossanto. République du Mali, Direction Nationale
de la Géologie et des Mines, Bamako. Klöckner Industrie-Anlagen,
Duisburg, 54– 81.
Deynoux, M., Proust, J. N. & Simon, B. 1991. Late Proterozoic glacially
controlled shelf sequences in western Mali (West Africa). Journal
African Earth Sciences, 12, 181–198.
Deynoux, M., Affaton, P., Trompette, R. & Villeneuve, M. 2006.
Pan-African tectonic evolution and glacial events registered in
Neoproterozoic to Cambrian cratonic and foreland basins of West
Africa. Journal of African Earth Sciences, 46, 397– 426.
Evans, D. A. D. 2000. Stratigraphic, geochronological, and paleomagnetic
constraints upon the Neoproterozoic climatic paradox. American
Journal of Science, 300, 347– 433.
Eyles, N. & Januszczak, N. 2004. ‘Zipper-rift’: a tectonic model for
Neoproterozoic glaciations during the breakup of Rodinia after 750
Ma. Earth Science Reviews, 65, 1– 73.
graphic shifts in carbon isotopes from Proterozoic stromatolitic car-
bonates (Mauritania): influences of primary mineralogy and
diagenesis. American Journal of Science, 290A, 46– 79.
Halverson, G. P., Maloof, A. C. & Hoffman, P. 2004. The Marinoan
glaciation (Neoproterozoic) in northeast Svalbard. Basin Research,
16, 297– 324.
Hoffman, P. F. & Schrag, D. P. 2002. The snowball Earth hypothesis:
testing the limits of global change. Terra Nova, 14, 129–155.
Hoffmann, K.-H., Condon, D. J., Bowring, S. A. & Crowley, J. L.
2004. U– Pb zircon date from the Neoproterozoic Ghaub Formation,
Namibia: constraints on Marinoan glaciation. Geology, 32, 817– 820,
doi:10.1130/G20519.l.
Kennedy, M. J. 1996. Stratigraphy, sedimentology, and isotopic geo-
chemistry of Australian Neoproterozoic postglacial cap dolostones:
deglaciation, d13C excursions and carbonate precipitation. Journal
of Sedimentary Research, 66, 1050– 1064.
Lahondère, D., Roger, J. et al. 2005. Notice explicative des cartes géo-
logiques à 1/200,000 et 1/500,000 de l’extrème sud de la Mauritanie.
DMG, Ministère des mines et de l’industrie, Nouakchott, Rapport
BRGM/RC-54273-FR, 610.
Leprun, J.-C. & Trompette, R. 1969. Subdivision du Voltaı̈en du
massif de Gobnangou (République de Haute-Volta) en deux séries
discordantes séparées par une tillite d’âge éocambrien probable.
Comptes Rendus de l’Académie des Sciences, Paris, 269,
2187– 2190.
Moussine-Pouchkine, A. & Bertrand-Sarfati, J. 1997. Tectonosedi-
mentary subdivisions in the Neoproterozoic to Early Cambrian
cover of the Taoudenni Basin (Algeria, Mauritania, Mali). Journal
African Earth Sciences, 24, 425– 443.
Nedelec, A., Affaton, P., France-Lanord, C., Charriere, A. &
Alvaro, J. 2007. Sedmentology and chemostratigraphy of the
Bwipe Neoproterozoic cap dolostones (Ghana, Volta Basin): a
record of microbial activity in a peritidal environment. Comptes
Rendus Geoscience, 339, 223– 239.
Porter, A. M., Knoll, A. H. & Affaton, P. 2004. Chemostratigraphy of
Neoproterozoic cap carbonates from the Volta Basin, West Africa.
Precambrian Research, 130, 99 – 112.
Proust, J. N. & Deynoux, M. 1994. Marine to non-marine sequence
architecture of an intracratonic glacially related basin. Late Protero-
zoic of the West African platform in western Mali. In: Deynoux,
M., Miller, J. M. G., Domack, E. W., Eyles, N., Fairchild, I. J.
& Young, G. M. (eds) Earth’s Glacial Record. Cambridge University
Press, Cambridge, 121– 145.
Rooney, A. D., Selby, D., Houzay, J.-P. & Renne, P. R. 2010.
Re –Os geochronology of Mesoproterozoic sediments from
the Taoudeni basin, Mauritania: implications for basin-wide corre-
lations, supercontinent reconstruction and Re– Os systematics of
organic-rich sediments. Earth and Planetary Science Letters, 289,
486– 496.
Rossi, P., Deynoux, M. & Simon, B. 1984. Les formations glaciaires
du Précambrien terminal et leur contexte stratigraphique
(formations pré et post-glaciaires et dolérites permiennes(?) du
massif du Kaarta (dans le bassin de Taoudéni au Mali occidental
(Afrique de l’Ouest). Sciences Géologiques Bulletin, Strasbourg,
37, 91 – 106.
Shields, G. A., Deynoux, M., Culver, S. J., Brasier, M. D., Affaton,
P. & Vandamme, D. 2007a. Neoproterozoic glaciomarine and cap
dolostone facies of the southwestern Taoudéni Basin (Walidiala
Valley, Senegal/Guinea, NW Africa). Comptes Rendus de l’Acadé-
mie des Sciences: Geosciences, 339, 186– 199.
Shields, G. A., Deynoux, M., Strauss, H., Paquet, H. & Nahon, D.
2007b. Barite-bearing cap carbonates of the Taoudéni Basin, north-
west Africa: sedimentary and isotopic evidence for methane
seepage from permafrost after a Neoproterozoic glaciation. Precam-
brian Research, 153, 209–235.
Simon, B. 1979. Essai de synthèse sur les formations sédimentaires de la
partie occidentale du Mali. Rapport Inédit Laboratoire Géologie
Dynamique Univ. Aix-Marseille III, 133.
 NEOPROTEROZOIC GLACIATION IN THE TAOUDÉNI BASIN
Trompette, R. 1973. Le Précambrien supérieur et le Paléozoique inférieur
de l’Adrar de Mauritanie (bordure occidentale du bassin de Taoudeni,
Afrique de l’Ouest). Un exemple de sédimentation de craton, Etude
stratigraphique et sédimentologique. Travaux du Laboratoire des
Sciences de la Terre St-Jerome, Marseille (B), 7.
Veizer, J., Compston, W., Clauer, N. & Schidlowski, M.
1983. 87Sr/86Sr in Late Proterozoic carbonates: evidence for a
‘mantle’ event at 900 Ma ago. Geochimica et Cosmochimica
Acta, 47, 295– 302.
Villeneuve, M. 1988. Evolution comparée du bassin de Taoudéni et de la
chaine des Mauritanides en Afrique de l’Ouest. Comptes Rendus Aca-
démie des Sciences, Paris, 307, 663– 668.
171
Villeneuve, M. & Cornée, J. J. 1994. Structure, evolution and palaeo-
geography of the West African craton and bordering belts during
the Neoproterozoic. Precambrian Research, 69, 307– 326.
Yoshioka, H., Asahara, Y., Tojo, B. & Kawakami, S. 2003.
Systematic variations in C, O, and Sr isotopes and elemental
concentrations in Neoproterozoic carbonates in Namibia: impli-
cations for glacial to interglacial transition. Precambrian Research,
124, 69– 85.
Zimmermann, M. 1960. Nouvelle subdivision des séries antégothlan-
diennes de l’Afrique occidentale (Mauritanie, Soudan, Sénégal),
Rap. 21st International Geological Congress, Copenhagen, 8,
26– 36.
 Chapter 12

Neoproterozoic glaciogenic diamictites of the Katanga Supergroup, Central Africa

SHARAD MASTER1* & MAREK WENDORFF2

1
Economic Geology Research Institute, School of Geosciences, University of the Witwatersrand, P. Bag 3, WITS
2050, Johannesburg, South Africa
2
AGH University of Science and Technology, Faculty of Geology, Geophysics and Environmental Protection,
al. A. Mickiewicza 30, 30-059 Krakow, Poland
*Corresponding author (e-mail: sharad.master@wits.ac.za)

Abstract: Glaciogenic sediments of the Katanga Supergroup are represented by two units. The syn-rift Grand Conglomerat Formation
(,765 + 5 Ma to .735 + 5 Ma) occurs within the Nguba Group, and the Petit Conglomerat Formation defines the base of the Kunde-
lungu Group deposited in the earliest foreland basin of the Lufilian orogenic belt located between the Congo and Kalahari cratons. Their
glacial origin is inferred on the basis of the following features: the common and widespread occurrence of thick polymictic conglomerates
and diamictites with faceted and striated clasts, massive structure, abundant poorly sorted fine-grained matrix, and the presence of planar-
laminated shales (laminites) with dropstones. Glaciomarine facies associations prevail over most of the geographic extent of both units,
but at the northern periphery of the depository, continental glacial facies are present. The glaciomarine units are succeeded by carbonates:
the Kakontwe Limestone and ‘Calcaire Rose’ respectively. The clasts in the glaciogenic units are of extrabasinal and intrabasinal pro-
venance. Lower boundaries, conformable in the basin centre, evolve to unconformities in the marginal areas to the N and S. The palaeo-
magnetic evidence suggests deposition in low latitudes.

The Neoproterozoic – Lower Palaeozoic Katanga Supergroup
of Central Africa (Katanga Province of Democratic Republic of
Congo, Zambia, and eastern Angola) is exposed in the Lufilian
belt (part of the continental system of the Pan-African orogenic
belts of Africa), and also forms a less deformed plateau
molasse/foreland sequence over the Congo Craton (Fig. 12.1).
Katangan sediments were initially deposited in intracratonic
rifts related to early Neoproterozoic extension along the southern
margin of the Congo Craton. Continental break-up accompanied
by mafic volcanism led to the formation of major unconformities,
and development of a passive margin (Nguba Group). Subsequent
deformation in late Neoproterozoic to early Palaeozoic orogenic
belts (Lufilian-Zambezi), and syntectonic sedimentation in associ-
ated foreland basins, was related to the collision of the Congo and
Kalahari Cratons during the amalgamation of the Gondwana
Supercontinent (Hanson 2003; Wendorff 2005a, b; Master et al.
2005).
Regional mapping of Katanga (at a scale of 1:500 000) first
commenced with the work of Franz Edward Studt, Jules Cornet
and Henri Buttgenbach (Studt et al. 1908; Studt 1908, 1913).
The first identification of glacial diamictites in Katanga was
made by Stutzer (1911, 1913a) and by Grosse (1912), but both
authors misidentified these Katangan diamictites as belonging to
the late Carboniferous Dwyka glaciation. Robert (1912a, b) first
identified glacial diamictites belonging to the Kundelungu
System in the Katanga region on the basis of striated pebbles.
Hennig (1915) regarded the glacial beds of Central Africa ident-
ified by Stutzer (1911, 1913a) and Grosse (1912) as being much
older than the Dwyka. Different facies were identified in the
Katangan diamictites by Delhaye (1920). As a consequence of
more detailed regional mapping, at a scale of 1:200 000, by van
Doorninck (1928) and by Maurice Robert from 1926 to 1931, in
which diamicitites were used extensively as easily recognized
regional marker horizons, it was finally established by Robert
(1933), Gysin (1934) and Grosemans (1935) that there are two dia-
mictites, the Grand Conglomerat and Petit Conglomerat, which
were regarded as forming, respectively, the base of the Lower
and Upper ‘Series’ (‘Serie Inférieur’ and ‘Serie Supérieur’), con-
stituting the upper part of the Kundelungu ‘System’. Van Door-
ninck (1928) and Robert (1940a, b, 1947) argued for a glacial
origin of these diamictites based on a number of characteristics,
such as the great lateral continuity and thickness of these beds,
and the extrabasinal origin and faceted and striated nature of
the clasts.
The Grand Conglomerat Formation has also been designated
as unit Ki1.1 (François 1973) and as the Mwale Formation, or
Ng1.1 (Batumike et al. 2007), while the Petit Conglomerat has
been referred to as the Ks1.1 (François 1973), and the Kyandamu
Formation or Ku1.1 (Batumike et al. 2007). The two diamictites
are developed over a very extensive area, covering at least
65 000 km2, occurring roughly in the region between latitudes 88
and 138S, and longitudes 248 and 298E (Figs 12.2 & 12.3).
Structural and stratigraphic framework
The Katanga Supergroup was deposited on a basement comprising
the Palaeoproterozoic Lufubu Metamorphic Complex and the
Bangweulu Block (Rainaud et al. 2005a; De Waele et al. 2006),
and the Mesoproterozoic Kibaran Belt (Kokonyangi et al. 2006).
Traditionally, the Katanga Supergroup has been subdivided into
the Roan, Lower and Upper Kundelungu Groups (Cailteux et al.
1994; Kampunzu & Cailteux 1999). According to the recently
revised stratigraphy (Wendorff 2003, 2005b, c; Wendorff & Key
2009), the Katanga Supergroup is subdivided into the following
five groups: Roan (basal group), Nguba, Kundelungu, Fungurume
and Plateau (Fig. 12.2). Not all researchers in the region have
accepted the revised stratigraphy (e.g. Cailteux et al. 2007;
Batumike et al. 2006, 2007).
The Roan and Nguba groups record two distinct rifting stages
resulting from early Neoproterozoic extension within Rodinia.
The Kundelungu, Fungurume and Plateau groups were deposited
in the succeeding foreland basins related to collision of the
Congo and Kalahari cratons during Gondwana assembly. The
Grand Conglomerat Formation occurs within the Nguba Group
and records syn-rift glaciation after initial basaltic volcanism.
The succeeding Petit Conglomerat Formation formed after the
first orogenic event to affect the southern part of the Lufilian belt
in what is now Zambia.

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 173– 184. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.12
174 S. MASTER & M. WENDORFF
Fig. 12.1. Position of the Lufilian arc in the Pan-African orogenic belts system
in central and southern Africa. Box outlines the area shown in Fig. 12.2.
The Roan Group (Fig. 12.4a), deposited in the first Katangan rift
basin (,880 Ma; Armstrong et al. 2005), nonconformably over-
lies variably eroded pre-Katangan basement and forms a continu-
ous transgressive succession from terrigenous clastic sediments
Fig. 12.2. Regional geology of the Lufilian belt (modified from Porada 1989; Porada & Berhorst 2000; Wendorff 2003). Localities discussed in the text: L, Lwaio
Mission; M, Mwinilunga; Ks, Kansanshi; MF, Musonda Falls; Ma, Mansa; M-K, Makonga-Kibambale. 1– 10, positions of stratigraphic logs shown in Fig. 12.5.
(siliciclastic unit, or the Mindola Subgroup) at the base, grading
upwards into a mixed association of siliciclastic and carbonate
strata, the Kitwe Subgroup, succeeded by a carbonate platform
sequence, the Bancroft Subgroup, which prograded from the
south (Binda 1994).
Major uplift (765 Ma) in the southern part of the Roan rift
basin (Zambia) terminated deposition of the Roan Group platform
carbonates and led to the opening of the Nguba rift (Wendorff
2005b). Wendorff (2005b, c) has interpreted the breccias of the
Mufulira Formation (lower Mwashya) as syn-rift olistostromes,
derived from uplifted Roan strata as products of mass-wasting
and deposited by sediment-gravity flows at the base of the
Mwashya Subgroup. Northward expansion of the Nguba rift,
beyond the northern margin of the Roan rift, resulted in prograda-
tion of the olistostromes and their nonconformable deposition
upon pre-Katangan basement in what is now the fold-thrust
belt region in the Democratic Republic of Congo (DRC). The suc-
ceeding Mwashya strata are composed of terrigenous siliciclastic
rocks, silicified oolitic/pisolitic grainstones, algal dolomites and
ironstones (middle Mwashya), overlain by shales with siltstones
grading upwards to black shales deposited under increasingly
anoxic conditions (upper Mwashya). Basaltic volcaniclastic
rocks and lavas mostly form subordinate interbeds in the
Mwashya Subgroup, but are locally significant. For example, a
broad belt of basaltic lava can be traced in a NE direction
from eastern Angola across NW Zambia and into southern DRC
(Unrug 1987, 1988; Kampunzu et al. 1993, 2000; Tembo et al.
1999; Key et al. 2002). In the centre of the basin, the upper
 NEOPROTEROZOIC GLACIOGENIC DIAMICTITES OF THE KATANGA SUPERGROUP 175

Fig. 12.3. Distribution of the Grand

Conglomerat and Petit Conglomerat
diamictites in Katanga (Democratic
Republic of Congo) and in adjacent areas of
Zambia (modified after Cahen &
Lepersonne 1979).

Fig. 12.4. (a) Simplified version of revised stratigraphy of the Katangan succession based upon syntectonic conglomerate complexes and unconformity-bounded
megasequences (Wendorff 2005b; Wendorff & Key 2009). Ages after Key et al. (2002) and Master et al. (2005), and references therein. Wavy lines indicate major
unconformities. (b) Previous/traditional subdivisions of the Katangan and lithostratigraphic correlation between the DRC and Zambia (simplified from Cailteux et al.
1994). Ore horizons: *Cu in Lower Roan in Zambia; **Cu– Co in the Mines Group in the DRC. Glaciogenic horizons: GC, Grand Conglomerat; PC, Petit Conglomerat.
Note that the Mines Group in the DRC consists of allochthonous megablocks, which resulted from dismemberment of nappes/thrust sheets with the Roan Group strata
and gravity-driven emplacement of the mineralized Roan megablocks into the foreland basin (the nappes originated in the south of the orogen (now Zambia) and were
thrust towards the foreland region in the north (today’s DRC) during the Pan-African orogenesis). Therefore the Mines Group does not exist as an individual stratigraphic
unit (Wendorff 2003, 2005b), and the megablocks are embedded in the Fungurume Group strata, as shown in (a).
176 S. MASTER & M. WENDORFF
boundary of the Mwashya Subgroup is transitional into the Grand
Conglomerat (Binda & Van Eden 1972; Bodiselitsch et al. 2005)
and is defined by the appearance of dropstones in massive or lami-
nated black shales typical of the underlying upper division of
Mwashya Subgroup. Mafic igneous rocks that occur within the
Roan, Mwashya and Grand Conglomerat have been interpreted
as indicators of rifting continuing throughout the deposition of the
Roan and Nguba groups (Tembo et al. 1999; Kampunzu et al.
2000). The northward propagation of the rifting is attributed
to the formation of the Kundelungu Aulacogen, which was
filled with up to 7 km of Neoproterozoic sediments, including the
two glacial diamictites, the Grand and Petit Conglomerates
(Dumont & Hanon 1997; Master 2007). Sedimentary strata that
overlie the Grand Conglomerat include the Kakontwe Limestone
Formation. Haematitic and jaspilitic banded Fe-formations are
known to occur both in the Mwashya Subgroup below the Grand
Conglomerat and in Nguba Group strata above it (Robb et al.
2003, 2004).
Inversion from an extensional to a compressional tectonic
regime occurred in the south of the Lufilian belt after the depo-
sition of the Grand Conglomerat but before the Petit Conglomerat
was deposited (Wendorff 2005b). West of the Kafue Anticline in
Zambia, the Petit Conglomerat rests unconformably on a folded
succession of the Roan-Nguba groups (Wendorff & Key 2009).
It defines the base of the Kundelungu Group, which is an infill of
the first foreland basin (Wendorff 2005b), itself superposed on
the former Kundelungu aulacogen (Master 2007), and is composed
of marine sandy shales, shales and dolomites, with thick proximal
conglomerate complexes occurring in the south of the Lufilian belt.
The Petit Conglomerat Formation is overlain by a pink dolomite
(the Calcaire Rose).
The succeeding Fungurume Group fills the second foreland
basin formed in the northern part of the fold-thrust region of the
Lufilian belt in the DRC (Wendorff 2003, 2005b) and contains
synorogenic conglomerates and megablocks derived from nappes
composed of the older Katangan strata uplifted to the south and
thrust northwards (compare Fig. 12.4a, b). The lower boundary
is unconformable on folded Kundelungu strata, and the succession
evolves from sedimentary olistostromes through transitional
shallow marine and continental redbeds to shallow marine
sequences of siliciclastic and carbonate strata. The degree of defor-
mation gradually decreases between the northern marginal part of
the external fold-thrust belt and the succeeding undeformed
Plateau Group (also known as the Biano Group) further to the
north. Continental arkoses and shales of the latter unit were depos-
ited in the youngest foreland basin, extending to the north of
the Fungurume Group foreland (Figs 12.2 & 12.4). 40Ar/39Ar
dating of detrital muscovite from the Plateau Group shows that
these sedimentary rocks were deposited after 573 + 5 Ma
(Master et al. 2005).
The deformation of the Katanga Supergroup to form the Lufilian
arc occurred during the collision of the Kalahari and Congo
cratons, in the time interval from c. 590 to c. 512 Ma, followed
by a protracted period of post-orogenic uplift and cooling extend-
ing to c. 483 Ma (Porada & Berhorst 2000; John et al. 2003, 2004;
Rainaud et al. 2005b). During the Lufilian Orogeny, up to 150 km
of crustal shortening occurred (Porada & Berhorst 2000; Jackson
et al. 2003). Thick-skinned thrusting resulted in the formation of
basement-cored domes in the Domes Region (Daly et al. 1984)
(Fig. 12.2), while thin-skinned thrusting resulted in the external
fold-thrust belt (Jackson et al. 2004). Former Roan rift basins suf-
fered tectonic inversion during basin deformation (Selley et al.
2005). In the folded Lufilian fold-thrust belt, the metamorphic
overprint ranges from greenschist to amphibolite grade, with
local development of talc-kyanite whiteschists (John et al. 2004),
and the diamictites are deformed and cleaved, with recrystalliza-
tion of matrix constituents to biotite-bearing schists (Gysin 1934;
Key et al. 2002; Broughton et al. 2002; Master et al. 2005;
Rainaud et al. 2005b). However, in the external parts of the
Lufilian arc, and in the northern parts of the Katangan basin, the
metamorphic grade is very low, and sedimentary structures are
very well preserved (Master et al. 2005).
Glaciogenic deposits and associated strata
Grand Conglomerat Formation, Nguba Group
The diamictite of the Grand Conglomerat has long been interpreted
as glaciogenic (Cahen 1978). The evidence for a glacial origin rests
on the common and widespread occurrence (over an area of around
65 000 km2) of thick (up to a maximum of 1200 m) polymictic
deposits containing subrounded to subangular faceted clasts
(ranging from gravel-sized fragments to boulders .1 m3 in
volume) with striations, sometimes in multiple sets (in more than
20 localities); the generally massive, poorly sorted nature of the dia-
mictite with a fine-grained matrix supporting the coarser clasts; and
the presence of associated planar-laminated shales with dropstones
(van Doorninck 1928; vanden Brande 1936; Cahen 1954, 1963,
1978; François 1973; Binda & van Eden 1972; Dumont & Cahen
1977; Wendorff & Key 2009). Striated clasts have been described
by many authors and illustrated by Stutzer (1911), Studt (1913),
Grosse (1918), Robert (1946) and Cahen (1963).
The clasts in the Grand Conglomerat diamictite are overwhel-
mingly composed of quartzite and vein quartz, but also include a
variety of other lithologies, such as granites, gneisses, mica-schists,
porphyries and basic rocks (van Doorninck 1928; Robert 1940a;
Master et al. 2005). François (1973), working in the western end
of the Lufilian arc, recorded clasts in the Grand Conglomerat
made up predominantly (.75%) of quartzites (fine-grained,
coarse-grained, rarely micro-conglomeratic) derived from the
Kibaran Belt. In the Makonga and Kibambale regions (Fig. 12.3)
in the NW Katangan Basin, adjacent to the Kibaran Belt, Dumont
& Cahen (1977) recorded clasts of quartzites, biotite granite, phyl-
lite, graphic pegmatite, tourmalinite, quartz, calcic marbles, jasper,
carnelian agate, black silicified pisolites, black chert, amygdaloidal
spilitic lavas and dolerite. Clasts made of oolitic cherts were
recorded by Cornet (1897), Studt (1913), Stutzer (1913b), Guille-
main (1913), Grosse (1918) and Dumont & Cahen (1977),
especially in the vicinity of Mwashya in the northern part of the
Lufilian arc.
In the central Lufilian arc, most clasts have rounded edges, but
completely rounded clasts are absent (van Doorninck, 1928). In
the Luapula beds, clasts in clast-supported cobble and boulder
conglomerates are well rounded, but clasts in matrix-supported
poorly sorted pebbly sandstones and conglomerates are subangular
to subrounded (Abraham 1959). A study of clast shapes in the
Grand Conglomerat diamictites from two localities in the Likasi
district was made by Museu (1987), who found that the quartzite
clasts (c. 70% of the population) show a very poor sorting, but
the median clast size diminished from NE towards the SW.
More than 80% of the quartzite clasts were rounded or subrounded,
and a flattening index showed values typical for torrential environ-
ments; hence these diamictites were interpreted by Museu (1987)
to have been deposited in fluviatile or glaciofluvial environments.
Petrographic studies of the Grand Conglomerat were initiated
by Beck and Wagner, as reported by Stutzer (1911). Grosse
(1918) and van Doorninck (1928) described the tillite as having
a matrix of extremely fine-grained, ash-grey, almost opaque
clayey material, with small (up to 0.75 mm), splinters and more
rounded fragments of predominantly quartz, with sporadic rare
fine-grained fragments of granite, graphic granite, quartz por-
phyry, feldspars, muscovite, chlorite, quartzite, magnetite, horn-
blende and other minerals. The petrography of the Grand
Conglomerat at Kipushi indicates that it contains numerous
clasts derived from a mixed plutonic (granitic and amphibolitic)
and metavolcanic (quartz porphyry schist) terrain, such as the
Lufubu Metamorphic Complex (Master et al. 2005). Detailed
 NEOPROTEROZOIC GLACIOGENIC DIAMICTITES OF THE KATANGA SUPERGROUP
modern petrographic studies of the Grand Conglomerat diamictites
have found microfabrics and microstructures in these diamictites
that are comparable with those found in modern glaciogenic tills
(Delpomdor 2007; Delpomdor et al. 2008).
Facies in the Grand Conglomerat
The Grand Conglomerat sediments represent a broad range of con-
tinental, shallow marine and deep marine facies controlled by the
sedimentary environment, ongoing tectonic evolution of the
depository, climatic variations and changes in ice dynamics.
177
Proximal facies of a .400-m-thick gravelly fan-delta succes-
sion at Mushishima, west of the Kafue Anticline in Zambia
(Figs 12.5 & 12.6), contain two interlayers of the glaciogenic sedi-
ments (Wendorff & Key 2009). The fan-delta conglomerates are
massive, matrix-supported and form thick amalgamated beds
that locally contain dish structures testifying to rapid deposition
and early post-depositional fluid escape. Clast-supported conglom-
erate layers are rare. Together with pebbles, cobbles and solitary
boulders of quartzite, arkose and siltstone, there occur clasts of car-
bonate rocks derived from the Roan Group.
A more distal facies association of a proglacial fan delta occurs
at Kansanshi to the west (Fig. 12.2), where a c. 200-m-thick
Fig. 12.5. Lithostratigraphic sections
discussed in this chapter (from Wendorff &
Key 2009, and references therein). Sections
2– 5 are modified from Wendorff (2005b),
section 6 from Andersen & Unrug (1984),
section 7 after Key et al. (2002), sections 8
and 9 from François & Cailteux (1981), and
sections 10 and 11 after Dumont & Cahen
(1977). The thicknesses of
lithostratigraphic units in section 7 are
approximate due to variable dips and
poor exposure.
178 S. MASTER & M. WENDORFF
glaciogenic succession is composed of coarse glacial rain-out
(dropstones up to 10 cm across) of Roan and Mwashya derivation
and muddy-silty suspension deposits laid down simultaneously
with intermittent sediment-gravity flows ranging from gravelly
debris flows to sandy turbidites, and associated with subordinate
traction currents (Wendorff & Key 2009). Bed thickness usually
ranges from a few centimetres to a few tens of centimetres. There
are also rare sequences of c. 1-m-thick matrix- to clasts-supported
conglomerate beds grading upwards to intervals of pebbly mud-
stone over 1 m thick, which may represent very thick pebbly turbi-
dites/megaturbidites (Postma et al. 1988). Reworking of bottom
sediment by intermittent weak traction currents is expressed by
solitary silty current ripplemarks embedded in massive or lami-
nated mudstone beds. Dropstones may occur in any of these
sedimentary facies; their origin as ice-rafted melt-out debris is
especially obvious when they occur as outsized clasts loaded into
very thin beds and finely laminated siltstones and mudstones, the
clast diameter often exceeding the thickness of the encompassing
strata. The textural and structural sedimentary features of this
facies association suggest subaqueous deposition as a proglacial
fan delta or apron (Eyles & Eyles 1992; Wendorff & Key 2009).
The facies associations and regional facies gradients between
Mushishima and Kansanshi suggest the presence of a prominently
uplifted southern margin of the Nguba rift in what is now Zambia
(Wendorff 2005b). This uplifted margin is thought to have been
the source area supplying the basin adjacent to the north with
large amounts of detritus produced by glacial erosion of the
Roan and Mwashya lithologies (Wendorff & Key 2009).
In the Chambishi Basin, in borehole MJZC/9, the 26-m-thick
Grand Conglomerat is conformable upon black shales and
Fig. 12.6. Schematic SW to NE
cross-sections through the Katangan Basin
during deposition of the Grand
Conglomerat, showing the Nguba Rift
superposed on the earlier Roan Rift (see
Fig. 12.2 for locations). (a) Eastern region –
Kafue Anticline and Bangweulu Block
(Fig. 12.5), from Mushishima to
Mansa-Lueba on the Luapula River. (b)
Western region – Zambia and DRC
(Fig. 12.5), from Lwaio-Mwinilunga to
Kibambale-Makonga.
turbidites (109-m-thick) of the underlying Mwashya Subgroup.
Diamictites of the Grand Conglomerat are interbedded with turbi-
dites, and are interpreted to have formed by sediment-gravity flow
processes in a glaciomarine basin (Master et al. 2005).
In the Itawa area near Ndola, Zambian Copperbelt, there are
great thickness variations of the glaciomarine facies of the Grand
Conglomerat (Binda & Van Eden 1972). On an east –west section
line the thickness varies from 6 m to a maximum of 67 m, and back
to 21 m, over a distance of just 1.5 km. Isopach contours of the
Grand Conglomerat, based on its thickness in the 14 boreholes
given in fig. 1 of Binda & Van Eden (1972), indicate the existence
of a .5-km-long, c. 1-km-wide, trough-like feature trending
SSE –NNW, which may have been a submarine canyon. In
borehole section IT 28 at Itawa (Fig. 12.5, Section 5) the Grand
Conglomerat is c. 150 m thick and contains subangular to sub-
rounded rain-out clasts varying from granules to boulders of silici-
clastic and dolomitic sedimentary rocks derived from the
Roan-Mwashya succession. Characteristic for this section is a
rain-out diamictite composed of grey and black silty and sandy
mudstone often modified by remobilization and redeposition
within the basin as debris flows (Wendorff & Key 2009). Numer-
ous slump-folds (Binda & Van Eden 1972) and slump-generated
debris flows that occur in both dropstone-rich and dropstone-
devoid intervals suggest deposition on a palaeoslope unstable
because of rapid deposition or syndepositional tectonic move-
ments of the basin floor, or both processes combined (Eyles &
Januszczak 2004; Wendorff & Key 2009). The uppermost part
of the Grand Conglomerat Fm. at Itawa is a C –U sequence
from laminite to thin turbidites representing Ta,e and Tc,e
sequences of Bouma intervals. It is interpreted as distal facies
 NEOPROTEROZOIC GLACIOGENIC DIAMICTITES OF THE KATANGA SUPERGROUP
deposited in a non-ice contact region of the basin (Wendorff &
Key 2009). A 26-m-thick massive diamictite represents the
Grand Conglomerat Fm. in borehole Ks 17 at Mokambo, to the
east of Itawa.
At the NE margin of the depository, in the Luapula River valley
(Bangweulu Block region; Fig. 12.5, Section 6), glaciofluvial
sandstone and pebbly sandstone with large-scale cross-bedded
sets, indicating palaeocurrent flow towards the SW and south, is
interbedded within a massive diamictite.
In the NW part of the external fold-thrust belt in Zambia
between Mwinilunga and Lwaio Mission (Fig. 12.5, Section 7;
Fig. 12.6), a sequence of marine conglomerates, laminites and
dropstone-bearing siltstones and shales occurs. The clasts rep-
resent basement granite and gneiss and fragments derived from
the Roan units (Wendorff & Key 2009).
At Lufunfu, in the NW part of the Lufilian Belt in the DRC
(Fig. 12.5, Section 8; Fig. 12.6), the glaciomarine massive diamic-
tite with a carbonaceous, slightly calcareous mudstone matrix
reaches a thickness of 950 m (François 1973). The thickness of
the glaciogenic strata decreases southwards, as dominant clast
size decreases from pebble to granule and a conglomerate and
several sandstone interbeds wedge out. These facies trends indi-
cate that the source of the glaciogenic material supplied to the
northern part of the basin was located to the north of the Nguba
rift. At Tombolo, some 100 km north of Lufunfu (Fig. 12.5,
Section 9; Fig. 12.6), the diamictite reaches a maximum thickness
of 1200 m.
In the Makonga-Kibambale area located within the Plateau
region, which was a graben/aulacogen basin during deposition of
the Nguba Group (Unrug 1987; Wendorff 2005b; Master 2007;
Fig. 12.3, Fig. 12.5 (Sections 10 and 11), Fig. 12.6), c. 150 km
north of the Lufilian fold-thrust belt, the Grand Conglomerat is rep-
resented by a massive diamictite with interbeds of coarse-grained,
cross-bedded, feldspathic wacke (Dumont & Cahen 1977). Faceted
and striated pebbles and cobbles of igneous and metamorphic rocks
are derived from the Kibaran basement to the north and clasts of
sedimentary rocks originated locally by scouring of the underlying
Mwashya Subgroup. The glacial strata are intercalated with and
overlain by basaltic pillow lavas and volcanic breccias, and
intruded by dolerite sills. The pillow lavas imply a generally sub-
aqueous environment, but it is not certain whether the glacial depo-
sition between Makonga and Kibambale occurred in a marine or
continental setting (Dumont & Cahen 1977).
In summary, the nature of the sedimentary facies and their trends
suggest that the Grand Conglomerat glaciogenic sediments were
derived from both the southern and northern margin of the basin.
Stratigraphy in the southern region (at the southern margin of the
Nguba rift) is characterized by interbeds of the glacial sediments
within a coarse-clastic, high-energy fan-delta association and
abundance of clasts derived from the Roan and Mwashya units
(Fig. 12.6). The presence of the fan-delta facies both below and
above the glaciogenic strata, and the deposits of a proglacial fan
delta, imply a strong uplift of the southern shoulder of the
Nguba rift. This is consistent with the existence of a possible sub-
marine channel and considered here as the reason why the Grand
Conglomerat in the southern region has characteristics of a deep
marine deposit adjacent to a steep palaeoslope, for example, a
considerable proportion of mass-flow phenomena, ranging
from coarse clastic debris flows to megaturbidites to small-scale
low-density turbidites, as well as abundant slump beds and
slump-initiated debris flows composed entirely of fine-grained
intrabasinal components.
At the northern margin, on the other hand, the continental gla-
ciogenic deposits grade southward towards the open marine
basin into glaciomarine melt-out sediments (Fig. 12.6). These
interfinger laterally with, and pass upwards into, proximal sand-
stone interbeds of marginal marine origin. The proximal, postgla-
cial clastic (terrigenous) facies become progressively finer within
the marine basin further to the south, and in the distal regions of
179
the basin, grade into carbonates of the Kakontwe Limestone
Formation. This pattern suggests that the marginal marine strata
originated in a low-topography graded shelf during the syn-
and immediately post-glacial period. On the other hand, the
maximum thicknesses of the Grand Conglomerat in the NW
region of the Nguba rift at Tombolo suggest subsidence rates
three times higher than in the south and several times higher
than in the rifted margin adjacent to the north. This situation
suggests extremely high sedimentation rates, which must have
balanced the high subsidence to maintain a low-gradient palaeo-
topography of the shelf. Such high sedimentation rates may be
explained by extremely effective supply of the clastic material
during deglaciation, derived from the Kibaran Belt further to the
NW, where considerable topographic relief remained (Cahen &
Lepersonne 1979).
Petit Conglomerat Formation, Kundelungu Group
The Petit Conglomerat Formation forms the base of the Kunde-
lungu Group, and may overlie the Nguba Group with an erosional
unconformity (Wendorff 2003). The Petit Conglomerat diamicitite
is, like the Grand Conglomerat, thought to be of glacial origin,
based on the abundant and widespread presence of faceted and
striated clasts of both intrabasinal and extrabasinal origin
(Vanden Brande 1936; Cahen 1978). It is overlain consistently
by a carbonate unit, the ‘Calcaire Rose’ or ‘Dolomie Rose de
Lusele’, which consists of a 5–10-m-thick finely and regularly
bedded pink dolomite (François 1973; Dumont & Cahen 1977).
The Petit Conglomerat Formation has also been called the Kayan-
damu Formation (Ku1.1; Batumike et al. 2007).
Cahen (1978) distinguished two facies in the Petit Conglomerat:
a southern diamictite facies with small (,2 cm) clasts, and a north-
ern mixed diamictite and conglomerate facies, with large clasts
(up to 1 m granite clasts described by Grosemans 1935) of
varied compositions. The thickness of the Petit Conglomerat also
varies from north to south: it is thickest in the north central part
of the Katangan basin (up to 80 m in the Lukafu area; Vanden
Brande 1936). In the northern part of the western Lufilian Arc,
François (1973) records its thickness varying from 50 to 35 m,
while in the southern part of the same area, its thickness varies
from 40 to 35 m. In the Kipushi area further southeast, it is only
24 m (Master et al. 2005). The Petit Conglomerat formation
thins towards the northwestern margin of the Katangan basin: it
is 25 m thick in the Makonga area, adjacent to the Kibaran Belt.
On the northeastern margin of the basin, where they onlap
against the basement rocks of the Bangweulu Block, the equivalent
rocks of the Petit Conglomerat Formation in the Luapula Beds are
up to 100 m thick, but here they consist of a series of alternating
conglomerates and pebbly sandstones (Abraham 1959). The con-
glomerates have a coarse, poorly sorted angular to subangular
arkosic sandy matrix, with well-rounded cobbles and boulders
(up to 40 cm in diameter) made of granites and rhyolite porphyries
from the Bangweulu Block basement, as well as quartzites and
shales from the Mporokoso Group. The pebbly sandstones are
trough crossbedded, having a poorly sorted subangular arkosic
matrix, with scattered angular to subangular pebbles, many of
which exhibit a triangular form and faceted appearance, similar
to glaciogenic clasts described by Von Engelin (1930). Because
of their textural and compositional immaturity, and the evidence
that the pebbles have undergone little transport, these beds were
interpreted by Abraham (1959) as periglacial outwash gravels.
Immature, feldspathic, pebbly sandstones, which are interbedded
within the conglomerate complex are trough crossbedded with
ripple marks and peculiar cracks with polygonal outlines in plan
view, resemble patterned ground typical of periglacial regions
(Washburn 1969). These were interpreted by Daily & Cooper
(1976) to be a relict three-dimensional network of ice veins in a
permafrost environment.
180 S. MASTER & M. WENDORFF
In the northern regions, clasts recorded in the Petit Conglom-
erat consist of quartz, granites, basic rocks, agates, amygdaloidal
lavas, rhyolites, quartzites, siliceous oolites, sandstones and
shales, including many faceted and striated clasts (Grosemans
1935; Vanden Brande 1936; Batumike et al. 2006). Many of
the clasts originated from the adjacent Kibaran Belt to the
NW and from the Kibambale volcanic complex to the north
(Dumont & Cahen 1977). In the Kapulo area of NE Katanga,
just north of Lake Mweru, a distinctive and heterogeneous
suite of clast types (maximum diameter, 30 cm) includes quart-
zite, rhyolite, porphyries, alaskites and rare clasts of gneiss,
mica schists, metaconglomerates and pisolitic black cherts
(André 1976; Cahen 1978) derived from the adjacent Bang-
weulu Block to the east (Master et al. 2005). Recent petro-
graphic and geochemical studies on the Petit Conglomerat and
other sedimentary rocks of the Nguba Group (Batumike et al.
2006, 2007) support the north –south facies variations, and a
derivation from the Kibaran Belt and Bangweulu Block to the
NW and NE of the Katangan basin.
Petrographic studies of the Petit Conglomerat from Kipushi
Mine were made by Master et al. (2005). Here the Petit
Conglomerat consists of a fine-grained biotitic siltstone with a
few scattered clasts, averaging about 0.5 mm across, but ranging
up to a maximum size of 5 mm. The clasts consist of quartz, car-
bonate, shale, chert and altered orthoclase. These clasts are
mainly of intrabasinal derivation, with some contribution from
basement granitoids (orthoclase). The presence in this rock of acri-
tarchs (of planktonic origin), similar to acritarchs previously
described from Kundelungu beds, indicates that the rock was
deposited in glaciomarine environment (Master et al. 2005).
Recent detailed petrographic studies of the Petit Conglomerat
diamictites from the Kiaka anticline indicate that they show
microfabrics and microstructures comparable to those found
in modern glaciogenic tills (Delpomdor 2007; Delpomdor
et al. 2008).
Boundary relations with overlying and underlying
non-glacial units
The lower contact between the Grand Conglomerat and the
underlying strata of the Mwashya Subgroup varies from being
completely conformable, for example, in the Itawa area (Binda
& Van Eden 1972), in the central part of the Chambishi Basin
(Bodiselitsch et al. 2005) and in the central Lufilian arc (van
Doorninck 1928), to disconformable onto the lower or middle
Mwashya Subgroup in the western part of the Chambishi
Basin (Wendorff & Key 2009), in the northern Lufilian arc
(Robert 1940a), and in the Kibambale area (Dumont & Cahen
1977). In the Luapula Beds, the Grand Conglomerat rests non-
conformably on the crystalline granitic basement of the Bang-
weulu Block, and with an angular unconformity on metavolcanic
rocks of the Luapula Porphyries and quartzites of the Mporo-
koso Group (Thieme 1970, 1971; Andersen & Unrug 1984).
The Grand Conglomerat is overlain conformably by carbonate
strata of the Kakontwe Limestone Formation or by correlative
terrigenous ‘fines’ in the regions proximal to the south and
north margins of the Nguba Rift (Fig. 12.5).
The Petit Conglomerat Formation rests with a disconformity
or low-angle unconformity on upper Nguba Group strata in
the DRC (Robert 1940a) and with a pronounced angular unconfor-
mity on the strongly folded Nguba succession in Zambia
(Wendorff & Key 2009). In the Luapula Beds, it rests with a dis-
conformity or low-angle unconformity on the Nguba Group and
also transgressively onlaps onto the crystalline granitic basement,
which it overlies nonconformably. The Petit Conglomerat For-
mation is overlain conformably by carbonate of the Lusele For-
mation (Dolomie Rose), or by shales where the carbonates are
not developed.
Chemostratigraphy
Reconnaissance carbon and oxygen isotope studies of Katangan
carbonate rocks have been made by Master & Verhagen (unpub-
lished data, 1993), and by Bodiselitsch (2004). Carbon isotopic
values (d13C) in the Roan Group recorded by Master & Verhagen
(unpublished data, 1993) range between þ2.2 and þ4.2‰ Vienna
Pee Dee Belemnite (VPDB), while the values in the Mwashya Sub-
group show a negative trend with stratigraphic height, ranging
from þ2.8 to – 1.2‰ VPDB. A sample of carbonate from the
matrix of the Grand Conglomerat had a d13C value of –4.6‰
VPDB. In the Kakontwe carbonate overlying the Grand Conglom-
erat, the d13C values show a stratigraphic upward positive trend,
increasing from þ2.7 to þ5.5, while in the upper Nguba Group,
values are between –0.2 to þ1.8‰ VPDB (Master & Verhagen,
unpublished data, 1993). These isotopic values correlate well
with the secular trends for the composite Neoproterozoic d13C
record both preceding and following the Sturtian glaciation
(Halverson et al. 2005). The carbon and oxygen isotopic data of
Bodiselitsch (2004) remain unpublished; however, Bodiselitsch
et al. (2005) reported that d13C values in the Calcaire Rose and
other carbonate rocks overlying the Petit Conglomerat at Kipushi
(DRC) were consistently in the range of –2 to –4‰ VPDB.
Bodiselitsch et al. (2005) obtained detailed major and trace
element geochemical profiles from drill cores that intersect the
Grand and Petit Conglomerat diamictites and overlying carbonate
rocks. They found significant Ir anomalies at or near the base of the
carbonate rocks immediately overlying both the Grand Conglom-
erat and Petit Conglomerat diamictites at Kipushi, and also above
the Grand Conglomerat at Chambishi. They interpreted the Ir
anomaly to be due to platinum group elements (PGE)-enriched
meteoritic debris (mainly in the form of interplanetary dust par-
ticles, or IDPs), which had accumulated on ice caps on a frozen
earth during prolonged Cryogenian glaciations (i.e. during the
Snowball Earth model of Kirschvink 1992; Hoffman et al.
1998), and which were released into the sediment during melting
of the ice. They used the magnitude of the sharp Ir anomaly
above the Petit Conglomerat diamictite to estimate a minimum
time of 12 Ma for the duration of the ice cap, assuming a meteoritic
flux similar to that which prevailed in the later Phanerozoic.
Other characteristics
Economic deposits
The Katangan glaciogenic diamictites and associated over- and
underlying strata contain several economic mineral deposits,
some of which are currently being exploited. Within the
Mwashya strata below the Grand Conglomerat diamictite, there
are two horizons of haematite and jaspilite, some outcrops of
which have been mined for iron (Jamotte 1947). A mafic pyroclas-
tic horizon within the Mwashya Subgroup is the host rock for stra-
tabound copper – cobalt mineralization at the Shituru Mine
in Katanga (Lefebvre 1974; Cailteux et al. 2007). The Grand Con-
glomerat diamictite itself is mineralized with disseminated copper
sulphides and oxides in a few places in DRC: at Tombolo, NNE of
Kolwezi; at the Lonshi mine in the Pedicle of Katanga and at
Kamoa, 20 km west of Kolwezi, where a giant high-grade strati-
form orebody was discovered in 2009. The periglacial conglomer-
ates associated with the Grand Conglomerat in the Luapula Beds of
northern Zambia contain disseminated copper mineralization in
several places (Thieme 1970). The Kakontwe carbonate, which
overlies the Grand Conglomerat, is mineralized with epigenetic
breccia-fill Zn –Pb– Cu –Ge –Ga mineralization at the Kipushi
Mine, DRC, as well as in the much smaller Kengere and Lombe
deposits, which were exploited in the past (see review in Batumike
et al. 2007). There is also a stratiform Cu –Co deposit, Tantara,
consisting primarily of secondary copper carbonates and silicates,
 NEOPROTEROZOIC GLACIOGENIC DIAMICTITES OF THE KATANGA SUPERGROUP
which is hosted by this stratigraphic unit (Reintjens 1935). In
South Katanga, near the Zambian border, there are important
iron deposits hosted by the Kakontwe limestone. The giant Kan-
sanshi Cu –Au deposit in northern Zambia is hosted by flat-lying
carbonates and schists of the Kakontwe Formation (Broughton
et al. 2002). Ba –Fe mineralization occurs in the Kapumba
deposit NW of Likasi, DRC, where barite-specularite veins cut
the Grand Conglomerat and overlying dolomites and shales of
the Nguba Group (Intiomale & Mbuyi 1997); these appear
similar to the barite veins cutting cap carbonates in Mauritania
(Shields et al. 2007). The Lusele or Calcaire Rose carbonate
above the Petit Conglomerat diamictite contains copper mineraliz-
ation at the Sokoroshi I and II deposits (Reintjens 1935). The
Lusele carbonate in the Luapula Beds contains Pb–Zn mineraliz-
ation (Thieme 1970). The metallic mineralizations associated with
the Neoproterozoic glacial intervals in the Katangan and elsewhere
have been speculatively linked to global anoxia, and the generation
of reduced metalliferous fluids from associated reduced strata
(Robb et al. 2003, 2004).
Characteristics of carbonate rocks overlying
the diamictites
The Kakontwe dolostone overlying the Grand Conglomerat con-
tains several very distinctive features. At its base, where it is
locally referred to as the ‘serie tigré’, it has a laminated appearance,
resembling algal lamination, and at Kipushi it also shows roll-up
structures identical to those which have been described from the
base of the Rasthof carbonate overlying the Chuos diamictite in
Namibia (Hoffman et al. 1998). These basal carbonates also com-
monly show soft-sediment deformation features, such as convolute
bedding and isoclinal folds, which have been observed at Kipushi
and at Musoshi (Master, unpublished data). These structures
resemble similar features seen in cap carbonates overlying Neopro-
terozoic glacial diamictites in Brazil (Nogueira et al. 2003) and in
Mauritania (Master 2004, unpublished data; Shields et al. 2007).
Palaeolatitude and palaeogeography
Meert (2003) inferred that at c. 750 Ma, the present-day southern
part of the Congo Craton was at palaeolatitudes of between 5
and 108S, based on a palaeopole for the Mbozi Complex in
southern Tanzania, for which a SHRIMP U –Pb zircon age of
748 + 6 Ma has been obtained (Mbede et al. 2004). Wingate
et al. (2010) obtained palaeomagnetic data on the 800-m-thick
765 + 5 Ma (SHRIMP U –Pb zircon age) Luakela volcanics of
the Lwavu Formation, Mwashya Subgroup (Key et al. 2002),
based on alternating field (AF) and thermal demagnetization of
65 samples from nine sites. Of three components found, the one
component, carried by magnetite and regarded as primary,
coincides with the Mbozi palaeopole obtained by Meert et al.
(1995). Because the overlying Grand Conglomerat diamictite is
bracketed in age between 765 + 5 and 735 + 5 Ma (see below),
it was most likely deposited in near-equatorial low latitudes.
There is abundant palaeogeographic evidence from around the
Congo Craton, in the form of extensive stromatolitic carbonate
platforms, that it was situated in tropical palaeolatitudes during
much of the Neoproterozoic Era (Verbeek 1970; Cahen 1973b,
1982; Alvarez 1993; Evans 2000). Thus although there are no
direct palaeomagnetic studies of the Petit Conglomerat, it was
also likely deposited in tropical palaeolatitudes.
Geochronological constraints
Key et al. (2002) obtained a SHRIMP U –Pb zircon crystallization
age of 765 + 5 Ma for the Luakela lavas from the Lwavu
181
Formation, Mwashya Subgroup. This age provides a maximum
age constraint on the overlying diamictites correlated with the
Grand Conglomerat Formation. Key et al. (2002) also obtained a
SHRIMP U –Pb zircon age of 735 + 5 Ma from a volcanic
breccia within strongly deformed strata at Mwinilunga (M in
Fig. 12.2, & Fig. 12.5, Section 7), interpreted as being part of
the Nguba beds overlying the Grand Conglomerat. Although its
stratigraphic position is not entirely certain, the dated volcanic
breccia is spatially associated with carbonate rocks that are not
found beneath the Grand Conglomerat in the undeformed succes-
sion immediately to the west (LM, Lwaio Mission in Fig. 12.2).
These carbonate rocks are correlated with the Kakontwe Lime-
stone Formation (Wendorff & Key 2009). It is therefore concluded
that the Grand Conglomerat was most likely deposited prior to
735 + 5 Ma and definitely after the 765 + 5 Ma Mwashya vol-
canism. The youngest detrital zircons from the Grand Conglomerat
diamictite at Kipushi Mine, DRC, yielded an imprecise age of
729 + 50 Ma (Master et al. 2005), consistent with the above
age constraints.
From the available radiometric data, the age of the Petit Con-
glomerat is not yet well constrained, and is only bracketed
between 735 + 5 Ma, the age of volcanics in the West Lunga For-
mation in the Nguba Group (Key et al. 2002), and c. 620 Ma, the
age of uraninites from veins in thrust zones that affect the Katan-
gan stratigraphy to the top of the Kundelungu Group (Cahen
1973a). However, on the basis of regional correlations and their
shared tectonic history as Neoproterozoic rift to passive margin
sequences on the southern edge of the Congo craton, the Petit
Conglomerat and the overlying Calcaire Rose may be equivalent
to the Ghaub diamictite and Maieberg Formation in the Otavi
Group of the Damara Orogen, Namibia, which are dated at
635.5 + 1.2 Ma (Hoffmann et al. 2004).
Palynological work on the Nguba Group, including laminated
dropstone-bearing shales of the Grand Conglomerat, was carried
out by Binda (1972a, b, 1977). Although several microfossil
forms were identified (including three new varieties of the genus-
form Fibularix and seven new types of sphaeromorph acritarch),
these had little biostratigraphic usefulness, except in indicating a
Neoproterozoic age. Vavrdova & Utting (1974) described actri-
tarchs from the Luapula beds, which they ascribed to a Lower
Palaeozoic age; however, their age attribution is also considered to
be unreliable, and of little stratigraphic significance (Binda 1977).
Discussion
No striated pavements have ever been observed beneath the two
Katangan diamictites thought to be of glacial origin, mainly
because the underlying rocks consisted largely of unconsolidated
sediments. On the basis of clast composition, and detrital zircon
ages, the glaciers appear to have fed into the Katangan Basin
from the flanking basement regions to the NW (Kibaran Belt)
and to the NE (Bangweulu Block) (Master et al. 2005). Palaeocur-
rent data from associated periglacial (glaciofluvial) sandstones
in the Luapula Beds of north Zambia indicate palaeocurrents
trending towards the SW. Distinctive feldspar-phyric porphyritic
granitoid clasts within the Grand Conglomerat in Kipushi district
are glacial indicators that can be traced to a source in the Luina
Dome 150 km away, indicating a WSW pre-tectonic transport
direction for these glacial erratics (Master et al. 2005). Possible
submarine channels within the glaciomarine facies of the Grand
Conglomerat in the Itawa area of Zambia trend NNW. The distri-
bution and thickness variations of facies, and indications of source
areas, show that transport of the glaciers was globally generally
north to south, with many local fluctuations.
The size distribution and nature of clasts in the Petit Conglom-
erat indicate a north –south transport direction, corresponding to
the diminution in size and abundance of extrabasinal clasts
(derived from the Kibaran Belt and the Bangweulu Block).
182 S. MASTER & M. WENDORFF
However, a facies gradient and clast composition in the syn-glacial
fan-delta facies to the west of the Kafue Anticline in Zambia
suggest clast derivation and palaeotransport direction towards
the north, away form the uplifted southern shoulder of the
Roan-Nguba rift basin (Wendorff & Key 2009).
Conclusions
The Grand and Petit Conglomerat diamictites of the Katanga
Supergroup in Central Africa, which were deposited in a variety
of continental, shallow and deep marine facies, show abundant evi-
dence of a glaciogenic origin. This evidence includes the occur-
rence of (i) poorly sorted subangular to subrounded clasts of
intra- and extrabasinal origin, ranging in size from millimetric
grains to boulders with a volume of several cubic metres, sitting
in a fine-grained matrix; (ii) faceted and striated clasts from numer-
ous localities; (iii) microfabrics and microstructures diagnostic of
processes associated with modern tillites; and (iv) associated sedi-
mentary rocks indicative of glacial or periglacial conditions,
including laminated facies with dropstones, and sandstones with
‘patterned ground’ orthogonal crack networks. Their widespread
occurrence (over an area of .65 000 km2) and great thickness
(up to 1200 m) make the Katangan diamictites some of the most
important Neoproterozoic glaciogenic units. Despite their rela-
tively poor geochronological constraints, the Grand and Petit Con-
glomerat diamictites can be reasonably correlated, respectively,
with the Chuos and Ghaub glaciogenic diamictites of the Damara
orogenic belt in Namibia, especially considering similarities in
terms of C-isotope chemostratigraphy, textures in overlying car-
bonate units, and their common tectonic positions on the southern
rifted margin of the Congo Craton. The presence of the Katangan
diamictites within a major metallogenic district, the Central
African Copperbelt, has resulted in many drill cores being made
available through the diamictites and their over- and underlying
carbonate strata. These cores have facilitated the very detailed
geochemical studies that resulted in the discovery of Ir anomalies
in the carbonate units overlying the diamictites (Bodiselitsch
et al. 2005), which has contributed significantly to invigorating
the debate on the Snowball Earth model for Neoproterozoic
glaciations.
We thank P. Binda, R. Key, P. Hoffman and F. Delpomdor for discussions;
J. Cailteux, F. Mbuyi and B. Ngoie for guidance in the field; and J. Batumike
for reprints. We are indebted to R. Hanson, J. Arthurs and G. Halverson for
their comprehensive reviews, which have helped considerably in improving the
manuscript. We are grateful to G. Halverson for his invitation to produce this
review, and for his patience and encouragement during the editorial process.
This represents a contribution of the IUGS- and UNESCO-funded IGCP
(International Geoscience Programme) Project #512.
References
Abraham, D. 1959. The stratigraphical and structural relationship of the
Kundelungu System, Plateau Series and basement rocks in the Mid-
Luapula valley, Northern Rhodesia. D. Phil. thesis, University of
Leeds.
Alvarez, Ph. 1993. Un modèle de lagune d’âge Protérozoı̈que supérieur:
le Schisto-calcaire du Congo. Journal of African Earth Sciences, 17,
75 –87.
Andersen, L. S. & Unrug, R. 1984. Geodynamic evolution of the Bang-
weulu Block, northern Zambia. Precambrian Research, 25, 187– 212.
André, L. 1976. Etude aérophotomorphologiques et pétrographique du
Katangien de la mosaı̈que contrôlée de Kapulo au Shaba. Mémoire
de licence, Univ. Libre de Bruxelles, Belgium.
Armstrong, R. A., Master, S. & Robb, L. J. 2005. Geochronology of the
Nchanga Granite, and constraints on the maximum age of the Katanga
Supergroup, Zambian Copperbelt. Journal of African Earth Sciences,
42, 32 –40.
Batumike, M. J., Kampunzu, A. B. & Cailteux, J. L. H. 2006. Petrology
and geochemistry of the Neoproterozoic Nguba and Kundelungu
Groups, Katangan Supergroup, southeast Congo: implications for
provenance, palaeoweathering and geotectonic setting. Journal of
African Earth Sciences, 44, 97 – 115.
Batumike, M. J., Cailteux, J. L. H. & Kampunzu, A. B. 2007. Litho-
stratigraphy, basin development, base metal deposits, and regional
correlations of the Neoproterozoic Nguba and Kundelungu rock suc-
cessions, central African Copperbelt. Gondwana Research, 11,
432– 447.
Binda, P. L. 1972a. Preliminary observations on the palynology of the
Precambrian Katanga Sequence. Geologie en Mijnbouw, 51,
315– 319.
Binda, P. L. 1972b. Microfossils from the Lower Kundelungu (Late
Precambrian) of Zambia. 24th International Geological Congress,
Montreal, Section 1, Precambrian Geology, 179–186.
Binda, P. L. 1977. Microfossils from the Lower Kundelungu (Late
Precambrian) of Zambia. Precambrian Research, 4, 285–306.
Binda, P. L. 1994. Stratigraphy of Zambian Copperbelt orebodies.
Journal of African Earth Sciences, 19, 251– 264.
Binda, P. L. & Van Eden, J. G. 1972. Sedimentological evidence for the
origin of the Precambrian Great Conglomerate (Kundelungu Tillite),
Zambia. Palaeogeography, Palaeoclimatology, Palaeoecology, 12,
151– 168.
Bodiselitsch, B. 2004. Geochemical and stable isotope investigations on
‘Snowball Earth’ samples from the Lufilian tectonic arc, D. R. Congo
and Zambia, and on the late Eocene sediment samples from Mas-
signano, Italy. PhD thesis (unpublished), University of Vienna,
Vienna, Austria.
Bodiselitsch, B., Koeberl, C., Master, S. & Reimold, W. U. 2005.
Estimating duration and intensity of Neoproterozoic snowball glacia-
tions from Ir anomalies. Science, 308, 239–242.
Broughton, D. W., Hitzman, M. W. & Stephens, A. J. 2002.
Exploration history and geology of the Kansanshi Cu –(Au) deposit,
Zambia. Society of Economic Geologists, Special Publication, 9,
141– 153.
Cahen, L. 1954. Géologie du Congo Belge. Vaillant-Carmanne, Liège.
Cahen, L. 1963. Glaciations anciennes et dérive des continents. Annales
de la Société géologique de Belgique, 86, B19 –B84.
Cahen, L. 1973a. L’uraninite de 620 m.a. post-date tout le Katangien,
mise au point. Musée Royale d’Afrique Centrale, Tervuren (Belgi-
que), Département de Géologie et Minéralogie, Rapport Annuel,
1972, 35– 38.
Cahen, L. 1973b. Corrélations du certains séries du Précambrien supér-
ieur du Zaı̈re à la lumière de l’étude des stromatolithes et des données
géochronologie radiométrique. Musée Royale d’Afrique Centrale,
Tervuren (Belgique), Département de Géologie et Minéralogie,
Rapport Annuel, 1972, 38 – 51.
Cahen, L. 1978. Les mixtites anté-cambriennes de l’est du Zaı̈re: mise
au point interimaire. Musée Royale d’Afrique Centrale, Tervuren
(Belgique), Département de Géologie et Minéralogie, Rapport
Annuel, 1977, 33 –64.
Cahen, L. 1982. Geochronological correlation of the Late Precambrian
sequences on and around the stable zones of Equatorial Africa.
Precambrian Research, 18, 73 –86.
Cahen, L. & Lepersonne, J. 1979. Upper Proterozoic diamicitites
of Shaba (formerly Katanga) and neighbouring regions of
Zambia. In: Hambrey, M. J. & Harland, W. B. (eds) Earth’s
Pre-Pleistocene Glacial Record. Cambridge University Press,
Cambridge, 162–166.
Cailteux, J., Binda, P. L. et al. 1994. Lithostratİgraphical correlation of
the Neoproterozoic Roan Supergroup from Shaba (Zaire) and
Zambia, in the central African copper-cobalt metallogenic province.
Journal of African Earth Sciences, 19, 265– 278.
Cailteux, J. L. H., Kampunzu, A. B. & Lerouge, C. 2007. The Neopro-
terozoic Mwashya-Kansuki sedimentary rock succession in the
central African Copperbelt, its Cu –Co mineralisation, and regional
correlations. Gondwana Research, 11, 414–431.
Cornet, J. 1897. Observations sur le terrains ancien du Katanga faites au
cours de l’expédition Bia-Francqui (1891– 1893). Annales de la
Société Géologique de Belgique, 24, Mémoires, 1896–1897,
25 – 191.
 NEOPROTEROZOIC GLACIOGENIC DIAMICTITES OF THE KATANGA SUPERGROUP
Daily, B. & Cooper, M. R. 1976. Clastic wedges and patterned ground in
the Late Ordovician–Early Silurian tillites of South Africa. Sedimen-
tology, 23, 271– 227.
Daly, M. C., Chakraborty, S. K. et al. 1984. The Lufilian arc and
Irumide belt of Zambia: results of a traverse across their intersection.
Journal of African Earth Sciences, 4, 311– 318.
Delhaye, F. 1920. Les variations de facies du conglomérat inférieur
du Système du Kundelungu au Katanga. Annales de la Société
Géologique de Belgique, Publications relatives au Congo Belge et
aux régions voisins, 43, 1919–1920, 19– 27.
Delpomdor, F. 2007. Etude des dépôts diamictitiques du Néoprotérozoı̈-
que supérieure en République Démocratique du Congo et au
sud-ouest du Rwanda. Unpublished report, Musée Royal d’Afrique
Centrale, Tervuren, Belgium.
Delpomdor, F., Tack, L. & Préat, A. 2008. Microstructures in the Neo-
proterozoic tillites around the Congo River Basin (CRB), Democratic
Republic of the Congo (DRC) – comparison with the Karoo tillites
from the Dekese borehole in the CRB. Extended Abstract, 22nd
Colloqium of African Geology, 2 –6 November 2009, Hammamet,
Tunisia.
De Waele, B., Liégeois, J.-P., Nemchin, A. A. & Tembo, F. 2006. Isoto-
pic and geochemical evidence of Proterozoic episodic crustal rework-
ing within the Irumide belt of south-central Africa, the southern
metacratonic boundary of an Archaean Bangweulu Craton. Precam-
brian Research, 148, 225–256.
Dumont, P. & Cahen, L. 1977. Les complexes conglomeratiques de la
bordure sud-orientale de la Chaine Kibarienne et leurs relations
avec les couches Katangiennes de l’Arc Lufilienne, Rapport Annuelle
1977. Museé royal de l’Afrique centrale, Départment de Géologie et
Minéralogie, Tervuren, Belgique, 111–135.
Dumont, P. & Hanon, M. 1997. Le plateau des Kundelungu, paléograben
ou aulacogène. In: Charlet, J.-M. (ed.) Proceedings of the Inter-
national Cornet Symposium ‘Strata-bound Copper Deposits and
Associated Mineralizations’ (Mons, 5 – 9 September 1994). Académie
Royale des Sciences d’Outre-Mer, 51– 69.
Evans, D. A. D. 2000. Stratigraphic, geochronological, and paleomagnetic
constraints upon the Neoproterozoic climatic paradox. American
Journal of Science, 300, 347–433.
Eyles, N. & Eyles, C. H. 1992. Glacial depositional systems. In:
Walker, R. G. & James, N. P. (eds) Facies Models: Response to
Sea Level Change. Geological Association of Canada, St. John’s,
73 – 100.
Eyles, N. & Januszczak, N. 2004. ‘Zipper-rift’: a tectonic model for
Neoproterozoic glaciations during the breakup of Rodinia after
750 Ma. Earth-Science Reviews, 65, 1– 73.
François, A. 1973. L’extremité Occidentale de l’Arc Cuprifère Shabien,
Etude Géologique. Gécamines, Lubumbashi.
François, A. & Cailteux, J. 1981. La couverture katangienne entre les
socle de Nzilo et de la Kabompo, République du Zaı̈re, région de
Kolwezi. Annales de la Musée royal d’Afrique centrale, Sciences
géologiques, 87, 50.
Grosemans, P. 1935. Contribution à l’etude du conglomérat de base (petit
conglomérat) du Kundelungu supérieur. Annales de Services des
Mines, Comité Spécial du Katanga, 5, 38 –57.
Grosse, E. 1912. Dwykakonglomerat und Karroosystem im Katanga.
Zeitschrift der Deutsche Geologischer Gesellschaft, Monatsberichte,
64, 320– 321.
Grosse, E. 1918. Grundlinien der Geologie und Petrographie des östlichen
Katanga. Neues Jahrbuch für Mineralogie, Geologie und Paläontolo-
gie Beilage, 42, 272–419.
Guillemain, C. 1913. Zur Geologie von Katanga. Zeitschrift der
Deutsche Geologischer Gesellschaft, Monatsberichte, 65, 304– 328.
Gysin, M. 1934. Les tillites métamorphiques du Kundelungu de la Haute
Lufira (Congo Belge). Compte Rendu séances, Société de Physique et
de l’Histoire naturelle, Genève, 51, 218–221.
Halverson, G. P., Hoffman, P. F., Schrag, D. P., Maloof, A. C. &
Rice, A. H. N. 2005. Toward a Neoproterozoic composite
carbon-isotope record. Geological Society of America Bulletin, 117,
1181– 1207.
Hanson, R. E. 2003. Proterozoic geochronology and tectonic evolution of
southern Africa. In: Yoshida, M., Windley, B. F. & Dasgupta, S.
(eds) Proterozoic East Gondwana: Supercontinent Assembly,
183
Breakup. Geological Society, London, Special Publications, 206,
427– 463.
Hennig, E. 1915. Die Glazialerscheinerungen in Äquatorial- und Süd-
afrika. Geologische Rundschau, 6, 154–165.
Hoffman, P. F., Kaufman, A. J., Halverson, G. P. & Schrag, D. P.
1998. A Neoproterozoic snowball earth. Science, 281, 1342– 1346.
Hoffmann, K.-H., Condon, D. J., Bowring, S. A. & Crowley, J. L.
2004. U–Pb zircon dates for the Neoproterozoic Ghaub Formation,
Namibia: constraints on Marinoan glaciation. Geology, 32, 817– 820.
Intiomale, M. M. & Mbuyi, K. 1997. Le gisement Ba –Fe de Kapumba
(Shaba, Zaı̈re). Musée Royale d’Afrique Centrale, Tervuren (Belgi-
que), Département de Géologie et Minéralogie, Rapport Annuel
1995–1996, 183– 186.
Jackson, M. P. A., Warin, O. N., Woad, G. M. & Hudec, M. R. 2003.
Neoproterozoic allochthonous salt tectonics during the Lufilian
orogeny in the Katangan Copperbelt, central Africa. Geological
Society of America Bulletin, 115, 314– 330.
Jamotte, A. 1947. Esquisse géologique de la région de Kasenga. Bulletin
de l’Institut Royale Coloniale Belge, 18, 461– 476.
John, T., Schenk, V., Haase, K., Scherer, E. & Tembo, F. 2003. Evi-
dence for a Neoproterozoic ocean in south central Africa from
MORB-type geochemical signatures and P –T estimates of
Zambian eclogites. Geology, 31, 243– 246.
John, T., Schenk, V., Mezger, K. & Tembo, F. 2004. Timing and PT
evolution of whiteschist metamorphism in the Lufilian arc-Zambezi
belt orogen (Zambia): implications for the assembly of Gondwana.
Journal of Geology, 112, 71– 90.
Kampunzu, A. B. & Cailteux, J. 1999. Tectonic evolution of the Lufilian
Arc (Central Africa Copperbelt) during the Neoproterozoic
Pan-African orogenesis. Gondwana Research, 2, 401– 421.
Kampunzu, A. B., Kanika, M., Kapenda, D. & Tshimanga, K. 1993.
Geochemistry and geotectonic setting of late Proterozoic Katangan
basic rocks from Kibambale in Central Shaba (Zaire). Geologische
Rundschau, 82, 619– 630.
Kampunzu, A. B., Tembo, F., Matheis, G., Kapenda, D. & Huntsman-
Mapila, P. 2000. Geochemistry and tectonic setting of mafic igneous
units in the Neoproterozoic Katangan basin, Central Africa: impli-
cations for Rodinia breakup. Gondwana Research, 3, 125– 153.
Key, R. M., Liyungu, A. K., Njamu, F. M., Somwe, V., Banda, J.,
Mosley, P. M. & Armstrong, R. A. 2002. The western end of the
Lufilian arc in NW Zambia and its potential for copper deposits.
Journal of African Earth Sciences, 33, 503–528.
Kirschvink, J. L. 1992. Late Proterozoic low-latitude glaciation: the
snowball Earth. In: Schopf, J. W. & Klein, C. (eds) The Proterozoic
Biosphere. Princeton University Press, New York, 51 – 52.
Kokonyangi, J. W., Kampunzu, A. B., Armstrong, R., Yoshida, M.,
Okudaira, T., Arima, M. & Ngulube, D. A. 2006. The Mesoproter-
ozoic Kibaride belt (Katanga, SE D.R. Congo). Journal of African
Earth Sciences, 46, 1– 35.
Lefebvre, J. J. 1974. Mineralisations cupro-cobaltiferes associées aux
horizons pyroclastiques situés dans le faisceau supérieur de la Serie
de Roan, à Shituru, Shaba, Zaire. In: Bartholomé, P., de
Magnée, I, Evrard, P. & Moreau, J. (eds) Gisements stratiformes
et provinces cuprifères. Société Géologique de Belgique, Liège,
103– 122.
Master, S. 2007. Neoproterozoic evolution of the Kundelungu Plateau,
Katanga Supergroup, Central Africa: from aulacogen to foreland
basin. Extended Abstract, IGCP 485 and IGCP 497 Joint Conference,
Chouaı̈b Doukkali University, El Jadida, Morocco, 28 November – 5
December 2007, 62 – 65.
Master, S., Rainaud, C., Armstrong, R. A., Phillips, D. & Robb, L. J.
2005. Provenance ages of the Neoproterozoic Katanga Supergroup
(Central African Copperbelt), with implications for basin evolution.
Journal of African Earth Sciences, 42, 41 –60.
Mbede, E. I., Kampunzu, A. B. & Armstrong, R. A. 2004. Neoproter-
ozoic inheritance during Cainozoic rifting in the western and south-
western branches of the East African Rift System: evidence from
carbonatite and alkaline intrusions. International Commission on
Earth Sciences in Africa Conference: The East African Rift, Addis
Ababa, Abstracts.
Meert, J. 2003. A synopsis of events related to the assembly of eastern
Gondwana. Tectonophysics, 362, 1– 40.
184 S. MASTER & M. WENDORFF
Meert, J. G., van der Voo, R. & Ayub, S. 1995. Paleomagnetic investi-
gation of the Neoproterozoic Gagwe lavas and Mbozi complex, Tan-
zania and the assembly of Gondwana. Precambrian Research, 74,
225– 244.
Museu, M. 1987. Considérations sur l’origine du Grand Conglomérat de
base du Kundelungu inférieur au Shaba (République du Zaı̈re).
Musée royale de l’Afrique centrale, Tervuren (Belgique), Départe-
ment de Géologie et Minéralogie, Rapport Annuel 1985–1986,
165– 168.
Nogueira, A. C. R., Riccomini, C., Sial, A. N., Moura, C. A. V. & Fair-
child, T. R. 2003. Soft-sediment deformation at the base of the Neo-
proterozoic Puga cap carbonate (southwestern Amazon craton,
Brazil): confirmation of rapid icehouse to greenhouse transition in
snowball Earth. Geology, 31, 613– 616.
Porada, H. 1989. Pan-African rifting and orogenesis in southern to equa-
torial Africa and eastern Brazil. Precambrian Research, 44, 103– 136.
Porada, H. & Berhorst, V. 2000. Towards a new understanding of the
Neoproterozoic – Early Palaeozoic Lufilian and northern Zambezi
Belts in Zambia and the Democratic Republic of Congo. Journal of
African Earth Sciences, 30, 727–771.
Postma, G., Nemec, W. & Kleinspehn, K. L. 1988. Large floating clasts
in turbidites: a mechanism for their emplacement. Sedimentary
Geology, 58, 47– 61.
Rainaud, C., Master, S., Armstrong, R. A. & Robb, L. J. 2005a. Geo-
chronology and nature of the Palaeoproterozoic basement in the
Central African Copperbelt, with regional implications. Journal of
African Earth Sciences, 42, 1 –31.
Rainaud, C., Master, S., Armstrong, R. A., Phillips, D. & Robb, L. J.
2005b. Monazite U– Pb dating and 40Ar/39Ar thermochronology of
metamorphic events in the Central African Copperbelt during the
Pan-African Lufilian orogeny. Journal of African Earth Sciences,
42, 183– 199.
Reintjens, E. 1935. Les gisements cuprifères du Katanga et de la Rhodé-
sie septentrionale. Comité Special du Katanga, Annales des Services
des Mines, 6, 10 – 19.
Robb, L. J., Master, S., Armstrong, R. A., Rainaud, C. & Greyling, L.
2003. Timing of Cu–Co and Pb –Zn mineralisation in the Central
African Copperbelt: a link to Neoproterozoic glaciations? Trans-
actions of the Institution of Mining and Metallurgy, Section B,
Applied Earth Sciences, 112, B164 –B166.
Robb, L., Master, S., Armstrong, R. A., Greyling, L. & Rainaud, C.
2004. Neoproterozoic glaciations and the link to Cu –Co and Pb –Zn
mineralization in the Central African Copperbelt. Geoscience Africa
2004 Conference, Abstracts Volume 2, University of the Witwaters-
rand, Johannesburg, 12 –16 July 2004, 550– 551.
Robert, M. 1912a. La stratigraphie du système du Kundelungu au
Katanga. Annales de la Société Géologique de Belgique, Publications
relatives au Congo Belge et aux régions voisins, 39, 1911– 1912, fasc.
I, 5 – 8.
Robert, M. 1912b. Le système du Kundelungu au Katanga. Annales de la
Société Géologique de Belgique, Publications relatives au Congo
Belge et aux régions voisins, 40, 1912–1913, 213– 275.
Robert, M. 1933. Le système du Kundelungu au Katanga. Bulletin de
l’Institut royale colonial belge, 4, 436 –440.
Robert, M. 1940a. La glaciation du Kundelungu au Katanga (Congo
Belge). Report of the 17th International Geological Congress,
Moscow, USSR, 1937, 6, 99 –113.
Robert, M. 1940b. Contribution à la géologie du Katanga. Le Système du
Kundelungu et le Système Schisto-dolomitique (1re partie). Mémoire
de l’Institut royale colonial belge, Science Naturelle et Médecine,
Coll. in-4o, 6, 108.
Robert, M. 1946. Le Congo Physique, troisième édition, revue et complé-
tée. Vaillant-Carmanne, Liège.
Robert, M. 1947. Les traces de glaciation et les périodes climatiques
glaciaires au Katanga et en Afrique australe. Bulletin de la Société
belge de Géologie, Paléontologie et Hydrologie, 56, 62 – 76.
Selley, D., Broughton, D. et al. 2005. A New Look at the Geology
of the Zambian Copperbelt. Society of Economic Geologists, Inc.,
Tulsa, 100th Anniversary Volume, 965– 1000.
Shields, G., Deynoux, M., Strauss, H., Paquet, H. & Nahon, D. 2007.
Barite-bearing cap dolostones of the Taoudéni Basin, northwest
Africa: sedimentary and isotopic evidence for methane seepage after
a Neoproterozoic glaciation. Precambrian Research, 153, 209–235.
Studt, F. E. 1908. Carte géologique du Katanga (1/500.000e) et notice
explicative. Annales du Musée de Congo belge, sér. 2, 1, 5– 16.
Studt, F. E. 1913. The geology of Katanga and Northern Rhodesia: an
outline of the geology of South Central Africa. Transactions of the
Geological Society of South Africa, 16, 44– 106.
Studt, F. E., Cornet, J. & Buttgenbach, H. 1908. Carte géologique du
Katanga et notes descriptives. Annales Musée du Congo, Bruxelles,
Géologie, Géophysique, Minéralogie & Paléontologie, série 2,
Katanga, 1, 94.
Stutzer, O. 1911. Über Dwyka Konglomerat im Lande Katanga, Bel-
gisch Kongo. Zeitschrift der Deutsche Geologischer Gesellschaft,
Monatsberichte, 63, 626– 629.
Stutzer, O. 1913a. Über glaziale Konglomerate im Lande Katanga, Bel-
gisch Kongo. Zeitschrift der Deutsche Geologischer Gesellschaft,
Monatsberichte, 65, 114– 117.
Stutzer, O. 1913b. Über den geologischen Aufbau der südöstlichen
Katanga. Jahresbericht der Freiberger Geologischer Gesellschaft,
VI, 41 – 47.
Tembo, F., Kampunzu, A. B. & Porada, H. 1999. Tholeiitic magmatism
associated with continental rifting in the Lufilian Fold Belt of Zambia.
Journal of African Earth Sciences, 28, 403– 425.
Thieme, J. G. 1970. The geology of the Mansa area: Explanation of
Degree Sheet 1128, parts of NW Quarter and NE Quarter. Report
of the Geological Survey of Zambia, Lusaka, 26, 37.
Thieme, J. G. 1971. The geology of the Musonda Falls area: Explanation
of Degree Sheet 1028, SE Quarter. Report of the Geological Survey of
Zambia, Lusaka, 32, 25.
Unrug, R. 1987. Geodynamic evolution of the Lufilian arc and the Kun-
delungu aulacogen: Angola, Zambia and Zaire. In: Matheis, G. &
Schandelmeier, H. (eds) Current Research in African Earth
Sciences; 14th Colloquium of African Geology, Extended Abstracts.
Balkema, Rotterdam, 117–120.
Unrug, R. 1988. Mineralization controls and source of metals in the Lufi-
lian fold belt, Shaba (Zaire), Zambia and Angola. Economic Geology,
83, 1247– 1258.
Vanden Brande, P. 1936. Etudes géologiques dans le feuille Lukafu.
Comité Special du Katanga, Annales des Services des Mines, 6, 51–69.
van Doorninck, N. H. 1928. De Lufilische Plooing in den Boven Katanga
(Belgischen Congo). G. Naeff, ’s-Gravenhage.
Vavrdova, M. & Utting, J. 1974. Lower Paleozoic microfossils from the
Luapula beds of the Mansa area. Records of the Geological Survey
of Zambia, 12, 81– 89.
Verbeek, T. 1970. Géologie et lithologie du Lindien (Précambrien supér-
ieur du Nord de la République Démocratique du Congo). Annales de
la Musée Royale d’Afrique Central, séries In-88, Sciences géologi-
ques, 66, 311.
Von Engelin, O. D. 1930. Type form of facetted and striated glacial
pebbles. American Journal of Science, 19, 9– 16.
Washburn, A. L. 1969. Patterned ground in the Mesters Vig district,
northeast Greenland. Biuletyn Peryglacjalny, Łódź, 18, 259–330.
Wendorff, M. 2003. Stratigraphy of the Fungurume Group– evolving
foreland basin succession in the Lufilian fold-thrust belt, Neoproter-
ozoic –Lower Palaeozoic, Democratic Republic of Congo. South
African Journal of Geology, 106, 17 –34.
Wendorff, M. 2005a. Coarse clastic markers of Rodinia breakup and
Gondwana assembly in the Lufilian belt, Pan-African orogen of
Central Africa. In: Wingate, M. T. D. & Pisarevsky, S. A. (eds)
Supercontinents, Earth Evolution Symposium. Geological Society
of Australia Abstracts, Perth, 101.
Wendorff, M. 2005b. Evolution of Neoproterozoic – Lower Palaeozoic
Lufilian arc, Central Africa: A new model based on syntectonic con-
glomerates. Journal of the Geological Society, London, 162, 5– 8.
Wendorff, M. 2005c. Sedimentary genesis and lithostratigraphy of
Neoproterozoic megabreccia from Mufulira, Copperbelt of Zambia.
Journal of African Earth Sciences, 42, 61– 81.
Wendorff, M. & Key, R. M. 2009. The relevance of the sedimentary
history of the Grand Conglomerat Formation (Central Africa) to the
interpretation of the climate during a major Cryogenian glacial
event. Precambrian Research, 172, 127–142.
Wingate, M. T., Pisarevsky, S. A. & De Waele, B. 2010. Paleomagnet-
ism of the 765 Ma Luakela volcanics in Northwest Zambia and
implications for Neoproterozoic positions of the Congo Craton.
American Journal of Science, 310, 1333–1344.
 Chapter 13

Neoproterozoic sequences of the West Congo and Lindi/Ubangi Supergroups

in the Congo Craton, Central Africa

JENNY TAIT1*, FRANCK DELPOMDOR2, ALAIN PRÉAT2, LUC TACK3, GIJS STRAATHOF1 & VALENTIN KANDA NKULA4
1
School of Geosciences, University of Edinburgh, Edinburgh EH9 3JW, UK
2
Department of Earth Sciences and Environmental Sciences, University of Brussels, 1050 Brussels, Belgium
3
Department of Geology and Mineralogy, Royal Museum for Central Africa (RMCA), Tervuren, Belgium
4
Département des Sciences de la Terre, Faculté des Sciences, Université de Kinshasa (UNIKIN), Democratic Republic of the Congo
*Corresponding author (e-mail: jenny.tait@ed.ac.uk)

Abstract: The focus of this chapter is the West Congo Supergroup in the West Congo Belt (WCB), which extends along the western
margin of the Congo Craton from Gabon in the north to northern Angola in the south, and the Lindi/Ubangi Supergroup of the
Lindian and Fouroumbala – Bakouma Basins exposed on the northern margin of the craton. In both regions, up to two distinct diamictite
horizons have been recognized, the younger of which is often associated with carbonate rocks. Geochronological constraints are gener-
ally rather poor, many of the deposits lack modern sedimentological analysis, and the glacial versus non-glacial genesis of the diamictites
is a matter of debate in the literature. However, recent studies suggest a periglacial influence of diamictite deposition, particularly for the
sequences in the WCB. The stratigraphy of the various basins is described, available geochemical and geochronological information col-
lated, and recent work regarding the periglacial nature of the diamictites discussed. Finally, an updated chronostratigraphic correlation
between the basins is presented. However, much more work is required, particularly in the Neoproterozoic basins on the northern margin
of the Congo Craton, and more accurate geochronological constraints are required before the Neoproterozoic palaeogeography and
depositional environments of the western and northern Congo Craton can be fully understood.

The Congo Craton (Fig. 13.1), which is defined here as the central
African landmass that amalgamated at the time of Gondwana
assembly (c. 550 Ma; De Waele et al. 2008), is encircled by Neo-
proterozoic sedimentary basins (Fig. 13.1, see also Hoffman 2011;
Master & Wendorff 2011).
Delhaye & Sluys (1923– 1924) first identified Precambrian
diamictites of assumed glacial origin in the Democratic Republic
of Congo (DRC), which Lepersonne (1951) subsequently inter-
preted in the West Congo Belt (WCB) as representing two distinct
diamictite horizons termed the Lower and Upper Diamictite
Formations. The stratigraphy of the WCB was first described in
detail in Gabon by Hudeley (1966), in the Popular Republic of
Congo by Dadet (1969), in the DRC by Cahen (1978) and in
Angola by Stanton et al. (1963), and all available data were
synthesized by Trompette (1994). The glacial origin of the late
Neoproterozoic WCB diamictites was the centre of much debate
in the early 1970s (see Kröner & Carreira 1973; Schermerhorn
1974; Cahen & Lepersonne 1981). More recently, these deposits
have been the subject of facies and microstructural analysis
(Delpomdor 2007a, b, c; Delpomdor et al. 2008) and a periglacial
influence suggested.
Geochronological and palaeomagnetic data for the West Congo
Supergroup and Lindi/Ubangi Supergroup are scant, and ages and
palaeolatitudinal constraints are largely derived indirectly by cor-
relation (e.g. Evans 2000). Recent chemostratigraphic studies on
the Haut-Shiloango and Schisto-Calcaire Subgroups by Frimmel
et al. (2006), Poidevin (2007) and Préat et al. (2011) support
general lithological correlation between the Upper Diamictite and
Lower Diamictite Formations with the glaciogenic Chuos and
Ghaub Formations in the Kaoko-Damara belt to the south,
and the Petit and Grand Conglomérat Formations in the Zambezi-
Lufilian belt to the SE.
Structural framework and basin setting
The Congo Craton comprises several Archaean nuclei (Fig. 13.1),
which welded together during the Eburnean orogeny c. 2.1– 1.8 Ga
(Pinna et al. 1996; De Waele et al. 2006, 2008; Noce et al. 2007;
Delor et al. 2008), and subsequently remained stable and a coher-
ent block throughout Late Palaeoproterozoic and Mesoproterozoic
times (Tack et al. 2006, 2008, 2009).
The Early Neoproterozoic was marked by rifting along the
western margin of the Congo Craton related to the break-up
of Rodinia and opening of the Adamastor Ocean (Tack et al.
2001), followed by the passive margin-type sedimentation of the
West Congolian Group. The western margin of the Congo
Craton collided with the active São Francisco margin, thus
forming the Araçuai – West Congo Orogen (AWCO), including
the Brasiliano Araçuai belt now preserved adjacent to the São
Francisco Craton in Brazil (Pedrosa-Soares et al. 2008), and the
WCB in central Africa. Late Neoproterozoic deformation in the
WCB involved thrusting of Palaeoproterozoic basement rocks
(the c. 2.1 Ga Kimezian Supergroup) onto the Neoproterozoic
sequences of the West Congo Supergroup, while to the east the
external foreland basin sequences of the West Congo Supergroup
lie unconformably on the Archaean cratonic basement. Opening
of the Atlantic in the Cretaceous split the Araçuai – WCB into
two parts, the Brazilian side of which inherited two-thirds of
the AWCO, including all Neoproterozoic ophiolitic slivers, the
entire magmatic arc, the suture zone and syn- (c. 585–560 Ma)
to post- (c. 530–490 Ma) collisional magmatism (Pedrosa-Soares
et al. 2008).
The northern margin of the Congo Craton is marked by the
Late Neoproterozoic east – west trending Sergipano-Central
African Belt, which stretches from northern Brazil, through
southern Cameroon and the Central African Republic (CAR).
This south-verging belt marks the boundary between the
Central African Mobile Zone to the north, and the northern
margin of the Congo and São-Francisco cratons. This northwes-
ternmost part of the Congo Craton is represented by the Archaean
Ntem complex, which is overlain by Proterozoic metasedimen-
tary and volcanic rocks (Lower Dja Series), and has been over-
thrust from the north by the Pan African Oubanguide Nappe
of the Central African Belt. Nappe emplacement in southern
Cameroon is constrained by 620– 610 Ma ages obtained from

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 185– 194. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.13
186 J. TAIT ET AL.

0°E 10°E lt 20°E 30°E 40°E

an Be
ntra l Afric Fouroumbala-Bakouma
Serg ipe-Ce Basin

CUB LA

(East African Orogen)

Ubangui
N

Mozambique Belt
Lindi Ruw.
0°S Gabon Belt
LV
craton NE
Kibaran
Itombwe Belt
Sao Francisco
CRB

Ru
Mbuji- Tanzania

zis
craton

ian
West
West Congo
Mayi craton

Be
Congo

lt
Araçuai Belt Kibaran LT Ub
Bambui

lt
Belt en

Be
Belt dia
Kimezian Kasai

ran
nB
Belt el

ga
10°S t
Block erb
e l Bangweulu t

a
Us
p p L
Co Block
M
Katanga
B elt
Angola Lufilian m i de
Block Belt Iru h e r n elt
o ut e B Lurio
Legend S d
mi Foreland Fig. 13.1. Geological sketch map of
rI u S.M.
Phanerozoic to Recent cover Zambezi Belt Sub-Saharan Africa (modified after De
Neoproterozoic MB Waele et al. 2008) and showing the
CKB
Mesoproterozoic belts Congo-Uganda Block (CUB) and the
20°S
Palaeoproterozoic belts Zimbabwe Neoproterozoic sedimentary basins in the
Palaeoproterozoic with possible
Archaean basement (unexposed) Damara &
block NE. LA, Lake Albert; LV, Lake Victoria;
Archaean cratons Kaoko Belts km CKB, Choma Kalomo Belt; MB, Magondi
Limpopo 0 500
Belt.

plutonic rocks (Toteu et al. 2006), while a minimum age of Stratigraphy of the West Congolian Group
628 + 12 Ma has been obtained from syn-collisional granites in
the Sergipano Belt of Brazil (Oliveira et al. 2006; Bueno et al. The West Congo Supergroup is exposed along the western margin
2009). Pin & Poidevin (1987) suggest that the Central African of the Congo Craton in the Bas-Congo Basin of the DRC, and the
Belt represents a deeply eroded orogenic belt, whereas Abdelsa- Nyanga Basin of SW Gabon (Fig. 13.2). Up to two Neoproterozoic
lam et al. (2002) interpret it as the result of continent – continent diamictite units, interpreted as being glaciogenic in origin, have
collision between the northern Congo Craton and the Saharan been identified. The stratigraphy of these two sub-basins of the
Metacraton. WCB are described below.

8°E 10°E 12°E 14°E 10°E 12°E 14°E 16°E 18°E 20°E
0° LIBREVILLE 2°N

0°
CONGO

N 2°S

4°S
GABON

2°S
SW Gabon DRC
6°S
Basin N ATLANTIC
YA 8°S
ATLANTIC N OCEAN
BA GA ANGOLA
OCEAN SI -NI
N AR
4°S I Bas-Congo
Basin KINSHASA

LEGEND

POST KAROO 6°S

KAROO SUPERGROUP
INKISI GROUP 100 1 200 300 km
WEST CONGOLIAN GROUP
MAYUMBIAN GROUP
ZADINIAN GROUP
8°S
KIMEZIAN
ARCHAEAN
NOQUI-TYPE GRANITES Fig. 13.2. Geological sketch map of the
LUANDA
West Congo Belt (after Frimmel et al.
2006).
 THE WEST CONGO AND LINDI/UBANGI SUPERGROUPS 187

The Bas-Congo Basin, DRC
The Neoproterozoic West Congo Supergroup of the DRC is divided
into the Zadinian, Mayumbian and West Congolian groups. The
most complete Neoproterozoic sequence of the West Congolian
Group, with two diamictite units, is exposed in the Bas-Congo
Basin (Fig. 13.3). Overlying the c. 1000–910 Ma volcanoclastic
Zadinian and Mayumbian Groups (Tack et al. 2001; Fig. 13.2),
the West Congolian Group (originally described in detail by
Cahen 1978) corresponds to passive margin siliciclastic and

West-Congolian Group West-Congolian Group Lindi Supergroup Lindi/Ubangi Supergroup

SW Gabon Bas-Congo, DRC Fouroumbala-Bakouma Lindi, DRC
CAR
† 566 Ma

SCHISTO- MPIOKA
Dialanga Formation
GRESEUX SUBGROUP
GROUP BANALIA
Bili Formation GROUP
BAKOUMA
SC4 C5 SERIES
SCHISTO-
SC3
CALCAIRE †† 575 Ma
SC2 GROUP Bakouma Formation
SC1
NIARI TILLITE BONDO TILLITE Alolo Formation
C4 FORMATION
BOUENZA SCHISTO-
FORMATION ARUWIMI
CALCAIRE Mbiana Formation
x x x x x x x
GROUP
x x x x x x
SUBGROUP
x x x x x x x C3 Galamboge Formation
x x x x x x
x x x x x x x KEMBE-
NAKANDO Kole Formation
SERIES Mamungi LOKOMA
C2
Formation GROUP
C1
Bobwamboli Formation
UPPER DIAMICTITE (?)
FORMATION AKWOKWO TILLITE
Asoso Formation
†† 645 Ma (Sh8) †† 730-750 Ma
BOUGBOULOU Lenda ITURI
Formation
SERIES x
x
x
x
x
x
x
x
x
x
x
x
x GROUP
x x x x x x x Penge Formation
x x x x x x
x x x x x x x

† 650 Ma (Sh6) LOWER TILLITE

x x FORMATION
HAUT x
x x
x
x
x
x
x

SHILOANGO x
x
x
x
x
x
x
x
x
x
x
x
x

SUBGROUP

LEGEND

DOLOMITE

v v LOWER LIMESTONE/ ARGILLACEOUS

DIAMICTITE LIMESTONE
v v v
FORMATION v DIAMICTITE/ BASALT-DOLERITE SILL
SHALE
† 923 ± 43 Ma
SANDSTONE-QUARTZITE/
CONGLOMERATE
SANSIKWA x x x x
x x x x
SUBGROUP x x x x PALAEOPROTEROZOIC/ARCHAEAN

v v v
OOLITE, CYANOBACTERIAL MATS,
STROMATOLITE
††† 910 Ma PANAFRICAN TECTONIC
MAYUMBIAN UNCONFORMITY
† Frimmel et al. (2006)
GROUP
†† Poidevin (2007)
††† Tack et al. (2001)

Fig. 13.3. Stratigraphic logs of Neoproterozoic sequences from the western and northern margins of the Congo Craton. Modified from Delpomdor et al. (2011).
188 J. TAIT ET AL.
carbonate platform deposits and is particularly well developed in
the foreland part of the WCB where it is only gently folded and
unmetamorphosed. The sedimentary rocks have been subdivided
into four subgroups: the Sansikwa, Haut-Shiloango, Schisto-
Calcaire and the Mpioka Subgroups as described below. The
uppermost Inkisi ‘Subgroup’ is no longer considered to be part
of the Neoproterozoic sequences (Frimmel et al. 2006 and refer-
ences therein) and considered post-Pan African in age (see discus-
sion further below).
The Sansikwa Subgroup forms the base of the West Congolian
Group and comprises siliciclastic rocks representing a continental
rift depositional environment. The Lower Diamictite Formation
marks the top of the Sansikwa Subgroup, and contains up to
400 m of diamictite with interbedded sands and shales (Cahen &
Lepersonne 1981). Tholeiitic basalts, sometimes including pillows
and hyaloclastic breccia, are interlayered in the diamictite (De
Paepe et al. 1975; Kampunzu et al. 1991) with accompanying
feeder dykes and sills intruded in the underlying Sansikwa Sub-
group. Their emplacement age has yet to be confirmed, as the
40Ar/39Ar resetting age of 566 + 42 Ma is thought to represent
regional greenschist facies metamorphism during Pan African
orogeny and amalgamation of western Gondwana (Frimmel
et al. 2006).
The Sansikwa Subgroup is overlain by 700–800-m-thick, pre-
dominantly siliciclastic rocks of the Haut-Shiloango Subgroup.
Conglomerates, quartzarenites, argillites and shales make up the
lower two-thirds of the Haut-Shiloango Subgroup. The Sekelolo
Limestone forms the upper part and consists of dominantly
subtidal, medium- to fine-bedded limestones with a minor clastic
component. The overlying Upper Diamictite Formation, which is
recognized throughout the length of the WCB, from southern
Gabon (the Niari Formation) through to northern Angola, is up
to 200 m thick in the Bas-Congo Basin (Cahen & Lepersonne
1981). It is overlain by a sequence of up to 12-m-thick, finely lami-
nated pink dolomitic carbonates, forming the lowermost part of
the Schisto-Calcaire Subgroup, which share many lithological
characteristics with basal Ediacarian carbonates elsewhere. The
remainder of the Schisto-Calcaire Subgroup consists of a range
of dominantly shallower-water carbonate shelf facies, including
oolites, diverse stromatolites, evaporites and cherts (Alvarez
1995). The Schisto-Calcaire Subgroup shows a transition from
east to west between dolomitic intertidal facies with stromatolitic
biostromes to supratidal facies with evaporitic needles, ooids and
cross-bedding (Delpomdor 2007c; Préat et al. 2010, 2011).
These two facies are separated by a stromatolitic reef barrier
(Trompette 1994).
The Schisto-Calcaire Subgroup is overlain by the Mpioka
Subgroup, c. 1000-m-thick siliciclastic succession with conglom-
erates, quartzoarenites and argillites, interpreted as being a
late-orogenic molasse deposit. The Mpioka Subgroup is recog-
nized throughout the WCB, and the uppermost subgroup has
been locally affected by Pan African Late Neoproterozoic/Early
Cambrian deformation. In this regard, it is analogous to the
Mulden Group along the southwestern margin of the Congo
Craton.
The SW Gabon Basin (or the Nyanga-Niari Basin)
The West Congo Supergroup of Gabon is exposed in the Nyanga
synclinorium (Gérard 1958; Dadet 1969), and is subdivided into
several informal units. The oldest sedimentary rocks comprise
the Bouenza Formation (Fig. 13.3), a fluvial sequence up to 100 m
thick (Prian 2008), which records erosion of the Palaeoprotero-
zoic and Mesoarchaean basement. The sandstones are massive
with clear cross-bedding and ripple cross-lamination, suggesting
palaeocurrents from SW to NE. These are overlain by the
fluvio-glacial deposits of the Niari Group (or Niari Tillite,
Fig. 13.3). The Bouenza Formation and the Niari Group record
lacustrine-fluvio-glacial infilling of extensional zones in the
Central African basement (Alvarez 1995). On the east flank of
the basin, the Niari Group rests on the Bouenza Formation with
a non-angular unconformity.
The Niari diamictites are overlain by the Schisto-Calcaire Group
(or Carbonate Subgroup; Prian 2008) which is correlated with the
Schisto-Calcaire Subgroup of the Bas-Congo Basin (DRC) and
thought to be Late Neoproterozoic in age. Again, precise ages
from the Gabon sequences are lacking, and the most reliable
ages and correlations are summarized in Figure 13.3. The Schisto-
Calcaire Group of the SW Gabon Basin is predominantly a carbon-
ate sequence composed of four formations consisting of calcareous
to dolomitic shales with an uppermost sandy shale-siltstone unit
with interbedded limestones. Carbonate deposition of the upper
part was under similar sedimentological environments to those
of the Bas-Congo (Préat et al. 2010), indicating hypersaline
shallow water sub-supratidal conditions, in possible lagoonal
setting with development of lithoherms and cyanobacterial mats.
In SW Gabon, the Schisto-Calcaire Group is overlain by the
Schisto-Gréseux Group (Chevalier et al. 2002), which is correlated
with the Mpioka Subgroup of the Bas-Congo Basin (Tack et al.
2001; Frimmel et al. 2006). The series is thinner in Gabon (up
to 100 m thick) and of Late Neoproterozoic age (Thiéblemont
et al. 2009). The Schisto-Gréseux Group is divided into two
subgroups: the lower subgroup is predominantly composed of
siltstones and argillites with, in places, a 6– 15-m-thick basal con-
glomerate, and the upper subgroup contains feldspathic sandstones
and argillites.
Stratigraphy of the Lindi/Ubangi Supergroup
Neoproterozoic sedimentary rocks exposed on the northern Congo
Craton are recognized in the Bangui Basin of southwestern CAR,
the Fouroumbala-Bakouma Basin of south central CAR and the
Lindian Basin of northeastern DRC (Fig. 13.1). Trompette (1994)
considered all these sub-basins to be part of the Lindian Basin.
Correlation is problematic due to the limited number of studies
that have been carried out, poor or non-existent age constraints,
poor outcrop conditions and problems concerning accessibility.
Diamictite units have been recognized in all three basins
(Fig. 13.4). The best studied is the Lindian Basin of northeastern
DRC, which was described in detail by Verbeek (1970), and the
stratigraphy of which is summarized below.
The Lindian Basin
The Lindian Basin is an intracratonic basin that developed on
the northern Congo Craton in the Neoproterozoic. The rocks of
the Lindi Supergroup, which are up to 2500 m thick, rest uncon-
formably upon crystalline Archaean and/or Early Proterozoic
basement rocks and have undergone little or no metamorphism.
To the west of the Lindian Basin, the continuation of this sequence
in the DRC has been termed the Ubangi Supergroup. As summar-
ized by Verbeek (1970) and Trompette (1994), the Lindi
Supergroup is divided into three groups: the basal Ituri, the
Lokoma and the Aruwimi groups, the last of which is subdivided
into the Galamboge, Alolo and Banalia formations. However, in
Figure 13.3 the Aruwimi Group as originally defined (Verbeek
1970) is here split into two units, with the uppermost Banalia For-
mation ranked as a separate, younger ‘Group’ (see below). Only
one diamictite horizon, occurring at the base of the Lokoma
Group and termed the Akwokwo Tillite, is interpreted as being
glacial in origin (Fig. 13.3).
The basal clast-supported conglomerate with coarse-grained
arkoses and quartzarenites (30 –50-m-thick Penge Formation,
Ituri Group) is overlain by thick oolitic and stromatolitic limestone
with dolomite layers (the Lenda Formation). Poidevin (2007)
 THE WEST CONGO AND LINDI/UBANGI SUPERGROUPS 189

West-Congo West-Congo Lindi Ubangi/Lindi Lindi, DRC

SW Gabon Bas-Congo, DRC Fouroumbala- Bangui, CAR
Bakouma
CAR
POST-PRECAMBRIAN INKISI BANALIA
SUBGROUP GROUP
Panafrican
E Tectonic †566 Ma

D SCHISTO-
I GRESEUX MPIOKA ARUWIMI
A GROUP SUBGROUP GROUP
C BAKOUMA
Post-Marinoan SUBGROUP
A
R ††575 Ma
SCHISTO-
I 600 SCHISTO- BANGUI LOKOMA
CALCAIRE
A Ma CALCAIRE GROUP
GROUP
N SUBGROUP BIMBO

C Marinoan NIARI UPPER BONDO CONGLO. AKWOKWO

R Glaciation TILLITE DIAMICTITE TILLITE TILLITE
Y ††645 Ma KEMBE
O NAKANDO
G SUBGROUP ITURI
E Post-Sturtian †650 Ma GROUP
BOUENZA
N FORMATION HAUT- BOUG- ††730-750 Ma
I SHILOANGO BOULOU ?
700
A SUBGROUP SUBGROUP
N
Sturtian LOWER LOWER
Glaciation DIAMICTITE TILLITE

†923 ± 43 Ma

800

T LEGEND
O Pre-Sturtian
N Diamictites/Mixtites/Tillites
I SANSIKWA
A SUBGROUP
Sandstones/Quartzites
N Cap carbonates/Dolomites
Limestones/Dolomites
900 Shales/Argillites/Pelites
†††910 Ma

† Frimmel et al. (2006)

†† Poidevin (2007) Fig. 13.4. Stratigraphic correlation of
††† Tack et al. (2001) Neoproterozoic sequences of the western and
northern Congo Craton. Correlations based on
Sr isotopic data, radiometric age constraints
and revised lithological relationships.
1000 Modified after Poidevin (2007).

reported the presence of early Neoproterozoic microfaunas in these
horizons. The top of the Ituri Group is marked by c. 50 m of mica-
ceous shales, carbonates and sandstones of the Asoso Formation.
The entire Ituri Group represents lagoonal or shallow marine plat-
form type sedimentary rocks (Daly et al. 1992) and is restricted to
the eastern part of the Lindian Basin. It has been tentatively corre-
lated with both the Haut-Shiloango (Trompette 1994; Daly et al.
1992) and the Sansikwa Subgroup (Poidevin 2007) of the West
Congolian Group.
The Lokoma Group (.500 m thick) overlies the Ituri Group
with an angular unconformity, and begins with the Akwokwo
Tillite, which has only limited exposure in the eastern parts of the
basin where it sits in erosional contact on the underlying Ituri
Group. It is up to 40 m thick, with diverse and poorly sorted
clasts (Verbeek 1970; Cahen 1978) in a grey-green clay matrix.
The rest of the Lokoma Group consists of arkoses, conglomerates
and argillites.
In the original definition of Verbeek (1970), the Aruwimi Group
is up to 1500 m thick and subdivided into three formations. The
Galamboge Formation (c. 100 m thick) consists of cross-bedded
medium-grained quartzoarenites interbebbed with argillites. The
quartzoarenites are marked by a rapid transition to the Alolo
Formation, and comprise 400-m-thick clayey to calcareous argil-
lites, often stratified, interbedded with dark carbonaceous argillite
layers, fine-grained sandstones, zoned pink and oolitic limestones
and dolomitic lenses. The uppermost Banalia sequences are
composed of .1000 m of massive or finely stratified and fre-
quently cross-bedded fine-grained reddish arkoses interbedded
with reddish micaceous and clayey argillites. The contact between
the Banalia and the Alolo Formations is apparently conformable.
In terms of thickness and depositional facies, the deltaic Banalia
Redbeds have been correlated with the similar Inkisi ‘Subgroup’
of the Bas-Congo Basin and interpreted as evidence of post-
Gondwana amalgamation (550 Ma) and pre-Karoo break-up in
central Africa (Tack et al. 2008, 2009a). For this reason, they
are ranked here, in both Figures 13.3 and 13.4, as (new) separate
groups. On the other hand, new geophysical data in and around
the DRC Cuvette Centrale suggest that not only the Banalia
Redbeds but the complete Aruwimi Group (thus including also
the lowermost Galamboge and overlying Alolo Formations) are
190 J. TAIT ET AL.
post-Pan African in age (Kadima et al. 2011). The precise age of
the Galamboge and Alolo Formations (i.e. upper Neoproterozoic
versus Phanerozoic) clearly needs to be determined using appro-
priate radiometric dating.
The Fouroumbala-Bakouma Basin
The Fouroumbala-Bakouma Basin is located in the south central
region of the CAR (Fig. 13.1). Three main sedimentary groups
have been recognized (Fig. 13.3): the Bougboulou, Kembé-
Nakando and Bakouma Subgroups (or Series). The stratigraphy
of the Fouroumbala-Bakouma Basin was described in detail by
Bigotte & Bonifas (1968), Poidevin et al. (1981) and Poidevin
(1985) with recognition of two diamictite horizons. The base of
the Bougboulou Subgroup rests with an angular unconformity on
the Archaean Bangui-Ketté basement. A basal diamictite, termed
the Lower Tillite, has been identified in the extreme SW of the
basin. The Bougboulou Subgroup comprises predominantly inter-
bedded shales and sandstones, although in the south of the Four-
oumbala Basin, silicified stromatolitic carbonates (the Kassa
Formation) have been identified (Mestraud 1952, 1953). The
Bougboulou Subgroup is unconformably overlain by the deltaic
to neritic Kembé-Nakando Subgroup comprising sandstones and
black and red shales of the Mbiana Formation. The base of the
overlying Bakouma Subgroup is marked by the Bondo Tillite,
which is thought to be glacial in origin, containing varves and
dropstones. The Bondo Tillite is overlain by the red and grey
Bakouma Formation dolomites, which are similar to other post-
glacial (cap) carbonates of the Congo Craton (Poidevin 2007).
This association, however, is only known from drill cores with
no surficial exposure. The Bakouma Formation is overlain by the
Bili and Dialanga Formations (Fig. 13.3). The Bili Formation
comprises basal shales overlain by white and grey limestones.
The clastic Dialanga Formation is the youngest member of the
Bakouma Subgroup and comprises black shales and quartzo-
arenites. Poidevin (2007) correlated the Bakouma Group with
the Aruwimi Group (original definition of Verbeek, 1970) of the
Lindi Supergroup. If this is correct, the upper tillites (Bondo Til-
lites) are missing in the Aruwimi Group of the Lindi Supergroup.
The Bangui Basin
The Bangui Basin was described lithostratigraphically by Babet
(1935), Legoux & Hourcq (1943), Gerard & Gerard (1952), Bes-
soles & Trompette (1980) and Poidevin (1976, 1979a, 1979b,
1985, 2007). Neoproterozoic sedimentary rocks are exposed in
SW CAR around the capital Bangui, and extend southwards to
the Congo River and overlie the Nola Group quartzopelitic meta-
sedimentary rocks, which are thought to be Palaeoproterozoic in
age. The Nola Group sequences have been overthrust in the
north by the Pan-African Oubanguides belt (i.e. Central African
Belt), and intruded by the Nola dolerite dykes. These dykes are
undated but yield a pan-African resetting Ar/Ar age of 571 Ma
(Moloto-A-Kenguemba et al. 2008). Correlation with 950 Ma
dolerite dykes in Gabon has been suggested (Vicat et al. 1996)
but remains to be demonstrated.
The Neoproterozoic sedimentary sequences of the Bangui
Basin are subdivided into the Kembe conglomerates, the Bimbo
sandstones and the Bangui carbonates (Fig. 13.4). The basal
Kembe conglomerates (up to 200 m thick and exposed near the
village of Bimbo along the Ubangui River), are tectonized and
comprise fine-grained polymict conglomerates interbedded with
mylonitized quartzitic sandstones and schistose shales. The over-
lying Bimbo sandstones (130 m thick) are a thick fluviatile
sequence of massive dark shales interstratified with thick beds
of conglomerate, sandstones and quartzarenites with cross-
stratification, which, according to Alvarez (1995) pass eastwards
into the fluvio-lacustrine Kembé-Nakando sandstones of the
Fouroumbala-Bakouma Basin in eastern CAR, and may be glacial
in origin (Alvarez 2000).
The Bimbo sandstones are unconformably overlain by the
Bangui Series (Fig. 13.4), a thick carbonate sequence exposed
mainly along the Ubangui River to the south of Bangui. The
Bangui Series carbonates have been affected by Pan-African
related deformation and, therefore, are considered to be Late Neo-
proterozoic in age. They have been correlated with the Schisto-
Calcaire Subgroup of the West Congolian Supergroup (Alvarez
1995). They have been subdivided into the distal Lessé Formation,
the proximal outer shelf Bobassa Formation, the offshore barrier
facies Mboma Formation and the lagoonal Fatima Formation.
These carbonates comprise laminated microbial limestones, rhyth-
mically bedded carbonates with clay and silt-rich interval inter-
preted as shallow subtidal or intertidal deposits, more massively
bedded limestones, and are capped by grey microbial limestones
intercalated with black dolomites indicating deposition in a
lagoonal environment (Alvarez 1992). Alvarez (1995) interpreted
these sedimentary rocks as characterizing an extensive carbonate
ramp that developed along the northern edge of the Central
Africa belt. However, it is now thought that they developed in
local intracratonic basins and they have been correlated with the
Bakouma Subgroup in more easterly regions of CAR.
Glaciogenic character of the WCB diamictites
Until 1970, the diamictites of the West Congolian Group were gen-
erally considered to be glacial in origin, and in 1986 the Upper Dia-
mictite of the Congo was described as a marine tillite associated
with an ice cap on the Chaillu massif (Trompette & Boudzoumou
1988). In the Congo, however, Cahen & Lepersonne (1981) con-
sidered the Lower Diamictite Formation to be a tectonically
related gravity flow deposit. Subsequently, Trompette (1994) con-
sidered the Upper Diamictite Formation to be a gravity-flow
deposit originating from reworking of glacial material along the
margins of the basin by small mountain glaciers developed on per-
ipheral highlands. In Angola, the discussion was more acute. Both
diamictites were interpreted as being tilloids, that is, non-glacial
and deposited as debris flows (Schermerhorn & Stanton 1963),
although the possibility of reworking of mountain glacier deposits
(i.e. Alpine glacial deposits) was suggested for the lower diamic-
tites (Schermerhorn 1974). Kröner & Correia (1973) reinterpreted
both diamictite levels in Angola as being true glaciomarine dia-
mictites with local intercalations of continental tillites. Schermer-
horn (1981) continued to ascribe the deposits to a deep-water
submarine turbiditic setting, arguing against a glacial origin. Simi-
larly, Vellutini & Vicat (1983) argued against any glacial influence
and considered both diamictites of the West Congo Belt as basal
conglomerates. In his synthesis of available data and taking all
the arguments into account, Trompette (1994) surmises that
there was most likely some glacial input, even if they do not rep-
resent true glaciomarine deposits.
More recently, Tack et al. (2006) critically reappraised the
origin of Neoproterozoic diamictites from central Afria. As a result,
detailed micro-structural and -textural analyses (based on tech-
niques applied to recent unconsolidated tills by several authors
such as Evans & Benn 2004; Benn & Evans 1996; Menzies
2000a, b, 2006; Phillips 2006; Phillips et al. 2007; Van der Meer
1993, 1997, 2003) were carried out on samples of the Lower and
Upper Diamictite of the Bas-Congo Basin, the Akwokwo Tillite
of the Lindian Basin, and the Niari Formation of the Nyanga
Basin (Delpomdor 2007c, Delpomdor et al. 2008, Préat 2008,
Préat et al. 2008). These observations provide important new con-
straints on the various sedimentological and depositional environ-
ments. The diamictites from the SW Gabon, Bas-Congo and the
Lindian Basin are principally composed of massive and stratified
mud-supported matrix with rare intercalation of varved mudstones
 THE WEST CONGO AND LINDI/UBANGI SUPERGROUPS
and sandstones. Clasts within the diamictites are only rarely
striated, do not display crescent or conchoidal fractures, and no
glacial pavements have been observed. Thus, a directly subglacial
environment cannot be identified. The combination of observed
microstructures demonstrates that all these diamictites were depos-
ited under high strain rates with moderate to high stress conditions,
influenced by water pressure or clays within the beds. The cumu-
lative evidence of overprinting microstructures suggests that all
these diamictites were deposited in fluvioglacial or glaciomarine
conditions in proximal subglacial and/or more distal subaqueous
environments.
Chemostratigraphy
The only chemostratigraphic data available for the WCB are from
a handful of recent studies (Frimmel et al. 2006; Poidevin 2007;
Straathof et al. 2008; Préat et al. 2010, 2011). Strontium isotope
ratios measured in carbonate rocks of the West Congo Supergroup
range from 0.7066 to 0.7109 (Poidevin et al. 2007), although the
highest ratios are surely overprints. The least-altered and most
Sr-rich samples in the Haut Shiloango Subgroup have 87Sr/86Sr
ratios of c. 0.7068 –0.7072, compared to ratios of 0.7074–
0.7075 in the lower Schisto-Calcaire Subgroup (Frimmel et al.
2006; Poidevin 2007). Ten d13Ccarbonate values in the Haut
Shiloango Subgroup range from 3.2 to 8‰. The Schisto-Calcaire
Subgroup defines a larger range in d13Ccarbonate, from – 5.4 to
9‰. Values in the lowermost part of the subgroup are consistently
negative (Frimmel et al. 2006; Straathof et al. 2008; Préat et al.
2011), whereas those in the upper part of the subgroup are
erratic, probably due to temporarily elevated evaporation rates
(Frimmel 2009). The near-shore depositional environment for
the Bas-Congo carbonates is also demonstrated geochemically
by Frimmel (2009) who concluded that carbon isotopes obtained
from successions developed in such environments, particularly
when elevated evaporation has occurred, should be treated
with great care and are not suitable for chemostratigraphic
correlation.
Palaeolatitude and palaeogeography
The Precambrian palaeomagnetic database for the Congo Craton is
extremely sparse, with very few of the available data meeting
modern reliability criteria. Nevertheless, the latest Mesoprotero-
zoic position of the Congo can be constrained by poles obtained
from the São Francisco Craton (D’Agrella-Filho et al. 1990).
These data place the western margin of the craton (present day
co-ordinates) at intermediate palaeolatitudes (45 –508) for the time
period 1.1 –1.0 Ga (De Waele et al. 2008). Neoproterozoic palaeo-
geography is constrained by two poles from the 795 + 7 Ma
Gagwe lavas of Tanzania (Piper 1972; Meert et al. 1995, Ar/Ar
age after Deblond et al. 2001) and the 743 + 30 Ma Mbozi intru-
sive rocks (Meert et al. 1995, K –Ar biotite age after Brock 1968).
According to these data, by the early Cryogenian the Congo craton
had moved into tropical latitudes, positioning the present-day
western margin close to the equator, and by 755 Ma, had rotated
some 908 while remaining in the tropics. Collisional deformation
related to consolidation of western Gondwana was initiated c.
600 Ma in the western Congo and 580 –530 Ma in the south and
southeast. Palaeomagnetic data from the Sinyai dolerites of
Kenya (Meert & Van der Voo 1996) and geological evidence indi-
cate that proto-Gondwana had assembled by the Early Cambrian.
The implications of the available palaeomagnetic data are that
the Neoproterozoic sedimentary rocks of the West Congo and
Lindi Supergroups were deposited in tropical latitudes. In terms
of more detailed palaeogeography, it is important to point out
that the paucity of Precambrian data for the Congo Craton
means the polarity of the palaeopoles is uncertain.
191
Geochronological constraints
Radiometric data from the West Congo and Lindi/Ubangi Super-
groups are sparse and geochronological constraints are based
largely on bulk detrital zircon analyses and correlation with bette-
dated successions elsewhere on the Congo Craton. The Lower
Diamictite Unit is constrained only to be younger than c. 910 Ma
based on the youngest U –Pb ages from the Mayumbian Group
volcanic rocks, which underlie the Sansikwa Subgroup (Tack
et al. 2001). The youngest detrital zircon ages obtained from the
overlying Haut-Shiloango Subgroup yield ages of 547 + 45,
671 + 20 and 709 + 20 Ma (1s errors, 10% discordance)
from which, given the errors, a possible age of 650 Ma is inter-
preted by Frimmel et al. (2006). Clearly, more reliable age
data are required to more accurately constrain maximum and
minimum depositional ages for both the Lower and Upper Diamic-
tite Formations. The 40Ar/39Ar age of 566 + 42 Ma obtained from
a sill intruding the Lower Diamictite Unit and dating greenschist
metamorphism during west Gondwana amalgamation (Frimmel
et al. 2006) provides a loose minimum age constraint for the
entire sedimentary package.
Correlation between the West Congolian Group and the
Lindi Supergroup
Updated lithostratigraphic correlations are summarized in
Figure 13.4 and are based on Sr isotope ratios obtained from
carbonates, SHRIMP data (Tack et al. 2001; Frimmel et al.
2006; Poidevin 2007) and revised lithological relationships. The
Neoproterozoic West Congolian Group sedimentary rocks are
relatively well constrained in comparison to those of the Lindian
Supergroup in the northern part of the Congo Craton. In the Bas-
Congo Basin, the West Congolian Group starts with siliciclastic
deposits of the Sansikwa Subgroup. The lower age of the Sansikwa
Subgroup is constrained by the underlying 912 + 7 Ma rhyolites
(Tack et al. 2001). The upper age of the Sansikwa Subgroup is
less well known, but has been inferred by Frimmel et al. (2006)
to be c. 750 Ma based on correlation of the overlying Lower Dia-
mictite unit with the Sturtian glacial event (assuming the Sturtian
to have occurred at 720– 750 Ma, although this age is controver-
sial). No carbonate rocks are observed in the Sansikwa Subgroup;
the oldest carbonate rocks occur at the top of the post-Sturtian
Haut-Shiloango Subgroup (Fig. 13.3). However, Poidevin (2007)
proposed that the oldest limestones in the Lindi Supergroup
(the Lenda Formation, Ituri Group) are pre-Sturtian and 730–
755 Ma in age on the basis 87Sr/86Sr ratios and correlation with
the chemostratigraphic compilation of Halverson et al. (2007).
However, the isotopic ratios reported by Poidevin (2007) are extre-
mely variable (0.70663 –0.71090), making correlation difficult.
Given the lack of any carbonate rocks prior to deposition of the
lower diamictite unit (considered to be Sturtian in age in the Bas-
Congo), the limestones of the Lenda Formation (Ituri Group) are
considered to be post-Sturtian in age, as traditionally presented
in the literature (Trompette 1994), and correlated with the Haut-
Shiloango Subgroup of the Bas-Congo Basin and the Bougboulou
Series in the Fouroumbala-Bakouma Basin (Fig. 13.4). In Gabon,
the Bouenza Group is a correlative of Haut-Shiloango Subgroup
(Chevallier 2002).
The Ituri Group is overlain by the Akwokwo Tillite which, as
mentioned above, Poidevin (2007) correlated with the Lower Dia-
mictite Formation of the West Congolian Group. However, recent
field observations of Neoproterozoic diamictites in SW Gabon
and/or in the DRC Bas-Congo and Katanga regions show that car-
bonate rocks comprising massive zoned pink to greyish dolomites
immediately cap the upper diamictites in these regions, but are not
found in association with the lower diamictites. These ‘cap carbon-
ates’ are also recorded in the Fouroumbala-Bakouma Basin (the
Bakouma Formation capping the Bondo Tillite) (Fig. 13.3). In
192 J. TAIT ET AL.
the Lindian Basin, large lenses of zoned pink dolomites are
embedded in the Mamungi Formation overlying the Akwokwo
Tillites. The close association of the Akwokwo Tillite with
massive zoned pink to greyish dolomites is also confirmed from
the RMCA (Tervuren) sample collection (Tack, pers. comm.).
Hence, the occurrence in DRC and adjacent CAR of the cap car-
bonate is thought to be indicative of the Marinoan glacial event
(Fig. 13.4). No carbonate is observed capping the diamictites of
the Bangui Basin, but the Kembé Formation conglomerates are
commonly correlated with the Bondo Tillite of the Fouroumbala-
Bakouma Basin (Poidevin 1976, 1985, 2007; Bessoles & Tromp-
ette 1980; Cornacchia & Giorgi 1986). Finally, oolitic carbonate
rocks of the Lindi Supergroup occur in the Alolo Formation
(Aruwimi Group overlying the Lokoma Group; Fig. 13.3). Com-
parable oolitic carbonate rocks of post-Marinoan age are also
known up sequence in the West Congo and Katanga regions.
In the Bili and Dialinga Formations (upper part of the Bakouma
Series) of the Fouroumbala-Bakouma Basin, some metabasalt
layers are intercalated. In the Aruwimi Group of the Lindian
Basin, Verbeek (1970) considers the Kaparata Breccia to be a con-
glomerate, possibly including some clasts of a volcanic rock. In
fact, the RMCA (Tervuren) Kaparata samples are a hyaloclastic
breccia, and are reminiscent of rocks observed in the Bas-Congo
Basin, suggesting the occurrence in the Lindian Basin of
pillow lavas and an up to now radiometrically undated tholeiitic
basaltic event.
In conclusion, unlike Poidevin (2007), we correlate the Akwokwo
Tillite (Lindian Basin) with the Bondo Tillite (Fouroumbala-
Bakouma Basin), the Upper Diamictite (Bas Congo) and Niari
Tillite (SW Gabon), and conclude a Marinoan age. However, as
already suggested by Poidevin (2007), the Upper Bakouma
Series are correlated with the Aruwimi Group, as redefined in
this paper (Fig. 13.3). This correlation is tentatively extended to
the Mpioka Subgroup of the WCB (in SW Gabon and DRC Bas-
Congo). Finally, the Redbeds of the Inkisi ‘Group’ of the WCB,
with a maximum depositional age of c. 550 Ma (i.e. post-Pan
African) are correlated with the Banalia ‘Group’ of NE DRC (for-
merly considered as the uppermost unit of the Neoproterozoic
Lindi Supergroup).
Without new geochronological and palaeomagnetic constraints
for these Neoproterozoic basins it will be difficult to refine
these correlations.
In Gabon, the fieldwork was carried out under the terms of the ‘Programme
Sysmin, 8ème Fonds Européen du Développement au groupement BRGM-
CGS-SANDER-MRAC’. J.T. and G.S. gratefully acknowledge financial
support from the European Commission FP6 Programme (Marie Curie Excel-
lence Grant). H. Frimmel and U. Zimmermann are gratefully thanked for
their helpful and constructive reviews. This represents a contribution of the
IUGS- and UNESCO-funded IGCP (International Geoscience Programme)
Project #512.
References
Abdelsalam, M. G., Liegeois, J. P. & Stern, R. J. 2002. The Saharan
Metacraton. Journal of African Earth Sciences, 34, 119– 136.
Alvarez, P. 1992. Répartition de la sédimentation dans le golfe Protéro-
zoı̈que supérieur du Schisto-calcaire au Congo et Gabon. Implications
en Afrique centrale. Palaeogeography, Palaeoclimatology, Palaeo-
ecolology, 96, 281– 297.
Alvarez, P. 1995. Evidence for a Neoproterozoic carbonate ramp on the
northern edge of the Central African Craton: relations with Late
Proterozoic intracratonic troughs. Geologische Rundschau, 84,
636– 648.
Alvarez, P. 2000. A quantitative method for the study of non-
fossiliferous clastic formations: pre-Pan-African sandstones from
central Africa and the northern Democratic Republic of Congo
[ex-Zaire]. Journal of African Earth Sciences, 31, 263– 284.
Babet, V. 1935. Esquisse géologique provisoire de la région comprise
entre Bangui et la frontière du Cameroun. Chronique Mineralogique
Col. Paris, 38, 160– 164.
Benn, D. I. & Evans, D. J. A. 1996. The interpretation and classification
of subglacially-deformed materials. Quaternary Science Reviews, 15,
23 – 52.
Bessoles, B. & Trompette, R. 1980. Géologie de l’Afrique: la chaı̂ne
panafricaine ‘zone mobile d’Afrique Centrale (partie sud) et zone
mobile soudanaise’. Mémoire du Bureau de la Recherche Geleolgi-
que et Minière, 92, 398.
Bigotte, G. & Bonifas, G. 1968. Faits nouveaux sur la géologie de la
région Bakouma. Chronique de Mines et Recherche Minerale, 36,
43 – 46.
Bueno, J. F., Oliveira, E. P., McNaughton, N. J. & Laux, J. H. 2009.
U –Pb dating of granites in the Neoproterozoic Sergipano Belt,
NE-Brazil: implications for the timing and duration of continental
collision and extrusion tectonics in the Borborema Province. Gond-
wana Research, 15, 86– 97.
Cahen, L. 1978. La stratigraphie et la tectonique du Supergroupe Ouest-
Congolien dans les zones médiane et externe de l’orogénèse Ouest-
Congolien (Pan-African) au Bas-Zaı̈re et dans les régions voisines.
Annales de la Musée Royale de l’Afrique Centrale, Tervuren, in 88,
Science Geologique, 83, 150.
Cahen, L. & Lepersonne, J. 1981. Proterozoic diamictites of Lower
Zaire. In: Hambrey, M. A. & Harland, H. W. (eds) Earth’s
Pre-Pleistocene Glacial Record. Cambridge University Press,
Cambridge, 153–157.
Chevallier, L., Makanga, J. F. & Thomas, R. J. 2002. Carte géologique
de la république gabonaise, 1:1,000,000. Notice explicative. Council
for Geoscience, South Africa.
Cornacchia, M. & Giorgi, L. 1986. Les séries Précambriennes d’origine
sédimentaire et volcano-sédimentaire de la République Centrafi-
quaine. Annales de la Musée Royale de l’Afrique Centrale, Tervuren,
in 88, Science Geologique, 93, 51.
Dadet, P. 1969. Notice explicative de la carte géologique de la république
du Congo-Brazzaville au 1:500 000. Mémoire du Bureau de la
Recherche Geleolgique et Minière, 40, 103.
D’Agrello-Filho, M. S., Pacca, I. G., Renne, P. R., Onstott, T. R. C.
& Teixeira, W. A. 1990. Paleomagnetism of middle Proterozoic
(1.01– 1.08 Ga) mafic dykes in Southeasterrn Bahia State –São Fran-
cisco Craton, Brazil. Earth and Planetary Science Letters, 101,
332– 348.
Daly, M. C., Lawrence, S. R., Diemu-Tshiband, K. & Matouana, B.
1992. Tectonic evolution of the Cuvette Centrale, Zaire. Journal of
the Geological Society, London, 149, 539–546.
Deblond, A., Punzalan, L. E., Boven, A. & Tack, L. 2001. The Mala-
garazi supergroup of southeast Burundi and its correlative Bukoba
supergroup of northwest Tanzania: Neo- and Meso-proterozoic con-
straints from Ar – Ar ages of mafic intrusive rocks. Journal of African
Earth Science, 32, 435– 449.
Delhaye, F. & Sluys, M. 1923– 1924 and 1928– 1929. Observations
ayant servi a l’elaboration de l’ ‘Esquisse geologique du Congo occi-
dental’. Etude du systeme Schisto-Calcaire. 1er– 3e memoire.
Annales de la Société Belge de Géologie, 1923– 1924, C. 50– 91 et
1928– 1929, C. 69– 114.
Delor, C., Theveniaut, H. et al. 2008. New insights into the Precam-
brian Geology of Angola: basis for an updated lithochronological
framework at 1:2,000,000 scale. 22nd Colloquium African Geology,
Hammamet, Tunisia, Abstracts volume, 52 –53.
Delpomdor, F. 2007a. Lithostratigraphie et sédimentologie du faisceau
de Sekelolo Sh8 (Sous-Groupe du haut-Shiloango), de la Formation
de la Diamictite supérieure et de la Dolomie rose C1 du fiasceau du
Kwilu CI (Sous-Groupe du Schisto-Calcaire), anticlinal de Congo
Dia Kati, Bas-Congo, République Démocratique du Congo (RDC),
unpublished Internal report of Royal Museum of Central Africa, Ter-
vuren (Belgium).
Delpomdor, F. 2007b. Etude des dépots diamictitiques du Néoprotérozoı̈-
que supérieure en République Démocratique du Congo et au
sud-ouest du Rwanda, unpublished internal report of Royal
Museum of the Central Africa, Tervuren (Belgium).
Delpomdor, F. 2007c. Lithostratigraphie et sédimentologie de la chaı̂ne
Ouest Congolienne du Néoprotérozoı̈que supérieur (Formation de
 THE WEST CONGO AND LINDI/UBANGI SUPERGROUPS
la Diamictite supérieure et Sous-groupe du Schisto-Calcaire) Bas-
Congo, République Démocratique du Congo, unpublished MSc
thesis, Free University of Brussels.
Delpomdor, F., Tack, L. & Préat, A. 2008. Microstructures in the Neo-
proterozoic tillites around the Congo River Basin (CRB), Democratic
Republic of the Congo (DRC)– Comparison with the Karoo tillites
from the Dekese borehole in the CRB. 22nd Colloquium African
Geology, Hammamet, Tunisia, Abstracts volume.
Delpomdor, F., Tait, J., Tack, L. & Préat, A. 2011. Neoproterozoic
lithostratigraphy in the Democratic Republic of Congo (DRC). 23rd
Colloquiom of African Geology, Johannesburg.
De Paepe, P., Hertogen, J. & Tack, L. 1975. Mise en évidence de laves
en coussin dans les faciès volcaniques basiques du massif de Kibungu
(Bas-Zaı̈re) et implications pour le magmatisme ouest-congolien.
Annales de la Société Belge de Géologie, 98, 251– 270.
De Waele, B., Liégeois, J.-P., Nemchin, A. A. & Tembo, F. 2006. Isoto-
pic and geochemical evidence of Proterozoic episodic crustal rework-
ing within the Irumide Belt of south-central Africa, the southern
metacratonic boundary of an Archaean Bangweulu Craton. Precam-
brian Research, 148, 225–256.
De Waele, B., Johnson, S. P. & Pisarevsky, S. A. 2008. Palaeoproter-
ozoic to Neoproterozoic growth and evolution of the eastern Congo
Craton: its role in the Rodinia puzzle. Precambrian Research, 160,
127– 141.
Evans, D. A. D. 2000. Stratigraphic, geochronological, and paleomagnetic
constraints upon the Neoproterozoic climatic paradox. American
Journal of Science, 300, 347–433.
Evans, D. J. A. & Benn, D. I. 2004. A Practical Guide to the Study of
Glacial Sediments. Arnold Publishers, London.
Frimmel, H. 2009. Trace element distribution in Neoproterozoic
carbonates as palaeoenvironmental indicator. Chemical Geology,
258, 338– 353.
Frimmel, H. E., Tack, L., Basei, M. S., Nutman, A. P. & Boven, A.
2006. Provenance and chemostratigraphy of the Neoproterozoic
West Congolian Group in the Democratic Republic of Congo.
Journal of African Earth Sciences, 46, 221– 239.
Gérard, G. 1958. Carte géologique de l’Afrique Equatoriale Française
au 1/2 000 000 avec notice explicative. Brazzaville Directorate des
Mines et Géologie A.E.F., 198, 4 feuilles.
Gérard, G. & Gérard, J. 1952. Stratigraphie du Précambrien
d’Oubangui-Chari occidental (A.E.F.) et essai de corrélations avec
les territoires voisins. 19th International Geological Congress
(Algiers), A.S.G.A., 145– 153.
Halverson, G. P., Dudas, F. O., Maloof, A. C. & Bowring, S. A.
2007. Evolution of the 87Sr/86Sr composition of Neoproterozoic sea-
water. Palaeogeography Palaeoclimatology Palaeoecology, 256,
103– 129.
Hoffman, P. F. 2011. Glacigenic and associated strata of the Otavi car-
bonate platform and foreslope, northern Namibia: evidence for
extreme glacioeustatic fluctuation. In: Arnaud, E., Halverson,
G. P. & Shields-Zhou, G. (eds) The Geological Record of Neo-
proterozoic Glaciations. Geological Society, London, Memoirs, 36,
195– 210.
Hudeley, H. & Belmonte, Y. 1966. Carte géologique de la République
Gabonaise, échelle 1/1 000 000. Bureau de Recherches Géologiques
et Minières (B.R.G.M.).
Kadima, K. E., Mwene, N. S. S. & Francis, L. 2011. A Proterozoic-rift
origin for the structure and the evolution of the cratonic Congo basin.
Earth and Planetary Science Letters, 304, 240–250.
Kampunzu, A. B., Kapenda, D. & Manteka, B. 1991. Basic magmatism
and geotectonic evolution of the Pan African belt in central Africa:
evidence from the Katangan and West Congolian segments. Tectono-
physics, 190, 363–371.
Kroner, A. & Correia, H. 1973. Further evidence for glaciogenic origin
of Late Precambrian mixtites in Angola. Nature –Physical Science,
246, 115– 117.
Legoux, P. & Hourcq, V. 1943. Esquisse géologique de l’Afrique Occi-
dental Française, Bulletin du Service des Mines de l’Afrique Occiden-
tale Française.
Lepersonne, J. 1951. Données nouvelles sur la stratigraphie des terrains
anciens du Bas-Congo. Bulletin de la Société Belge de Géologie, 60,
169– 189.
193
Master, S. & Wendorff, M. 2011. Neoproterozoic glaciogenic diamic-
tites of the Katanga Supergroup, Central Africa. In: Arnaud, E.,
Halverson, G. P. & Shields-Zhou, G. A. (eds) The Geological
Record of Neoproterozoic Glaciations. Geological Society of
London, Memoir, 36, 173– 184.
Meert, J. G. & Van der Voo, R. 1996. Paleomagnetic and 40Ar/39Ar
study of the Sinyai Dolerite, Kenya: implications for Gondwana
assembly. Journal of Geology, 104, 131– 142.
Meert, J. G., Van der Voo, R. & Ayub, S. 1995. Paleomagnetic investi-
gation of the Neoproterozoic Gagwe lavas and Mbozi complex, Tan-
zania and the assembly of Gondwana. Precambrian Research, 74,
225– 244.
Menzies, J. 2000a. Micromorphological analyses of microfabrics and
microstructures, indicative of deformation processes, in glacial sedi-
ments. In: Maltman, A. J., Hubbard, B. & Hambrey, M. J. (eds),
Deformation of Glacial Materials. Geological Society, London,
176, 245–248.
Menzies, J. 2000b. Microstructures in diamictites of the lower Gondwana
Formation (Huronian), near Elliot Lake Ontario: evidence for
deforming-bed conditions at the grounding line? Journal of Sedimen-
tary Research, 70, 210–216.
Menzies, J., van der Meer, J. J. M. & Rose, J. 2006. Till as a glacial ‘tec-
tomict’, its internal architecture, and the development of a ‘typing’
method for till differentiation. Geomorphology, 75, 172– 200.
Mestraud, J.-L. 1952. Formation du socle en Oubangui-Chari
central. 19th International Geological Congress (Algiers) A.S.G.A.,
155– 162.
Mestraud, J.-L. 1953. Notice explicative sur la feuille Bangassou Ouest.
Carte géologique de reconnaissance au 1/500.000. Direction Mines
et Géologie, l’Afrique Equatoriale Française, Brazzaville.
Moloto-A-Kenguemba, G. R., Trindade, R. I. F., Monie, P., Nedelec,
A. & Siqueira, R. 2008. A late Neoproterozoic paleomagnetic pole
for the Congo Craton: tectonic setting, paleomagnetism and geochro-
nology of the Nola dike swarm (Central African Republic). Precam-
brian Research, 164, 214– 226.
Noce, C. M., Pedrosa-Soares, A. C., da Silva, L. C., Armstrong, R. &
Piuzana, D. 2007. Evolution of polycyclic basement complexes in
the Araçuai Orogen, based on U–Pb SHRIMP data: implications
for Brazil– Africa links in Paleoproterozoic time. Precambrian
Research, 159, 60– 78.
Oliveira, E. P., Toteu, S. F. et al. 2006. Geologic correlation
between the Neoproterozoic Sergipano belt (NE Brazil) and the
Yaounde belt (Cameroon, Africa). Journal of African Earth Sciences,
44, 470– 478.
Pedrosa-Soares, A. C., Alkmin, F. F., Tack, L., Noce, C. M., Babinski,
M., Silva, L. C. & Martins-Neto, M. A. 2008. Similarities and
differences between the Brazilian and African counterparts of the
Neoproterozoic Araçuai– West Congo Orogen. In: Pankhurst,
R. J., Trouw, R. A. J., Brito Neves, B. B. & De Wit, M. J. (eds)
West Gondwana: Pre-Cenozoic Correlations Across the South Atlan-
tic Region. Geological Society, London, Special Publications, 294,
153– 172.
Phillips, E. 2006. Micromorphology of a debris flow deposit: evidence of
basal shearing, hydrofacturing, liquefaction and rotational defor-
mation during emplacement. Quaternary Science Reviews, 25,
720– 738.
Phillips, E., Merrit, J., Auton, C. & Golledge, N. 2007. Microstruc-
tures in subglacial and proglacial sediements: understanding faults,
folds and fabrics, and the influence of water on the style of defor-
mation. Quaternary Science Reviews, 26, 1499–1528.
Pin, C. & Poidevin, J. L. 1987. U– Pb zircon evidence for a Pan-African
granulite facies metamorphism in the Central-African-Republic – a
new interpretation of the high-grade series of the northern border of
the Congo Craton. Precambrian Research, 36, 303– 312.
Pinna, P., Cocherie, A., Thieblemont, D., Feybesse, J.-L. & Lagny, P.
1996. Evolution géodynamique du Craton Est-Africain et détermi-
nisme gı̂tologique. Geodynamic evolution and metallogenic controls
in the East-African Craton (Tanzania, Kenya, Uganda). BRGM,
Chronique Recherche Miniere, 525, 33 – 43.
Piper, J. D. A. 1972. Paleomagnetic study of Bukoban System, Tanzania.
Geophysical Journal of the Royal Astronomical Society, 28,
111–127.
194 J. TAIT ET AL.
Poidevin, J.-L. 1976. Les formations du Précambrien supérieur de la
région de Bangui (R.C.A.). Bulletin de la Societé Géologique de
France, 18, 999– 1003.
Poidevin, J.-L. 1979a. Echelle stratigraphie des formations précam-
briennes de Centrafrique (R.C.A.). Résumé 10ème Colloque de Géolo-
gie Africaine, Montpellier.
Poidevin, J.-L. 1979b. Les basaltes et dolérites Précambrien supérieur de
la région de Bakouma (Empire centraficain). 7ème Réuion Annuelle
des Sciences de la Terre, Lyon, 374.
Poidevin, J.-L. 1985. Le Protérozoı̈que supérieur de la République centra-
fricaine. Annales de Muséum Royal d’Afrique Centrale Tervuren,
Belgique, Série in 8ème, Sciences Géologiques, 91, 75.
Poidevin, J.-L. 2007. Stratigraphie isotopique du strontium et datations
des formations carbonatées et glaciogéniques néoprotérozoı̈ques du
Nord et de l’Ouest du Craton du Congo. Comptes Rendus Geoscience,
339, 259– 273.
Poidevin, J.-L., Alabert, J. & Miauton, J. D. 1981. Géologie des séries
du Précambrien supérieur de la région de Bakouma (République
centrafricaine). Bulletin du Bureau de Recherches Geologiques et
Minieres, Section IV, 4, 313–320.
Préat, A. 2008. Etude au microscope polarisant de 165 lames minces de
roches carbonatées du synclinal de la Nyanga (feuilles à 1:200 000 de
Fougamou et de N’Dendé). Rapport Université Libre de Bruxelles,
Tervuren, Belgique.
Préat, A., Bouton, P., Kolo, K., Prian, J.-P., Simo Ndounze, S. &
Thiéblemont, D. 2008. Sédimentologie et isotopes (carbone
et oxygène) des carbonates Précambriens du Gabon: rapport au
mode de fonctionnement des bassins Néo- et Paléoprotérozoı̈ques.
Communication Spec. Session Geologica Belgica, Namur, Belgique.
Préat, A., Delpomdor, F., Kolo, K., Gillan, D. & Prian, J.-P. 2011.
Stromatolites and cyanobacterial mats in peritidal evaporitive
environments in the Neoproterozoic of Bas-Congo (Democratic
Republic of Congo) and South of Gabon. In: Tewari, V. C. & Seck-
back, J. (eds) Stromatolites: Interactions of Microbes with Sediment,
Cellular Origin and Life in Extreme Habitats and Astrobiology.
Springer Verlag, 43 –63.
Préat, A., Kolo, K., Prian, J.-P. & Delpomdor, F. 2010. A peritidal eva-
porite environment in the Neoproterozoic of South Gabon (Schisto-
Calcaire Subgroup, Nyanga Basin). Precambrian Research, 177,
253– 265.
Prian, J.-P. 2008. Notice géologique et ressources minérales de la carte
de N’Dendé à 1/200 000. Document provisoire, carte géologique
du Gabon.
Schermerhorn, L. J. G. 1974. No evidence for glacial origin of Late
Precambrian tilloids in Angola. Nature, 252, 114–115.
Schermerhorn, L. J. G. 1981. Late Precambrian Tilloids of northwest
Angola. In: Hambrey, M. J. & Harland, W. B. (eds) Earth’s
Pre-Pleistocene Glacial Record. Cambridge University Press,
Cambridge, 158– 161.
Schermerhorn, L. J. G. & Stanton, W. I. 1963. Tilloids in the West
Congo geosyncline. Quarterly Journal of the Geological Society of
London, 119, 201–234.
Stanton, W. I., Schermerhorn, L. J. G. & Korpershoek, H. R. 1963.
The West Congo System. Boletim dos Serviços de Geologia e Minas
de Angola, 8, 69 –78.
Straathof, G. B., Tait, J., Cibambula, E., Kanda Nkula, V. &
Zimmermann, U. 2008. Constraints on the glacial events from the
Neoproterozoic West Congolian Group, 33rd International Geologi-
cal Congress, Oslo.
Tack, L., Wingate, M. T. D., Liégeois, J.-P., Fernandez-Alonzo, M.
& Deblond, A. 2001. Early Neoproterozoic magmatism
(1000 –910 Ma) of the Zadinian and Mayumbian Groups (Bas-
Congo), onset of Rodinia rifting at the western edge of the Congo
Craton. Precambrian Research, 110, 277–306.
Tack, L., Fernandez-Alonso, M., Kanda Nkula, V., Mpoyi, J.,
Delvaux, D., Trefois, Ph. & Baudet, D. 2006. Neoproterozoic Dia-
mictites around the Congo River Basin: a critical reappraisal of their
origin. 21st Colloquium African Geology (CAG21), 3– 5 July 2006,
Maputo, Mozambique, Abstract Book, 152– 153.
Tack, L., Delvaux, D. et al. 2008. The 1000 m thick Redbeds sequence
of the Congo River Basin (CRB): a generally overlooked testimony in
Central Africa of post-Gondwana amalgamation (550 Ma) and pre-
Karoo break-up (320 Ma). 22nd Colloquium on African Geology,
Hammamet, Tunisia, Abstracts volume, 86 –88.
Tack, L., Delvaux, D. et al. 2009a. The 1000 m thick Redbeds sequence
of the Congo River Basin (CRB): a generally overlooked testimony in
Central Africa of post-Gondwana amalgamation (550 Ma) and pre-
Karoo break-up (320 Ma). Geological Society of London Fermor
Meeting, Edinburgh.
Tack, L., Wingate, M. T. D. et al. 2009b. The Mesoproterozoic
‘Kibaran Event’ in Central Africa: a 1375 Ma intracratonic emplace-
ment of a Large Igneous Province (LIP). Geological Society of
London Fermor Meeting, Edinburgh.
Thiéblemont, D., Castaing, C., Billa, M., Bouton, P. & Préat, A. et
collaborateurs. 2009. Notice explicative de la carte géologique et des
Ressources minérales de la République gabonaise à 1:1 000 000. Edi-
tions DGMG, Ministère des Mines, du Pétrole, des Hydrocarbures.
Libreville.
Toteu, S. F., Fouateu, R. Y. et al. 2006. U– Pb dating of plutonic rocks
involved in the nappe tectonic in southern Cameroon: consequence
for the Pan-African orogenic evolution of the Central African fold
belt. Journal of African Earth Sciences, 44, 479– 493.
Trompette, R. 1994. Geology of Western Gondwana (2000– 500 Ma).
Pan-African-Brasiliano aggregation of South America and Africa.
Balkema, Rotterdam.
Trompette, R. & Boudzoumou, F. 1988. Paleogeographic significance
of stromatolitic buildups on Late Proterozoic Platforms– the
example of the West Congo Basin. Palaeogeography, Palaeoclima-
tology, Palaeoecology, 66, 101–112.
Van der Meer, J. J. M. 1993. Microscopic evidence of subglacial defor-
mation. Quaternary Science Reviews, 12, 553– 587.
Van der Meer, J. J. M. 1997. Particle and aggregate mobility in till:
microscopic evidence of subglacial processes. Quaternary Science
Reviews, 16, 827–831.
Van der Meer, J. J. M. 1999. Particle and aggregate mobility in till:
microscopic evidence of subglacial processes. Quaternary Science
Reviews, 22, 1659–1685.
Van der Meer, J. J. M., Menzies, J. & Rose, J. 2003. Subglacial till:
the deforming glacier bed. Quaternary Science Reviews, 22,
1659– 1685.
Vellutini, P. & Vicat, J. P. 1983. On the origin of basal conglomeratic
formations of the West-Congolan Geosyncline (Gabon, Congo,
Zaire, Angola). Precambrian Research, 23, 87– 101.
Verbeek, T. 1970. Géologie et lithostratigraphie du Lindien (Précambrien
supérieur du nord de la République Démocratique du Congo). Annale
de la Musée Royale de l’Afrique Centreale, série n88, 66, 309.
Vicat, J.-P., Léger, J. M., Nsifa, E., Piguet, P., Nzenti, J. P., Tcha-
meni, R. & Pouclet, A. 1996. Distinction au sein du craton congolais
du Sud-Ouest du Cameroun, de deux épisodes doleitiques initiant les
cycles orogeniques eburneé en (Paléoprotérozoique) et panafricain
(Néoprotérozoique). Comptes Rendus de l’Académie des Sciences.
Série 2. Sciences de la terre et des planètes, 323, 575– 582.
 Chapter 14

Glaciogenic and associated strata of the Otavi carbonate platform and foreslope, northern
Namibia: evidence for large base-level and glacioeustatic changes

PAUL F. HOFFMAN1,2
1
Department of Earth and Planetary Sciences, Harvard University, Cambridge, MA 02138, USA
2
School of Earth and Ocean Sciences, University of Victoria, Victoria, BC V8W 2Y2, Canada
(e-mail: hoffman@eps.harvard.edu)

Abstract: Two discrete, mappable, glaciogenic formations occur within the Otavi Group, a 3 + 1-km-thick carbonate-dominated
platform of late Neoproterozoic age, developed on the SW promontory of the Congo craton in northern Namibia and exposed in bordering
late Ediacaran fold belts. Each is overlain abruptly by an expanded postglacial carbonate sequence, the younger of which begins with a
globally-correlative transgressive dolopelarenite. The older Chuos glaciation (,746 Ma) occurred during a time of north-south crustal
stretching. Debris derived from upturned older rocks collected in structural depressions. The younger Ghaub glaciation (635 Ma)
occurred, after stretching ceased, on a thermally-subsiding marine platform and its distally-tapered foreslope. A continuous ice
grounding-zone wedge (GZW) occurs on the distal foreslope, while the upper foreslope and outer platform are devoid of glacial
debris and only small pockets of lodgement facies exist on the inner platform. Debris in the GZW is derived from a distinctive
falling-stand wedge that is unique to the foreslope and from immediately older strata mined preferentially from the inner platform.
The GZW rests on a smooth surface that includes a transverse steep-walled trough presumably cut by an ice-stream, within which is
a towering doubly-crested moraine composed of composite, massive, carbonate diamictite. The surface suggests that the ice-sheet
was grounded on the distal foreslope, implying a large fall in base level at a glacial maximum that predates the GZW. The glacial
record ends with Fe-stained beds, rich in ice-rafted debris, that are notably absent from the moraine, upper foreslope and platform,
which were apparently above sea-level at that time.

The existence of glaciogenic diamictite within the Neoproterozoic
carbonate succession of the Otavi Mountains in north-central
Namibia (Fig. 14.1a) was first recognized by le Roex (1941),
who cited as evidence of glacial action the presence of faceted
and striated clasts, absence of sorting or clast gradation, uniformity
of gritty fine-grained matrix, and import of granitic and arenaceous
clasts into a conformable carbonate succession. Because its floor is
not morphologically exposed, le Roex (1941) forthrightly admitted
that the diamictite could have been deposited by grounded ice or
floating ice. Martin (1965a, b) argued that the regional extent of
this unit and the existence of similar diamictites of roughly the
same age throughout southern Africa and on other continents
favoured a glacial origin. He interpreted the Otavi Tillite (le
Roex 1941) as a glaciomarine unit on account of its stratigraphic
interposition between marine carbonate formations. Kröner &
Rankama (1972) and Hedberg (1979) also favoured a glaciomarine
interpretation. Martin (1965b) noted the close association of
banded Fe-formation with diamictite in the Otavi Group and
attributed ‘this peculiar combination of sediments’ to ‘oxygen
deficiency in stagnating bottom waters caused by an ice cover’.
A glacial origin for the carbonate-clast diamictite (now Ghaub
Formation) on the Fransfontein foreslope (Fig. 14.1b) was rejected
by Frets (1969), who favoured submarine landsliding in response
to a fall in base level. This view was taken up by Martin et al.
(1985), who at age 75 (and 20 years removed from Africa) recanted
his earlier conclusions and hypothesized that a major eustatic
event, possibly caused by glaciation in other parts of the globe,
had triggered submarine landslides simultaneously throughout
Namibia. The discovery of large submarine landslides of Quatern-
ary age on the modern African margin (Dingle 1977) may have
prompted this reinterpretation. Most present-day workers,
however, favour the glaciomarine interpretation (Hoffmann &
Prave 1996; Hoffman et al. 1998a, 2007; Evans 2000; Condon
et al. 2002; Hoffman 2005, 2011; Domack & Hoffman 2011).
Although structureless carbonate-clast diamictite (with rare
basement-derived clasts) is the predominant lithology, thin inter-
vals of well-stratified detrital carbonate with variable amounts
of rafted debris occur at the top and bottom of the formation and
discontinuously within it. Condon et al. (2002) argue that this evi-
dence for ice-line instability and iceberg-rafting is incompatible
with an ice-covered ocean at the time of Ghaub deposition.
In rejecting a glacial origin altogether, Eyles & Januszczak
(2004a, b) revived the ideas of Frets (1969) and Martin et al.
(1985). In their opinion, the Ghaub Formation consists ‘entirely
of a wide variety of mass flow facies, . . . derived from faulted
underlying carbonate platform strata’. This interpretation is diffi-
cult to reconcile with the regionally consistent stratigraphic pos-
ition of the Ghaub Formation with respect to the overlying
Keilberg Member (Hoffmann & Prave 1996; Hoffman et al.
1998a, 2007). It turns a blind eye to prior stratigraphic mapping,
sedimentology and chemostratigraphy showing that faulting
ended well before the Ghaub Formation was deposited (Soffer
1998; Hoffman 1999; Halverson et al. 2002). During Ghaub
time, no fault-related structural rotations occurred on the platform
or on its margin (Hoffman & Halverson 2008).
Structural framework
The Otavi Group is exposed in an arcuate fold-and-thrust belt
around the junction of the Late Ediacaran – Early Cambrian
Kaoko and Damara orogens (Fig. 14.1a). At the apex of the fold
belt is the Kamanjab inlier, a large basement-cored structural cul-
mination. For most of the twentieth century, studies focused on the
Otavi Mountains near the eastern terminus of the fold belt (Miller
2008; Bechstädt et al. 2009), where major carbonate-hosted Cu –
Pb– Zn –V mineralization occurs. Since Namibian independence
in 1990, scientifically motivated research has favoured the better
exposed sections around the western Kamanjab inlier. Weakly
folded Otavi Group strata extend across the Owambo Basin in
the subsurface, making the total area of the ancient carbonate plat-
form in Namibia .110 000 km2 (Miller 1997).
The western margin of the carbonate platform is complicated
by a major, sinistral-oblique thrust zone (Sesfontein thrust) bound-
ing the Kaoko Belt (Goscombe et al. 2003). The southern margin
is preserved in the east around the Otavi syncline (Smit 1962;

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London, Memoirs,
36, 195– 209. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.14
196 P. F. HOFFMAN
Hedberg 1979) and to the west along the autochthonous southern
flank of the Kamanjab inlier (Fig. 14.1b). Distal stratigraphic
equivalents of the Otavi Group occur on tectonically stretched and
later foreshortened Congo crust within the central and eastern zones
of the Kaoko Belt (Swart 1992; Stanistreet & Charlesworth 1999;
Goscombe & Grey 2007; Paciullo et al. 2007) and in the central
and northern zones of the Damara Belt (Hoffmann et al. 2004;
Johnson et al. 2006; Clifford 2008; Hoffman & Halverson 2008).
Stratigraphy
The Otavi Group is a 3 + 1-km-thick, carbonate-dominated suc-
cession of Cryogenian and early Ediacaran age, deposited
between c. 770 and 580 Ma (Halverson et al. 2005). It is conform-
ably underlain by fluviatile feldspathic quartzites and conglomer-
ates of the Nosib Group (Fig. 14.2), and overlain by marine and
non-marine, synorogenic clastics of the Mulden Group. The
Otavi and Mulden groups are separated by a paraconformity
with .1.0 km of karstic palaeotopographic relief (Frets 1969;
Hoffman & Hartz 1999; Hoffman & Halverson 2008).
Subgroups of the Otavi Group
The Otavi Group was originally divided into two subgroups,
Abenab and Tsumeb, at a surface corresponding to the floor of
what was then thought to be a single glaciogenic formation
Fig. 14.1. (a) Geological map of the Otavi
Group fold belt and its relation to the
Owambo Basin (Congo craton) and to the
Kaoko and Damara orogenic belts
(Hoffman 2005). Dashed white lines
indicate the southern and western edges of
the Otavi Platform. The black rectangle
shows the area of Figure 14.1b. (b)
Geological map of the Fransfontein Ridge
and its relation to the Otavi Platform of the
Upper Huab outlier and Danubé (after
Hoffman et al. 2007). Also shown is the
division of the foreslope into proximal
and distal zones, the location of the
Duurwater trough and moraine, and
district road C35.
(Hedberg 1979; SACS 1980), assumed correlative with the
glaciogenic Chuos Formation of the central Damara Belt (Gevers
1931). Hoffmann & Prave (1996) first established the existence
of two discrete and extensive glaciogenic formations in the
Otavi Group, the older of which (Varianto Formation of SACS
1980) is equivalent to the type Chuos Formation and the younger
of which (the Otavi Tillite of le Roex 1941) they renamed the
Ghaub Formation. This critical distinction has been amply con-
firmed by subsequent studies (Hoffman et al. 1998a, b; Kennedy
et al. 1998; Halverson et al. 2005; Hoffman & Halverson 2008).
Hoffmann & Prave (1996) divided the Otavi Group into three
subgroups at surfaces corresponding to the floors of the two glacio-
genic formations. Thus the Ombombo (om-bomb-boh) Subgroup
ends below the Chuos Formation, the Abenab (ah-ben-ab) Sub-
group below the Ghaub Formation, and the Tsumeb (tsoo-meb)
Subgroup below the Mulden Group. South of the Huab palaeoridge
(Fig. 14.2), the Ugab (oo-khab) Subgroup occupies a stratigraphic
position analogous to the Ombombo Subgroup, but U –Pb zircon
ages show it to be significantly younger (,746 + 2 Ma) than the
Devede (de-ved-ee) Formation of the Ombombo Subgroup,
which is .760 + 1 Ma (Halverson et al. 2005).
Crustal stretching and thermal subsidence in relation
to glaciation
Subsidence accommodating the Otavi Group was driven by north –
south crustal stretching, which lasted at least episodically from
 THE OTAVI CARBONATE PLATFORM AND FORESLOPE 197

δ13C (per mil)

South 50 Otavi Platform 200 –10 –5 0 5 10
Tsumeb
-0.5 "
Subgroup
Huttenberg
-1.0
Elandshoek
Fore-
slope Keilberg Maieberg
-2.0 Ombaatjie Abenab
Gruis Subgroup
km .
Huab . . Rasthof
Karibib Makalani
. Ridge Ridge . . Ugab Sbgp
Ombombo

Ghaub
. . . Subgroup Fig. 14.2. Stratigraphic restoration of the
Devede Otavi Group in a north–south cross-section
746 ± 2 Ma 759 ± 1 Ma
. . . based on measured sections and mapped
Chuos . . .
Naauw . –10 –5 0 5 10 stratigraphic cutoffs around the western
Ugab
-poort
Palaeoproterozoic metamorphic- . Nosib
. . δ 13C (per mil V-PDB) Kamanjab inlier (after Hoffman et al. 2007).
plutonic complex (Congo craton) Congo
Tanz. Note the existing U –Pb zircon age
Angola
constraints from volcanic lavas and tuffs
Zam.
shallow-water carbonate cannibalized clastics (syn-rift) Mal. (Hoffman et al. 1996; Halverson et al.
Nam. Bots.
Zim.
Moz. 2005), the composite d13C profile
deep-water carbonate bimodal alkaline volcanics Swaz. (Halverson et al. 2005) and the location map

. .
South
Africa showing the Kalahari and Congo cratons
glaciogenic diamictite feldspathic sandstone (palaeoflow)
(shaded).

c. 770 Ma (Halverson et al. 2005) until a rift-sag transition at the
base of the Ombaatjie (om-bye-gee) Formation (Fig. 14.2) of
the Abenab Subgroup (Halverson et al. 2002). Accordingly, the
Chuos glaciation occurred during a prolonged period of active
faulting, which is reflected by the diversity of its debris and by
a low-angle (1.58) structural unconformity at its base (Fig. 14.2)
that truncates .2 km of Ombombo and Nosib, or Ugab and
Naauwpoort (nauf-port) strata (Hoffman et al. 1998a). No causal
connection between glaciation and crustal stretching (Eyles &
Januszczak 2004a, b) is indicated: stretching began millions of
years before the glaciation and continued for millions of years
afterwards.
In contrast to the Chuos, the Ghaub glaciation occurred at a time
of broad regional subsidence, which is reflected by platform-wide
structural conformity between depositional cycles of the upper
Ombaatjie Formation and the Keilberg cap dolostone (Halverson
et al. 2002; Hoffman et al. 2007). The Ombaatjie Formation was
regionally flat-lying at the time of the Ghaub glaciation.
The Tsumeb Subgroup is a structurally conformable succession
on the regional scale and thermal subsidence is the inferred accom-
modation mechanism for the Ombaatjie through Hüttenberg for-
mations (Halverson et al. 2002).
Palaeogeography of the platform and the southern foreslope
The unfaulted southern margin of the platform (Figs 14.1a & 14.2)
is well defined lithologically in units above and below the Ghaub
Formation (Tsumeb Subgroup and Ombaatjie Formation, respect-
ively). To the north are stacked peritidal parasequences, commonly
bounded by subaerial exposure surfaces marked by tepee struc-
tures and tepee breccias (Kendall & Warren 1987). To the south
is a distally tapered foreslope wedge of progressively deeper-water
carbonate facies (Hoffman & Halverson 2008). More proximal
parts of the foreslope (i.e. 0– 5 km seaward of the slope break)
are dominated by ribbony lutites and arenites lacking subaerial
exposure surfaces. More distally (i.e. .5 km seaward of the
slope break), parallel-sided rhythmites, rhythmite slump breccias
and carbonate turbidites prevail. Overall, the Tsumeb Subgroup
tapers from c. 1400 m on the platform to 320 –540 m for isotopi-
cally correlative strata on the distal foreslope (Halverson et al.
2005). The Abenab Subgroup tapers from 500–800 m on the plat-
form to 70– 200 m on the distal foreslope (Hoffman 2005). Only
the Keilberg cap dolostone exhibits shallow-water bedforms on
the distal foreslope, reflecting large-amplitude glacioeustatic fluc-
tuation (Hoffman et al. 2007).
The foreslope wedge is best exposed on the 60-km-long Frans-
fontein Ridge (Fig. 14.1b), a present-day topographic feature
underlain by a simple, south- to SE-dipping, structural homocline
of Otavi Group carbonates. The strata dip 40–608 on average, with
near-vertical cleavage in limestones and fine clastics, and local
north-vergent thrust imbrication of major rheological interfaces
(e.g. dolomite against fine clastics at the Otavi-Mulden group
contact, and dolostone against marly limestone within the Maie-
berg Formation). The foreslope is also preserved in structural out-
liers 60 and 80 km west of Fransfontein (on the farms Bethanis and
Toekoms, and Vrede and Opdraend, respectively; see Hoffman &
Halverson 2008). It also occurs for 250 km east of Fransfontein
along the southern flank of the Kamanjab inlier and reportedly
on the southern limb of the Otavi Syncline (Smit 1962; Hedberg
1979). Thus, the foreslope wedge is discontinuously exposed for
330 km along strike.
The foreslope divides naturally on the basis of lithofacies (see
above) and stratigraphic architecture (Fig. 14.3b) into proximal
and distal zones, projected to lie 0–5 and .5 km seaward of the
slope break, respectively (Fig. 14.1b). The western two-thirds of
the Ridge strike east –west, subparallel to the inferred palaeoslope
contours. It exposes a longitudinal section of the distal slope.
The eastern third of the ridge strikes NE and exposes a section
that angles obliquely up the proximal slope to the edge of the plat-
form (or slope break) near its northernmost point on the farm
Danubé 59.
Comparison with recent fine-grained carbonate slopes like the
western (leeward) margin of the Great Bahama Bank (Eberli &
Ginsburg 1987; Adams & Schlager 2000) suggests that the distal
foreslope (where the glaciogenic Ghaub Formation occurs) lay at
palaeodepths .0.5 km below the rim of the platform (Hoffman
2005, 2011).
198 P. F. HOFFMAN

Glaciogenic and associated strata 1986), but the paucity of bed-scale stratification suggests that
debris flows were of secondary importance in the Chuos Formation
Chuos Formation of the Otavi Group.
The associated strata, in decreasing order of abundance, are con-
The Chuos Formation is highly variable in thickness up to 1.0 km glomerates and pebbly sandstones, laminated siltstones with rafted
and the predominant lithology, especially in the thicker sections, debris, and banded Fe-formation. The coarse clastics are
is structureless to weakly stratified, matrix-supported diamictite. co-derived with the diamictites but are grain-supported,
The diamictite forms composite bodies derived in varying pro- clast-imbricated and crudely stratified or cross-stratified. They
portions from basement rocks (metavolcanic, metaplutonic and were clearly deposited by running water, but whether the flow
metasedimentary rocks, notably orthoquartzite), sub-feldspathic was proglacial or subglacial is not known. The conglomerates
arenites of the Nosib Group, and dolomites, cherts and canniba- and pebbly sandstones generally occur at the top and/or bottom
lized clastics of the Ombombo and Ugab subgroups. Clasts of of the Chuos Formation, and are richer in dolomite clasts and/or
strongly porphyritic, mafic and felsic lava, derived from the felspathic basement debris compared with similar lithofacies in
Naauwpoort Formation (747 + 2 Ma), are locally abundant. The the underlying, non-glaciogenic, Okakuyu (oh-ka-koo-you) For-
matrices of the diamictites are foliated to structureless, mud-rich mation (Ombombo Subgroup). Relatively thin units of finely lami-
wackestones. Some of the foliated fabrics are pre-tectonic. Striated nated siltstone or silty mudstone with rafted lonestones occur
clasts are probably of glacial origin. Fe- and/or Mn-rich cements locally at the base or within the Chuos Formation. Discontinuous
coat the detrital grains, giving the matrix a strong colour – brick beds and lenses of hematitic Fe-formation are intimately interca-
red, olive green or black. Fabric analysis to determine emplace- lated with structureless or stratified diamictites in different areas
ment mechanisms are compromised by tectonic strains, but the (Martin 1965a). Discharges of oxygenated meltwater at the
diamictites appear to represent both ice-contact (foliated) and grounding lines of tidewater glaciers could have triggered the pre-
proximal rain-out deposits. Diamictites in the central zone of the cipitation of Fe-formation precursor in the presence of Fe(II)-
Damara orogen were emplaced as debris flows (Henry et al. laden seawater.

(a) SOUTH Distance inward from the glacial-age slope break (km) NORTH
–20 –10 0 20 40 60 80 100
Depth wrt glacial-age platform rim (m)

400 OUTER PLATFORM subaerial surface INNER PLATFORM

400 m
200 sea-floor
cements iso
che
m
FR AN SF ON TE IN

aeolianite
FO RE SL OP E

0
Trezona anomaly δ13C = 0 o/oo datum
–200
GREAT OTAVI CARBONATE BANK Fig. 14.3. (a) Stratigraphic –
palaeobathymetric cross-section of the
–400 POST- Maieberg Fm: shoalwater dolostone grainstone Otavi carbonate platform and foreslope,
showing showing the Ghaub diamictite
GLACIAL Maieberg Fm: deepwater limestone rhythmite
Ghaub wedge and its relations to the Frannis-aus
–600 prism SYN-DEGLACIAL Keilberg Mb: transgressive 'cap' dolostone falling-stand wedge and the Maieberg
SYN-GLACIAL Ghaub Fm: diamictite prism (ice grounding zone) cap-carbonate sequence. The 0‰
cross-over of the Trezona d13C anomaly
–800 ADVANCE GLACIAL Franni-aus Mb: oolite debris flows (lowstand wedge) serves as the datum for the platform.
Ombaatjie Fm: shelf-slope carbonate (post-Trezona 0 o/oo) Foreslope palaeodepths are estimated
PRE-GLACIAL assuming a profile (below – 200 m) like the
Ombaatjie Fm: shelf-slope carbonate (pre-Trezona 0 o/oo)
(b) western margin of the present Great
Bahama Bank (see text). Ghaub diamictite
Palaeodepth wrt permanent post-glacial flood

20 km inboard of the slope break is the

–0.2 Keilberg cap dolostone
0.0 ‘Otavi Tillite’ (le Roex 1941; Bechstädt
E

et al. 2009). Note parallelism between the

Palaeodepth wrt top of platform (km)

–0.3
OP

–0.1 Ghaub diamictite prism Trezona datum and the top of the Maieberg
SL

sequence, disproving structural rotations

AL

–0.4
Franni-aus Member
M

–0.2 due to faulting associated with the Ghaub

I

low-stand wedge
OX

–0.5 diamictites. (b) Proximal and distal

PR

–0.3 differentiation of the Fransfontein foreslope

–0.6 based on stratigraphic relations between the
–0.4 DISTAL SLOPE Franni-aus falling-stand wedge, the Ghaub
–0.7 diamictite wedge and the Keilberg cap
–0.5 Ombaatjie Fm dolostone. The palaeodepth scale with
–0.8 respect to permanent (i.e. after isostatic
–0.6
Narachaams Member adjustment) post-glacial flood assumes a
–0.9 siltstone of cycle b8 water depth on the outer platform of 140 m
–0.7 for the Maieberg Formation limestone
–1.0
rhythmite, which was deposited below
–0.8 10 9 8 7 6 5 4 3 2 1 0 –1
storm wave and accommodated c. 300 m of
South Horizontal distance from slope break (km) North additional cap-carbonate strata.
 THE OTAVI CARBONATE PLATFORM AND FORESLOPE
Ghaub Formation
The palaeogeographic distribution of the Ghaub Formation is criti-
cal to its interpretation. As mapped, roughly 99% of its mass is
contained within a linear wedge that tracks the distal foreslope,
.5 km downslope from the outer edge of the platform
(Fig. 14.1b). It pinches out slightly upslope from the zero-isopach
of the Franni-aus Member (Fig. 14.3), a glacioeustatic falling-
stand wedge composed of carbonate turbidites and debris-flows
of non-glacial origin (see below). Abundant Franni-aus-derived
debris within the Ghaub Formation indicates that the falling-stand
wedge originally extended farther upslope, where it was sub-
sequently eroded and redeposited downslope during the glaciation.
In the Summas Mountains, 30 km south of the platform, the Ghaub
Formation is c. 4.0 m thick and its glacial affinity is scarcely recog-
nizable (Clifford 2008; Hoffman & Halverson 2008). The geo-
metry of the Ghaub Formation is a wedge, continuous along
strike, pinching-out upslope and severely tapered downslope.
The Ghaub Formation is totally absent from the proximal fore-
slope and the raised outer zone of the platform (Fig. 14.3). There,
the Keilberg Member (cap dolostone) rests disconformably on the
upper Ombaatjie Formation (Figs 14.2 & 14.3). Pockets of carbon-
ate diamictite reappear beneath the Keilberg Member on the inner
zone of the platform, which was preferentially eroded by up to
80 m relative to the outer zone during the glaciation (Hoffman
et al. 2007). The type section of the Otavi Tillite (le Roex 1941;
Bechstädt et al. 2009) is an exceptional 60-m-thick body of
diamictite, situated c. 20 km north of the edge of the platform.
On the whole, however, the platform and proximal foreslope
were areas of erosion during the Ghaub glaciation, and the main
zone of sedimentation was on the distal foreslope.
In order to document the internal facies architecture and lateral
variability of the Ghaub Formation, 72 stratigraphic sections were
measured on the Fransfontein Ridge as a whole, including 40 sec-
tions in a 13-km-long stretch east of the town of Fransfontein
(Fig. 14.4). The sections examined by Condon et al. (2002) and
Eyles & Januszczak (2007) are among those measured, as are
the sections recently logged in greater detail (Domack &
Hoffman 2011).
Facies associations and stratal architecture. The Ghaub Formation is
composed of three facies associations. The first consists of weakly
stratified to structureless diamictites, individually ranging from
1 to 75 m in thickness. Pebble- to boulder-sized, rounded to
angular clasts of limestone and/or dolostone (rarely basement-
derived granitoids) ‘float’ in a structureless to faintly laminated
matrix of unsorted detrital carbonate mud, silt, sand and granules.
Because of middle greenschist-grade metamorphism, the clasts are
thoroughly welded to their carbonate matrix, making it impossible
to observe clast surface morphology or micromorphology (e.g.
striae). Isotopic data indicate that the clasts are derived from the
upper part of the Ombaatjie Formation (Fig. 14.2) and from the
Frannis-aus Member (see below), which contributed the very
coarse-grained oolite clasts and grains conspicuous in many of
the diamictites. Some diamictites gradationally overlie well-
stratified units, whereas others have erosive bases and internal
reactivation surfaces. Collectively, weakly stratified and structure-
less diamictites make up 68% of the Ghaub Formation west of the
Duurwater moraine (see below) and 82% overall.
The second facies association, amounting to 17% of the for-
mation as a whole, consists of well-stratified detritus and com-
prises suspension-plume fallout, turbidites, debrites, contourites
and rafted debris, in varying proportions. All the detritus is carbon-
ate, except to the west of Fransfontein town (particularly on Betha-
nis farm), where the basal stratified interval is composed of
siliciclastic siltstone with rare outsize quartz granules. This terrige-
nous sediment is likely reworked from the Narachaams (nahr-
a-khams) Member (Figs 14.3b & 14.5), a fine-grained siliciclastic
199
incursion in the upper Ombaatjie Formation. Plume fallout forms
laterally continuous drapes of laminated mud and fine silt with
both normal and reverse grain-size gradation. Plume fallout
hosts rafted debris of variable size and concentration. Turbidites
and debris flows consist of poorly sorted, sandy to pebbly material
in discrete tabular or channelized sedimentation units that are com-
monly grain-size graded towards the top and less commonly
reverse-graded near the base. Contourites are starved (isolated)
ripples with foreset laminae consistently indicating current flow
from ENE to WSW, subparallel to depositional strike. Rafted
debris are lonestones not carried by debris flows or bedload sedi-
mentation units. They are most abundant in slowly deposited sedi-
ments (plume fallout) and least abundant in rapidly deposited
material (turbidites and debris flows). Subjacent laminae are
pinched or punctured, adjacent ones are laterally ejected, and
superjacent ones onlap or drape the rafted debris.
The third facies association consists of current-rippled or cross-
bedded, sand- or granule-sized, detrital carbonate with sparse lone-
stones. Distal facies include fine sand with trains of climbing
ripples, implying combined bedload and suspension fallout. Palaeo-
flow directions are southwestward with large apparent dispersion.
Proximal facies involve gravelly sand organized in multi-metric
clinoforms that coarsen up-dip. Grain-supported pebble conglomer-
ate is the most proximal (and least abundant) component of this
facies association. The sandy facies occurs in units up to 5.5 m
thick that can rarely be traced laterally for more than a few 100 m.
They amount to less than 2% of the formation as a whole.
West of the Duurwater moraine (Fig. 14.4), the facies asso-
ciations are organized as follows. Each vertical section contains
2 to 7 (average 4.6) diamictite units, usually separated by thinner
intervals of well-stratified and/or sandy detritus, but occasionally
amalgamated diamictite-on-diamictite with erosive contacts. The
stratified intervals commonly coarsen upwards, and their con-
tacts with overlying diamictites are gradational or erosively
truncated. Compositionally, they normally herald the overlying
diamictite. These characteristics are all well illustrated by the
stratified interval at the base of the formation. The tops of
diamictites are commonly abrupt, in some cases associated with
coarse-grained debris flows and in others by a direct transition
into fine-grained, relatively distal sandy facies. In the 5-km
stretch east of Fransfontein town, correlation of diamictite bodies
from section to section suggests an overall imbricate structure,
younging westward in the line of section (Hoffman 2005). The
2D nature of the homoclinal exposure makes it impossible to deter-
mine directly if the imbrication represents lateral accretion, sub-
parallel to the inferred slope contours, or basinward accretion in
a highly oblique section.
The herein-defined Bethanis Member (Fig. 14.4) forms a con-
tinuous blanket, 5–15 m thick, of well-stratified detrital carbonate
choked with rafted debris of all sizes. It transgresses the imbricate
structure of the rest of the formation. In contrast to the older well-
stratified intervals, which coarsen upwards and are followed by
diamictite, the Bethanis Member becomes finer grained and
more distal upwards in all its major constituents (i.e. debrites, tur-
bidites and plume fallout) except for rafted debris. The latter
includes boulders up to 5 m in diameter composed of laterally
linked, hemispheroidal, stromatolitic dolostone. This clast type is
not observed in the underlying diamictites or rafted debris.
Boulder-sized rafted debris continues to the top of the Bethanis
Member, where the detrital components give way gradationally,
over a stratigraphic interval of only 2– 5 cm, to peloidal-
micropeloidal dolostone of the Keilberg Member (Hoffmann &
Prave 1996; Hoffman et al. 2007). The isotopic systematics of
the Keilberg Member (Hoffman et al. 2007) preclude a detrital
origin for the dolomite as postulated by Eyles & Januszczak
(2007). The Bethanis Member is dark brown in colour, apparently
due to Fe and Mn oxides. Starved ripples generated by westerly
directed contour currents occur in some sections. The Bethanis
200 P. F. HOFFMAN

Distance in kilometres east of district road C35

7.0 8.0 9.0 10.0 11.0 12.0 13.0
D5 D6 D7 D8 D9 D10

P7010
(a)
20
Keilberg
Bethanis
Bethanis
0

P6510
Ghaub

P7021
P7013

P7024
-20

P7022
P4027

P7026
P7025

P6514
P6508
P7009

P7011
P7012

P7027
P7023
-40 Franni-aus

P7028
P7014
Narachaams Member
Member
P8011

P6506

Distance in kilometres east of district road C35 Datum base Bethanis Mb

0.0 1.0 2.0 3.0 4.0 5.0 6.0
(b) D0 2 4
D1 D D3 D
20 Keilberg Member

Bethanis Member
0

-20

-40
Ghaub Formation
-60
P7004
P2093

P5003
P6537
P6535

-80

P7003
P7002
P7007

P8010

-100
P8007

P6536
P7006
E0801

Franni-aus Member
P7018

P8008
P7008

-120
P6538

Keilberg Member cap dolostone Ghaub Formation

P8012

marly dolostone turbidity flows weakly- and non-stratified limestone-dolomite diamictite

E0803

peloidal dolostone, giant wave ripples weakly- and non-stratified limestone diamictite

-140 dolostone, isopachous sheet-crack cements weakly- and non-stratified dolomite-chert diamictite
P7001

current-rippled carbonate sandstone, minor ice-rafted debris

D2 - drainage number east of C35 stratified diamictite (debris flows, contourites, ice-rafted debris)

Fig. 14.4. Facies associations of the Ghaub Formation on the distal foreslope of the Fransfontein Ridge: (a) between the western edge of Duurwater trough (at 7.5 km
east) and the western edge of Duurwater moraine (at 13.0 km east) and (b) west of the Duurwater trough.

Member is lithologically distinct from earlier stratified units of the 0.1 km deep, as defined with respect to the underlying stratigraphy
Ghaub Formation and represents the terminal collapse of the ice (Fig. 14.5). On the floor of the trough, carbonates of the falling-
sheet on the Otavi platform. stand wedge (Franni-aus Member) and terrigenous siltstone of
the underlying Narachaams Member are missing, and the Ghaub
Duurwater trough and moraine. In the central part of the Fransfon- Formation rests directly on older units of the Abenab Subgroup.
tein Ridge (Fig. 14.1b), the erosion surface on which the Ghaub The trough is roughly coaxial with a major complex of deepwater
Formation rests cuts out a steep-walled trough, 18 km wide and carbonate (oolitic) sand bodies (submarine channel and levee
 THE OTAVI CARBONATE PLATFORM AND FORESLOPE 201

0 WEST Duurwater Trough EAST

–100 DT Platform
N Otavi crystal fans
–200 above cap
dolostone
Palaeodepth in metres below top of platform

–300
DM Belt
Damara
–400
Karibib Formation
–500 (includes Keilberg cap dolostone) Fig. 14.5. Stratigraphic relations in the
Abenab Subgroup and Ghaub Formation on
Ghaub Formation moraine
–600 the Fransfontein Ridge, reconstructed to
falling-stand o o o o o ridge show the inferred palaeobathymetry of the
o o o wedge
o o o
–700 o Duurwater trough and moraine (see
Narachaams Member Fig. 14.1b for location) at the end of the
–800 Ghaub glaciation (simplified after Hoffman
2005). Inset shows a postulated
grainstone
–900 palaeogeography (see text) in which the
prism
crystalline Duurwater trough (DT) marks a tributary
–1000 crystalline basement Chuos basement ice streamlet to a trunk ice stream flowing
diamictite westward along the northern zone of the
Abenab Subgroup Damara Belt. Duurwater moraine (DM) is
–1100 slope carbonates interpreted as one of a set of terminal
moraines (black dots) formed by the
–1200
–5 0 5 10 15 20 km Abenab retreating trunk ice stream,
Subgroup contemporaneous with deposition of the
–1300 Distance east of district road C35 Bethanis Member.

deposits) in the lower Abenab Subgroup. It straddles a palaeoval-
ley on the crystalline erosion surface lined by basement-derived
boulder diamictite and minor iron-formation of the Chuos For-
mation (Fig. 14.5). Hoffman (2005) attributes the Ghaub-age
trough to a transverse, southward-flowing ice stream. It occupied
a preexisting drainage system developed long before the Ghaub
glaciation. Compared with Quaternary ice-stream troughs, the Duur-
water trough is quite modest in size, similar to the Isfjorden trough
on the European continental shelf west of Svalbard (Ottesen
et al. 2005). Like the Ghaub Formation itself, what makes the
Duurwater trough unusual is its location on the distal slope, well
below the depth of any known Quaternary ice-stream trough.
The Ghaub wedge thins on both sides of the Duurwater trough to
about half its average thickness (Fig. 14.5), without change in
facies. In the middle of the trough, it swells to a pair of thickness
maxima of 288 m (east) and 602 m (west). The maxima are 3 km
apart and separated by a 182-m-thick saddle. As the wedge thick-
ens, it loses its well-stratified facies and at both crests consists
exclusively of amalgamated, weakly stratified to structureless,
boulder diamictites.
For want of a datum in the upper Abenab Subgroup beneath
the trough (see above), the true palaeo-elevation of the morainal
crest above the rims of the trough is uncertain. That it was a
major bathymetric feature is reflected by the anomalous nature
of the directly overlying units. The Keilberg cap dolostone is
less than half its thickness off the moraine, and the normal fine-
grained micropeloids are winnowed away, leaving lags of unu-
sually coarse-grained peloids (Hoffman et al. 2007, fig. 5b). The
directly overlying Maieberg limestone hosts spectacular crystal
fans (Hoffman 2005, fig. 11a), pseudomorphous after aragonitic
sea-floor cements. These are elsewhere absent from the distal fore-
slope, but are modestly well-developed on the proximal foreslope
(e.g. Hoffman et al. 2007, section S5).
The orthogonal aspect ratio of 0.08 for the morainal buildup
(7.5 km wide at the base) in the line of exposure implies a trans-
verse orientation assuming a ridge-like form. It is termed a
medial moraine (Hoffman 2005) with respect to its location
within the Duurwater trough, and its double crest could reflect
the fact that true medial moraines consist of paired lateral moraines
back-to-back. On the other hand, morainal ridges associated with
Quaternary ice streams are mostly lateral moraines (e.g. Isfjorden
Svalbard, Ottesen et al. 2005) or terminal moraines oriented at
high angles to the direction of ice-stream flow (Ottesen et al.
2005; McMullen et al. 2007).
Boundary relations with overlying and underlying
non-glacial units
Chuos Formation
The Chuos Formation truncates the basement complex and all
older Neoproterozoic strata at a low-angle (,1.58) unconformity
(Fig. 14.2). Carbonates directly beneath the unconformity are
typically shattered and silicified. In contrast, the top of the Chuos
Formation is a sharp but conformable contact lacking evidence
for substantial hiatus and only locally of significant reworking
(Hoffman & Halverson 2008). The top of the Chuos Formation
is strongly enriched in hematite, but concentrations along tectonic
cleavage planes suggests secondary Fe migration. The basal
Rasthof facies is generally a dark grey, parallel-laminated dolo-
micrite with variable proportions of calcitic, allodapic, centimetric
density-flows (Yoshioka et al. 2003; Pruss et al. 2010). It appears
to be the deepest-water facies within the Rasthof Formation,
meaning that the Rasthof lacks a transgressive facies tract.
Hoffman & Schrag (2002) suggested that seawater failed to reach
critical oversaturation with respect to CaCO3 or [CaMg]CO3
until after the glacioeustatic rise accompanying Chuos deglacia-
tion. This failure could be related to insufficient availability of
exposed carbonate globally, as only carbonate weathering could
have supplied sufficient alkalinity on the timescale of glacio-
eustatic recovery (Higgins & Schrag 2003).
Ghaub Formation
Relation to the Franni-aus Member. The Ghaub Formation paracon-
formably overlies the Franni-aus Member (Fig. 14.3) on the distal
foreslope (Hoffman 2005). The contact is a smooth, planar, erosion
surface despite the coarse fragmental nature of the underlying
202 P. F. HOFFMAN
debris flows. The Franni-aus Member (Hoffman & Halverson
2008) is a coarsening-upward stack of limestone and dolomite
turbidites and debris flows, characterized in the upper part by
variably disaggregated slabs and grains of very coarse-grained
(,3 mm) oolite, typically heavily silicified. The contact with the
Ghaub Formation is marked by an abrupt decrease in grain size,
a change in colour from pale grey to medium brown, and the
gradual appearance of increasingly numerous and sizeable drop-
stones, which grade upwards into the first, non-stratified, rain-out
diamictite. The ooids are inferred to have originally formed in
highly agitated waters at the surf zone and were redeposited grav-
itationally before complete lithification. As there is no known
source for these ooids in the Ombaatjie or any other formation
on the platform, it is hypothesized that they originally formed
on the proximal foreslope during glacioeustatic regression
(Hoffman 2005; Hoffman et al. 2007). Continued base-level fall
hastened their redeposition more distally as debris flows. The
only strata of equivalent age on the platform are ,30 m of
aeolian(?) dolarenite sandwiched between the youngest marine
cycle of the Ombaatjie Formation and the Keilberg Member
close to the outer edge of the platform in the Upper Huab area
(Fig. 14.1b) (Soffer 1998; Halverson et al. 2002). The Franni-aus
Member is cut out by the Duurwater trough, where the Ghaub For-
mation rests directly on siliciclastic siltstone of the Narachaams
Member or underlying slope carbonates of the lower Abenab
Subgroup (Fig. 14.5). Reworked siltstone at the base of the
Ghaub Formation outside the trough is most likely derived from
the Narachaams Member, implying that the Duurwater trough or
others like it had been eroded before the Ghaub Formation was
deposited (Hoffman 2005). If the Duurwater trough was cut by
an ice stream, glacial maximum conditions are not represented in
the Ghaub Formation.
Relation to the Ombaatjie Formation on the platform. Lenticles of
Ghaub Formation overlie a continuous paraconformity on the plat-
form having ,80 m of stratigraphic relief with respect to the
Ombaatjie Formation (Halverson et al. 2002; Hoffman et al.
2007). The erosion surface must be the combined product of kar-
stification during the initial glacioeustatic fall and subsequent
erosion by the ice sheet after it developed. It can be reconstructed
in detail using the 0-per-mil intersection of the Trezona
carbon-isotope anomaly as a datum (Halverson et al. 2002;
Hoffman et al. 2007). At the end of the glaciation, the platform
had a raised rim and a bowl-shaped interior with local highs.
The Keilberg cap dolostone consistently swells over positive-relief
features on the platform (Hoffman et al. 2007) and sea-floor
cements (pseudomorphosed aragonite crystal fans) are localized
above them in the overlying Maieberg limestone rhythmite.
Relation to the Keilberg Member. The Ghaub Formation is every-
where overlain conformably and without significant hiatus by
the Keilberg Member (Hoffman et al. 2007). Well-stratified pro-
glacial deposits choked with dropstones (Bethanis Member)
precede the Keilberg except on the Duurwater moraine, where
the Bethanis Member is absent and the Keilberg Member rests
directly on non-stratified diamictite. The base of the Keilberg is
generally a laminated peloidal-micropeloidal dolostone with low-
angle cross-stratification. In distal-slope sections, however, the
cross-stratified facies is commonly preceded by ,1.0 m of deeper-
water dolomitic turbidites (Hoffman et al. 2007; Hoffman &
Macdonald 2010).
Chemostratigraphy
The Otavi Group is ideally suited for chemostratigraphy on account
of its carbonate-dominated lithology, ocean margin setting and
estimated duration of c. 190 Ma (Halverson et al. 2005).
Carbon isotopes
Pre-glacial negative anomalies. A composite d13Ccarb curve for the
Otavi platform is given in Figure 14.2. The major features of the
curve have been replicated in numerous sections and compare
favourably with correlative sections outside Namibia (Halverson
et al. 2005; Halverson 2006; Nogueira et al. 2007). Both glacia-
tions were preceded by precipitous declines in d13C, following
persistent (.500 m) intervals of strong 13C enrichment
(d13C ¼ 4 –8‰ PDB). The decline before the Ghaub glaciation
is deeper, from þ7.5‰ to –5.0‰ PDB; it represents the
‘Trezona anomaly’ (Halverson et al. 2005; Halverson 2006)
observed below the Elatina, Stelfox (Ice Brook) and Smalfjord gla-
ciations in Australia, Canada and Norway, respectively. The older
decline is observed at the top of the Ugab Subgroup, south of the
Huab palaeoridge, and its nadir of only 0‰ is probably truncated
by sub-Chuos erosion. It possibly represents the here-named ‘Islay
anomaly’, observed to precede the Port Askaig, Petrovbreen and
Julius River glaciations of Scotland, East Svalbard and Tasmania,
respectively (Calver 1998; Brasier & Shields 2000; Halverson
et al. 2005; Halverson 2006; McCay et al. 2006; Prave et al. 2009).
Post-glacial negative anomalies. Negative d13C anomalies are
associated with the post-glacial Rasthof and Maieberg formations
(Fig. 14.2), and differences between them are common to the pairs
of Cryogenian glaciations in Australia and Canada (Kennedy et al.
1998; Hoffman & Schrag 2002). The anomaly in the Rasthof
Formation has been documented in fine detail by Yoshioka et al.
(2003) and their findings are valid over most of the Otavi platform
(Pruss et al. 2010). The negative anomaly is limited to the lower
(abiotic rhythmite) member, which is 2–70 m thick regionally
and 14 m in its detailed section. Its top and bottom are dolomitic
but the middle 10 m is calcitic. The basal dolomite rhythmite
rises asymptotically in d13C from –4.0‰ towards –2.0‰, the
middle calcite rhythmite rises gradually by an additional 1‰,
and the upper dolomite rhythmite rises from –1.0‰ asymptoti-
cally to þ2‰. The middle (microbialaminite) and upper (grain-
stone) members of the Rasthof Formation have d13C values
between þ4 and þ6‰.
The negative d13C anomaly following the Ghaub glaciation
extends through the full ,400-m thickness of the Maieberg For-
mation (Fig. 14.2). Isotopic profiles of the Keilberg cap dolostone
differ significantly according to palaeogeography because of dia-
chroneity associated with glacioeustatic transgression: shallow-
water facies in distal slope sections are entirely older than the
same facies on the platform (Hoffman et al. 2007). Distal slope
sections begin at –0.5‰ and fall asymptotically towards
–3.0‰; platform sections begin near –3.0‰ and rise impercept-
ibly to –2.5‰ over most of their thickness, before falling steeply
to –4.5‰ at the dolostone-limestone transition. Proximal slope
sections occupy the middle part of the overall sigmoidal trend.
If the timescale of the global meltdown was on the order of 104
years (Peltier et al. 2004), severe constraints are placed on the
origin of the secular d13C change of 4‰ (Hoffman et al. 2007).
Independent of the secular changes, the absolute d13C values
vary systematically across the platform. Sections at the edge of
the platform and on the proximal slope are 1.0‰ and 1.5‰
enriched in 13C, respectively, compared with sections in the plat-
form interior. It is implausible, due to rapid mixing, that this
lateral isotopic gradient could be maintained in ocean surface
waters but, if the dissolved inorganic carbon (DIC) pool was domi-
nated by CO2 (Snowball Earth hypothesis), the gradient in d13Ccarb
could reflect a lateral temperature gradient resulting from strong
surface warming over the platform and upwelling of colder
waters at the slope (Hoffman et al. 2007). The isotopic gradient
would result from temperature-dependent isotopic fractionation
of CO23 – ions with respect to aqueous and atmospheric CO2.
Thus, the lateral gradient in d13C reflects spatial variation in
surface temperature, and the secular changes in d13C in all areas
 THE OTAVI CARBONATE PLATFORM AND FORESLOPE
reflect progressive warming over time as global ice cover recedes.
Importantly, this interpretation is contingent upon low-pH (,7.2)
conditions, unique to a snowball aftermath, in which CO2 is the
dominant carbon-bearing species.
The observed consistent spatial and temporal variations in d13C
(Hoffman et al. 2007) demonstrate that the Keilberg cap dolostone
does not consist of detrital dolomite (Eyles & Januszczak 2007).
Early diagenetic breccias and cements like those in presumed cor-
relative cap dolostones in South China and West Africa (Jiang
et al. 2003, 2006; Shields et al. 2007) are largely absent from
the Keilberg Member (Hoffman et al. 2007).
The lower limestone rhythmites of the Maieberg Formation
have rather uniform d13C values of –6.0 to –6.5‰ (Fig. 14.2).
The drop of c. 0.75‰ (and 2– 3‰ in d18O) at the dolostone – lime-
stone transition is a step function apparently related to mineralogy.
The combined d13C and d18O changes are consistent with
low-temperature equilibrium fractionation between coexisting
dolomite and calcite (Friedman & O’Neil 1977; Vasconcelos
et al. 2005). Reverse steps occur at the limestone-dolostone tran-
sition in the upper Maieberg Formation. The Maieberg is the
only formation in the Otavi Group (including the Rasthof For-
mation) exhibiting a systematic isotope fractionation between stra-
tigraphically associated calcite and dolomite (Hoffman &
Halverson 2008).
The rise in d13C with stratigraphic height through the middle and
upper Maieberg Formation varies laterally in its trajectory, rising
less near the edge of the platform and more in the platform interior.
This may reflect faster sedimentation rates on the outer platform
and diachronous shoaling from the edge of the platform toward
the interior. The ‘isochems’ in Figure 14.3a denote schematically
the surfaces of uniform d13C composition.
Strontium isotopes
87
Sr/86Sr ratios have been measured in limestones of the Otavi
Group deposited before, between and after the Cryogenian glacia-
tions. Because Sr is a mobile trace element in carbonate rocks,
samples with low Sr concentrations (,800 ppm), high Rb –Sr
ratios, low d18O values and high 87Sr/86Sr ratios are screened out.
No data from dolostones survive screening because of their low
Sr concentrations. According to these criteria, the best limestones
from the Ombombo Subgroup have 87Sr/86Sr ratios c. 0.7065
(P. F. Hoffman & A. J. Kaufman, unpublished data), the Rasthof
cap carbonate c. 0.7068 (Yoshioka et al. 2003), the lower Ombaat-
jie Formation c. 0.7073 (Halverson et al. 2007) and the Maieberg
cap carbonate c. 0.7072 (Halverson et al. 2007). These data are
broadly consistent with results from both western and eastern
Laurentia (Brasier & Shields 2000; Halverson et al. 2007) and
support the correlation of the pairs of glaciogenic formations in
each region.
Sulphur isotopes (CAS)
Secular variations in the concentration and sulphur isotopic com-
position (d34S) of carbonate-associated sulphate (CAS) have
been profiled in sections of the Abenab Subgroup and the Maieberg
cap-carbonate sequence (Hurtgen et al. 2002, 2006). CAS concen-
trations are mostly ,200 ppm, lower than Phanerozoic carbonates,
and this combined with large (.20‰) and apparently rapid isoto-
pic fluctuations suggests that seawater sulphate concentrations
were lower than in Phanerozoic oceans (Hurtgen et al. 2002).
Large, complex, but broadly similar d34S anomalies are associated
with the Rasthof and Maieberg cap-carbonate sequences. Their
lower and upper parts are isotopically enriched to 40 and 50‰
(CDT), respectively, while their middle part, corresponding
closely to the deepest-water facies on the platform, are isotopically
depleted to c. 15‰. In the Maieberg Formation, the steepest
203
isotopic change straddles the dolostone – limestone transition,
which also coincides with the disappearance of wave-generated
bedforms and is interpreted as the incursion due to flooding of
anoxic sulphidic waters onto the platform, accompanied by the
oxidation of sulphide to sulphate at the oxic – anoxic interface
(Hurtgen et al. 2006).
Reactive Fe and Mn concentrations
Hurtgen et al. (2006) report reactive Fe and Mn concentrations for
three, widely spaced, closely sampled sections of the Maieberg
cap-carbonate sequence. Fe and Mn concentrations from acid-
soluble (mainly Fe2þ and Mn2þ carbonates) and diothionite-
extractable (mainly oxides and oxyhydroxides) sources were
measured, but not sulphide-bound Fe. There is excellent agreement
between the sections. All concentrations rise from very low values
in the lower half of the Keilberg cap dolostone towards high values
in the middle Maieberg limestone member, before falling in the
upper Maieberg dolomitic grainstone. A sharp peak in acid-soluble
Fe (and Ba) concentration occurs at the cap dolostone – limestone
transition. Combined with d13C, d18O, d34Spyrite and d34SCAS data
from the same samples, Hurtgen et al. (2006) propose a model
involving progressive mixing between cold, euxinic (i.e. anoxic-
sulphidic), deep water and a lid of warm, oxic, brackish water,
during and after deglaciation (see also Shields 2005).
Oxygen isotopes
Oxygen isotopes in carbonate rocks are susceptible to alteration by
fluid flows under diagenetic and metamorphic conditions. d18O
values for relatively unaltered Otavi Group carbonates fall
mostly in the range of – 4 to –6‰ VPDB (Hurtgen et al. 2006,
unpublished data). Of the three formations containing both lime-
stone and dolostone, including fabric-retentive dolostone, one
(Maieberg Formation) displays consistent isotopic fractionations
of both d18O and d13C between the two phases, whereas the
others (Rasthof and Ombaatjie formations) show no pattern what-
soever on a cross-plot of d13C v. d18O (Hoffman & Halverson
2008). In the Maieberg Formation (including its Keilberg
Member), limestone is isotopically depleted by c. 1.7‰ in d13C
and c. 3.8‰ in d18O, relative to dolostone. This fractionation
remains locked despite the large secular change (5‰ in d13C)
the formation records. This analysis includes the upper dolostone
grainstone member as well as the Keilberg Member. The Maieberg
data are consistent with low-temperature (c. 50 8C) isotopic equi-
librium fractionation between dolomite and calcite (Friedman &
O’Neil 1977), and by implication that the dolomite formed in
contact with seawater.
If one ‘corrects’ the limestone values to bring them in line with
adjacent dolomites (e.g. add 3.8‰ in d18O), a well-defined nega-
tive d18O anomaly of 1 to 3‰ is evident in the first 130 m of the
Maieberg Formation (limestone from 13 to 330 m). This is poss-
ibly an isotopic record of the meltwater plume attending deglacia-
tion, before it was mixed with salty glacial deepwater (Hoffman
1999; Shields 2005).
Boron and calcium isotopes
Kasemann et al. (2005, 2010) report d11B and d44Ca measurements
for carbonates of the Ombaatjie and Maieberg formations as
proxies for pre- and post-glacial seawater. There is a large (c.
20‰) isotopic fractionation between borate ion (which is incorpor-
ated into carbonate) and boric acid in seawater, and the relative
proportions of these species change from nearly 0 to almost 1.0
over the pH range of 7.5– 9.5. Changes in seawater d11B over geo-
logical time (i.e. millions of years) may also reflect changes in
204 P. F. HOFFMAN
ocean inputs and outputs of boron. Because continental weathering
tends to lower seawater d11B and mid-ocean ridge hydrothermal
exchange to raise it, we may suppose that seawater should
become relatively more enriched if continental weathering was
held down for millions of years under snowball conditions.
Instead, post-glacial d11B values are more depleted, reaching a
nadir at the top of the Keilberg cap dolostone and in the directly
overlying Maieberg limestones. Using their d44Ca data to correct
for changes in Ca (and indirectly boron) ocean input and output,
Kasemann et al. (2005) calculate changes in the maximum levels
of atmospheric pCO2 over time. Their results allow for a rise in
pCO2 of up to 0.1 atm (100 000 ppm) in the glacial aftermath,
broadly consistent with the Snowball Earth hypothesis.
Palaeolatitudes and palaeogeography
Neoproterozoic sedimentary rocks in Namibia were extensively
and thoroughly remagnetized during the orogenic amalgamation
of Gondwanaland in the Cambrian. Palaeopoles from mafic
members of the post-tectonic Mbozi igneous complex in Tanzania
(Meert et al. 1995), felsic components of which give K –Ar cooling
ages of 755 + 25 Ma (recalculated from Cahen & Snelling 1966;
Evans 2000), suggest a palaeolatitude of 10 + 58 for the Otavi
platform c. 40 Ma before the Chuos glaciation, assuming a rigid
Congo craton (Evans 2000). Dual-polarity palaeopoles from the
Nola dyke swarm (Central African Republic) dating from its
metamorphism and remagnetization at 571 + 9 Ma (Moloto-A-
Kenguemba et al. 2008) place the Otavi platform near the palaeoe-
quator (04 + 58) c. 65 Ma after the Ghaub glaciation.
Preliminary palaeomagnetic data from the basal part of the
Keilberg cap dolostone place the Otavi platform in the (southern)
subtropics with its present southern margin facing the palaeoequa-
tor (R.I.P. Trindade, pers. comm.). These fragmentary results are
at least consistent with more robust data sets indicating that
other Neoproterozoic carbonate-dominated successions formed
at palaeolatitudes ,358 (e.g. Trindade et al. 2003; Macouin
et al. 2004; Kilner et al. 2005; Maloof et al. 2006), and therefore
that the meridional climatic gradient was not reversed as proposed
in the large orbital obliquity hypothesis for low-latitude glaciation
(Williams 1993; see also Evans 2006).
Geochronological constraints
Zircons were extracted from tuffaceous layers within the Chuos
Formation, from strata directly preceding the Ghaub glaciation
on the platform and slope, and from both cap carbonates. No
primary volcanogenic zircons were recovered. The glacial history
of the Otavi platform is currently constrained by just three U –Pb
zircon ages.
The first provides a maximum constraint for the Chuos For-
mation on the platform. An age of 760 + 1 Ma (Halverson et al.
2005) was obtained for a tuff near the top of the Devede Formation
of the Ombombo Subgroup (Fig. 14.2), nearly 350 m stratigraphi-
cally below the Chuos at that locality. The age is close to
756 + 2 Ma for the Oas quartz-syenite pluton (Hoffman et al.
1996), representing a suite of peralkaline igneous rocks associated
with crustal stretching.
The younger Naauwpoort bimodal (basalt-rhyolite) volcanics
provide a tighter maximum constraint on Chuos glaciation in the
Summas Mountains area (Fig. 14.1a). There, rhyolite lava that is
unconformably overlain by (tectonically overturned) Chuos dia-
mictite has an age of 747 + 2 Ma (Hoffman et al. 1996). An
ash-flow tuff with an indistinguishable age of 746 + 2 Ma in the
same area is separated from the Chuos by 720 m of shallow-water
mixed carbonates and clastics of the Ugab Subgroup (Hoffman
et al. 1996; Hoffman & Halverson 2008). Accordingly, the onset of
the Chuos glaciation must be substantially younger than 746 Ma.
No constraints on the Ghaub glaciation have been obtained
from the Otavi platform, but a tuff layer within carbonates of the
central Damara Belt carrying ice-rafted debris, believed correla-
tive with the Ghaub Formation, gives an age of 635.5 + 0.5 Ma
(Hoffmann et al. 2004; Condon et al. 2005). This is indistinguish-
able from the age of 635.2 + 0.6 Ma for a tuff lying upon the post-
glacial (Nantuo diamictite) cap dolostone in South China (Condon
et al. 2005). The Ghaub glaciation is therefore inferred to have
ended in 635 Ma.
Discussion
Tectonic/palaeogeographical setting and origin of the
Ghaub Formation
Eyles & Januszczak (2007; see also Eyles 2004; Eyles & Januszc-
zak 2004a, b) examined the Ghaub Formation in two sections, near
the western terminus of the Fransfontein Ridge (Narachaams) and
at the Fransfontein drainage gap (Fig. 14.1b). They conclude that it
consists of subaqueous mass-flow deposits of non-glacial origin,
coincident with rifting and breakup of the Congo craton. Accord-
ingly, the Ghaub Formation has no bearing on the Neoproterozoic
glacial record and provides no support for the Snowball Earth
hypothesis (Eyles & Januszczak 2007).
Their argument boils down to four interconnected points. First,
the Ghaub Formation lacks key glacial indicators such as faceted,
striated and bullet-shaped clasts. Second, the pre-, syn- and post-
Ghaub facies are indistinguishable, meaning that the identification
of a glaciogenic interval is entirely arbitrary. Third, the Ghaub dia-
mictites occur strictly in base-of-slope palaeoenvironments, con-
sistent with their origin as subaqueous mass-flows. Fourth, they
were deposited contemporaneously with rift faulting and continen-
tal break-up at the edge of the Congo craton, suggesting that active
faulting was the trigger for mass wasting. These conclusions are so
at variance with those favoured by other authors (e.g. Condon et al.
2002; Domack & Hoffman 2011) that a point-by-point discussion
is warranted.
Diagnostic glacial indicators. In the type area of the Ghaub diamic-
tite, le Roex (1941) reported that ‘a high percentage of the pebbles
show typical glacial faceting, particularly the hard quartzitic
types’. Some also show striations, ‘which are much better pre-
served on the quartzitic than the calcareous types’ (le Roex
1941). Quartzitic pebbles do not occur in the Ghaub Formation
on the Fransfontein foreslope, only carbonate clasts (along with
rare granitic detritus at Bethanis). The problem is that the carbon-
ate clasts and the carbonate matrix were thoroughly welded
together during middle greenschist facies metamorphism. Facets
and striae might exist on every clast, or on none: it would be
impossible to tell the difference.
In the absence of clast morphology, dropstones are the most
diagnostic glacial feature (along with the overall facies associ-
ations and stratal architecture). Classic attributes of dropstones
are well-developed in the Ghaub Formation, including pinched and
punctured substrata, thickened and ejected side-strata, draped or
onlapping superstrata, and independence from bottom sediment
transport events. Ejection folds are directly analogous to the ‘over-
turned flap’ on the rim of Meteor Crater, Arizona (Shoemaker
1963). Eyles & Januszczak (2004a, 2007) emphasize that drop-
stones also occur in non-glacial settings, but the frequency of
their occurrence in the Ghaub Formation is unmatched by any non-
glacial deposit. With a conservative average of one dropstone per
1000 ppm3, the Bethanis Member alone held 50 trillion (5  1013)
dropstones given its original dimensions of 0.1  5  100 km3.
Pre-, syn- and post-glacial facies. The lower and upper contacts of
the Ghaub Formation can be unambiguously mapped on the
basis of lithology with an accuracy of centimetres along the
 THE OTAVI CARBONATE PLATFORM AND FORESLOPE
length of the Fransfontein foreslope. The same is true of the out-
liers to the west (Bethanis-Toekoms and Vrede-Opdraend). The
carbonate turbidite-hosted debris flows of the underlying
Franni-aus Member and overlying Maieberg Formation are oligo-
mictic and intraformational; those of the Ghaub Formation are
polymictic and extraformational. Weakly stratified to structureless
diamictites make up 82% of the Ghaub Formation (based on 72
measured sections) but are uncommon or absent in the adjacent
formations. Similarly, dropstones are profligate in the Ghaub
but rare or non-existent in the Franni-aus and Maieberg. In short,
the Ghaub is a true formation, lithologically mappable on the
regional scale.
How could Eyles & Januszczak (2007) conclude that the same
essential features occur within and below the Ghaub Formation?
The answer is the following. At Narachaams (see their figs 4 and
5), they place the base of the Ghaub Formation at the base of the
Bethanis Member, which constitutes only the top 2 m of the
54-m-thick Ghaub Formation in that section. At Fransfontein
(their figs 3 and 5), everything they call Ghaub Formation lies
below the Bethanis Member. In other words, their ‘Ghaub
Formation’ at Fransfontein is stratigraphically equivalent to
their ‘Abenab Subgroup’ at Narachaams. Small wonder their
‘pre-Ghaub’ and ‘syn-Ghaub’ facies are similar; they are one
and the same.
Palaeoenvironmental setting of the Ghaub Formation. Eyles &
Januszczak (2007) concur with the interpretation of the distal
Fransfontein foreslope (Fig. 14.1b) as a deep-water marine
environment (Henry et al. 1990; Halverson et al. 2002, 2005;
Hoffman 2005). Their interpretation of the Ghaub Formation as
a stack of submarine mass flows is consistent with this setting,
but it does not account for sedimentary structures (low-angle cross-
lamination and giant wave ripples) in the overlying Keilberg cap
dolostone, which demonstrate that the distal foreslope was above
storm wave-base at the glacial – deglacial transition (Hoffman
et al. 2007). The diachronous nature and extent of the cap dolo-
stone (same-depth deposits on the platform and distal foreslope)
support a large-amplitude glacioeustatic fluctuation, as a result
of which the ice grounding-line advanced to the distal foreslope
before retreating at the glacial termination (Hoffman et al. 2007).
The facies associations and stratal architecture of the Ghaub
Formation have much in common with ice grounding-line wedges
of Quaternary age on the continental shelves and upper slopes of
polar seas (e.g. Alley et al. 1989; Boulton 1990; Powell 1990;
King et al. 1991; King 1993; Powell & Domack 1995). The well-
stratified facies association represents proglacial marine deposits.
The sandy facies association resembles ice grounding-line fans
(Powell 1990). Weakly-stratified to structureless diamictites rep-
resent ice-proximal and ice-contact deposits. Diamictites that
grade downwards into stratified proglacial deposits probably accu-
mulated as rain-out deposits immediately seaward of the ice
grounding line. Ice-contact deposits include diamictites with
erosive bases, internal reactivation surfaces and discontinuous
stringers of finely laminated siltstone. The silt-stringers are inter-
preted as ice-bed separation cracks beneath stagnant grounded
ice. They were susceptible to deformation associated with recur-
rent ice flowage.
Eyles & Januszczak (2007) fail to mention the widespread relics
of Ghaub diamictite on the Otavi platform, including the 60 m
of diamictite with granitic and quartzitic clasts in the Otavi
Mountains type section (le Roex 1941; Hoffmann & Prave
1996). As these diamictites are underlain and overlain by shallow-
water carbonates (Ombaatjie Formation and Keilberg Member,
respectively), it is improbable that they originated as submarine
mass-flows.
Role of rift faulting. Le Roex (1941) argued that the quartzitic and
granitic clasts in his diamictite could not be locally derived (e.g.
through faulting) because the diamictite is separated from quartzite
205
and granitic basement in the Otavi Mountains by hundreds of
metres of shallow-water (stromatolitic and oolitic) carbonate
strata forming a conformable stratigraphic succession. He con-
cluded that the clasts must have been transported from distant
sources, where quartzite and granitic basement were exposed in
Ghaub time.
Halverson et al. (2002) greatly strengthened le Roex’s (1941)
argument with detailed isotopic chemostratigraphy of the upper
Ombaatjie Formation. They documented a steep decline in d13C
from þ5‰ or higher down to – 5‰ in the final pair of depositional
cycles (parasequences) at 17 locations distributed from the edge of
the platform far into its interior. This isotopic shift has been corre-
lated globally and is named the Trezona anomaly after the for-
mation in South Australia where it was first encountered
(Halverson et al. 2002, 2005). Plotting all the sections using the
interpolated 0‰ intersection as the datum provides a quantitative
basis for reconstructing the palaeotopography of the erosion
surface on the platform beneath the Ghaub diamictite, or the Keil-
berg Member where the Ghaub is absent (Halverson et al. 2002;
Hoffman et al. 2007). There is up to 80 m of local relief on the
erosion surface but no evidence of uplift or back-rotation of the
outer platform. If a basinward-dipping normal fault was active at
the edge of the platform (Eyles & Januszczak 2007), uplift and
back-rotation would occur as an isostatic response to tectonic
unloading of the footwall. The resulting unconformities as well
as cannibalistic clastic deposits on the back-rotated dip-slopes
are precisely the criteria successfully used to determine the
location and timing of rift faulting in the lower Otavi Group,
prior to the Ombaatjie Formation (Soffer 1998; Hoffman 1999;
Halverson et al. 2002; Hoffman & Halverson 2008). However,
there is no evidence of such activity at the time of the Ghaub
Formation. The edge of the platform preserves more upper
Ombaatjie strata than does the interior, not less, and over the plat-
form as a whole the upper Ombaatjie cycles and the Keilberg
Member are parallel. This is a very sensitive test because even a
small angular rotation of a rift ‘shoulder’ of modest dimensions
would result in significant stratigraphic truncation in a shallow
marine environment. None is observed. Before Eyles & Januszc-
zak (2004a, 2007) began their study, the history of rift faulting
on the Otavi platform was known to have ended before the
Ombaatjie Formation was deposited, millions of years before the
Ghaub glaciation (Hoffman et al. 1998a; Halverson et al. 2002;
Hoffman 2002).
Significance of ice-rafted dropstones for maximum
ice-shelf extent
The presence of ice-rafted dropstones in the Ghaub Formation on
the Fransfontein Ridge was taken as evidence for the existence of
free-floating icebergs (Condon et al. 2002), which would be incon-
sistent with a globe-encircling ice shelf as predicted by the Snow-
ball Earth hypothesis (Hoffman & Schrag 2002; Warren et al.
2002; Goodman & Pierrehumbert 2003). Two issues are raised
by this interpretation. Were the dropstones carried by icebergs or
were they released by basal melting of ice-shelf ice carrying eng-
lacial debris across the grounding line? Free-floating icebergs
transport debris farther from the grounding line than ice-shelf
ice. Iceberg transport is favoured for the Bethanis Member, but
the high proportion (88%) of weakly stratified and structureless
diamicitites in the rest of the Ghaub Formation supports deposition
close to the grounding line. Thus, in the absence of iceberg
‘dumps’ (Condon et al. 2002), the origin of dropstones below
the Bethanis Member is ambiguous with respect to the continuity
of ocean ice cover.
The second issue is whether the Ghaub Formation represents the
maximum glacial conditions, or only the waning stages of glacia-
tion. If the smooth, planar and laterally continuous surface at the
base of the Ghaub Formation on the Fransfontein Ridge represents
206 P. F. HOFFMAN
a subglacial erosion surface, then the overlying glacial deposits do
not represent the first, or even necessarily the maximum, glacial
stage. Likewise the Duurwater trough, if it was cut by an ice
stream, must represent a glacial maximum that is unrepresented
by deposits in the Ghaub Formation itself (Hoffman 2005).
Accordingly, any evidence for open water within the Ghaub For-
mation on the Fransfontein Ridge is ambiguous with regards to
the maximum extent of the ice. A complete glacial record will
only be found seaward of the maximum ice grounding line.
Significance of grounding-line oscillations
Condon et al. (2002) also suggested that the evidence in the
Ghaub Formation for grounding-line oscillations was inconsistent
with an ice-covered ocean. The logic behind this assertion is
unclear: changes in the mass-balance of an ice sheet (driven,
for example, by orbital forcing) or, alternatively, episodicity
in ice-sheet dynamics (e.g. the binge – purge cycles of
MacAyeal 1993) might just as well occur if the hydrological
cycle was driven by the sublimation of sea ice as by the evapor-
ation of seawater. Relative to grounding-line oscillations on Qua-
ternary continental shelves, however, the oscillations recorded in
the Ghaub Formation appear to have been more limited in scale,
possibly because of the inclination of the Otavi foreslope
(Pollard & deConto 2007; but see also Alley et al. 2007).
Magnitude of base-level and glacioeustatic changes
The existence of grounded ice on the distal foreslope at an
inferred palaeodepth of c. 0.5 km below the rim of the platform,
combined with shallow-water structures (low-angle cross-strata
and giant wave ripples) in the directly overlying cap dolostone
and post-glacial flooding of the platform below storm wave-base
in the succeeding limestone (Fig. 14.3a) imply a base-level
change of .0.5 km at the glacial termination. The glacioeustatic
change must have been far greater than this because of the iso-
static rebound of the platform due to the removal of the ice
sheet, the hydroisostatic sinking of the sea floor (and therefore
of sea level) due to the addition of glacial meltwater, and the
additional lowering of sea level in the vicinity of a retreating
ice sheet due to the weakening of its gravitational attraction on
the adjacent ocean (Clark 1976; Farrell & Clark 1976). The
time scale over which the platform was flooded, given the thick-
ness of the deepwater deposits (Fig. 14.3a), must have far
exceeded the timescale of the isostatic adjustments. If the thick-
ness of the ice sheet near the edge of the Otavi platform was
c. 1.0 km, comparable to the edge of the East Antarctic ice
sheet today (Lythe et al. 2001), then the isostatic rebound
would have been c. 0.27 km (rice/rmantle). If the global average
thickness of ice sheets on all continents was .2.5 km, as
suggested by climate models of a Snowball Earth (Donnadieu
et al. 2003; Pollard & Kasting 2004), then .1.25 km of meltwater
would have been added to the ocean because its area is roughly
twice that of the continents. The resulting hydroisostatic sinking
of the seafloor would have been .0.38 km (1.25/rmantle). The
additional lowering of sea level in the vicinity of the former ice
sheet could be as great as c. 0.1 km (Farrell & Clark 1976).
Accordingly, a glacioeustatic change of .1.25 km is fully compa-
tible with the inferred base-level change of .0.5 km (i.e.
1.25–0.27–0.38 –0.1 ¼ 0.5 km).
Significance of the associated carbonates
The ‘reverse’ solubility of calcium carbonate accounts for the
preferential accumulation of carbonate sediments in the warmer
parts of the surface ocean. Carbonate dominates shallow-water
sediments forming today only in latitudes ,358 (Köppen &
Wegener 1924, p. 19; Rodgers 1957). The zonal distribution of car-
bonate sediments did not change between the warm (non-glacial)
and cool (glacial – interglacial) periods of the Phanerozoic (Black-
ett 1961; Briden & Irving 1964; Briden 1970; Ziegler et al. 1984;
Opdyke et al. 1990; Witzke 1990; Kiessling 2001). This is because
carbonate production and preservation reflect relative, not absolute
temperatures. Moreover, it implies that the presence or absence
of polar ice sheets caused no perceptible change in the flux of
alkalinity entering the ocean. It is therefore a reasonable inference
that the Otavi carbonate platform occupied one of the warmer parts
of the global ocean before, between and after the Cryogenian
glaciations.
If one accepts that an ice sheet on the Otavi platform had a con-
tinuous tidewater grounding line in one of the warmer parts of the
Cryogenian ocean, then contemporaneous ice sheets must have
existed at higher elevations and palaeolatitudes, wherever there
was net meteoric precipitation or glacial influx.
Research on the Otavi Group was initially supported by the Geological Survey
of Namibia and the Canadian National Science and Engineering Research
Council (NSERC). Since 1994, support was generously provided by the United
States National Science Foundation (NSF), NASA Astrobiology Institute,
Harvard University Center for the Environment (HUCE), and the Canadian
Institute for Advanced Research (CIFAR). Among the many who have contribu-
ted to this work, I am particularly indebted to S. A. Bowring, E. W. Domack,
G. P. Halverson, J. A. Higgins, M. T. Hurtgen, A. J. Kaufman, F. A. Macdonald,
A. C. Maloof, D. P. Schrag, G. Soffer and R. Trindade. Roy McG. Miller gave
valuable advice on stratigraphic names. G. Jiang is thanked for his perceptive
comments on the manuscript. This represents a contribution of the IUGS- and
UNESCO-funded IGCP (International Geoscience Programme) Project #512.
References
Adams, E. W. & Schlager, W. 2000. Basic types of submarine slope
curvature. Journal of Sedimentary Research, 70, 814– 828.
Alley, R. B., Blankenship, D. D., Rooney, S. T. & Bentley, C. R.
1989. Sedimentation beneath ice shelves– the view from ice stream
B. Marine Geology, 85, 101– 120.
Alley, R. B., Anandakrishnan, S., Dupont, T. K., Parizek, B. R. &
Pollard, D. 2007. Effect of sedimentation on ice-sheet grounding-
line stability. Science, 315, 1838–1841.
Bechstädt, T., Jäger, H., Spence, G. & Werner, G. 2009. Late Cryo-
genian (Neoproterozoic) glacial and post-glacial successions at the
southern margin of the Congo Craton, northern Namibia: facies,
paleogeography and hydrocarbon perspective. In: Craig, J.,
Thurow, J., Thusu, B., Whitham, A. & Abutarruma, (eds)
Global Neoproterozoic Petroleum Systems: The Emerging Potential
in North Africa. Geological Society, London, Special Publications,
326, 255– 287.
Blackett, P. M. S. 1961. Comparison of ancient climates with the ancient
latitudes deduced from rock magnetic measurements. Proceedings of
the Royal Society of London, Series A, 263, 1 –30.
Boulton, G. S. 1990. Sedimentary and sea-level changes during
glacial cycles and their control on glacimarine facies architecture.
In: Dowdeswell, J. A. & Scourse, J. D. (eds) Glacimarine Environ-
ments: Processes, Sediments. Geological Society of London, Special
Publications, 53, 15 –52.
Brasier, M. D. & Shields, G. 2000. Neoproterozoic chemostratigraphy
and correlation of the Port Askaig glaciation, Dalradian Supergroup
of Scotland. Journal of the Geological Society, London, 157, 909–914.
Briden, J. C. 1970. Palaeolatitude distribution of precipitated sediments.
In: Runcorn, S. K. (ed.) Palaeogeophysics. Academic Press,
London, 437– 444.
Briden, J. C. & Irving, A. 1964. Palaeolatitude spectra of sedimentary
palaeoclimatic indicators. In: Nairn, A. E. M. (ed.) Problems in
Palaeoclimatology. John Wiley & Sons, New York, 199–224.
Cahen, L. & Snelling, N. J. 1966. The Geochronology of Equatorial
Africa. North-Holland Publishing Co., Amsterdam.
 THE OTAVI CARBONATE PLATFORM AND FORESLOPE
Calver, C. R. 1998. Isotope stratigraphy of the Neoproterozoic Togari
Group, Tasmania. Australian Journal of Earth Sciences, 45,
865– 874.
Clark, J. A. 1976. Greenland’s rapid postglacial emergence: A result of
ice-water gravitational attraction. Geology, 4, 310– 312.
Clifford, T. N. 2008. The geology of the Neoproterozoic Swakob-Otavi
transition zone in the Outjo District, northern Damara Orogen,
Namibia. South African Journal of Geology, 111, 117– 140, 3 maps.
Condon, D. J., Prave, A. R. & Benn, D. I. 2002 Neoproterozoic
glacial rainout intervals: observations and implications. Geology,
30, 35 – 38.
Condon, D., Zhu, M., Bowring, S. A., Wang, W., Yang, A. & Jin, Y.
2005. U– Pb ages from the Neoproterozoic Doushantuo Formation,
China. Science, 308, 95– 98.
Dingle, R. V. 1977. The anatomy of a large submarine slump on a sheared
continental margin (SE Africa). Journal of the Geological Society
(London), 134, 293 –310.
Domack, E. W. & Hoffman, P. F. 2011. An ice grounding-line wedge
from the Ghaub glaciation (635 Ma) on the distal foreslope of
the Otavi platform, Namibia, and its bearing on the Snowball
Earth hypothesis. Geological Society of America Bulletin, 123,
1448– 1477.
Donnadieu, Y., Fluteau, F., Ramstein, G., Ritz, C. & Besse, J. 2003.
Is there a conflict between the Neoproterozoic glacial deposits and the
snowball Earth interpretation: an improved understanding with
numerical modeling. Earth and Planetary Science Letters, 208,
101– 112.
Eberli, G. P. & Ginsburg, R. N. 1987. Segmentation and coalescence of
Cenozoic carbonate platforms, northwestern Great Bahama Bank.
Geology, 15, 75 –79.
Evans, D. A. D. 2000. Stratigraphic, geochronological, and paleomagnetic
constraints upon the Neoproterozoic climatic paradox. American
Journal of Science, 300, 347–433.
Evans, D. A. D. 2006. Proterozoic low orbital obliquity and axial-dipolar
geomagnetic field from evaporite palaeolatitudes. Nature, 444,
51 – 55.
Eyles, N. 2004. Frozen in time: concepts of ‘global glaciation’ from 1837
(die Eiszeit) to 1998 (the Snowball Earth). Geoscience Canada, 31,
157– 166.
Eyles, N. & Januszczak, N. 2004a. ‘Zipper-rift’: a tectonic model for
Neoproterozoic glaciations during the breakup of Rodinia after
750 Ma. Earth-Science Reviews, 65, 1– 73.
Eyles, N. & Januszczak, N. 2004b. Interpreting the Neoproterozoic
glacial record: the importance of tectonics. In: Jenkins, G. S.,
McMenamin, M. A. S., McKey, C. P. & Sohl, L. (eds) The
Extreme Proterozoic: Geology, Geochemistry, and Climate. Geophy-
sical Monograph 146. American Geophysical Union, Washington,
DC, 125–144.
Eyles, N. & Januszczak, N. 2007. Syntectonic subaqueous mass flows
of the Neoproterozoic Otavi Group, Namibia: where is the evidence
of global glaciation? Basin Research, 19, 179–198, doi: 10.1111/
j.1365-2117.00319.x
Farrell, W. E. & Clark, J. A. 1976. On postglacial sea level. Geophysi-
cal Journal of the Royal Astronomical Society, 46, 647–667.
Frets, D. C. 1969. Geology and structure of the Huab-Welwitschia area,
South West Africa. Precambrian Research Unit Bulletin, 5, 235, 2
maps, University of Cape Town, South Africa.
Friedman, I. & O’Neil, J. R. 1977. Compilation of stable-isotope frac-
tionation factors of geochemical interest. In: Fleischer, M. (ed.)
Data of Geochemistry, 6th edn. United States Geological Survey,
Washington, DC, Professional Paper 44-KK.
Gevers, T. W. 1931. An ancient tillite in South-West Africa. Transactions
of the Geological Society of South Africa, 34, 1 –17.
Goodman, J. & Pierrehumbert, R. T. 2003. Glacial flow of floating
marine ice in ‘Snowball Earth’. Journal of Geophysical Research,
108, doi: 10.1029/2002JC001471.
Goscombe, B. & Gray, D. R. 2007. The Coastal Terrane of the Kaoko
Belt, Namibia: outboard arc-terrane and tectonic significance. Pre-
cambrian Research, 155, 139–158.
Goscombe, B., Hand, M. & Gray, D. 2003. Structure of the Kaoko Belt,
Namibia: progressive evolution of a classic transpressional orogen.
Journal of Structural Geology, 25, 1049– 1081.
207
Halverson, G. P. 2006. A Neoproterozoic chronology. In: Xiao, S. &
Kaufman, A. J. (eds) Neoproterozoic Geobiology and Paleobiology.
Springer, Dordrecht, 231– 271.
Halverson, G. P., Hoffman, P. F., Schrag, D. P. & Kaufman, J. A.
2002. A major perturbation of the carbon cycle before the Ghaub
glaciation (Neoproterozoic) in Namibia: prelude to snowball
Earth? Geophysics, Geochemistry, Geosystems, 3, doi: 10.1029/
2001GC000244.
Halverson, G. P., Hoffman, P. F., Schrag, D. P., Maloof, A. C. &
Rice, A. H. N. 2005. Toward a Neoproterozoic composite
carbon-isotope record. Geological Society of America Bulletin, 117,
1181–1207, doi: 10.1130/B25630.1
Halverson, G. P., Dudás, F. Ö., Maloof, A. C. & Bowring, S. A.
2007. Evolution of the 87Sr/86Sr composition of Neoproterozoic
seawater. Palaeogeography, Palaeoclimatology, Palaeoecology,
256, 103–129.
Hedberg, R. M. 1979. Stratigraphy of the Owamboland Basin, South
West Africa. Precambrian Research Unit Bulletin, 24, 325, 6 maps,
University of Cape Town, South Africa.
Henry, G., Stanistreet, I. G. & Maiden, K. J. 1986. Preliminary results
of a sedimentological study of the Chuos Formation in the central
zone of the Damara Orogen: evidence for mass flow processes and
glacial activity. Communications of the Geological Survey of South
West Africa/Namibia, 2, 75 –92.
Henry, G., Clendenin, C. W., Stanistreet, I. G. & Malden, K. J.
1990. A multiple detachment model for the early rifting stage of
the Late Proterozoic Damara Orogen in Namibia. Geology, 18,
67– 71.
Higgins, J. A. & Schrag, D. P. 2003. Aftermath of a snowball Earth. Geo-
physics, Geochemistry, Geosystems, 4, doi: 10.1029/2002GC000403.
Hoffman, P. F. 1999. The break-up of Rodinia, birth of Gondwana, true
polar wander, and the snowball Earth. Journal of African Earth
Sciences, 28, 17 –33.
Hoffman, P. F. 2002. Carbonates bounding glacial deposits: evidence
for Snowball Earth episodes and greenhouse aftermaths in the Neo-
proterozoic Otavi Group of northern Namibia. Excursion Guide,
16th International Sedimentological Conference, Auckland Park,
South Africa, 39.
Hoffman, P. F. 2005. 28th DeBeers Alex Du Toit Memorial Lecture: On
Cryogenian (Neoproterozoic) ice-sheet dynamics and the limitations
of the glacial sedimentary record. South African Journal of Geology,
108, 557–576.
Hoffman, P. F. 2011. Strange bedfellows: glacial diamictite and cap
carbonate from the Marinoan (635 Ma) glaciation in Namibia. Sedi-
mentology, 58, 57– 119.
Hoffman, P. F. & Halverson, G. P. 2008. The Otavi Group of the North-
ern Platform and the Northern Margin Zone. In: Miller, R. McG.
(ed.) The Geology of Namibia, vol. 2. Geological Survey of
Namibia, Windhoek, 13.69– 13.136.
Hoffman, P. F. & Hartz, E. H. 1999. Large, coherent, submarine
landslide associated with Pan-African foreland flexure. Geology,
27, 687– 690.
Hoffman, P. F. & Macdonald, F. A. 2010. Sheet-crack cements and
early regression in Marinoan (635 Ma) cap dolostones: regional
benchmarks of vanishing ice-sheets? Earth and Planetary Science
Letters, 300, 374– 384.
Hoffman, P. F. & Schrag, D. P. 2002. The snowball Earth hypothesis:
testing the limits of global change. Terra Nova, 14, 129– 155.
Hoffman, P. F., Hawkins, D. P., Isachsen, C. E. & Bowring, S. A. 1996.
Precise U– Pb zircon ages for early Damaran magmatism in the
Summas Mountains and Welwitschia Inlier, northern Damara belt,
Namibia. Communications of the Geological Survey of Namibia,
11, 47– 52.
Hoffman, P. F., Kaufman, J. A. & Halverson, G. P. 1998a. Comings
and goings of global glaciations on a Neoproterozoic carbonate plat-
form in Namibia. GSA Today, 8, 1 – 9.
Hoffman, P. F., Kaufman, A. J., Halverson, G. P. & Schrag, D. P.
1998b. A Neoproterozoic snowball Earth. Science, 281, 1342– 1346.
Hoffman, P. F., Halverson, G. P., Domack, E. W., Husson, J. M.,
Higgins, J. A. & Schrag, D. P. 2007. Are basal Ediacaran (635 Ma)
post-glacial ‘cap dolostones’ diachronous? Earth and Planetary
Science Letters, 258, 114–131.
208 P. F. HOFFMAN
Hoffmann, K.-H. & Prave, A. R. 1996. A preliminary note on a revised
subdivision and regional correlation of the Otavi Group based on gla-
ciogenic diamictites and associated cap dolomites. Communications
of the Geological Survey of Namibia, 11, 77 –82.
Hoffmann, K.-H., Condon, D. J., Bowring, S. A. & Crowley, J. L.
2004. U– Pb zircon date from the Neoproterozoic Ghaub
Formation, Namibia: constraints on Marinoan glaciation. Geology,
32, 817– 820.
Hurtgen, M. T., Arthur, M. A., Suits, N. S. & Kaufman, A. J. 2002.
The sulfur isotopic composition of Neoproterozoic seawater sulfate:
implications for a snowball Earth? Earth and Planetary Science
Letters, 203, 413–429.
Hurtgen, M. T., Halverson, G. P., Arthur, M. A. & Hoffman, P. F.
2006. Sulfur cycling in the aftermath of a 635 Ma snowball glacia-
tion: evidence for a syn-glacial sulfidic deep ocean. Earth and Plane-
tary Science Letters, 245, 551– 570.
Jiang, G., Kennedy, M. J. & Christie-Blick, N. 2003. Stable isotopic
evidence for methane seeps in Neoproterozoic postglacial cap car-
bonates. Nature, 426, 822– 826.
Jiang, G., Kennedy, M. J., Christie-Blick, N., Wu, H. & Zhang, S.
2006. Stratigraphy, sedimentary structures, and textures of the late
Neoproterozoic Doushantuo cap carbonate in South China. Journal
of Sedimentary Research, 76, 978–995.
Johnson, S. D., Poujol, M. & Kisters, A. F. M. 2006. Constraining the
timing and migration of collisional tectonics in the Damara Belt,
Namibia: U– Pb zircon ages for the syntectonic Salem-type Stinkbank
granite. South African Journal of Geology, 109, 611– 624.
Kasemann, S. A., Hawkesworth, C. J., Prave, A. R., Fallick, A. E. &
Pearson, P. N. 2005. Boron and calcium isotope composition in Neo-
proterozoic carbonate rocks from Namibia: evidence for extreme
environmental change. Earth and Planetary Science Letters, 231,
73 –86.
Kasemann, S. A., Prave, A. R., Fallick, A. E., Hawkesworth, C. J. &
Hoffmann, K.-H. 2010. Neoproterozoic ice ages, boron isotopes,
and ocean acidification: implications for a snowball Earth. Geology,
38, 775– 778.
Kendall, C. G. St. C. & Warren, J. 1987. A review of the origin and
setting of tepees and their associated facies. Sedimentology, 34,
1007– 1028.
Kennedy, M. J., Runnegar, B., Prave, A. R., Hoffmann, K.-H. &
Arthur, M. A. 1998. Two or four Neoproterozoic glaciations?
Geology, 26, 1059–1063.
Kiessling, W. 2001. Paleoclimatic significance of Phanerozoic reefs.
Geology, 29, 751–754.
Kilner, B., Mac Niocaill, C. & Brasier, M. 2005. Low-latitude glacia-
tion in the Neoproterozoic of Oman. Geology, 33, 413– 416.
King, L. H. 1993. Till in the marine environment. Journal of Quaternary
Science, 8, 347–358.
King, L. H., Kokoengen, K., Fader, G. B. J. & Gunleiksrud, T. 1991.
Till-tongue stratigraphy. Geological Society of America Bulletin, 103,
637– 659.
Köppen, W. & Wegener, A. 1924. Die Klimate der geologischen Vorzeit.
Gebrüder Borntraeger, Berlin.
Kröner, A. & Rankama, K. 1972. Late Precambrian glaciogenic sedi-
mentary rocks in southern Africa: a compilation with definitions
and correlations. Precambrian Research Unit Bulletin, 11, 37,
University of Cape Town, South Africa.
Lythe, M. B. & Vaughan, D. G.the BEDMAP Consortium. 2001.
BEDMAP: a new ice thickness and subglacial topographic model
of Antarctica. Journal of Geophysical Research, 106, 11335–11351.
MacAyeal, D. R. 1993. Binge-purge oscillations of the Laurentide ice
sheet as a cause of the North Atlantic’s Heinrich events. Paleoceano-
graphy, 8, 775–784.
Macouin, M., Besse, J., Ader, M., Gilder, S., Yang, Z., Sun, Z. &
Agrinier, P. 2004. Combined paleomagnetic and isotopic data
from the Doushantuo carbonates, South China: implications for the
‘snowball Earth’ hypothesis. Earth and Planetary Science Letters,
224, 387– 398.
Maloof, A. C., Halverson, G. P., Kirschvink, J. L., Schrag, D. P.,
Weiss, B. P. & Hoffman, P. F. 2006. Combined paleomagnetic,
isotopic and stratigraphic evidence for true polar wander from the
Neoproterozoic Akademikerbreen Group, Svalbard. Geological
Society of America Bulletin, 118, 1099–1124, doi: 10.1130/
B25892.1.
Martin, H. 1965a. The Precambrian geology of South West Africa and
Namaqualand. Precambrian Research Unit Bulletin, 1, 159, 11
plates, 2 maps, University of Cape Town, South Africa.
Martin, H. 1965b. Beobachtungen zum Problem der jung-präkam-
brischen Glazialen Ablagerungen in Südwestafrika (Observations
concerning the problem of the late Precambrian glacial deposits in
South West Africa). Geologische Rundschau, 54, 115– 127.
Martin, H., Porada, H. & Walliser, O. H. 1985. Mixtite deposits of the
Damara Sequence, Namibia, problems of interpretation. Palaeogeo-
graphy, Palaeoclimatology, Palaeoecology, 51, 159–196.
McCay, G. A., Prave, A. R., Alsop, G. I. & Fallick, A. E. 2006. Glacial
trinity: Neoproterozoic Earth history within the British-Irish Caledo-
nides. Geology, 34, 909–912, doi: 10.1130/G22694A.1
McMullen, K., Domack, E., Leventer, A., Olson, C., Dunbar, R. &
Brachfeld, S. 2006. Glacial morphology and sediment formation in
the Metz Trough, East Antarctica. Palaeogeography, Palaeoclima-
tology, Palaeoecology, 231, 169–180.
Meert, J. G., Van der Voo, R. & Ayub, S. 1995. Paleomagnetic investi-
gation of the Neoproterozoic Gagwe lavas and Mbozi complex,
Tanzania, and the assembly of Gondwana. Precambrian Research,
74, 225– 244.
Miller, R. McG. 1997. The Owambo Basin of northern Namibia. In:
Selley, R. C. (ed.) African Basins. Sedimentary Basins of the
World, 3. Elsevier, Amsterdam, 237– 268.
Miller, R. McG. 2008. Otavi Group of the Otavi Mountainland (OML),
the region west of the OML, and the eastern Kaokoveld. In: Miller,
R. McG. (ed.) The Geology of Namibia, 2. Geological Survey of
Namibia, Windhoek, 13.52– 13.69.
Moloto-A-Kenguemba, G., Trindade, R. I. F., Monié, P., Nédélec, A.
& Siqueira, R. 2008. A late Neoproterozoic paleomagnetic pole for
the Congo craton: tectonic setting, paleomagnetism and geochronol-
ogy of the Nola dike swarm (Central African Republic). Precambrian
Research, 164, 214– 226.
Nogueira, A. C. R., Riccomini, C., Sial, A. N., Moura, C. A. V., Trin-
dade, R. I. F. & Fairchild, T. R. 2007. Carbon and strontium isotope
fluctuations and paleoceanographic changes in the late Neoprotero-
zoic Araras carbonate platform, southern Amazon craton, Brazil.
Chemical Geology, 237, 168– 190.
Opdyke, B. N. & Wilkinson, B. H. 1990. Palaeolatitude distribution of
Phanerozoic marine ooids and cements. Palaeogeography, Palaeocli-
matology, Palaeoecology, 78, 135–148.
Ottesen, D., Dowdeswell, J. A. & Rise, L. 2005. Submarine landforms
and the reconstruction of fast-flowing ice streams within a large
Quaternary ice sheet: the 2500-km-long Norwegian– Svalbard
margin (578 – 808N). Geological Society of America Bulletin, 117,
1033– 1050.
Paciullo, F. V. P., Ribeiro, A., Trouw, R. A. J. & Passchier, C. W.
2007. Facies and facies association of the siliciclastic Brak River
and carbonate Gemsbok formations in the Lower Ugab River
valley, Namibia, W. Africa. Journal of African Earth Sciences, 47,
121– 134.
Peltier, W. R., Tarasov, L., Vettoretti, G. & Solheim, L. P. 2004.
Climate dynamics in deep time: modeling the ‘snowball bifurcation’
and assessing the plausibility of its occurrence. In: Jenkins, G. S.,
McMenamin, M. A. S., McKey, C. P. & Sohl, L. (eds) The
Extreme Proterozoic: Geology, Geochemistry, and Climate. Geophy-
sical Monograph 146, American Geophysical Union, Washington,
DC, 107–124.
Pollard, D. & DeConto, R. M. 2007. A coupled ice-sheet/ice-shelf/
sediment model applied to a marine margin flowline: forced and
unforced variations. In: Hambrey, M. J., Christoffersen, P.,
Glasser, N. F. & Hubbard, B. (eds) Glacial Marine Processes and
Products. International Association of Sedimentologists Special
Publication No. 39, Blackwell Publishing, Malden, MA, 37– 52.
Pollard, D. & Kasting, J. F. 2004. Climate-ice sheet simulations of Neo-
proterozoic glaciation before and after collapse to Snowball Earth.
In: Jenkins, G. S., McMenamin, M. A. S., McKey, C. P. & Sohl,
L. (eds) The Extreme Proterozoic: Geology, Geochemistry, and
Climate. Geophysical Monograph 146, American Geophysical
Union, Washington, DC, 91 –105.
 THE OTAVI CARBONATE PLATFORM AND FORESLOPE
Powell, R. D. 1990. Glacimarine processes at grounding-line fans and
their growth to ice-contact deltas. In: Dowdeswell, J. A. &
Scourse, J. D. (eds) Glacimarine Environments: Processes and Sedi-
ments. Geological Society Special Publication No. 53, London,
53 – 73.
Powell, R. D. & Domack, E. 1995. Modern glacimarine environments.
In: Menzies, J. (ed.) Glacial Environments, vol. 1: Modern Glacial
Environments: Processes, Dynamics and Sediments. Butterworth-
Heinemann, Oxford, 445– 486.
Prave, A. R., Strachan, R. A. & Fallick, A. E. 2009. Global C cycle
perturbations recorded in marbles: a record of Neoproterozoic
Earth history within the Dalradian succession of the Shetland
Islands, Scotland. Journal of the Geological Society, London, 166,
129– 135.
Pruss, S. B., Bosak, T., Macdonald, F. A., McLane, M. & Hoffman,
P. F. 2010. Microbial facies in a Sturtian cap carbonate, the Rasthof
Formation, Otavi Group, northern Namibia. Precambrian Research,
181, 187– 108.
Rodgers, J. 1957. The distribution of marine carbonate sediments: a
review. In: Le Blanc, R. J. & Breeding, J. G. (eds) Regional
Aspects of Carbonate Deposition. Society of Economic Paleontolo-
gists and Mineralogists (SEPM) Special Publication No. 5, Tulsa,
Oklahoma, 2– 14.
le Roex, H. D. 1941. A tillite in the Otavi Mountains, S.W.A. Trans-
actions of the Geological Society of South Africa, 44, 207–218, 2
plates, 1 map.
SACS (South African Committee for Stratigraphy) 1980. Stratigraphy of
South Africa. Part 1 (comp. L.E. Kent): Lithostratigraphy of the
Republic of South Africa, South West Africa/Namibia and the repub-
lics of Bophuthatswana, Transkei and Venda. Handbook, Geological
Survey of South Africa, 8, 415– 437.
Shields, G. A. 2005. Neoproterozoic cap carbonates: a critical appraisal
of existing models and the plumeworld hypothesis. Terra Nova, 17,
299– 310.
Shields, G. A., Deynoux, M., Strauss, H., Paquet, H. & Nahon, D.
2007. Barite-bearing cap dolostone of the Taoudéni Basin, northwest
Africa: sedimentary and isotopic evidence for methane seepage after
a Neoproterozoic glaciation. Precambrian Research, 154, 209– 235.
Shoemaker, E. M. 1963. Impact mechanics at Meteor Crater, Arizona.
In: Middlehurst, B. M. & Kuiper, G. P. (eds.) The Moon Meteorites
Comets 4. University of Chicago Press, Chicago, 301– 336.
209
Smit, J. M. 1962. Stratigraphy and metamorphism of the Otavi Series
southeast of Otavi, South West Africa. Transactions of the Geological
Society of South Africa, 65, 63 –78.
Soffer, G. 1998. Evolution of a Neoproterozoic continental margin
subject to tropical glaciation. BA thesis, Harvard College,
Cambridge, USA.
Stanistreet, I. G. & Charlesworth, E. G. 1999. Damaran basement-
cored fold nappes incorporating pre-collisional basins, Kaoko Belt,
Namibia, and controls on Mesozoic supercontinental breakup. Econ-
omic Geology Research Unit, Information Circular, 332, University
of the Witwatersrand, Johannesburg, 14.
Swart, R. 1992. Facies analysis of late Proterozoic carbonate turbidites in
the Zerrissene Basin, Damara Orogen, Namibia. Journal of African
Earth Sciences, 14, 283–294.
Trindade, R. I. F., Font, E., D’Agrella-Filho, M. S., Nogueira,
A. C. R. & Riccomini, C. 2003. Low-latitude and multiple geomag-
netic reversals in the Neoproterozoic Puga cap carbonate, Amazon
craton. Terra Nova, 15, 441– 446, doi: 10.1046/j.1365-3121.2003.
00510.x.
Vasconcelos, C., McKenzie, J. A., Warthmann, R. & Bernasconi, S.
M. 2005. Calibration of the d18O paleothermometer for dolomite
precipitated in microbial cultures and natural environments.
Geology, 33, 317– 320. doi: 10.1130/G20992.1.
Warren, S. G., Brandt, R. E., Grenfell, T. C. & McKay, C. P. 2002.
Snowball Earth: ice thickness on the tropical ocean. Journal of Geo-
physical Research, 107, doi: 10.1029/2001JC001123.
Williams, G. E. 1993. History of the Earth’s obliquity. Earth-Science
Reviews, 34, 1 –45.
Witzke, B. J. 1990. Palaeoclimatic constraints for Palaeozoic palaeo-
latitudes of Laurentia and Euramerica. In: McKerrow, W. S. &
Scotese, C. R. (eds) Palaeozoic Palaeogeography and Biogeo-
graphy. Geological Society of London Memoir, 12, 57 –73.
Yoshioka, H., Asahara, Y., Tojo, B. & Kawakami, S. 2003.
Systematic variations in C, O, and Sr isotopes and elemental concen-
trations in Neoproterozoic carbonates in Namibia: implications for
a glacial to interglacial transition. Precambrian Research, 124,
69– 85.
Ziegler, A. M., Hulver, M. L., Lottes, A. L. & Schmachtenberg,
W. F. 1984. Uniformitarianism and paleoclimates: inferences from
the distribution of carbonate rocks. In: Brenchley, P. J. (ed.)
Fossils and Climate. John Wiley & Sons, New York, 3 –25.
 Chapter 15

The Witvlei Group of East-Central Namibia

A. R. PRAVE1*, K.-H. HOFFMANN2, W. HEGENBERGER2 & A. E. FALLICK3

1
Department of Earth Sciences, University of St Andrews, St Andrews KY16 9AL, UK
2
Geological Survey of Namibia, 1 Aviation Road, Windhoek, Namibia
3
Scottish Universities Environmental Research Centre, East Kilbride G75 0QF, UK
*Corresponding author (e-mail: ap13@st-andrews.ac.uk)

Abstract: The Witvlei Group is preserved in two regional synclinoria in the Gobabis-Witvlei area of east-central Namibia and as isolated
outcrops 90 km SW of Rehoboth, itself some 200 km south of that area. It consists of mixed, coarse- to fine-grained siliciclastic and
carbonate strata deposited in deep- to shallow-marine, and locally non-marine, settings along the post-rift continental margin of the
Kalahari Craton prior to the onset of foreland basin sedimentation recorded by the overlying terminal Neoproterozoic–Cambrian
Nama Group. No direct age constrains exist for the Witvlei Group, but it post-dates c. 800 Ma rift-related rocks and pre-dates the
c. 548 Ma base of the Nama Group, thereby placing it as Cryogenian to Ediacaran in age.
The Witvlei Group consists of three main units, from oldest to youngest, the Blaubeker, Court and Buschmannsklippe Formations. The
Blaubeker Formation is highly variable in thickness and can be as much as 1000 m thick. It consists mostly of massive, polymict diamictite
and, in the area of the type locality, contains conglomerate and pebbly sandstone beds. The diamictic strata combined with the presence of
numerous faceted and striated clasts provide the evidence for glaciogenic influences on sedimentation. The highly variable thickness
pattern likely reflects the infill of palaeo-valleys formed by the deep erosion and scouring of bedrock by ice, and the conglomerates
and pebbly sandstones record glacial outwash processes. The Tahiti Formation is a locally developed, fine-grained sandstone above
the Blaubeker Formation. It is poorly exposed and its exact stratigraphic relationship to the Blaubeker rocks and overlying Court Formation
remains to be determined.
The Blaubeker rocks are overlain sharply by the basal unit of the Court Formation, the Gobabis Member. This Member is from
20 to 60 m thick and consists mostly of dark and light grey laminated dolostones that display a d13Ccarbonate profile that rises
from values of 24‰ in the lowermost beds to values of 5‰ in the topmost. The Gobabis Member is conformably overlain by the
shales, marls and thin limestones of the Constance Member followed by quartzites of the uppermost unit of the Court Formation, the
Simmenau Member.
The basal unit of the Buschmannsklippe Formation is the light to tan and pink grey dolostone of the Bildah Member. Its basal contact
is sharp everywhere, and it is gradationally overlain by a coarsening (shoaling) upward succession from shales, thin limestones (some
exhibiting formerly aragonitic fans) and fine sandstones of the La Fraque Member, to interbedded quartzites and stromatolitic and
cherty dolostones of the Okambara Member. The d13Ccarbonate profile for the Buschmannsklippe rocks shows that the basal beds of the
Bildah Member begin at –4‰, followed by a decline to –6‰ in the lower La Fraque limestones and then a rise to –3‰ in the dolostones
of the Okambara Member before being truncated by the base of the regionally unconformably overlying basal Weissberg Quartzite
Member of the lower Nama Group. Although no glacial sediments have been recognized below the Bildah Member, its lithofacies char-
acter, stratigraphic position and C-isotopic profile are compatible with and strikingly similar to younger Cryogenian cap carbonates. Thus,
the Witvlei Group arguably contains both the older and younger cap carbonates of Neoproterozoic time, but only the older Cryogenian
glacial deposit.

Glaciogenic rocks of the Witvlei Group form part of a mixed
siliciclastic and carbonate sedimentary succession of Late Neopro-
terozoic (Cryogenian – early Ediacaran) age that is preserved along
the NW margin (present-day coordinates) of the Kalahari Craton
(Fig. 15.1). It averages 200 –400 m in thickness (but has a compo-
site thickness of more than 1000 m) and rests unconformably on a
varied substrate comprised of thick sequences of pre-glacial, early –
middle Neoproterozoic (Tonian – early Cryogenian) rift-related
arkosic sandstones, conglomerates and argillites of the Eskadron
and Doornpoort Formations (Tsumis Group) and the Kamtsas For-
mation (Nosib Group) and, locally, older Palaeo/Mesoproterozoic
metasedimentary and igneous rocks. The Witvlei Group records
sedimentation in a range of shallow- to deep-marine and locally
non-marine settings that developed following rifting and antece-
dent to the onset of foreland basin development represented by
the unconformably overlying terminal Neoproterozoic (late Edia-
caran) – Cambrian Nama Group. Hegenberger (1993) provides the
most complete description and synthesis of the Witvlei Group
rocks, while Hoffmann (1989) presents lithostratigraphic corre-
lations of the Witvlei Group succession within a regional strati-
graphic framework encompassing the Naukluft Nappe Complex,
the southern and northern Damara Belt (including the Otavi
Group) and the Gariep Belt of southern Namibia.
Only one glaciogenic unit has been recognized in the Witvlei
Group, the diamictic Blaubeker Formation, but two ‘cap
carbonates’ are present (Fig. 15.2). The older of the two is the
Gobabis Member of the Court Formation; it caps the Blaubeker
diamictite. The younger one is the basal unit of the Buschmannsk-
lippe Formation, the Bildah Member; it sits sharply on and trans-
gresses a variable substrate of the Court Formation and, where
absent, older Kamtsas and Eskadron-Doornpoort Formation
rocks, but no glaciogenic rocks have been recognized.
Structural framework
The Witvlei Group is preserved in two NE– SW trending regional
synclinoria located within the southern foreland zone of the
Damara Belt in east-central Namibia (Fig. 15.1; Hegenberger
1993). The western one, the Witvlei Synclinorium, averages
about 20 km in width and is 180 km in length. It comprises
several open to tight second- and numerous third-order folds,
and exposures are good along many of the moderately to steeply
dipping limbs. The eastern structure is termed the Gobabis Syncli-
norium and covers an area of c. 80 km by 100 km; dips are rela-
tively shallow and exposure is poor. Minor faulting disrupts the
stratigraphy in both synclinoria, but displacements are small and
typically no more than a few tens of metres. Several larger steep
faults are associated with the Witvlei Synclinorium but they do
not compromise the stratigraphic coherence of the succession.

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 211– 216. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.15
212
Folding was associated with very low-grade metamorphism,
decreasing from NW to SE, and is attributed to the late Precam-
brian –Cambrian Damaran orogeny (Miller 1983).
The original extent of the Witvlei Group is not known but was
likely much greater than the currently preserved area of outcrop.
To the NW, the Witvlei Group outcrop belt is truncated by the
Fig. 15.2. Stratigraphy of the Witvlei Group (after Hoffmann 1989;
Hegenberger 1993). Glaciogenic rocks (Blaubeker Formation) and cap
carbonates (Gobabis and Bildah Members) are highlighted in bold and with
shaded ornament. See text for discussion.
A. R. PRAVE ET AL.
Fig. 15.1. Generalized geological map of
the southern Damara foreland showing the
distribution of Witvlei Group rocks in the
Witvlei-Gobabis area (after Hegenberger
1993). Inset map shows location of main
figure and the major tectonic regions and
sedimentary basins. B, Blaubeker; GB,
Gariep Belt; NNC, Naukluft Nappe
Complex; R, Rehoboth.
Damaran structural front, defined by thrust faults involving pre-
Witvlei Group rocks including crystalline basement. To the east,
south and SW, the Witvlei Group rocks are buried beneath Phaner-
ozoic strata (Karoo and Kalahari Groups) and how far they extend
under that cover sequence is not known, although drill holes some
40 km south of the southwestern tip of the Witvlei Synclinorium
have intersected Witvlei Group rocks (Hegenberger 1993).
Further SW, within the western part of the southern Damara fore-
land, glacial sediments of the Blaubeker Formation are preserved
as local erosional remnants on Blaubeker farm, the designated
type locality (Schalk 1970) some 90 km SW of Rehoboth, with
thick strata of the Nama Group of the northern Nama (Zaris)
basin resting directly on the older Neoproterozoic Tsumis and
Nosib Group rift-related sediments, and, still further west,
Palaeo- and Mesoproterozoic basement. Absence of rocks equival-
ent to the Court and Buschmannsklippe Formation is either
due to non-deposition or erosional removal prior to deposition of
the Nama.
In the Gobabis-Witvlei area, the lower units of the Witvlei
Group, the Blaubeker and Court Formation, display lateral thick-
ness and facies variations and a major intraformational unconfor-
mity at the base of the Simmenau Member that has been
interpreted as evidence for influence by synsedimentary tectonism
(Hegenberger 1993). No unequivocal basin-bounding faults and/
or growth structures have been identified, but the Nina Anticline
(which separates the Gobabis and Witvlei Synclinoria) was
actively up-warped during this time, resulting in erosion of
lower units of the Court Formation across the crest of that struc-
ture. The upper unit of the Witvlei Group, the Buschmannsklippe
Formation, exhibits systematic thickness variations and similar
facies characteristics throughout the region, and displays an
overall westward thickening (Hegenberger 1993). This suggests
that sedimentation occurred during a phase of relatively uniform
subsidence.
Stratigraphy
The rocks that make up the Witvlei Group, including the locally
preserved diamictite beds in the Gobabis-Witvlei area, were first
 THE WITVLEI GROUP
described collectively as the Buschmannsklippe Formation by
Martin (1965) and regarded for many years as a local, northeastern
facies of the Nama Group (e.g. Hegenberger & Seeger 1980).
Since then, lithostratigraphic subdivision and nomenclature of
the succession has undergone several revisions, the most important
of which is the recognition that the lower and middle parts of the
original Buschmannsklippe Formation are older than the Nama
Group, which led to the establishment of the Witvlei Group as a
major new division for these units (Hoffmann 1989; see Hegenber-
ger 1993 for a review). Here we adopt the subdivision by Hegen-
berger (1993), but include the Blaubeker Formation at the base
as recommended by Hoffmann (1989). The Witvlei Group thus
consists of three main units, the Blaubeker, Court and Busch-
mannsklippe Formations (Fig. 15.2). The Blaubeker Formation is
patchily preserved, the Court Formation is most complete in the
Gobabis Synclinorium, and the Buschmannsklippe Formation is
best developed and exposed in the Witvlei Synclinorium.
The Blaubeker Formation is characterized by polymict diamic-
tite. At the type locality on Blaubeker farm and surrounding areas,
it overlies the Kamtsas Formation with an angular unconformity
and consists of a basal unit of interbedded conglomerate and
quartzite followed by massive polymict diamictite, which is
unconformably overlain by the Nama Group; both the Court and
Buschmannsklippe Formations are absent. Total thickness varies
between 250 and 300 m (Schalk 1970). In the Gobabais-Witvlei
area, the Blaubeker Formation is present in two main outcrop
areas where it unconformably overlies the Kamtsas Formation.
The thickest and best-developed exposures are in the south-central
portions of the Gobabis Synclinorium (particularly in the valleys of
the White and Black Nossob Rivers) and it is also present along the
easternmost edge of the south-central flanks of the Witvlei Syncli-
norium (Hegenberger 1993). In these areas the Blaubeker For-
mation is composed entirely of massive polymict diamictite and
is highly variable in thickness, ranging from as little as 10 m to
as much as 1000 m (Hegenberger & Seeger 1980).
The Court Formation consists of four units (Hegenberger
1993). The oldest is the Tahiti Member, a 1– 10-m-thick fine-
grained feldspathic arenite. This member is poorly exposed and
few sedimentological details exist. It is restricted in its occurrence
to a few outcrops along the extreme east-southeastern edge of
the Gobabis Synclinorium (and may well be a facies of the
Blaubeker Formation, but this remains to be proven). The first, well-
developed and laterally extensive unit of the Court Formation is
the Gobabis Member. It mostly consists of grey laminated
dolostones and local stromatolitic limestone, and varies between
20 and 60 m in thickness. The youngest units of the Court Formation
are the Constance Member, a recessive, poorly exposed 100–
200-m-thick succession of pale-coloured shale, marl and thin-
bedded limestone, and the Simmenau Member, a 100 –150-m-
thick fine- to medium-grained quartzite to feldspathic arenite.
The Buschmannsklippe Formation is divided into three units,
from oldest to youngest, the Bildah, La Fraque and Okambara
Members (Hegenberger 1993). The Formation as a whole defines
a west-SW-thickening wedge, from about 50 m thick in the east-
ernmost sections of the Gobabis Synclinorium to more than
300 m thick in the Witvlei Synclinorium. The Bildah Member is
typically 40– 60 m thick and consists of light grey to pale tan
and pink dolostone. Its base defines a regional unconformity that
oversteps the underlying Court Formation to rest unconformably
on older Neoproterozoic rocks. Overlying the Bildah is the La
Fraque Member, a succession of interbedded shale, thin-bedded
limestone and fine sandstone. It thickens to the west and reaches
a maximum thickness of c. 150 m in the central portions of the
Witvlei Synclinorium. The Okambara Member, like the La
Fraque, similarly thickens westward attaining a maximum thick-
ness of about 125 m. It is characterized by cream to pink-grey
coloured commonly stromatolitic dolostone and cherty dolostone
and subordinate fine, white quartzite. In the thicker sections in
the central portions of the Witlvlei Synclinorium, Hegenberger
213
(1993) subdivided the Okambara into an upper and lower
dolostone-dominated unit separated by a quartzite-dominated unit.
The top of the Witvlei Group is marked by the regional uncon-
formity at the base of the Weissberg Member (Dabis Formation) of
the lower Nama Group.
Glaciogenic deposits and associated strata
The Blaubeker Formation
No detailed modern sedimentological investigations of the Blaube-
ker Formation have been carried out, and the following summary is
based on descriptions by Martin (1965), Schalk (1970), Kröner &
Rankama (1972), Hegenberger & Seeger (1980) and field obser-
vation by the authors. In the type locality and adjacent areas
west of Rehoboth, exposures are patchy, but show that the Blaube-
ker Formation consists of a lower unit of pebbly quartzite with rare
shale interbeds. This unit is several tens of metres thick and is
sharply overlain by massive polymict diamictite between 200
and 300 m thick. In the Gobabis-Witvlei area, the Blaubeker con-
sists entirely of massive, non-graded, non-stratified, polymict dia-
mictite; the best exposures are at Farm Tahiti in the Black Nossob
River valley. In both synclinoria, clasts are dispersed in a grey-
green to reddish-grey, fine siliciclastic matrix; the matrix is
massive and no sandstone lenticules or laminae have been
reported. Clasts typically range between 0.05 and 0.5 m in the
longest dimension, but can be as large as 1 m. They consist
mainly of quartzite, lesser felsic and mafic volcanics, granitoids,
gneiss and schist. Shapes vary from angular to sub-rounded and
many of the clasts display variably orientated striations and facet-
ing. This is the primary observational evidence to infer a glacially
influenced origin for the Blaubeker diamictite. At Gobabis, a 0.1 –
0.5-m-thick, crumbly weathering, poorly sorted, massive brown
sandstone containing dispersed quartzite clasts and cobbles
occurs between the underlying Kamtsas Formation and the sharp
base of the Gobabis Member. This sandstone may be a facies of
the Blaubeker Formation, but this remains to be proven.
The Doornpoort/Eskadron and Kamtsas Formation
These rocks are pre-glacial in age but are briefly described here
because many of the clasts in the Blaubeker diamictites were
reworked from these older units. All three formations can reach
thicknesses measured in many thousands of metres and are domi-
nated by feldspathic quartzite, pebbly quartzite and arenite and
local conglomerate. The conglomeratic portions contain clasts
ranging in size from pebbles to boulders that consist of quartz,
quartzite, felsic and mafic volcanic rocks, granites and granitic
gneiss (Hegenberger & Seeger 1980).
The Court Formation
Resting sharply on the Blaubeker Formation is the Gobabis
Member of the Court Formation (locally, the Tahiti Formation
sandstones occur on top of the Blaubeker rocks). In the vicinity
of the type locality, the Gobabis Member is 50–60 m thick and
consists of two main units. The lower unit is of variable thickness
(10 –30 m) and characterized by thin-bedded, planar- to wavy-
parallel laminated dolomicrite. The centimetre-thick laminae alter-
nate between dark grey and lighter grey in colour, giving the
Gobabis Member a distinctive striped pattern. In places, syn-
sedimentary (or early lithification) disruption of bedding has
resulted in locally formed intraformational breccias; these,
however, are rare. Much of the remainder of the Member is com-
posed of similarly dark –light coloured dolomicrite, but the lami-
nation is crinkly-wrinkly in character and exhibits microbial
214 A. R. PRAVE ET AL.
roll-up structures. Columnar stromatolites are developed locally,
and some of these are in limestone rather than dolostone.
The Buschmannsklippe Formation
At its type locality, the Bildah Member rests with sharp contact on
quartzites of the Kamtsas Formation and consists of four units,
from the base upward these are: (i) light grey to tan coloured,
flat to wavy-parallel laminated dolomicrite containing locally
developed soft-sediment slumping and irregularly shaped
calcite-spar-filled vugs; (ii) microbially laminated to stromatolitic
light grey dolomicrite with quartz- and calcite-spar-filled vugs and
tubes (developed in variable densities); (iii) fine-grained dolo-
stone, light grey in colour, exhibiting cross-bedding and flat lami-
nation; and (iv) thin-bedded to laminated, fine-grained, grey
dolostone with mudstone partings. The Bildah cap dolostone
varies in thickness from 60 to 100 m, and much of this variability
is due to the lesser or greater development of the stromatolitic unit.
Gradationally overlying the Bildah dolostones is the reddish-
brown shale and thin-bedded limestone of the lower part of the
La Fraque Member; some of the limestones exhibit centimetre-
scale aragonite fan structures. These beds pass transitionally
upward into tan-brown coloured, cross-bedded and hummocky
cross-stratified, commonly calcareous-cemented, sandstone marking
the upper part of the La Fraque Member. Gradationally overlying the
La Fraque rocks are the light-coloured quartzites and cream-coloured
dolostones of the Okambara Member. The Okambara rocks contain a
variety of sedimentary features including crinkly-wrinkly laminated
dolomicrite, columnar, branching stromatolites, as well as laterally
linked heads, imbricated (edgewise)-, wave-rosette- and flat-pebble
intraclastic limestones and dolostones, cross-bedded dolograin-
stones, rare oolitic dolostone, chicken-wire-fabric quartz nodules,
cauliflower chert and enterolithic bedding. The tops of some beds
exhibit desiccation cracks.
Boundary relations with overlying and underlying
non-glacial units
The Blaubeker Formation everywhere rests unconformably on the
Kamtsas Formation. The highly varied thickness pattern of the
Blaubeker (from 10 m or less to more than 1000 m) is evidence
that its base must be deeply erosive. In the area of the eponymous
type locality, strata of the lower Nama Group (Dabis Formation)
rest unconformably on the Blaubeker rocks. In the Gobabis-
Witvlei area, with the exception of where the areally restricted
Tahiti Member is present, the diamictic strata of the Blaubeker
Member are everywhere sharply capped by the basal laminated
dolostone of the Gobabis Member. The Gobabis Member is over-
lain by the shales and thin limestones of the Constance Member,
but the contact between these two Members is poorly exposed
and commonly obscured by calcrete, so we are uncertain if it is
sharp or gradational.
The base of the Bildah Member rests unconformably on a variable
substrate consisting of the Court and/or Kamtsas or Eskadron-
Doortpoort Formations and is everywhere sharp. The contact
between the Bildah Member and the overlying La Fraque Member
is gradational over a few metres; the topmost several metres of the
Bildah Member exhibit an upward increase in the proportion and
thickness of mudstone partings concomitant with a thinning
upward and proportional decrease in dolostone beds.
Chemostratigraphy
Kaufman et al. (1991, 1997) Saylor et al. (1998) and Kennedy et al.
(1998) did reconnaissance C-isotopic studies on the Witvlei Group
rocks and Gorjan et al. (2003) performed S-isotopic analyses. We
have expanded upon the C-isotopic work and the following is a
preliminary description of new, unpublished, data from the
carbonate-bearing units of the Witvlei Group by the authors. A
total of 193 stable isotopic analyses have been carried out, but
additional work is required before more robust trends can be con-
structed, especially for the less well exposed interglacial units
The C-isotopic profile of the older cap carbonate, the Gobabis
Member, displays a progressively rising trend: the basal layer
has d13Ccarbonate values of –4‰ to – 3‰ and these rise to –2‰
to –1‰ in the overlying thin-bedded and laminated dolomicrites.
The contact between the lower planar- to wavy-laminated unit and
the upper crinkly-wrinkly laminated unit with roll-up structures is
everywhere covered and/or mantled with calcrete, but the lowest
exposed beds of the upper unit begin with d13Ccarbonate values of
5‰, and these stay between 3‰ and 5‰ to the top of the
Gobabis Member. Because the contact is covered, we are uncertain
if the rise to positive values is gradational or abrupt.
The inter-glacial succession consists of the Constance and
Simmenau Members of the Court Formation. The former is
poorly exposed and the latter does not contain carbonate rocks,
so we have few data on these units and more work needs to be
carried out before anything substantial can be stated regarding
their chemostratigraphy.
The basal laminated unit of the Bildah cap dolostone consistently
has d13Ccarbonate values around –4‰; these rise slightly to values
around –3‰ in the middle and upper units of the cap dolostone.
At the contact with, and through the transition into, the La Fraque
Member, C-isotopic values decline to –5‰ to –6‰ through the
limestone rhythmite and thin-bedded limestone units (it is note-
worthy that at this stratigraphic level, fans considered to be formerly
aragonitic in composition are present). Above this and through
the Okambara Member, the d13Ccarbonate profile rises to values
between –3‰ and –2‰ before being truncated by the basal uncon-
formity of the overlying Weissberg Member of the Nama Group.
Other characteristics (e.g. economic deposits, biomarkers)
No biomarker data have been obtained on the Witvlei Group rocks,
nor are these rocks known to contain any major deposits or miner-
alizaton of economic importance. The Bildah Member has been
quarried locally for dimension stone.
Palaeolatitude and palaeogeography
Kröner et al. (1980) obtained palaeomagnetic data from three sites
of the Blaubeker Formation at its type locality. The data indicated
that the magnetization post-dated folding and was attributed to
thermal events associated with Damaran orogenesis. No other
palaeomagnetic data exist on the Witvlei Group rocks, but it is gen-
erally assumed that they, as part of the Kalahari craton, were
deposited in low-latitudinal positions during most of the Cryogen-
ian (e.g. Li et al. 2008).
Geochronological constraints
No direct geochronological data exist on the Witvlei Group rocks.
The Witvlei Group rests unconformably on rocks that are correla-
tive with those elsewhere in central and southern Namibia con-
sidered as c. 1000– 800 Ma in age; it is overlain unconformably
by the late Ediacaran – Cambrian Nama Group (Hegenberger
1993). Age constraints are good for the lower part of the Nama
Group and an ash bed within the Zaris Formation in the main
Nama basin to the SW has yielded a U – Pb zircon age of 548 Ma
(Grotzinger et al. 1995). These broad constraints show that the
Witvlei Group is entirely Neoproterozoic (Cryogenian to Edia-
caran) in age.
 THE WITVLEI GROUP
The lithological and chemostratigraphic similarity of the Bildah
Member to the Keilberg cap dolostone (Hoffmann 1989; Kennedy
et al. 1998) can be used to infer an age of 635 Ma, that is, the
timing of meltback associated with the Marinoan-equivalent
glacial and cap-carbonate rocks in northern Namibia and southern
China (Hoffmann et al. 2004; Condon et al. 2005). The post-
Marinoan d13C crossover to positive values (e.g. Halverson et al.
2005) is not recorded in the Buschmannsklippe C-isotopic data,
most likely due to erosional truncation along the unconformity
surface defining the base of the Nama Group. If so, then the time
gap between the base of the Nama Group and top of the Witvlei
Group could be considerable and on the order of many tens of
millions of years.
Direct age constraints are also lacking for the base of the Witvlei
Group, but can be inferred from the proposed correlation of the
Gobabis Member and Blaubeker Formation with the Rasthof cap
carbonate and Chuos Formation of northern Namibia (Hoffmann
1989; Kennedy et al. 1998), which overlies felsic volcanic rocks
with a U –Pb zircon age of 746 Ma (Hoffman et al. 1996). A still
younger age has been suggested as c. 710 Ma (Halverson et al.
2005). It is reasonable, then, to infer a similar age for the
Blaubeker-Gobabis glacial –cap couplet. If this is correct, then
the Witvlei Group likely spans some 100 Ma in time, from the
Cryogenian through to at least the early –middle Ediacaran.
Discussion
The Blaubeker Formation is the only unit in the Witvlei Group suc-
cession that can be attributed to recording glacial influences on
sedimentation in the form of diamictic rocks containing striated
and faceted clasts. It occurs in narrow outcrop belts displaying
highly variable thicknesses, and these features most likely are
the result of glacial scouring and carving of valleys and channels.
The pebbly sandstones and conglomerates associated with the
diamictites represent glacial outwash.
The Blaubeker is sharply capped by the Gobabis Member dolo-
stones. The overall facies development of this Member is identical
to that documented and described from the Rasthof Formation, the
older of the two cap carbonates in the Otavi succession of northern
Namibia (Hoffmann 1989; Hoffmann & Prave 1996; Kennedy
et al. 1998; Hoffman & Schrag 2002). This has been used in com-
bination with C-isotopic data to correlate the Gobabis dolostones
and the Blaubeker diamictite with the Rasthof Formation and
Chuos Formation diamictite (Hoffmann 1989; Kennedy et al.
1998; Hoffman & Schrag 2002). The overlying Constance and
Simmenau Members form the inter-glacial package and are
typified by fine-grain sizes, flat lamination, low-angle cross-
stratification and current ripples. Rare desiccation cracks are
preserved in the Constance Member and the Simmenau Member
contains mud-chip conglomerates in the base of broad, shallow-
channelled beds. These features have been used to infer that the
Constance Member was deposited in a low-energy intermittently
exposed, shallow-water setting and that the Simmenau Member
records a distal, fine-grained braidplain (Hegenberger 1993). Com-
bined, the Court Formation records a post-glacial transgression
from deeper marine (Gobabis Member) through shoaling into shal-
lower marine (Constance Member) settings.
Nowhere along the outcrop belt of the Witvlei Group have rocks
been recognized as glacial in origin beneath the Bildah Member.
Based on striking similarities in lithology and facies, it was pro-
posed to be equivalent to the characteristic pink-grey dolostones
that cap diamictites in the Naukluft Nappe Complex and the
Gariep Belt (Hoffmann 1989). This established, for the first
time, that the Bildah, even given the absence of an underlying dia-
mictite, was the same (cap) carbonate unit overlying a regional
glacial unconformity surface. This is similar to what is now recog-
nized from many places worldwide where glacial rocks are only
patchily preserved and it is the overlying cap carbonates that are
215
the most extensively and consistently developed (e.g. Hoffman
& Schrag 2002; Fairchild & Kennedy 2007). Since then, the C-
isotopic profile and overall facies of the Bildah Member has
been shown to be identical to those known to cap the younger
Cryogenian glaciation elsewhere and is now interpreted as being
equivalent to the Keilberg cap dolostone of the Otavi Group in
northern Namibia (Hoffmann & Prave 1996; Kennedy et al.
1998; Hoffman & Schrag 2002). The overlying La Fraque –
Okambara succession records the post-glacial transgression
through shoaling sequence: from the deeper-marine settings of
the lower La Fraque (interbedded shale and originally aragonite-
fan-bearing limestone rhythmites), to storm-dominated inner
shelf settings (hummocky cross-stratified sandstones) of the
upper La Fraque, to shallow subtidal and intertidal (quartz-
replacement evaporite fabrics and desiccated microbial laminites)
depositional settings of the Okambara rocks (Hegenberger 1993).
K.-H. H. and W. H. thank the Geological Survey of Namibia for support. A. R. P.
and A. E. F. acknowledge NERC and The Carnegie Trust for The Universities of
Scotland for support. This represents a contribution of the IUGS- and UNESCO-
funded IGCP (International Geoscience Programme) Project #512.
References
Condon, D., Zhu, M., Bowring, S., Wang, W., Yang, A. & Jin, Y. 2005.
U–Pb ages from the Neoproterozoic Doushantuo Formation, China.
Science, 308, 95– 98.
Fairchild, I. & Kennedy, M. J. 2007. Neoproterozoic glaciation in the
Earth System. Journal of the Geological Society, London, 164,
895– 921.
Gorjan, P., Walter, M. R. & Swart, R. 2003. Global Neoproterozoic
(Sturtian) post-glacial sulfide – sulfur isotope anomaly recognized in
Namibia. Journal of African Earth Sciences, 36, 89 –98.
Grotzinger, J. P., Bowring, S., Saylor, B. Z. & Kaufman, A. J. 1995.
Biostratigraphic and geochronologic constraints on early animal
evolution. Science, 270, 598– 604.
Halverson, G. P., Hoffman, P. F., Schrag, D. P., Maloof, A. C. &
Rice, A. H. N. 2005. Towards a Neoproterozoic composite carbon
isotope record. Geological Society of America Bulletin, 117,
1181–1207.
Hegenberger, W. 1993. Stratigraphy and sedimentology of the Late
Precambrian Witvlei and Nama Groups, East of Windhoek.
Geological Survey of Namibia, Memoir 17, 82.
Hegenberger, W. & Seeger, K. G. 1980. The geology of the Gobabis
area. Explanation of Sheet 2218, scale 1:250,000. Geological
Survey of South West Africa/Namibia Windhoek, 11.
Hoffman, P. F. & Schrag, D. P. 2002. The snowball Earth hypothesis:
testing the limits of global change. Terra Nova, 14, 129– 155.
Hoffman, P. F., Hawkins, D. P., Isachsen, C. E. & Bowring, S. A. 1996.
Precise U– Pb zircon ages for early Damaran magmatism in the
Summas Mountains and Welwitschia Inlier, northern Damara belt,
Namibia. Communications of the Geological Survey of Namibia,
11, 47– 52.
Hoffmann, K. H. 1989. New aspects of lithostratigraphic subdivision and
correlation of late Proterozoic to early Canbrianrocks of the southern
Damara Belt and their correlation with the central and northern
Damara Belt and the Gariep Belt. Communications of the Geological
Survey of Namibia, 5, 59 –67.
Hoffmann, K. H. & Prave, A. R. 1996. A preliminary note on a revised
subdivision and regional correlation of the Otavi Group based on gla-
ciogenic diamictites and associated cap dolostones. Communications
of the Geological Society of Namibia, 11, 81 –86.
Hoffmann, K. H., Condon, D. J., Bowring, S. A. & Crowley, J. L.
2004. A U– Pb zircon date from the Neoproterozoic Ghaub For-
mation, Namibia: constraints on Marinoan glaciation. Geology, 32,
817– 820.
Kaufman, A. J., Hayes, J. M., Knoll, A. H. & Germs, G. J. B. 1991. Iso-
topic compositions of carbonates and organic carbon from upper Pro-
terozoic successions in Namibia: stratigraphic variation and the
effects of diagenesis and metamorphism. Precambrian Research,
49, 301– 327.
216 A. R. PRAVE ET AL.
Kaufman, A. J., Hayes, J. M., Knoll, A. H. & Narbonne, G. M.
1997. Isotopes, ice ages, and terminal Proterozoic earth
history. Proceedings of the National Academy of Science, 94,
6600– 6605.
Kennedy, M. J., Runnegar, B., Prave, A. R., Hoffmann, K. H. &
Arthur, M. 1998. Two or four Neoproterozoic glaciations?
Geology, 26, 1059–1063.
Kröner, A., McWilliams, M. O., Germs, G. J. B., Reid, A. B. & Schalk,
K. E. L. 1980. Paleomagnetism of late Precambrian to early Paleozoic
mixtite-bearing formations in Namibia (South West Africa): the
Nama Group and Blaubeker Formation. American Journal of
Science, 280, 942–968.
Kröner, A. & Rankama, K. 1972. Late Precambrain glaciogenic sedi-
mentary rocks in southern Africa: a compilation with definitions
and correlations. Bulletin of the Precambrian Research Unit, Uni-
versity of Cape Town, 11, 37.
Li, Z. X., Bogdanova, S. V. et al. 2008. Assembly, configuration, and
break-up history of Rodinia: a synthesis. Precambrian Research,
160, 179– 210.
Martin, H. 1965. The Precambrian Geology of South West Africa and
Namaqualand. Precambrian Research Unit, University of Cape
Town, Cape Town.
Miller, R. McG. 1983. The pan-African Damara Orogen of South West
Africa/Namibia. In: Miller, R. McG. (ed.) Evolution of the Damara
Orogen of Sout West Africa/Namibia. Special Publication of the Geo-
logical Survey of South Africa, 11, 431– 515.
Saylor, B. Z., Kaufman, A. J., Grotzinger, J. P. & Urban, F. 1998.
A composite reference section for terminal Proterozoic strata of
southern Namibia. Journal of Sedimentary Research, 68, 1223– 1235.
Schalk, K. E. L. 1970. Some late Precambrian formations in central South
West Africa. Annals of the Geological Survey of South Africa, 8,
29 – 47.
 Chapter 16

The Chameis Gate Member, Chameis Group, Marmora Terrane, Namibia

HARTWIG E. FRIMMEL1,2
1
Department of Geological Sciences, University of Cape Town, Rondebosch 7701, South Africa
2
Present address: Geodynamics & Geomaterials Research Division, Institute of Geography and Geology, University of Wuerzburg,
Am Hubland, D-97074 Wuerzburg, Germany
(e-mail: hartwig.frimmel@uni-wuerzburg.de)

Abstract: The Chameis Gate Member is a poorly exposed and poorly investigated diamictite in the Chameis Subterrane of the Marmora
Terrane, which forms the western, completely allochthonous part of the Pan-African Gariep Belt (southwestern Namibia). Its significance
lies in its position in an entirely oceanic unit, the Dernburg Formation, which is dominated by mafic volcanic rocks. The diamictite con-
tains exotic dropstones in a mafic volcaniclastic matrix, thus providing evidence for transport by ice away from the continental margin
into an oceanic environment. No direct age data are available and stratigraphic relationships are obscured by limited outcrop and intense
syn-orogogenic deformation. Preliminary chemostratigraphic data obtained on carbonate rocks below and above the diamictite, impre-
cise Pb–Pb age data on the largely volcaniclastic silicate fraction within associated stromatolitic reef carbonates, and imprecise Ar–Ar
data on early hornblende related to sea-floor metamorphism of the associated volcanic rocks all point to an age loosely constrained
between 640 and 580 Ma. Based on a comparison between the tectono-stratigraphic units of the Marmora Terrane with the continental
Port Nolloth Group on the one side and the Rocha Group of the Punta del Este Terrane in Uruguay on the other side of the terrane, it is
suggested that the diamictite was deposited in a back-arc basin that developed in response to the 640 –590 Ma volcanic arc of the Dom
Feliciano Belt in southeastern Brazil and eastern Uruguay.

The Neoproterozoic Marmora Terrane is an internal, entirely
allochthonous tectonic unit in the western part of the Gariep Belt
(Fig. 16.1). Although the Marmora Terrane stretches along the
Atlantic coast of Namibia from 27.468S southwards into South
Africa to 28.768S, just south of the Orange River mouth, exposures
of the Chameis Gate Member are restricted to a narrow coastal
strip in the so-called Sperrgebiet (‘restricted area’) of southwestern
Namibia. As all outcrops of this member are effectively located
within the mining lease area of the alluvial diamond operations
of Namdeb Pty Ltd. along the coast NW of Oranjemund, access
to these outcrops is restricted. Moreover, the outcrops are in a pre-
dominantly sandy desert devoid of infrastructure, and much of the
rocks of the Chameis Gate Member are covered by shifting sand
dunes. As a consequence, this unit is only very poorly investigated.
The area was first mapped by Kaiser (1926), but he did not dis-
tinguish the diamictite that characterizes the Chameis Gate
Member. The largest exposures of the member can be found on
wind-blown surfaces to the west and south of a security gate,
named Chameis Gate, at 27.8378S, 15.7278E. Minor outcrops
can be found at Bakers Bay at 27.6558S, 15.5368E and further
NW at 27.4538S, 15.4258E (Fig. 16.2). The Chameis Gate
Member is so far the only glaciogenic unit that has been recognized
in that subterrane, and it was first described by Frimmel (2000b).
Preliminary lithogeochemical and chemostratigraphic data for
the bounding strata have been presented by Frimmel & Jiang
(2001). No further follow-up studies have been carried out, and
much of its stratigraphic interpretation hinges on circumstantial
evidence.
The main significance of the Chameis Gate Member lies in its
palaeodepositional environment. In contrast to most, if not all,
other diamictite units described so far, this diamictite occurs
associated with oceanic rocks with very little continental influence
except for pre-Gariep basement-derived clasts and dropstones in
the diamictite (Frimmel & Jiang 2001).
Structural framework and tectonic evolution
The Chameis Gate Member forms part of the largely oceanic
Marmora Terrane, which represents an allochthonous nappe
complex that was thrust in an easterly to southeasterly direction
onto the para-autochthonous continental Port Nolloth Zone along
the Schakalsberge Thrust (Fig. 16.1). The Marmora Terrane has
been subdivided into three tectono-stratigraphic units, the Scha-
kalsberge, Oranjemund and Chameis complexes (Hartnady &
Von Veh 1990) or subterranes (Frimmel 2000b). The tectonically
lowest of these units, the Schakalsberge Subterrane, consists
mainly of mafic metavolcanic rocks of the Grootderm Formation
with a dolomitic, in places stromatolitic to oolitic carbonate on
top, interpreted as a former guyot that evolved from an oceanic
seamount (Frimmel et al. 1996). The tectonically intervening
Oranjemund Subterrane consists of predominantly siliciclastic,
commonly turbiditic, metasedimentary rocks, with only very minor
carbonate and ferruginous chert, originally referred to as the Oran-
jemund Formation (Frimmel 2000b) but since raised to the rank of
group (Basei et al. 2005). The Chameis Subterrane, which litho-
stratigraphically comprises the Chameis Group and contains the
Chameis Gate Member, is the tectonically highest unit. It is inten-
sely deformed, resembling a tectonic melange zone, which makes
a stratigraphic subdivision difficult. The lower part consists of pre-
dominantly mafic metavolcanic rocks, similar to those of the
Grootderm Formation, whereas the upper part consists of carbon-
ates and siliciclastic metasedimentary rocks that resemble those of
the Holgat Formation in the Port Nolloth Zone and those of the
upper Oranjemund Group.
No basement has been identified in the Chameis Subterrane.
Available geochemical and petrological data for the mafic rocks
(both extrusive and intrusive) across the Marmora Terrane are
most compatible with an oceanic intra-plate setting (Frimmel
et al. 1996). For some metagabbros in the Chameis Subterrane,
geochemical characteristics of mid-ocean ridge basalt are indi-
cated. Although exact ages for the time of magmatism are lacking,
some crude constraints on the timing of sea-floor metamorphism
(Frimmel & Frank 1998) and of oceanic volcanism (Frimmel &
Fölling 2004), detailed below, suggest an interval sometime
between 640 and 580 Ma for the time of oceanic crust formation.
Based on zircon provenance data from the Oranjemund Group
siliciclastic rocks and comparable rocks from the Rocha Group
in the Punta del Este Terrane in eastern Uruguay, a back-arc
basin to the 640–590 Ma volcanic arc of the Dom Feliciano Belt
in southeastern Brazil and eastern Uruguay has been suggested
as the most likely depositional environment for the lower

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 217– 221. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.16
218 H. E. FRIMMEL

Fig. 16.1. Tectonic sub-division of the Marmora Terrane in the western Gariep
Belt (modified after Hartnady & von Veh 1990).

stratigraphic units of the Marmora Terrane (Basei et al. 2005).

During amalgamation of SW Gondwana, the basin was closed,
resulting in syn-orogenic sedimentation in a foredeep in front
and on top of already stacked thrust sheets of the Marmora
Terrane. This model explains the remarkable similarity in the
younger sedimentary deposits across tectonic boundaries in both
the Port Nolloth Zone and the Marmora Terrane (Frimmel &
Fölling 2004).
A penetrative foliation (s1) dips to the west and is axial –planar
relative to tight to isoclinal, east-vergent F1 folds. Locally, s1 was
refoliated around open to tight, NE-trending F2 kink bands. The
contact with the underlying Port Nolloth Zone is a major thrust
fault with top-to-SE transport (Schakalsberge Thrust). The inten-
sity of folding increases towards the Schakalsberge Thrust. At
the northern boundary of the Marmora Terrane, F1 folds within
the Chameis Subterrane are truncated by the Schakalsberge
Thrust, highlighting its syn-D2 timing. The contact between the
Oranjemund and the Schakalsberge Subterranes is a west- to
NW-dipping fault plane. The contact between the Chameis and
Oranjemund Subterranes is a zone of intense F2 folding with
top-to-SE transport and subsequent back-folding that resulted in
locally steeply SE-dipping crenulation.
The preferred kinematic model for the evolution of the main
orogenic structures involves two stages. The first is eastward
transport of the oceanic rocks of the Marmora Terrane, internal
stacking of these, and eventually obduction onto the continental
margin of the Port Nolloth Zone. The variable angle between
principal direction of compression and the pre-existing arcuate
north- to NW-trending continental margin led to strain partitioning
Fig. 16.2. Map showing the distribution of the Chameis Gate Member within
the Chameis Subterrane, Marmora Terrane (Gariep Belt), in southwestern
Namibia.
with NE-, east- and SE-directed transport along the northeastern,
eastern and southeastern basin margin, respectively (Frimmel
2000a). This was followed by the second stage, sinistral transpres-
sion with overall top-to-SE transport. The resulting folds and
thrusts are east-verging along the northeastern lateral ramps and
SE-verging along the southern frontal ramps (in South Africa)
within the outer Port Nolloth Zone (Gresse 1994).
The Marmora Terrane differs from the Port Nolloth Zone not
only with regard to stratigraphy and deformation, but also in its
metamorphic history. In the Port Nolloth Zone, only one stage
of regional metamorphism is discerned, the peak of which
(c. 520 8C and 3.0 –3.5 kbar) was reached after D2 but prior
to D3; it is ascribed to crustal thickening following overthrusting
of the Marmora Terrane onto the Port Nolloth Zone. The
 THE CHAMEIS GATE MEMBER
metamorphic evolution of the Marmora Terrane is more complex.
There, the mafic and ultramafic rocks experienced polyphase
metamorphism that is not recorded by their sedimentary cover.
As many as three metamorphic amphibole generations therein
reflect different metamorphic conditions of formation (Frimmel
& Hartnady 1992) and yield distinctly different Ar –Ar age
spectra (Frimmel & Frank 1998). They reflect three metamorphic
stages: (i) sea floor metamorphism, (ii) a relatively high-pressure
metamorphic event, probably related to first accretion of mafic
crust and (iii) regional metamorphism as a result of continental
collision.
Stratigraphy
The Chameis Gate Member forms part of the Chameis Group, the
rocks of which constitute the Chameis Subterrane of the Marmora
Terrane. The stratigraphic subdivision of this group given here
(Fig. 16.3) follows a preliminary subdivision presented by
Frimmel (2000b). It should be noted that the intensity of syn-
orogenic tectonic overprint of the entire subterrane makes it vir-
tually impossible to provide reliable constraints on stratigraphic
thickness and, in places, even on stratigraphic boundaries.
A thick pile of mafic metavolcanic rocks at the base of the
succession (Dernburg Formation) with minor, local intercalations
of carbonates (Sholtzberg Member) and diamictite (Chameis Gate
Member) is overlain by a carbonate sequence (Dreimaster Mem-
ber) and a siliciclastic, mainly arenitic succession with turbiditic
affinities. The latter two units constitute the Bogenfels Formation.
The Dernburg Formation was intruded by a series of tectonically
dismembered metagabbro bodies (Bakers Bay Suite).
Fig. 16.3. Generalized lithostratigraphy of the Marmora Terrane and correlation with that of the Port Nolloth Zone, all of which constitutes the Gariep Supergroup
(modified after Frimmel & Fölling 2004).
219
Glaciogenic deposits and associated strata
The Chameis Gate Member in the Dernburg Formation
The Dernburg Formation, which contains in its upper part the
glaciogenic unit of interest here, consists predominantly of thinly
laminated greenschist (metatuff), mafic hyaloclastites, metabasalt
and serpentinized metapicrite. Compositional layering on a
millimetre- to centimetre-scale reflects variations in magma
composition from mafic to intermediate. Acidic layers are rare.
The metabasalt displays pillow structures in strain-protected
domains. In places, mica-rich tuffitic to pelitic beds are inter-
calated. Locally, the mafic rocks show evidence of intense meta-
somatism, leading to epidosite, albitite, hornblendite and
ophicarbonate.
The Chameis Gate Member is distinguished from the remainder
of the formation by its diamictite lithology. No stratification is dis-
cernable. The matrix of the diamictite is similar to the laminated
greenschist (metatuff and -tuffite) and intercalated metapelite
above, but the clast composition is distinguishing. It contains
exotic lonestones and dropstones, up to 1.5 m in diameter. They
comprise various granitoids, gneisses, quartzite and dolomite,
and display a variable degree of rounding. In less intensely
deformed places, the dark grey pelitic or green chloritic beds are
deflected only on the lower side of a given larger clast, thus attest-
ing to its emplacement as a dropstone. Apart from the exotic clast
content, no other glacial indicators, such as glacial striations, have
been observed.
Associated with the metavolcanic and volcaniclastic rocks of the
Dernburg Formation are also distinct, in places Fe-rich, dolostone
outcrops, distinguished as Sholtzberg Member (Frimmel 2000b).
220 H. E. FRIMMEL
Their exact stratigraphic position is unclear because of tectonic
contacts, but they appear in the upper part of the formation,
close to the diamictite of the Chameis Gate Member. At one
locality (27.6078S, 15.0408E), relics of circular stromatolite
mounds are preserved. Elsewhere, the dolostone is associated
with magnesian metapelite that contains layers or lenses of tour-
malinite and albitite. Sodic metasomatism is widespread, resulting
in the crystallization of abundant magnesioriebeckite in all litho-
types in and around the Sholtzberg Member. An evaporitic
origin of this member is indicated by mineralogical and fluid
inclusion data as well as B isotopic evidence (Frimmel & Jiang
2001). For the lack of any continental influence in these deposits
(Rb, Zr, Nb, Th and U concentrations are an order of magnitude
lower than in normal marine carbonates) an atoll setting on an
oceanic seamount with temporary evaporation of seawater in the
central basin is envisaged for the Sholtzberg Member depositional
environment (Frimmel 2000b).
The Bogenfels Formation
The term ‘Bogenfels Formation’ was introduced by Martin (1965)
for those rocks that had previously been described as ‘Folded
Nama’ (Kaiser 1926). At that time, no distinction was made
between the Marmora Terrane and the Port Nolloth Zone. To
conform to the current tectonic nomenclature, it has been sug-
gested that the Bogenfels Formation be redefined as the unit of car-
bonate and siliciclastic rocks above the Dernburg Formation
within the Chameis Subterrane (Frimmel 2000b). It thus represents
the upper part of the Chameis Group. Its base is a regionally exten-
sive, laterally continuous, laminated, medium-grey limestone,
followed by a massive, thick-bedded, fine-grained, light creamy-
white to grey dolostone and finally a dolomitic breccia in which
the immediately underlying carbonates appear re-worked. This
carbonate succession of the Dreimaster Member is overlain by a
siliciclastic succession of alternating feldspathic and quartz
arenite, chlorite schist and calcpelite. The feldspathic arenite, a
greywacke to arkose, displays rare graded bedding of turbiditic
appearance.
In contrast to the Dernburg Formation, which is entirely differ-
ent to any unit in the Port Nolloth Zone, the Bogenfels Formation
is lithologically comparable to the Holgat Formation of the Port
Nolloth Zone (Fig. 16.3).
Boundary relations with overlying and underlying
non-glacial units
Upper and lower boundaries of the Chameis Gate Member are
either tectonic or covered by sand. The spatial distribution of the
member is confined to areas made up of metavolcanic and
-volcaniclastic rocks of the Dernburg Formation. The lower con-
tact of that formation is not exposed. The upper contact of the
Dernburg Formation with the Dreimaster Member (Bogenfels
Formation) is sharp and conformable, defined by the first appear-
ance of limestone. The stratigraphic relationship of the Chameis
Gate Member with the Sholtzberg Member within the Dernburg
Formation is unclear and only inferred because of a lack of
primary contacts.
Chemostratigraphy
Limited trace element concentrations as well as C, O and Sr isotope
data have been reported for carbonates of the Sholtzberg and Drei-
master Members in the stratigraphic vicinity of the glaciogenic
Chameis Gate Member (Frimmel 2000b; Frimmel & Jiang 2001;
Frimmel & Fölling 2004). 87Sr/86Sr ratios of 0.7075– 0.7078
were obtained on evaporitic dolostone of the Sholtzberg Member
with very low Rb –Sr ratios (,0.0005). The presence of perfectly
preserved dolomitic ooids in the succession and the lack of geo-
chemical evidence of diagenetic alteration suggest that the above
Sr isotope ratios are very close to primary. That dolostone has posi-
tive d13CCarb ratios (as much as 2.82‰ relative to the PDB stan-
dard) and d18OCarb as high as –4.3‰ (relative to PDB), whereas
the limestone of the younger Dreimaster Member is characterized
by consistently negative d13CCarb ratios slightly below zero, with
the lowest ratios ( –2.6‰) determined within the first few metres
above the bottom contact. A further distinctive feature of this lime-
stone is its high Sr content (as much as 1412 ppm).
Other characteristics
No economic ore deposits are known from the Chameis Subter-
rane, but base metal anomalies have been noted, particularly in
association with Na-metasomatism that affected most of the area
(H. E. Frimmel, unpublished data). The fluids involved were
highly oxidizing, which is evident from the widespread occurrence
of hematite veins. The metasedimentary rocks of the Marmora
Terrane still await a micropalaeontological investigation, which
might help in further constraining their age.
Palaeolatitude and palaeogeography
No reliable palaeomagnetic data are available for the entire
Gariep Belt and thus also for the Marmora Terrane. The low-grade
regional metamorphic overprint most likely led to the remagneti-
zation of the rocks during the Gariepian orogeny. Thus, any
assumption about palaeolatitude and palaeogeography is based
only on circumstantial evidence and comparison with continen-
tal-scale models (e.g. Collins & Pisarevsky 2005). The presence
of former evaporite deposits in the Marmora Terrane speak
against a high-latitude position at the time of sedimentation.
Geochronological constraints
The minimum age is provided by the age of metamorphism related
to the accretion of the Marmora Terrane, which has been con-
strained at 575 + 2 Ma by Ar –Ar data on a second, relatively
high-pressure metamorphic amphibole generation in metagabbro
of the Bakers Bay Suite (Frimmel & Frank 1998). Slightly older
Ar –Ar ages of 600–610 Ma, obtained in the same study on an
early, very low-pressure metamorphic amphibole generation
related to sea floor metamorphism, might set a minimum constraint
on the timing of oceanic crust formation, but are subject to a large
uncertainty. An attempt to date the dolostone of the Sholtzberg
Member by the double-spike Pb– Pb technique did not yield a
reliable result for the carbonate fraction. Its silicate fraction, con-
trolled by minute, presumably volcanogenic zircon grains, gave an
imprecise age of 598 + 60 Ma (Frimmel & Fölling 2004), which is
in agreement with the mentioned Ar –Ar data.
Discussion
The presence of exotic dropstones in marine, volcaniclastic rocks,
the massive, unbedded nature of the diamictite and lack of graded
bedding support a glaciomarine origin of the Chameis Gate
Member, but no definitive glaciogenic features have been
observed. Deposition distal from a glaciated continental margin
by processes of settling from suspension and ice rafting is
suggested. The Dreimaster Member carbonates are interpreted as
a cap-carbonate sequence based on the stratigraphic position and
C isotopic compositions equivalent to those of the post-Numees
Bloeddrif Member in the Port Nolloth Zone.
 THE CHAMEIS GATE MEMBER
The lack of precise radiometric age control for the Chameis
Gate Member precludes a reliable stratigraphic correlation with
glaciogenic deposits elsewhere. Although the available data base
is rather limited, tentative correlation with other glaciogenic
deposits in the region can be attempted on the basis of Sr and
C isotope data on over- and underlying carbonates and the
limited radiometric age data. The C isotopic evidence for the
Sholtzberg Member is ambiguous. The noted enrichment in 13C
therein could be merely a reflection of elevated evaporation
rates. However, its Sr isotopic composition is lower than in late
Ediacaran carbonates, but compares well with that reported region-
ally for the Pickelhaube Formation (Fölling & Frimmel 2002)
and globally for early to middle Ediacaran carbonates (Halverson
et al. 2007).
Fieldwork in the Sperrgebiet would have been impossible without the logistic
support and permissions to enter restricted areas granted by Namdeb Pty Ltd.
Funding by the South African National Research Foundation is gratefully
acknowledged. Comments from G. Halverson in particular helped in improving
the manuscript. This work represents a contribution of the IUGS- and
UNESCO-funded IGCP (International Geoscience Programme) Projects #512
and #478.
References
Basei, M. A. S., Frimmel, H. E., Nutman, A. P., Preciozzi, F. & Jacob,
J. 2005. A connection between the Neoproterozoic Dom Feliciano
(Brazil/Uruguay) and Gariep (Namibia/South Africa) orogenic
belts — evidence from a reconnaissance provenance study. Precam-
brian Research, 139, 195–221.
Collins, A. S. & Pisarevsky, S. A. 2005. Amalgamating eastern Gond-
wana: the evolution of the circum-Indian orogens. Earth-Science
Reviews, 71, 229–270.
Fölling, P. G. & Frimmel, H. E. 2002. Chemostratigraphic correlation of
carbonate successions in the Gariep and Saldania Belts, Namibia and
South Africa. Basin Research, 14, 69– 88.
221
Frimmel, H. E. 2000a. The Pan-African Gariep Belt in southwestern
Namibia and western South Africa. Communications of the Geologi-
cal Survey of Namibia, 12, 197–209.
Frimmel, H. E. 2000b. The stratigraphy of the Chameis Sub-terrane in the
Gariep Belt in southwestern Namibia. Communications of the Geo-
logical Survey of Namibia, 12, 179–186.
Frimmel, H. E. & Fölling, P. G. 2004. Late Vendian closure of the
Adamastor Ocean: timing of tectonic inversion and syn-orogenic
sedimentation in the Gariep Basin. Gondwana Research, 7,
685– 699.
Frimmel, H. E. & Frank, W. 1998. Neoproterozoic tectono-thermal evol-
ution of the Gariep Belt and its basement, Namibia/South Africa.
Precambrian Research, 90, 1 – 28.
Frimmel, H. E. & Hartnady, C. J. H. 1992. Blue amphiboles and
their significance for the metamorphic history of the Pan-African
Gariep belt, Namibia. Journal of Metamorphic Geology, 10,
651– 669.
Frimmel, H. E. & Jiang, S.-Y. 2001. Marine evaporites from an oceanic
island in the Neoproterozoic Adamastor ocean. Precambrian
Research, 105, 57– 71.
Frimmel, H. E., Hartnady, C. J. H. & Koller, F. 1996. Geochemistry
and tectonic setting of magmatic units in the Pan-African Gariep
Belt, Namibia. Chemical Geology, 130, 101–121.
Gresse, P. G. 1994. Strain partitioning in the southern Gariep Arc as
reflected by sheath folds and stretching lineations. South African
Journal of Geology, 97, 52– 61.
Halverson, G. P., Dudas, F. Ö., Maloof, A. C. & Bowring, S. A.
2007. Evolution of the 87Sr/86Sr composition of Neoproterozoic
seawater. Palaeogeography, Palaeoclimate, Palaeoecology, 256,
103– 129.
Hartnady, C. J. H. & von Veh, M. W. 1990. Tectonostratigraphic and
structural history of the Late Proterozoic –Early Palaeozoic Gariep
Belt, Cape Province, South Africa. Geological Society of South
Africa, Cape Town.
Kaiser, E. 1926. Die Diamanten Wüste. Dietrich Reimer, Berlin.
Martin, H. 1965. The Precambrian geology of South West Africa and
Namaqualand. Precambrian Research Unit, University of Cape
Town, Bulletin, 4, 1– 177.
 Chapter 17

The Kaigas and Numees formations, Port Nolloth Group, in South Africa and Namibia

HARTWIG E. FRIMMEL1,2
1
Department of Geological Sciences, University of Cape Town, Rondebosch 7701, South Africa
2
Present address: Geodynamics & Geomaterials Research Division, Institute of Geography and Geology, University of Wuerzburg,
Am Hubland, D-97074 Wuerzburg, Germany
(e-mail: hartwig.frimmel@uni-wuerzburg.de)

Abstract: The Port Nolloth Group makes up the eastern, external part of the Pan-African Gariep Belt (Port Nolloth Zone) in southern
Namibia and western South Africa. It contains two glaciogenic diamictite units, the older Kaigas Formation and the younger Numees
Formation, with intercalated and overlying carbonate-dominated units. Available chemostratigraphic information include O, C and Sr
isotope data. Micropalaeontological and geochronological data point to an early Cryogenian age (c. 750 Ma) of the Kaigas Formation
and possibly a middle Ediacaran age (c. 580 Ma) for the Numees Formation. The former was deposited in an evolving, but eventually
failed, continental rift on the western flank of the Kalahari Craton, probably at low latitude. The Numees Formation is a laterally con-
tinuous, up to 600-m-thick glaciomarine deposit for which a passive continental margin setting has been suggested. Alternatively, based
on more recent data, the depositional setting might have been a back-arc basin. The eroded remnants of the corresponding arc are present
in the Dom Feliciano Belt.

The Kaigas and Numees Formations represent the older and
younger glaciogenic diamictite-dominated units, respectively,
within the para-authochthonous Port Nolloth Zone of the Pan-
African Gariep Belt (Fig. 17.1). Owing to intense transpressional
overprinting of large parts of this zone, the distinction between
the two formations has been unclear for decades. This has led to
considerable confusion in the stratigraphic terminology used in
the older literature, with some workers even having argued for
the existence of only a single diamictite unit in the entire belt
(Hälbich & Alchin 1995; Jasper et al. 2000). Regional mapping
(von Veh 1993; Frimmel 2008), as well as the distinctly different
litho-, chemo-, chrono- and biostratigraphic characteristics in the
respective overlying carbonate-rich successions (Fölling et al.
2000; Fölling & Frimmel 2002; Frimmel & Fölling 2004;
Gaucher et al. 2005), however, have demonstrated the existence
of the two separate diamictite units at different stratigraphic pos-
itions. Here, the stratigraphic nomenclature as accepted by the
South African Commission on Stratigraphy (Frimmel 2003a) is
followed but with the understanding that from a sequence strati-
graphic perspective, it is in need of revision.
Rocks of both the Kaigas and Numees Formations are exposed at
several sites in southernmost Namibia and in the adjoining part of
South Africa (Fig. 17.2). The spatial distribution of outcrops
follows that of the outer zone of the Gariep Belt, which occurs in
an arcuate, south-trending, 20–60-km-wide zone stretching from
Lüderitz in Namibia, across the Orange River to the western Rich-
tersveld region of the Northern Cape Province in South Africa, and
striking out to sea south of Port Nolloth (Fig. 17.1). The Numees
Formation, including outcrops that are now assigned to the
Kaigas Formation, was the first Neoproterozoic unit in southern
Africa to have been recognized as being of glacial origin (Rogers
1916; Beetz 1926), an interpretation that has never been seriously
challenged by subsequent workers (Martin 1965a).
The original Kaigas Series of Rogers (1916) included most of
the strata now assigned to the Hilda Subgroup. The Kaigas For-
mation as redefined by De Villiers & Söhnge (1959) was limited
to beds along the Kaigas (also known as Gaigas) River, where
excellent, almost continuous exposures along a ridge to the east
of the river bed at 28.6318S, 17.0988E (Fig. 17.2) constitute the
holostratotype (Fig. 17.3). Diamictite in the type area for the
Numees Series of Rogers (1916) was later included by von Veh
(1993) in the Kaigas Formation because of lithological similarities.
Likewise, the diamictite exposures west and NW of Numees Peak
at the so-called Numees Prospect (28.29168S, 16.97018E) were
initially regarded as part of the Numees Formation (De Villiers
& Söhnge 1959) but since have been recognized to be older and
part of the Kaigas Formation (Kröner 1974; von Veh 1993).
Although the name Numees Formation is derived from a place
(Numees Prospect) that is now assigned to the Kaigas Formation,
the formation name is retained because the main occurrence of the
Numees Formation is only about 5 km to the west and the name is
firmly entrenched in the literature. The type area for the Numees
Formation is south of the Orange River at 28.18 –28.338S,
16.87 –16.938E.
Following the initial description of the South African portion of
the Port Nolloth Zone by Rogers (1916), the area was remapped in
greater detail by De Villiers & Söhnge (1959). Subsequent struc-
tural and lithostratigraphic studies by Kröner (1974) and von
Veh (1993) laid the foundation for the current lithostratigraphic
subdivision of the entire zone (Frimmel 2003b). The Namibian
sector was mapped in greater detail by McMillan (1968). Davies
& Coward (1982) provided a first interpretation of the Gariep
Belt’s structural evolution. This structural work was followed up
in the South African sector of the Port Nolloth Zone by Gresse
(1994) and in the Namibian sector by Hälbich & Alchin (1995)
and Jasper et al. (2000). An attempt to resolve the sedimentology
of parts of the stratigraphic succession was made by Jasper et al.
(1992 –1993). Metamorphism (Frimmel 1995), geochemistry and
tectonic setting of the various igneous bodies (Frimmel et al.
1996a), timing of tectonic events and the thermal evolution have
also been studied (Reid et al. 1991; Grotzinger et al. 1995;
Frimmel et al. 1996b; Frimmel & Frank 1998).
Attempts to place the rock record of the Port Nolloth Zone,
including the Kaigas and Numees Formations, into a regional
tectonic framework and to correlate units on a subcontinental
scale have been made by Frimmel et al. (2002). Our current
understanding of the tectonic settings for the deposition of the
glaciogenic deposits is based on work by Basei et al. (2005), and
the lithostratigraphic correlation has been updated by Gaucher
et al. (2005).
Structural framework and tectonic evolution
In the past, the entire Port Nolloth Group was interpreted as
having been deposited in a basin (Gariep Basin) that continuously

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 223– 231. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.17
224
Fig. 17.1. Position of the Port Nolloth Zone within the Gariep Belt (modified
after Hartnady & von Veh 1990).
evolved from a continental rift to a passive margin and eventually
became a foredeep prior to continental collision at c. 545 Ma
(Frimmel et al. 1996a; Frimmel & Frank 1998). An alternative
to the traditional Wilson Cycle model for the Gariep Belt has
emerged from a comparison with the tectono-thermal evolution
of the South American counterparts in southeastern Brazil and
eastern Uruguay (Basei et al. 2005; Frimmel & Basei 2006).
According to these more recent studies, the Gariep Basin might
have been a successor basin in which sediments had been deposited
in two different basins and were subsequently stacked on top of
each other.
The Kaigas Formation belongs to an older sequence that reflects
progressive continental rifting. Predominantly siliciclastic sedi-
ments of the Stinkfontein Subgroup were first deposited into a
widening rift graben, bounded by listric faults. The presence of
debris flow and turbidity fan deposits as well as large slump
masses indicates widespread syn-depositional deformation related
to continental rifting. Large, but only locally developed dolomitic
olistostromes near the northeastern basin margin are explained by
seismic activity along active, basin-bounding growth faults. These
normal faults were subsequently inverted during the Gariepian
orogeny, but the position of some of the major syn-rift faults is
still recognizable. A major growth fault must have existed along
today’s eastern to northeastern margin of the belt, separating base-
ment rocks to the east and NE from the rift graben fill. This fault is
now present as an inverted NW –SE trending thrust fault system,
with the Rosh Pinah Fault being a good example. Along those
faults, an elongated, roughly north –south trending basement
horst (Aurus Horst) developed, separating the so-called Rosh
H. W. FRIMMEL
Fig. 17.2. Map showing the distribution of the Kaigas and Numees Formations
within the Port Nolloth Zone (made up of rocks of the Port Nolloth Group),
Gariep Belt, in southern Namibia and westernmost South Africa (western
Richtersveld). Also shown are the localities of the type sections for the Kaigas
and Numees Formations.
Pinah Graben in the east, with a number of smaller, separate
basins, from a half-graben to the west. At an advanced rifting
stage, eustatic sea-level drop was accompanied by the entry of gla-
ciers into the basin along its eastern margin to the mainland. Debris
released from those glaciers makes up the bulk of the Kaigas
Formation. This stage was accompanied by bimodal, predomi-
nantly felsic, rift volcanism (Rosh Pinah Formation) on a growth
fault along the eastern basin margin. Although still in an overall
extensional regime, the eustatic drop in sea level led to the star-
vation of the Rosh Pinah Basin and consequently to anoxic
bottom waters, which provided a suitable environment for sedi-
mentary – exhalative base-metal mineralization (Rosh Pinah
mine and protore of the secondary Skorpion deposits). Renewed
transgression led to the drowning of parts of the basement horst
 KAIGAS AND NUMEES FORMATIONS 225

the emergence of the Dom Feliciano volcanic arc towards the

(north)west. Consequently, the upper Hilda Subgroup and the
overlying Numees Formation would have been deposited in a
back-arc basin behind the Dom Feliciano arc.
Inversion from extension to compression took place during
post-Numees sedimentation. A foredeep developed in which the
youngest sediments of the Port Nolloth Group (Holgat Formation)
were deposited (Frimmel & Fölling 2004). At the same time, first
stacking and accretion of the advancing oceanic rock pile in the
shrinking basin took place. This was followed by continental
collision in a SE-directed transpressional regime, which was
accompanied by regional low-grade metamorphism, east-vergent
folding, thrusting and back-thrusting. Many of the original
normal faults were reactivated at that stage (von Veh 1993;
Hälbich & Alchin 1995). In places, syn-tectonic foliation occurred
at a high angle to bedding, but primary sedimentary contacts
remained well preserved in strain-protected domains.

Stratigraphy

Both the Kaigas and Numees Formations form part of the Port
Nolloth Group (Fig. 17.4), which, in turn, is the part of the
Gariep Supergroup that makes up the bulk of the Port Nolloth
Zone. The Port Nolloth Group unconformably overlies rocks of
the Mesoproterozoic Namaqua Metamorphic Province and, in
places, alkaline intrusive rocks of the 833– 771 Ma Richtersveld
Suite (Frimmel et al. 2001). It is, in turn, unconformably overlain
by the Kuibis Subgroup of the Nama Group in the east, and towards
the west it is overlain tectonically by the Marmora Terrane, the
allochthonous internal sector of the Gariep Belt.
The stratigraphy of the Port Nolloth Group begins with a pre-
dominantly siliciclastic, minor volcanic to volcaniclastic continen-
tal rift graben fill, the Stinkfontein Subgroup. This is followed by
the stand-alone Kaigas Formation, which, in turn, is overlain by the
Hilda Subgroup. Within the Hilda Subgroup, four formations are
distinguished, some of which are time-equivalents of each other.

Fig. 17.3. Holostratotype of the Kaigas Formation (for location see Fig. 17.2;
modified from Frimmel et al. 2003a).

as well as the Kaigas Formation and the deposition of a varied,

carbonate-dominated succession (Pickelhaube Formation).
One of the great enigmas of the Gariepian stratigraphy has been
the lack of sediment in the post-rift successions. If a successful rift
is assumed for the tectonic evolution of the Gariep Basin (Adamas-
tor Ocean), as postulated by most workers in the past (Hartnady
et al. 1985; Germs 1995; Frimmel et al. 1996a), the post-Kaigas
period should have been marked by further transgression and the
development of a thick sedimentary passive margin sequence.
First regional drowning of the rift shoulders is indeed indicated
by the laterally continuous post-Kaigas carbonates, and the
various carbonate-dominated facies of the Hilda Subgroup
provide good evidence of the development of carbonate platforms.
Yet the total thickness of the Hilda Subgroup (except for the vol-
canically controlled Rosh Pinah area) is not more than 650 m.
This lack of sediment is explained by a major hiatus and/or
slope failure as the Hilda Subgroup is largely composed of turbi-
dites, olistostromes and potentially cannibalizing debris flows.
Fieldwork by the author has shown that the erosive surface
beneath the siliciclastic Wallekraal Formation within the Hilda
Subgroup extends across the entire Port Nolloth Zone and is a
far more widespread feature than previously recognized. It is a
first-order sequence boundary.
The temporal extent of the pre-Wallekraal hiatus remains un-
resolved. If the above back-arc model applies, the original basin- Fig. 17.4. Generalized lithostratigraphy of the Port Nolloth Group (Gariep
bounding faults would have become re-activated in response to Supergroup); unit thickness in metres given in parentheses.
226 H. W. FRIMMEL
In the volcanically influenced basin margin, the Kaigas Formation
diamictite is followed by the volcanic to volcaniclastic Rosh Pinah
Formation, whereas elsewhere it is followed by the carbonate-
dominated Pickelhaube Formation. The former comprises rhyolitic
lava flows, subaerial and submarine pyroclastic deposits of highly
variable grain size (from agglomerate to ash tuff and ignimbrite),
whereas the latter consists predominantly of carbonate rocks
(laterally discontinuous, massive dolomitic marble and laminated
to medium-bedded siliceous and graphite-bearing dolomitic,
partly allodapic limestone, with slump breccias) and subordinate
feldspathic arenite and graphitic metapelite alternating locally
with calcpelite. These are overlain paraconformably, or with an
erosional unconformity, by coarse-grained clastic sedimentary
rocks of the Wallekraal Formation in which parts of the older Pick-
elhaube carbonate platform appear re-worked. The Wallekraal
Formation consists predominantly of very coarse-grained felds-
pathic arenite and less abundant well-rounded polymict conglom-
erate, phyllitic to calcareous, locally graphitic, metapelite, isolated
lenses of dolostone, typically re-worked into olistostromes, and
thin-bedded allodapic limestone. The youngest unit of the Hilda
Subgroup is the Dabie River Formation, which consists essentially
of stromatolitic reef facies, including dolomite breccia and oolite.
The Hilda Subgroup is followed by diamicite and minor banded
iron-formation of the Numees Formation, which, in turn, is over-
lain by the Holgat Formation. The latter commences with a carbon-
ate sequence above the Numees Formation diamictite (Bloeddrif
Member), and in places a thick package of turbiditic, siliciclastic
metasedimentary rocks.
Because glaciogenic deposits are the emphasis of this paper, the
stratigraphic description of the two glaciogenic units and those that
are associated with them follow in more detail below.
Glaciogenic deposits and associated strata
Descriptions of the two principal glaciogenic units in the Port
Nolloth Group, the Kaigas and Numees Formations, have been
provided by von Veh (1993), Frimmel (2000, 2008) and
Frimmel et al. 2003a, b). To also obtain some information on
the rock record that provides possible information on the events
that led up to the glacial intervals and those that followed, the
bounding stratigraphic units below and above the glaciogenic
units will be also described briefly.
The Vredefontein Formation
Strata underlying the Kaigas Formation are known as the Vrede-
fontein Formation (Stinkfontein Subgroup). The Vredefontein
Formation is up to 300 m thick. Internal sedimentary structures,
accentuated by heavy mineral concentrations along laminations,
are well preserved in the dominantly medium-bedded feldspathic
arenite. These include various forms of cross-bedding, asymmetri-
cal ripple marks indicating palaeocurrent directions from the east
or SE, as well as various soft sediment deformation features
(von Veh 1993).
In the upper part of the formation, minor felsic, intermediate and
mafic metavolcanic rocks are intercalated (Middlemost 1963).
These rocks are only known from the Richtersveld in South
Africa, but they provide an important reference for the genesis
of more abundant volcanic rocks at a higher stratigraphic position
in Namibia (Rosh Pinah Formation). Sandy to gritty breccia that
occurs towards the top of the Vredefontein Formation contains
fragments of angular to sub-rounded vein quartz, granitoid, micro-
granite, gneiss, orthoquartzite, calcareous grit and felsic volcanic
rocks in a massive or indistinctly cross-bedded feldspathic
grit matrix.
The distribution of the Stinkfontein Subgroup in Namibia
remains enigmatic, because of the occurrence of similar
siliciclastic successions higher up in the stratigraphy, lateral
facies changes that make long-distance correlation with the type
sections in South Africa problematic, and the commonly tectonic
nature of its upper and lower contacts. Occurrences of the sub-
group are confined to the northeastern basin margin. Lithologi-
cally, the Stinkfontein Subgroup in Namibia is similar to the
Vredefontein Formation in the Richtersveld, whereas the relatively
clean quartz arenite, typical of the Lekkersing Formation in the
lower part of the Stinkfontein Subgroup, is only poorly developed
in Namibia.
The overall higher feldspar content in the siliciclastic rift depos-
its in Namibia might reflect an overall more proximal position of
the depositional environments relative to the area further south
in South Africa. This finds support from the observation that
along the NW –SE trending branch of the basin north of the
Aurus Horst, the entire Stinkfontein Formation is missing and
rocks of the next younger stratigraphic units, the Kaigas For-
mation, where present, or the Hilda Subgroup rest directly on
the basement.
The Kaigas Formation
The Kaigas Formation is only locally developed along the eastern
and northeastern margin of the basin. It is characterized by its
marked variations in thickness along strike, with individual occur-
rences pinching out over only several hundred metres. At its type
locality (Fig. 17.2) in the southern Richtersveld (South Africa), it
reaches a thickness of 115 m. The formation consists predomi-
nantly of medium- to thick-bedded diamictite and subordinate
massive, locally cross-bedded or graded bedded, feldspathic
arenite and argillite. Ripple cross laminations, indicating palaeo-
current directions from the east, rip-up of mudstone clasts, load
and flute casts, and sinuous ripple marks are locally present. The
diamictite beds display sharp upper and lower contacts and lack
internal structures except for crudely graded bedding. The clast
size distribution is strongly bimodal with generally larger clasts
present in thicker diamictite beds. In places, outsized erratic
blocks can also be found in pelitic beds. An overall decrease in
clast size away from the former basin margin is noted. Individual
clasts and blocks commonly appear sandwiched (cold deformation
by ice). Faceted clasts can be found in less deformed regions.
Unequivocal glacial striations are difficult to identify because of
the tectonic overprint under greenschist-facies temperatures,
which modified most clast surfaces.
Locally, dolostone is a common clast lithology in the southern
part of the Port Nolloth Zone, but elsewhere the Kaigas Formation
diamictite is notably free of carbonate clasts. The clast population
is generally dominated by basement-derived rock types, such as
granite, gneisses, amphibolite, schists and re-worked arenite
from the underlying Stinkfontein Subgroup. The general lack of
carbonate clasts (with the exception of the southern Port Nolloth
Zone) is a major feature by which the Kaigas Formation differs
from the younger Numees Formation. Further features that make
possible a distinction of the Kaigas from the Numees Formation
are the absence of a near-basal banded iron-formation or ferrugi-
nous arenite zone and the relative abundance of interbedded
arenite and pelite.
In Namibia, the formation reaches its maximum thickness in an
area east of the Aurus Mountains around Borehole 5 at 16.5278E,
27.6798S (Fig. 17.2), where it rests directly on basement. From
there it pinches out rapidly towards the SE, but re-emerges in
several small occurrences along the basement edge to the north
and SE of Rosh Pinah (Alchin et al. 2005; Frimmel 2008). On
the farm Spitzkop 111, the Kaigas Formation diamictite grades lat-
erally into a coarse-grained, proximal pyroclastic deposit (at
16.7068E, 27.8278S) that is assigned to the Rosh Pinah Formation,
indicating sedimentation that overlapped in time with volcanism.
Elsewhere, the background sedimentation towards the top of the
 KAIGAS AND NUMEES FORMATIONS
Kaigas Formation gradually changes from siliciclastic to calcar-
eous. An example of that can be found east of the Aurus Mountain
at 16.4868E, 27.6518S, where the matrix of the diamictite and
intercalated arenite becomes progressively dolomitic towards the
top of the formation, thus heralding a fundamental change in the
style of sedimentation that led to the next stratigraphic unit.
The Pickelhaube and Rosh Pinah Formations
Where the Kaigas Formation is developed, the Pickelhaube For-
mation rests on it conformably. Elsewhere, the Pickelhaube For-
mation occurs with an erosional unconformity on basement. It
starts with laminated, variably dolomitized, medium to dark grey
limestone, followed by a varied sequence of predominantly argil-
lite and marl with minor feldspathic quartz arenite in which
small-scale channels and asymmetric ripple marks indicate a
palaeocurrent direction from the SE. Intercalated are a number
of limestone and dolostone beds, which become more prominent
towards the top of the formation. The absence of conglomerate,
scarcity of cross-bedding, presence of thin bedding, and the
increasing carbonate content of the Pickelhaube Formation are-
nites indicate quieter, submerged conditions compared to the
underlying Stinkfontein Subgroup and Kaigas Formation, as can
be expected in the distal parts of a fan complex. Some of the
clastic beds may represent turbidites and minor debris flow
deposits. The original limestone deposits are largely dolomitized
in the more proximal positions but retained their syn-sedimentary
mineralogy farther away from the basin margin. Laminated alloda-
pic limestone (calcarenite) is a ubiquitous feature of this formation,
which reaches a maximum thickness of about 280 m.
Where the formation onlaps directly onto basement, basement-
derived clasts are embedded in allodapic limestone of the lower
Pickelhaube Formation. The carbonate beds in the middle of the
formation show extensive syn-sedimentary brecciation that is
reminiscent of tempestite. This is in contrast to slump breccias
and carbonate debris flow deposits, some of them containing extra-
basinal clasts, which are common in the thicker carbonate beds.
The Rosh Pinah Formation (Kapok Formation of Martin 1965b)
is distinguished from the Pickelhaube Formation by the presence
of felsic volcanic and volcaniclastic rocks in the former. The great-
est accumulation of felsic volcanic rocks, also referred to as Spitz-
kop Formation by some authors (Siegfried & Moore 1990), occurs
some 15 –20 km north of Rosh Pinah (on farm Spitzkop 111) and is
considered to represent the main volcanic centre. The Rosh Pinah
Formation displays repetitive sedimentary cycles that reflect quies-
cent periods intervening with rapid deposition related to reactiva-
tion on basin bounding faults with subsequent thermal and
mechanical subsidence (Alchin et al. 2005). Although deposition
of the volcanosedimentary succession was sustained in the more
active, proximal parts of the basin, distal background sedimen-
tation ranged from deposition of planar laminated allodapic lime-
stone to mudstone, followed by platform carbonates (Pickelhaube
Formation). In proximal positions, tilting of the rift shoulders led to
partial erosion of the Rosh Pinah Formation and thus to intraforma-
tional breccias and olistostromes.
Volcanic lithotypes in the Rosh Pinah Formation comprise
massive to flow banded quartz-alkali feldspar rhyolite to rhyoda-
cite, hyaloclastic breccias, often locally reworked (lapilli tuff brec-
cias), and a variety of volcaniclastic units, which reflect different
energy and distance from the eruptive centre. They range from
proximal agglomerate, to lapillistone, and distal coarse- and fine-
grained tuffs. Inversely graded lapillistone/tuff units are present
and indicate water-lain pumice ash flow deposits. A special
variety is an alkali feldspar-rich crystal tuff to tuffite. A bimodal
character of the Rosh Pinah volcanism is indicated by the local
presence of metabasalt and metagabbro and the lack of igneous
rocks of intermediate composition (Frimmel et al. 1996a).
Sedimentary rocks directly associated with the volcanic facies
227
comprise intercalated ripple-marked quartzite and dolostone,
which is, in places, highly ferruginous due to fumarolic activity
(Frimmel & Lane 2005).
The Numees Formation
In places where the stromatolitic reef facies of the Dabie River
Formation is not developed, the siliciclastic Wallekraal Forma-
tion grades conformably into the Numees Formation. The strato-
type section of the formation as accepted by the South African
Committee for Stratigraphy (Frimmel et al. 2003b) is located in
the Richtersveld region in westernmost South Africa (Fig. 17.5).
That area suffers, however, from considerable tectonic compli-
cation and less tectonized sections of the formation occur in the
Namibian part of the belt. The total thickness of the Numees For-
mation is estimated to be not more than 600 m, but decreases
towards the east where the formation onlaps directly onto pre-
Gariep basement. The dominant lithology is a generally massive,
very thick bedded (bedding poorly defined) diamictite with small
to very large (2 mm –10 m) subrounded clasts. The lithology of
the clasts comprises pegmatite, leucogranite, granitic-gneiss,
quartzite, schist, dolostone, minor limestone of the immediately
underlying Dabie River Formation and re-worked diamictite.
Carbonate clasts are particularly abundant in the lower parts of
the formation but become rare in the upper parts, where basement-
derived clasts dominate. Clast and matrix grain sizes decrease
upwards. Intercalated minor ferrugineous, feldspathic arenite is
coarse-grained, medium-bedded and forms the matrix of those
parts of diamictite that are dark grey to blue-green in colour. The
latter becomes dark brown on weathered surfaces. In addition,
thin-bedded, varve-like siltstone beds with intercalated coarser-
grained, graded sandstone beds occur. Some of these beds
contain dropstones with soft-sediment deformation around their
bottom contact.
A highly ferruginous unit, consisting of thinly laminated chlorite
schist, iron-formation and diamictite, is distinguished as the
Jakkalsberg Member near the base of the formation (Frimmel
et al. 2003b). This member is best developed in the southern
portion of the belt in South Africa (Fig. 17.5), but is largely
missing in Namibia (except for a few outcrops just north of the
Orange River). In places, exotic, mainly basement-derived drop-
stones are found within the iron-formation.
The Bloeddrif Member (Holgat Formation)
On top of the Numees Formation follows a distinct carbonate
unit that is distinguished as the Bloeddrif Member within the other-
wise siliciclastic Holgat Formation. The true thickness of the for-
mation is difficult to constrain because of intense folding and thrust
duplication. It appears, however, that stratigraphic onlap led to a
decrease in thickness from several hundred metres in the west to
less than 100 m in the east. The formation is best developed in
the western Richtersveld in South Africa, where the type sections
are located.
In less deformed regions near the basin margin, such as some
17 km to the NE of Rosh Pinah and on farm Nord Witputs 22 as
well as on farm Namuskluft east of Rosh Pinah, internal, vertical
tube-like structures of infilled micritic sediment and cement
form a conspicuous feature of the Bloeddrif Member. They
resemble structures that have been described from many other
post-glacial carbonates (Cloud 1974), and have been variably
interpreted as microbial in origin (Hegenberger 1993; Hoffman
et al. 1998; Corsetti & Grotzinger 2005) or by gas escape following
the destabilization of gas hydrate during warming of terrestrial
permafrost (Kennedy 2001). Some microbial activity in the Bloed-
drif Member is clearly indicated by the local presence of stromat-
olites. More distal, relatively pure limestone deposits that are
228 H. W. FRIMMEL
characteristically devoid of continental detritus can be distin-
guished from more proximal deposits that experienced more
intense, early diagenetic dolomitization and have intercalations
of thin arenite beds (Frimmel & Fölling 2004). A particularly
good example of a very shallow, proximal facies of this member
occurs to the north of Aurus Mountain at 16.2168E, 27.4538S,
where a thinly laminated dark grey to cream-coloured limestone
with numerous erosion surfaces and mud cracks grades upwards
into a chaotic dolomitic breccia that is unconformably overlain
by white, locally cross-bedded dolomitic arenite that became dis-
rupted and filled with sand dykes (Frimmel 2008). In most
places, the laminated limestone is overlain by a massive, dark
grey dolostone that is followed by a thickly bedded, distinctly
pink dolostone. It should be noted, however, that the distal sections
through the Holgat Formation appear continuous without breaks in
sedimentation.
Boundary relations with overlying and underlying
non-glacial units
The boundary relations of the Kaigas Formation with the under-
lying strata are marked by the laterally discontinuous distribution
of the formation. At the type locality in the southern part of the Port
Nolloth Zone, the contact of the Kaigas Formation with the under-
lying feldspathic arenite of the Vredefontein Formation is con-
formable and gradational, and is defined by the first appearance
of diamictite. Further north, the contact with the underlying Stink-
fontein Subgroup is, in many places, an angular discordance, or the
Kaigas Formation rests directly on the pre-Gariep basement, but a
conformable contact exists NE of Rosh Pinah (Fig. 17.4). On the
farm Spitzkop 111 north of Rosh Pinah, the Kaigas Formation dia-
mictite grades laterally into a coarse-grained, proximal pyroclastic
deposit of the Rosh Pinah Formation, indicating sedimentation that
overlapped in time with volcanism. Elsewhere, the background
sedimentation towards the top of the Kaigas Formation gradually
changes from siliciclastic to calcareous; the matrix of the diamic-
tite and intercalated arenite becomes progressively dolomitic
Fig. 17.5. Location of the holostratotypes
of the Numees Formation (a) and the
Jakkalsberg Member (b). The type section
of the Numees Formation is 380 m thick,
rests in fault contact on feldspathic arenite
of the Wallekraal Formation, and begins
with 30 m of thickly bedded diamictite.
Banded iron-formation, thinly laminated
Fe-rich metapelite and quartzite of the
Jakkalsberg Member, displaying graded
bedding, follow conformably and attain a
thickness of 5 m. They are overlain by 95 m
of Fe-rich, green, thickly bedded diamictite
and feldspathic arenite intercalations,
locally displaying horizontal lamination.
The diamictite is rich in dolomite clasts.
This is followed by 250 m of thickly
bedded, yellow-brown diamictite with only
few dolomite clasts but predominantly
basement-derived granite and gneiss clasts.
The upper contact of the formation is
tectonic, with a repetition of the
Jakkalsberg Member (from Frimmel et al.
2003b).
towards the top of the formation. The upper contact with the
Pickelhaube Formation is in many places conformable, but
locally (mostly in South Africa) unconformable.
The lower contact of the Numees Formation to the underlying
Hilda Subgroup is locally conformable (Kröner 1974), although
in most places it is tectonic. Wherever the Dabie River Formation
carbonates are not developed, the siliciclastic Wallekraal
Formation grades conformably into the Numees Formation. The
upper boundary of the Numees Formation with the carbonates of
the Bloeddrif Member is typically sharp, conformable or paracon-
formable. The stratigraphic relationships of the Jakkalsberg
Member within the Numees Formation with the bounding units
are quite clear from a regional perspective. The iron-formations
are always, without a single exception, near the bottom of the
Numees Formation diamictite, and with conformable contacts,
either sharp or gradational. To take those out and place them
into a different stratigraphic position altogether is not supported
by the field evidence.
Chemostratigraphy
The first stable isotopic compositional data for carbonates from the
Port Nolloth Zone were reported by Kaufman et al. (1991), but
those data suffered from poor stratigraphic control. Since then, a
number of chemostratigraphic profiles with isotope ratios carefully
screened for the effects of post-depositional alteration, through
various sections of carbonate-dominated successions in the Port
Nolloth Zone (Fölling & Frimmel 2002), have helped in dis-
tinguishing between post-Kaigas and post-Numees carbonates
because of distinctly different isotopic trends. The former are
characterized by an increase in d13CCarb from negative ratios
(,4‰ relative to PDB standard) above the Kaigas Formation dia-
mictite to values as high as þ8.7‰ further up-section. This trend is
accompanied by a decrease in 87Sr/86SrCarb from 0.7076 to 0.7071.
The Sr isotope ratios are considered near primary because of
high Sr (1828 –2916 ppm) and generally low Rb concentrations
(,13 ppm, Rb/Sr 0.007, Fölling & Frimmel 2002). In contrast,
 KAIGAS AND NUMEES FORMATIONS
the post-Numees carbonates of the Bloeddrif Member have signifi-
cantly higher, but consistent 87Sr/86SrCarb ratios of 0.7085 in the
limestone immediately above the diamictite and 0.7082 in the
remainder of the member. Considering Rb/Sr ratios of much
less than 0.001 (Sr and Rb concentrations of 1078 –2483 and
,5.3 ppm, respectively), the measured 87Sr/86SrCarb ratios are
again regarded as effectively primary (Fölling & Frimmel 2002).
The bottom of the Bloeddrif Member limestone is markedly
depleted in 13C (d13CCarb ¼ 24.6‰), whereas the rest has
d13CCarb consistently clustering within the narrow range between
21.0 and þ1.0‰. In a follow-up study (Frimmel & Fölling
2004) it could be shown, however, that the increase in d13CCarb
is strongly dependent on palaeoenvironment, with much higher
isotope ratios (þ4.0‰) achieved in the most proximal positions
for the same stratigraphic interval. Alternatively, deposition of
the cap carbonate could be diachronous from the slope up to the
shelf (e.g. Hoffman et al. 2007).
Although O isotopes are more susceptible to diagenetic altera-
tion, a distinct negative d18OCarb anomaly (decrease by as much
as 4‰) has been noted in the first 25 m of the post-Numees cap
carbonates in various outcrops and has been interpreted as reflect-
ing freshwater mixing with seawater during large-scale melting
of glaciers, in agreement with the slightly higher 87Sr/86SrCarb at
the same position (Fölling & Frimmel 2002).
By and large, the post-Numees carbonates are more enriched in
Sr than the post-Kaigas carbonates. Strontium concentrations of
several thousand ppm have been reported for the distal portions
(characterized by a lack of Rb, Y and Zr) of the Bloeddrif
Member limestone, whereas the near-shore equivalents contain
an order of magnitude less Sr (Frimmel & Fölling 2004).
Other characteristics
The Rosh Pinah Formation is economically significant for its
base-metal enrichment. Stratabound and largely stratiform sulphi-
dic Pb and Zn ore bodies are being mined at Rosh Pinah (past
production and resources are c. 50 Mt at an average grade of
7 wt% Zn, 2 wt% Pb, 0.1 wt% Cu and 11 ppm Ag). The ore is
hosted by a variably silicified carbonaceous mudstone that
reaches as much as 50 m in thickness, with intercalated carbonate
lenses, within an overall arkosic succession of the middle Rosh
Pinah Formation (Alchin et al. 2005). A protore in a similar strati-
graphic position is regarded as the source for the secondary oxi-
dized Zn deposit at Skorpion (25 Mt, average grade of 10.6 wt%
Zn; Borg et al. 2003), located c. 22 km NW of Rosh Pinah.
In spite of a low-grade metamorphic overprint, it has been
possible to recover a number of organic-walled microfossils that
help in setting some biostratigraphic constraints for units above
and below the Numees Formation (Gaucher et al. 2005). Poorly
preserved, highly carbonized acritarchs characterize units
beneath the Numees Formation. Of biostratigraphic significance
is the occurrence of Bavlinella faveolata in the Pickelhaube and
upper Wallekraal Formations. Acritarchs in the Bloeddrif Member,
Holgat Formation, are characterized by a low-diversity assemblage
consisting of Soldadophycus, Myxococcoides, Coniunctiophycus
and Leiosphaeridia, with a dominance of Soldadophycus bossii,
and the absence of acanthomorphs and large sphaeromorphs.
Associated with this acritarch assemblage is the agglutinated
foraminifer Titanotheca sp.
Palaeolatitude and palaeogeography
No reliable palaeomagnetic data are available for the Gariep Belt
or for the location of the Kalahari Craton in the Neoproterozoic
in general. Today, the Gariep Belt is positioned between the Kala-
hari and the Rio de la Plata Cratons. Both Pb and Nd isotopic evi-
dence (Frimmel et al. 2004; Frimmel & Basei 2006) point to a
229
provenance of the entire Port Nolloth Group on the Kalahari
Craton. This is supported by detrital zircon age spectra that
conform to the ages known from the immediate basement on the
Kalahari Craton but differ from those typical for the Rio de la
Plata Craton (Basei et al. 2005). Both the Kaigas and the
Numees Formation diamictites would have been deposited in a
marginal cratonic position for which, in most palaeogeographic
reconstructions, a low palaeolatitude is assumed, though on cir-
cumstantial evidence, throughout the Neoproterozoic time span
of interest here (Collins & Pisarevsky 2005).
Geochronological constraints
A maximum age constraint for the onset of sedimentation in the
Gariep Basin is given by the youngest age obtained on pre-Gariep
basement, that is, 771 + 6 Ma for the Lekkersing Granite
(Frimmel et al. 2001). The Kaigas Formation must be younger
than that age but cannot be younger than the volcanism evident
in the Rosh Pinah Formation. Felsic volcanic rocks above the
Kaigas Formation yielded Pb–Pb and U – Pb single zircon ages
of 741 + 6 Ma and 752 + 6 Ma, respectively (Frimmel et al.
1996b; Borg et al. 2003). This is supported by a double-spike
Pb– Pb carbonate age of 728 + 32 Ma obtained for the overlying
cap carbonates of the Pickelhaube Formation (Fölling et al.
2000), which has been interpreted as dating the time of diagenesis.
More problematic is the absolute age of the Numees Formation,
as no radiometric data are available for this unit. Minimum age
constraints are given by a double-spike Pb–Pb carbonate age of
555 + 28 Ma for the overlying limestone of the Bloeddrif
Member of the Holgat Formation (Fölling et al. 2000) and the
age of regional, syn-collisional metamorphism that is constrained
at 545 + 2 Ma by Ar –Ar data on syn-tectonic hornblende and
micas (Frimmel & Frank 1998).
Discussion
The sedimentary features of the Kaigas Formation diamictite, such
as sharp upper and lower bedding contacts, lack of internal struc-
tures (except for crudely graded bedding), correlation between
clast size and bedding thickness, rapid decrease in clast size
towards the basin centre, and the limited geographical distribution
suggest rapid deposition as debris flows. Transport by turbidity
currents in a subaqueous fan environment is indicated by com-
mon upwards fining beds in the intercalated feldspathic arenite.
The presence of westward-directed gravity flow deposits stepping
back onto the basement foreland suggests a phase of marine trans-
gression along an active fault scarp. This is also indicated by local
dolostone clasts in the diamictite, which implies the presence of
shallow-water, pre-Kaigas carbonate platform deposits, at least
in the south.
A component of melt-out glacial debris in the Kaigas Formation
diamictite, such as dropstones, bimodal clast distribution, sand-
wiched and faceted clasts, and outsized erratic blocks in shaly
beds, suggest a glaciomarine or fluvioglacial origin of parts of
the Kaigas Formation. Uplifted horsts along the margin of the
basin may have been the sites of mountain glaciers that contributed
some glacial material.
Although there is little doubt of a glacial influence during
Kaigas times, most of the Kaigas Formation reflects sediment
gravity flows in consequence of subaqueous slumping, probably
triggered by tectonic activity along the growing rift shoulders.
In contrast, the Numees Formation represents a truly glacial
deposit. Deposition adjacent to a glaciated continental margin by
processes of settling from suspension and ice rafting is indicated
by the extensive lateral distribution of the diamictite, its textural
homogeneity and largely unbedded nature, and the presence of
out-sized extrabasinal lonestones. The associated sediments, in
230 H. W. FRIMMEL
particular the banded iron-formation of the Jakkalsberg Member,
record marine rather than glaciofluvial depositional conditions.
Huge blocks, more than 10 m in length, of completely different
lithology (dolomite, gneiss) occurring next to each other within
the same diamictite bed (commonly found near the Orange
River) can only be explained reasonably by melt-out from drifting
icebergs. The Bloeddrif Member carbonates are interpreted as a
typical cap carbonate sequence.
Although the Kaigas and Numees Formations have been tenta-
tively correlated with the global Sturtian and Marinoan glaciations
(Frimmel et al. 2002), this interpretation has been questioned. The
age of the Kaigas Formation is c. 750 Ma, older than the Sturtian
glaciation and, purely based on evidence from the Gariep Belt,
there is insufficient evidence from the Gariep Belt to postulate a
global glacial event for the Kaigas Formation. On a regional
scale, the Kaigas glaciogenic deposits may be correlated with the
Chuos Formation of northern Namibia, based on similar radio-
metric ages for associated felsic volcanic rocks there (Hoffman
et al. 1996; Hoffman & Prave 1996). As evidence for glacial
conditions at that time exists also from other continents (Vieira
et al. 2007; Zheng et al. 2007), it may well represent a global
glacial event.
The lack of precise radiometric age control for the Numees
Formation precludes a reliable stratigraphic correlation with gla-
ciogenic deposits elsewhere. A precise age of 636 + 1 Ma has
been obtained on an ash bed in metamorphosed glaciomarine dia-
mictite from the Central Zone of the Damara Belt (Hoffmann et al.
2004). These authors consider the dated diamictite unit as equival-
ent to the Ghaub Formation in the northern platform and conse-
quently postulate a correlation with the Marinoan glaciation. A
correlation of the Numees Formation with the Ghaub Formation
appears intuitively probable (as suggested by Frimmel et al.
2002), but is not supported by more recent biostratigraphic evi-
dence (Gaucher et al. 2005). The microfossil assemblage in the
post-Numees rocks is markedly different from those in the pre-
Numees strata, and the former compares well with Ediacaran
assemblages elsewhere. An Ediacaran age of the Holgat Formation
is also supported by the Pb–Pb age of 555 + 28 Ma obtained for
the Bloeddrif cap carbonate (Fölling et al. 2000) and the compara-
tively high initial 87Sr/86SrCarb (Fölling & Frimmel 2002). A cor-
relation of the Numees Formation with the c. 580 Ma Gaskiers
glaciation is therefore preferred. Thus, if the successor basin
model is accepted for the Gariep Basin, the apparent absence of
any syn-Marinoan glacial deposits in the Port Nolloth Zone may
be explained by the hiatus between the continental rift stage
(770– 740 Ma) and the back-arc basin stage (c. 610– 580 Ma) in
the overall basin evolution.
The South African National Research Foundation is thanked for funding the
research on the Gariep Belt for many years. Reviews by F. MacDonald, R.
McMiller and G. Halverson significantly improved the original manuscript.
This work represents a contribution of the IUGS- and UNESCO-funded IGCP
(International Geoscience Programme) Projects #512 and #478.
References
Alchin, D. J., Frimmel, H. E. & Jacobs, L. E. 2005. Stratigraphic setting
of the metalliferous Rosh Pinah Formation and the Spitzkop and
Koivib Suites in the Pan-African Gariep Belt, southwestern
Namibia. South African Journal of Geology, 108, 19 –34.
Basei, M. A. S., Frimmel, H. E., Nutman, A. P., Preciozzi, F. & Jacob,
J. 2005. A connection between the Neoproterozoic Dom Feliciano
(Brazil/Uruguay) and Gariep (Namibia/South Africa) orogenic
belts – evidence from a reconnaissance provenance study. Precam-
brian Research, 139, 195–221.
Beetz, W. 1926. Über Glazialschichten an der Basis der Nama- und
Konkipformation in der Namib Südwestafrikas. Neues Jahrbuch f.
Mineralogie, Geologie und Paläontologie, Abteilung B, 56, 437– 481.
Borg, G., Kärner, K., Buxton, M., Armstrong, R. & Van der Merwe,
S. W. 2003. Geology of the Skorpion zinc deposit, southern Namibia.
Economic Geology, 98, 749–771.
Cloud, P. E. 1974. Giant stromatolites and associated vertical tubes from
the Upper Proterozoic Noonday Dolomite, Death Valley region,
eastern California. Geological Society of America, Bulletin, 85,
1869– 1882.
Collins, A. S. & Pisarevsky, S. A. 2005. Amalgamating eastern Gond-
wana: the evolution of the circum-Indian orogens. Earth-Science
Reviews, 71, 229–270.
Corsetti, F. A. & Grotzinger, J. P. 2005. Origin of tube structures in
Neoproterozoic post-glacial cap carbonates: example from
Noonday Dolomite, Death Valley, United Sates. Palaios, 20,
348– 362.
Davies, C. & Coward, M. P. 1982. The structural evolution of the
Gariep Arc in southern Namibia. Precambrian Research, 17,
173– 198.
De Villiers, J. & Söhnge, P. G. 1959. The geology of the Richtersveld.
Memoirs of the Geological Survey of South Africa, 48, 1– 295.
Fölling, P. G. & Frimmel, H. E. 2002. Chemostratigraphic correlation of
carbonate successions in the Gariep and Saldania Belts, Namibia and
South Africa. Basin Research, 14, 69– 88.
Fölling, P. G., Zartman, R. E. & Frimmel, H. E. 2000. A novel
approach to double-spike Pb –Pb dating of carbonate rocks: examples
from Neoproterozoic sequences in southern Africa. Chemical
Geology, 171, 97 –122.
Frimmel, H. E. 1995. Metamorphic evolution of the Gariep Belt. South
African Journal of Geology, 98, 176–190.
Frimmel, H. E. 2000. The Pan-African Gariep Belt in southwestern
Namibia and western South Africa. Communications of the Geologi-
cal Survey of Namibia, 12, 197–209.
Frimmel, H. E. 2003a. Gariep Supergroup. In: Johnson, M. R. (ed.)
Catalogue of South African Lithostratigraphic Units. Council for
Geoscience, Pretoria, 5 –7.
Frimmel, H. E. 2003b. Port Nolloth Group. In: Johnson, M. R. (ed.)
Catalogue of South African Lithostratigraphic Units. Council for
Geoscience, Pretoria, 35 – 37.
Frimmel, H. E. 2008. The Gariep Belt. In: Miller, R. M. (ed.) The
Geology of Namibia. Geological Survey of Namibia, Windhoek, 2,
14-1 –14-39.
Frimmel, H. E. & Basei, M. A. S. 2006. Tracking down the Neoprotero-
zoic connection between southern Africa and South America – a
revised geodynamic model for SW-Gondwana amalgamation. In:
Gaucher, C. & Bossi, J. (eds) V. South American Symposium on
Isotope Geology. 24 –27 August, Punta del Este, 94– 97.
Frimmel, H. E. & Fölling, P. G. 2004. Late Vendian closure of
the Adamastor Ocean: timing of tectonic inversion and
syn-orogenic sedimentation in the Gariep Basin. Gondwana
Research, 7, 685– 699.
Frimmel, H. E. & Frank, W. 1998. Neoproterozoic tectono-thermal
evolution of the Gariep Belt and its basement, Namibia/South
Africa. Precambrian Research, 90, 1 –28.
Frimmel, H. E. & Lane, K. 2005. Geochemistry of carbonate beds in the
Neoproterozoic Rosh Pinah Formation, Namibia: implications on
depositional setting and hydrothermal ore formation. South African
Journal of Geology, 108, 5 – 18.
Frimmel, H. E., Hartnady, C. J. H. & Koller, F. 1996a. Geochemistry
and tectonic setting of magmatic units in the Pan-African Gariep Belt,
Namibia. Chemical Geology, 130, 101– 121.
Frimmel, H. E., Klötzli, U. & Siegfried, P. 1996b. New Pb– Pb single
zircon age constraints on the timing of Neoproterozoic glaciation and
continental break-up in Namibia. The Journal of Geology, 104,
459– 469.
Frimmel, H. E., Zartman, R. E. & Späth, A. 2001. Dating Neoprotero-
zoic continental break-up in the Richtersveld Igneous Complex,
South Africa. The Journal of Geology, 109, 493– 508.
Frimmel, H. E., Fölling, P. G. & Eriksson, P. 2002. Neoproterozoic
tectonic and climatic evolution recorded in the Gariep Belt,
Namibia and South Africa. Basin Research, 14, 55 –67.
Frimmel, H. E., von Veh, M. W. & Fölling, P. G. 2003a. Kaigas For-
mation. In: Johnson, M. R. (ed.) Catalogue of South African Litho-
stratigraphic Units. Council for Geoscience, Pretoria, 17– 19.
 KAIGAS AND NUMEES FORMATIONS
Frimmel, H. E., von Veh, M. W. & Fölling, P. G. 2003b. Numees
Formation. In: Johnson, M. R. (ed.) Catalogue of South African
Lithostratigraphic Units. Council for Geoscience, Pretoria, 25– 28.
Frimmel, H. E., Jonasson, I. & Mubita, P. 2004. An Eburnean base metal
source for sediment-hosted zinc-lead deposits in Neoproterozoic units
of Namibia: lead isotopic and geochemical evidence. Mineralium
Deposita, 39, 328–343.
Gaucher, C., Frimmel, H. E. & Germs, G. J. B. 2005. Organic-walled
microfossils and biostratigraphy of the upper Port Nolloth Group
(Namibia): implications for the latest Neoproterozoic glaciations.
Geological Magazine, 142, 539– 559.
Germs, G. J. B. 1995. The Neoproterozoic of southwestern Africa, with
emphasis on platform stratigraphy and paleontology. Precambrian
Research, 73, 137– 151.
Gresse, P. G. 1994. Strain partitioning in the southern Gariep Arc as
reflected by sheath folds and stretching lineations. South African
Journal of Geology, 97, 52 – 61.
Grotzinger, J. P., Bowring, S. A., Saylor, B. Z. & Kaufman, A. J.
1995. Biostratigraphic and geochronologic constraints on early
animal evolution. Science, 270, 598–604.
Hälbich, I. W. & Alchin, D. J. 1995. The Gariep belt: stratigraphic-
structural evidence for obliquely transformed grabens and back-
folded thrust stacks in a combined thick-skin thin-skin structural
setting. Journal of African Earth Sciences, 21, 9 –33.
Hartnady, C., Joubert, P. & Stowe, C. 1985. Proterozoic crustal evol-
ution in southwestern Africa. Episodes, 8, 236– 244.
Hartnady, C. J. H. & von Veh, M. W. 1990. Tectonostratigraphic and
structural history of the Late Proterozoic – Early Palaeozoic Gariep
Belt, Cape Province, South Africa. Geological Society of South
Africa, Cape Town.
Hegenberger, W. 1993. Stratigraphy and sedimentology of the Late
Precambrian Witvlei and Nama Groups, east of Windhoek.
Memoirs of the Geological Survey of Namibia, 17, 1– 82.
Hoffman, P. F., Hawkins, D. P., Isachsen, C. E. & Bowring, S. A. 1996.
Precise U–Pb zircon ages for early Damaran magmatism in the
Summas Mountains and Welwitschia inlier, northern Damara belt,
Namibia. Communications of the Geological Survey of Namibia,
11, 47 – 52.
Hoffman, P. F., Kaufman, A. J. & Halverson, G. P. 1998. Comings and
goings of global glaciations on a Neoproterozoic tropical platform in
Namibia. GSA Today, 8, 1– 9.
Hoffman, P. F., Halverson, G. P., Domack, E. W., Husson, J. M.,
Higgins, J. A. & Schrag, D. P. 2007. Are basal Ediacaran
(635 Ma) post-glacial ‘cap dolostones’ diachronous? Earth and Pla-
netary Science Letters, 258, 114–131.
Hoffmann, K.-H. & Prave, A. R. 1996. A preliminary note on a revised
subdivision and regional correlation of the Otavi Group based on gla-
ciogenic diamictites and associated cap dolostones. Communications
of the Geological Survey of Namibia, 11, 77 – 82.
Hoffmann, K.-H., Condon, D. J., Bowring, S. A. & Crowley, J. L.
2004. A U– Pb zircon date from the Neoproterozoic Ghaub For-
mation, Namibia: constraints on Marinoan glaciation. Geology, 32,
817– 820.
Jasper, M. U., Stanistreet, I. G. & Charlesworth, E. G. 1992– 1993.
Report: preliminary results of a study of the structural and
231
sedimentological evolution of the late Proterozoic/early Palaeozoic
Gariep Belt, southern Namibia. Communications of the Geological
Survey of Namibia, 8, 99 –118.
Jasper, M. J. U., Stanistreet, I. G. & Charlesworth, E. G. 2000.
Neoproterozoic inversion tectonics, half-graben depositories and
glacial controversies, Gariep fold-thrust belt, southern Namibia.
In: Miller, R. M. (ed.) Henno Martin Commemorative Volume.
Communications of the Geological Survey of Namibia, Windhoek,
1887–1896.
Kaufman, A. J., Hayes, J. M., Knoll, A. H. & Germs, G. J. B. 1991.
Isotopic composition of carbonates and organic carbon from upper
Proterozoic successions in Namibia: stratigraphic variation and the
effects of diagenesis and metamorphism. Precambrian Research,
49, 301– 327.
Kennedy, M. J. 2001. Are Proterozoic cap carbonates and isotopic excur-
sions a record of gas hydrate destabilization following Earth’s coldest
intervals? Geology, 29, 443–446.
Kröner, A. 1974. The Gariep Group, Part I: Late Precambrian forma-
tions in the western Richtersveld, northern Cape Province. Pre-
cambrian Research Unit, University of Cape Town, Bulletin, 13,
1– 115.
Martin, H. 1965a. Beobachtungen zum Problem der jung-präkam-
brischen glazialen Ablagerungen in Südwestafrika. Geologische
Rundschau, 54, 115– 127.
Martin, H. 1965b. The Precambrian geology of South West Africa and
Namaqualand. Precambrian Research Unit, University of Cape
Town, Bulletin, 4, 1– 177.
McMillan, M. D. 1968. The geology of the Witputs-Sendelingsdrif area.
Precambrian Research Unit, University of Cape Town, Bulletin, 4,
1– 177.
Middlemost, A. E. K. 1963. Geology of the southeastern Richtersveld.
Unpublished PhD thesis, University of Cape Town, Cape Town.
Reid, D. L., Ransome, I. G. D., Onstott, T. C. & Adams, C. J. 1991.
Time of emplacement and metamorphism of Late Precambrian
mafic dykes associated with the Pan-African Gariep orogeny,
Southern Africa: implications for the age of the Nama Group.
Journal of African Earth Sciences, 13, 531–541.
Rogers, A. W. 1916. The geology of part of Namaqualand. Transactions
of the Geological Society of South Africa, 18, 72 – 101.
Siegfried, P. R. & Moore, J. M. 1990. The Rosh Pinah Zn– Pb– Cu– Ag
massive sulphide deposit–a product of early rift-related volcanism?
In: Geocongress 90. Geological Society of South Africa, Cape
Town, 512– 513.
Vieira, L. C., Trindade, R. I. F., Nogueira, A. C. R. & Ader, M. 2007.
Identification of a Sturtian cap carbonate in the Neoproterozoic Sete
Lagoas carbonate platform, Bambuı́ Group, Brazil. Comptes Rendus
Geoscience, 339, 240–258.
von Veh, M. W. 1993. The stratigraphy and structural evolution of the
Late Proterozoic Gariep Belt in the Sendelingsdrif-Annisfontein
area, northwestern Cape Province. Precambrian Research Unit,
University of Cape Town, Bulletin, 38, 1– 174.
Zheng, Y.-F., Wu, Y.-B., Gong, B., Chen, R.-X., Tang, J. & Zhao,
Z.-F. 2007. Tectonic driving of Neoproterozoic glaciations: evi-
dence from extreme oxygen isotope signature of meteoric water
in granite. Earth and Planetary Science Letters, 256, 196– 210.
 Chapter 18

The Karoetjes Kop and Bloupoort formations, Gifberg Group, South Africa

HARTWIG E. FRIMMEL1,2
1
Department of Geological Sciences, University of Cape Town, Rondebosch 7701, South Africa
2
Present address: Geodynamics & Geomaterials Research Division, Institute of Geography and Geology, University of Wuerzburg,
Am Hubland, D-97074 Wuerzburg, Germany
(e-mail: hartwig.frimmel@uni-wuerzburg.de)

Abstract: The Vredendal Outlier near the South African west coast lies in an intermediate position between the late Neoproterozoic
Gariep Belt further north and the Cambrian Saldania Belt further south. It consists of a sedimentary succession of siliciclastic and car-
bonate rocks, unified as the Gifberg Group, which contains two diamictite-bearing units, the Karoetjes Kop Formation at the base and the
Bloupoort Formation near the top of the group. The diamictite of the Karoetjes Kop Formation represents mainly debris flow deposits in a
continental rift setting, with some contribution from retreating glaciers. In contrast, the younger diamticite in the Bloupoort Formation is
glaciomarine, is associated with banded iron-formation, is underlain by stromatolitic reef carbonates and is overlain by carbonates. Most
of the Gifberg Group is poorly exposed and poorly investigated. In the absence of radiometric age data, stratigraphic interpretation and
correlation is based on lithological and chemostratigraphic evidence. The entire Gifberg Group is considered to be equivalent to the much
better investigated Port Nolloth Group in the Gariep Belt (sensu stricto). Whereas the Karoetjes Kop Formation is correlated with the
c. 750 Ma Kaigas Formation of the Gariep Belt, the diamictite and associated banded iron-formation of the Bloupoort Formation are
regarded as correlatives of the Numees Formation of the Gariep Belt. The entire Gifberg Group was subjected to transpressional defor-
mation and accompanying low-grade metamorphism during continental collision between the Rio de la Plata and Kalahari plates at the
end of the Neoproterozoic and again during the Cambrian accretionary orogeny along the southwestern margin of Gondwana, which led
to the development of the Saldania Belt further south.

The Karoetjes Kop and Bloupoort Formations both contain dia-
mictite. They occur in a succession of metasedimentary Neoproter-
ozoic rocks that are relatively poorly exposed along the South
African west coast between latitudes 308S and 328S and in the
plains that stretch from the coast up to 80 km inland in the area
around Vanrhynsdorp (Figs 18.1 and 18.2). This area was first
mapped by Rogers (Rogers & Schwarz 1904; Rogers 1911). Sub-
sequent studies (Brink 1950; Jansen 1960; Kröner 1968; Lamont
1947) led to rather contrasting stratigraphic interpretations. The
basis for the current stratigraphic subdivision is the work of Gresse
(1992), who re-mapped large parts of the area and studied the sedi-
mentology, structure and metamorphism. Originally, Gresse
included this metasedimentary succession as the Gifberg Subgroup
in the Vanrhynsdorp Group. The younger part of that group has
since been recognized as a correlative of the lower to middle
Nama Group in Namibia and represents the fill of a southern sub-
basin of the larger Nama foreland basin (Gresse & Germs 1993).
The Gifberg rocks occur unconformably below these foreland
basin deposits and bear strong lithological similarities to sedimen-
tary successions of the Neoproterozoic Port Nolloth Group in the
Gariep Belt further north. Subsequently, they were, therefore,
described as a separate group, the Gifberg Group (de Beer et al.
2002). Their area of distribution is referred to as the Vredendal
Outlier, which occupies an intermediate position between the
Gariep Belt in the north and the Saldania Belt in the south.
A glacial influence during deposition of the Karoetjes Kop
Formation was indirectly inferred by Gresse (1992), who was the
first to compare the diamicite of that formation with the glacio-
genic Kaigas Formation diamictite in the Port Nolloth Zone.
Less ambiguous is the glacial origin in the case of the diamictite
within the younger Bloupoort Formation. Although it was des-
cribed originally as agglomerate (Rogers & Schwarz 1904),
general agreement exists now on its glaciogenic character follow-
ing a more detailed description of the rock association by Buehr-
mann (1981). The correlation of individual formations between
the Port Nolloth and Gifberg Groups has remained speculative,
and uncertainty exists particularly with regard to the age of the
Bloupoort Formation diamictite (de Beer et al. 2002). Recent litho-
geochemical and isotopic data on carbonates that are positioned
between the two diamictite-bearing formations of the Gifberg
Group (Frimmel 2008) helped in better constraining the strati-
graphic position of that group.
Structural framework and tectonic evolution
Although a polyphase deformational history is evident from field
observations and structural data (Gresse 1992), the overall struc-
ture appears to be relatively simple with a gentle dip of primary
stratigraphic contacts towards the SW. The local distribution of
rock types can be largely explained by the interference between
two generations of folds (F2 and F3). Earlier F1 folds are rarely pre-
served. As all fold axes follow a similar trend that is subparallel to
the orogenic belt and colinear with the stretching direction, a trans-
pressive regime is inferred. The style of deformation is identical to
that in the Gariep Belt further north, where it has been explained
by sinistral transpression (Von Veh 1993; Gresse 1994). Similarly,
the deformation in the Vredendal Outlier can be explained by sinis-
tral transpression that reflects compression towards the east with
SSE-verging folds and thrusts (F1), west-verging back-folds and
back-thrusts (F2), ENE-verging F3 folds and conjugate kinks.
By analogy with findings from the Gariep Belt further north
(Frimmel & Frank 1998), an age of c. 545 Ma is assigned to F1
and F2. The F3 fold- and thrust-event also affected the siliciclastic
sedimentary rocks of the Vanrhynsdorp Group (Gresse 1992) and
is thus younger. Ar – Ar age data on syn-F3 micas are in the range
500– 480 Ma (Gresse et al. 1988; Frimmel & Frank 1998).
Syn-orogenic deformation was accompanied by regional meta-
morphism. The grade of metamorphism increases from the fore-
land in the east towards the west. Although the Vanrhynsdorp
Group rocks in the east have experienced only sub-greenschist
facies conditions, those near the thrust contact with the rocks of
the Gifberg Group reached greenschist-facies conditions. Conse-
quently, this phase of dynamic metamorphism must have taken
place after the deposition of the Vanrhynsdorp foreland basin fill
and it was syn-kinematic with respect to F3. An increase in the
metamorphic grade from lowest greenschist facies near Vanrhyns-
dorp to lowermost amphibolite facies at the Atlantic coast is

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 233– 237. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.18
234
Fig. 18.1. Map showing the location of the Vredendal Outlier within the
network of Pan-African orogenic belts of southwestern Africa.
evident from metapelites of the Gifberg Group, with kyanite-
staurolite-garnet-micaschist developed in the extreme NW of the
Vredendal Outlier.
Stratigraphy
The oldest unit of the Gifberg Group (Fig. 18.3) is the mixed sili-
ciclastic, diamictite-bearing Karoetjes Kop Formation. Variably
recrystallized limestone and dolostone, as well as minor meta-
greywacke, quartzite and phyllite of the Widouw Formation
follow above the Karoetjes Kop Formation. The overlying Aties
Formation is dominated by carbonaceous schist, with subordinate
phyllite to biotite-schist (in the westernmost exposures kyanite-
bearing), meta-arenite, and impure, thin carbonate beds. The
youngest formation is the Bloupoort Formation, which comprises
a variety of carbonate rocks, calcarenite, diamictite, pyritic and
graphitic phyllite, ferruginous chert and iron-formation.
Glaciogenic deposits and associated strata
The Karoetjes Kop Formation
This basal formation overlies high-grade metamorphic gneisses
of the Mesoproterozoic Namaqualand Metamorphic Complex. It
H. E. FRIMMEL
starts with a white to blue-grey basal conglomerate in which the
basement lithotypes appear reworked. This is followed by a grey,
fine-grained quartzitic greywacke in which thin, well-rounded
quartz (or quartzite) pebble beds are intercalated. In places, this
greywacke rests directly on the basement. A massive, poorly
bedded, laterally discontinuous diamictite follows gradationally
above the quartzitic greywacke. It reaches a maximum thickness
of a few tens of metres and contains poorly sorted pebbles,
cobbles and boulders of basement-derived gneiss, quartzite, vein
quartz and biotite schist, all of which are set in a metamorphosed
biotite-rich matrix. In places, the diamictite rests directly on base-
ment and grades laterally into the quartzitic greywacke.
Above the diamictite follows a reddish weathering, laterally
continuous feldspathic quartzite with intercalated metapelite (ser-
icite schist). Large trough cross-bedded sets are common, reaching
several metres in thickness. This lithotype grades upwards into
thick-bedded, coarse-grained, white to light blue-grey quartzite.
No complete section through the entire formation is exposed.
Consequently, no information on the true stratigraphic thickness
is available.
The Widouw Formation
Carbonate intersections of more than 200 m were recorded in bore-
holes through the central part of the Vredendal Outlier (Gresse
1992), but the true thickness may be less due to tectonic thickening
by isoclinal folding. The carbonate beds thin out towards the north
where a basement high probably existed throughout Neoprotero-
zoic to early Cambrian times (Kamieskroon Ridge, Fig. 18.1).
Variably coloured (from white, pink, ochre to black), flat lami-
nated, medium- to thick-bedded limestone dominates the lower
part of the formation. In several places, individual beds are later-
ally discontinuous, irregular and reflect former dissolution surfaces
(Frimmel 2008). Although the limestone unit is generally very
uniform, intra-formational syn-sedimentary limestone breccia
beds as well as stratiform black chert boudins occur in the upper
parts. Contorted dolomitized beds reach 20–30 cm in thickness
and contain white blades of calcite, several centimetres long,
which have been compared with herringbone calcite and inter-
preted as pseudomorphs after enterolithic gypsum (Frimmel
2008). Impure limestone contains as much as 40 vol% coarse-
grained, angular quartz clasts, which indicate an overall proximal
setting for the depositional environment of these carbonates.
Marked differences in grain size reflect different degrees of
metamorphic recrystallization. Several types of dolostone, ranging
in relative age from syn-sedimentary/early diagenetic to meta-
morphic, can be distinguished. The former includes stromatolitic
dolostone (Gresse 1992).
The Aties Formation
Although the lower contact of the Aties Formation with the under-
lying carbonates of the Widouw Formation is typically sheared, an
originally conformable contact is inferred. The Aties Formation is
made up of predominantly black carbonaceous, pyritic schist with
interbedded phyllite to biotite schist, meta-arenite and thin, impure
limestone and dolostone beds. The arenitic beds are quartz-
sandstone and greywacke in composition, but some of them are
highly feldspathic (arkose), coarse-grained and grade in places
into intraformational feldspathic conglomerate beds. Many of the
well-bedded greywacke beds display normal grading and trough
cross-bedding on a metre scale. Locally, monomictic quartz
pebble conglomerate occurs within white quartzite. The presence
of, inter alia, conglomerate and diamictite, associated with lime-
stone and dolostone has been reported as ‘interbedded’ with phyl-
lite of the Aties Formation (de Beer et al. 2002). It is suspected,
however, that this diamictite-bearing succession is part of the
overlying Bloupoort Formation.
 THE KAROETJES KOP AND BLOUPOORT FORMATIONS
The Bloupoort Formation
The distinction of this formation from the underlying Aties Forma-
tin has been problematic and prompted Gresse (1992) to include all
the Bloupoort rocks as a separate member within the Aties For-
mation. Subsequently, it was recognized that they are younger
and represent a relatively complex unit, and their stratigraphic
rank was raised to that of formation (de Beer et al. 2002).
The main distribution of this formation is to the SE of Vanrhyns-
dorp (Fig. 18.2), where a mixed carbonate unit grades into a
diamictite-dominated unit. The former is almost entirely dolomitic
and comprises laminated, stromatolitic (Conophyton-like), oolitic,
intraclastic, conglomeratic, and brecciated and siliceous dolo-
stone. Sandy to gritty dolostone is rich in variably rounded
quartz clasts and contains up to 15 vol% feldspar (Gresse 1992).
Oolites, pelletoids, pellets and lumps are common and were depos-
ited in carbonate mud (micrite). Dolomitic breccia consists of
micritic fragments that are cemented by sparry dolomite or micrite.
This dolomitic unit grades upwards into dolomitic arenite and
ferruginous diamictite, which constitute the Swartleikrans Bed
of Buehrmann (1981). The diamictite consists of very poorly
sorted, angular fragments and subrounded pebbles, cobbles and
boulders of basement-derived orthogneiss and granulite, as well
as basin-derived dolostone and arenite, and minor ferruginous
chert, jasper, vein quartz, limestone, oolitic dolostone and feldspar
grains, all of which are set in a ferruginous, hematite-rich argillitic
matrix. Bedding is only very poorly developed in the overall
massive diamictite, but in places very crude graded bedding
from diamictite into ferruginous sandstone and siltstone to argillite
can be observed on a metre scale. Overall, the proportion of inter-
bedded ferruginous argillite increases towards the top of the unit.
Laterally, the diamictite grades into banded iron-formation with
isolated dropstones and interbedded chert and jasper (Rooihoogte
Bed of Buehrmann 1981).
Locally, the diamictite is overlain by thinly laminated pink
dolostone and dolomitic calcarenite. Intense thrust-faulting and
folding in all of these units makes it virtually impossible to con-
strain their true sedimentary thickness.
235
Fig. 18.2. Areal distribution of the
individual formations that constitute the
Gifberg Group.
Boundary relations with overlying and underlying
non-glacial units
The lower contact of the diamictite in the Karoetjes Kop Forma-
tion with the underlying quartzitic greywacke is gradational.
Where the diamictite rests directly on basement gneiss, the
contact is an erosional unconformity. The upper contact of this
diamictite with the overlying quartzite is sharp and conformable.
The contact between the Karoetjes Kop Formation and the
Widouw Formation is not exposed, and the relative age relation-
ship between the two formations is only inferred from their
spatial distribution.
The lower contact of the diamictite-hosting Bloupoort For-
mation is typically tectonic. The upper contact of the formation
is either a thrust-fault, or it is covered by younger sedimentary
rocks of the Cape Supergroup or by aeolian sand. Within the for-
mation, the lower boundary of the diamictite with the underlying
dolostone is gradational, with an intervening dolomitic arenite,
whereas the upper boundary against the overlying dolostone is
sharp and appears conformable. The entire succession is,
however, strongly deformed. A fivefold rhythmical repetition of
an ‘interbedded sequence of dolomite, ironstone and ferruginous
mixtite’, each about 5 –15 m in thickness, described by Gresse
(1992), could well reflect tectonic repetition.
Chemostratigraphy
Limited trace element analyses as well as C, O and Sr isotope data
have been reported for carbonates of the Widouw Formation
(Frimmel 2008). Limestone of this formation is very rich in Sr
(as much as 4728 ppm) and has extremely low Rb/Sr ratios.
Their 87Sr/86Sr ratios can be taken as near-primary and they
range consistently throughout the succession between 0.70844
and 0.70847. The C and O isotope ratios in the limestone cover
a wide range of d13C between – 4.2 and þ4.8‰ (relative to
V-PDB) and d18O between –9.4 and –3.1 (relative to V-PDB).
Syn-sedimentary or early diagenetic dolostone is enriched in 18O
236 H. E. FRIMMEL

(e.g. Collins & Pisarevsky 2005). The depositional basin for the
Vanrhynsdorp Group, undifferentiated Gifberg Group was most likely the southern continuation of the
(c. 550–530 Ma) Gariep Basin, specifically the basins in which the sediments of
the Port Nolloth Group were deposited. Today the Gariep Belt is
positioned between the Kalahari and the Rio de la Plata Cratons.
Both Pb and Nd isotopic evidence (Frimmel et al. 2004; Frimmel
& Basei 2006) point to a provenance of the entire Gariep Super-
group, including the Gifberg Group, on the Kalahari Craton. For
Bloupoort
parts of the Widouw Formation, an evaporitic influence has been
Formation suggested based on petrographic observations, trace element distri-
bution and isotope data (Frimmel 2008), which speaks against a
high-latitude position at the time of sedimentation.
Gifberg Group

Geochronological constraints
Aties
Formation No radiometric age data that might help in constraining the absol-
ute age of the diamictite units in the Gifberg Group are available.
By analogy with the Port Nolloth Group in the eastern Gariep Belt,
Widouw the maximum age of sedimentation is given by the youngest age
Formation obtained on the pre-Gariep basement. This is a U –Pb single
zircon age of 771 + 6 Ma for the Lekkersing granite (Frimmel
et al. 2001). The minimum age for sedimentation is given by the
age of metamorphism. Ar – Ar age data obtained on syn-tectonic
muscovite and biotite from coastal outcrops in the west of the
Karoetjes Kop Vredendal Outlier (Frimmel & Frank 1998) range from 513
Formation to 491 Ma. This range includes a Ar – Ar whole-rock age of
496 + 2 Ma obtained on a phyllite from the Vanrhynsdorp
Group (Gresse et al. 1988). These ages have been interpreted to
date a late Pan-African tectonic pulse that led to NE-directed
thrusting of both the Gifberg Group and the younger Vanrhynsdorp
Basement (1.0 Ga) Group. It is therefore likely that the first regional metamorphism
that affected the Gifberg Group rocks took place earlier during
Limestone, the continental collision phase of the Gariepian orogeny. By
Arenite, rudite analogy with results obtained on the Port Nolloth Group (Frimmel
dolostone
& Frank 1998) and the Nama Group (Grotzinger et al. 1995)
Iron Formation Argillite further north, an age of c. 545 Ma is assumed for that stage of meta-
morphism and tectonism that caused the unconformity between the
Diamictite Gneiss, schist Gifberg and Vanrhynsdorp Groups.

Fig. 18.3. Generalized lithostratigraphy of the Gifberg Group (modified from

de Beer et al. 2002 and Frimmel & Fölling 2004). Discussion

No evidence exists for any marine influence in the Karoetjes Kop

relative to the limestone and it shows a similar, but narrower vari-
ation in d13C.
Other characteristics
Calcitic marble of the Widouw Formation is mined at several quar-
ries in the vicinity of Vredendal. The pyritic schists of the Aties
Formation are strongly weathered with numerous quartz veins
that carry Cu, Fe and Mn staining. Numerous surface Fe-oxide
deposits (‘gossans’) are derived from the palaeoweathering of
the pyrite-rich beds on a Paleocene palaeosurface, the so-called
African Surface (Partridge & Maud 1987).
No biostratigraphic data have been published so far for rocks
of the Gifberg Group.
Palaeolatitude and palaeogeography
No reliable palaeomagnetic data are available for the Vredendal
Outlier. The low-grade regional metamorphic overprint most
likely led to the remagnetization of the rocks during the
Pan-African/Gariepian orogeny. Thus, any assumption about
palaeolatitude and palaeogeography is based only on circumstan-
tial evidence and comparison with continental-scale models
Formation for which a narrow continental rift, dominated by allu-
vial fans, plains and fan deltas, is indicated. The depositional
environment envisaged for this formation is therefore similar to
that inferred for the Stinkfontein Subgroup and the Kaigas For-
mation in the Port Nolloth Zone (Chapter 17), except for a total
lack of any evidence for marine influence in the Karoetjes Kop
Formation. A number of very small outcrops of the Karoetjes
Kop Formation also exist along the coast further to the NW of
the Vredendal Outlier, thus suggesting a continuation of the
former basin margin from the Port Nolloth Zone southwards into
the area around Vredendal. Consequently, the current eastern
boundary of these Neoproterozoic rocks would be a good approxi-
mation of the original eastern basin margin. As with the Kaigas
Formation, the diamictite in the Karoetjes Kop Formation is
interpreted as mainly a debris-flow deposit because of its lack of
internal structure and very limited geographical distribution
(lateral discontinuity) along the basin edge. By analogy with the
Kaigas Formation, a fluvio-glacial origin for parts of the formation
is suggested.
The isotopic record of the Widouw Formation carbonates has
been interpreted as reflecting an overall 13C-depleted sea, with
the higher d13C ratios resulting from elevated evaporation rates
rather than changes in the global seawater composition (Frimmel
2008). These shallow-water marine deposits are therefore corre-
lated with the post-Kaigas transgression.
 THE KAROETJES KOP AND BLOUPOORT FORMATIONS
The regional and global correlation of the diamictite in the Blou-
poort Formation remains problematic. Little doubt exists about the
glacial nature of the diamictite. The association with, and lateral
transition into, banded iron-formation with occasional lonestones
and dropstones clearly indicate a glaciomarine origin. The appear-
ance of the underlying carbonates, which bear a remarkable resem-
blance to those of the Dabie River Formation in the Port Nolloth
Zone, the ferruginous matrix of the diamictite, its massive struc-
ture and its association with iron-formation are all features that
are identical to the Numees Formation in the Port Nolloth Zone.
In the absence of radiometric and biostratigraphic evidence, the
lower dolomitic unit of the Bloupoort Formation is correlated
with the Dabie River Formation, the diamictite and iron-formation
with the Numees Formation, and the upper carbonates and silici-
clastic rocks with the Holgat Formation.
Fieldwork in the Vredendal area was supported by Kumba Resources. Funding by
the South African National Research Foundation is gratefully acknowledged.
F. MacDonald and G. Halverson are thanked for helpful comments on the original
manuscript. This work represents a contribution of the IUGS- and UNESCO-
funded IGCP (International Geoscience Programme) Projects #512 and #478.
References
Brink, W. C. 1950. The geology, structure and petrology of the Nuwerus
area. Annals of the University of Stellenbosch, 16, 97 –221.
Buehrmann, H. T. 1981. The geology of the Lower Kobe Valley,
Vanrhynsdorp District. Annals of the University of Stellenbosch, 3,
Series A1, 67 –144.
Collins, A. S. & Pisarevsky, S. A. 2005. Amalgamating eastern Gond-
wana: the evolution of the circum-Indian orogens. Earth-Science
Reviews, 71, 229–270.
de Beer, C. H., Gresse, P. G., Theron, J. N. & Almond, J. E. 2002. The
Geology of the Calvinia Area, Explanation Sheet 3118 Calvinia
1:250000 scale. Council for Geoscience, Pretoria.
Frimmel, H. E. 2008. An evaporitic facies in Neoproterozoic post-glacial
carbonates: the Gifberg Group, South Africa. Gondwana Research,
13, 453– 468.
Frimmel, H. E. & Basei, M. A. S. 2006. Tracking down the Neoprotero-
zoic connection between southern Africa and South America – a
revised geodynamic model for SW-Gondwana amalgamation. In:
Gaucher, C. & Bossi, J. (eds) V. South American Symposium on
Isotope Geology. 24 –27 August, Punta del Este, 94 –97.
Frimmel, H. E. & Fölling, P. G. 2004. Late Vendian closure of the
Adamastor Ocean: timing of tectonic inversion and syn-orogenic
sedimentation in the Gariep Basin. Gondwana Research, 7, 685– 699.
237
Frimmel, H. E. & Frank, W. 1998. Neoproterozoic tectono-thermal
evolution of the Gariep Belt and its basement, Namibia/South
Africa. Precambrian Research, 90, 1 – 28.
Frimmel, H. E., Jonasson, I. & Mubita, P. 2004. An Eburnean base metal
source for sediment-hosted zinc-lead deposits in Neoproterozoic units
of Namibia: lead isotopic and geochemical evidence. Mineralium
Deposita, 39, 328–343.
Frimmel, H. E., Zartman, R. E. & Späth, A. 2001. Dating Neoprotero-
zoic continental break-up in the Richtersveld Igneous Complex,
South Africa. The Journal of Geology, 109, 493 – 508.
Gresse, P. G. 1992. The tectono-sedimentary history of the Vanrhyns-
dorp Group. Memoirs of the Geological Survey of South Africa, 79,
1– 163.
Gresse, P. G. 1994. Strain partitioning in the southern Gariep Arc as
reflected by sheath folds and stretching lineations. South African
Journal of Geology, 97, 52– 61.
Gresse, P. G., Fitch, F. J. & Miller, J. A. 1988. 40Ar/39Ar dating of the
Cambro-Ordovician Vanrhynsdorp tectonite in southern Namaqua-
land. South African Journal of Geology, 91, 257–263.
Gresse, P. G. & Germs, G. J. B. 1993. The Nama foreland basin:
sedimentation, major unconformity bounded sequences and
multisided active margin advance. Precambrian Research, 63,
247– 272.
Grotzinger, J. P., Bowring, S. A., Saylor, B. Z. & Kaufman, A. J.
1995. Biostratigraphic and geochronologic constraints on early
animal evolution. Science, 270, 598–604.
Jansen, H. 1960. The geology of the area around Nieuwoudtville. In: Von
Backström, J. W. (ed.) Explanation Sheet 241 (Nieuwoudthville).
Geological Survey of South Africa, Pretoria, 49.
Kröner, A. 1968. The gneiss-sediment relationship northwest of
Vanrhynsdorp, Cape Province. Bulletin, Precambrian Research
Unit, University of Cape Town, 3, 1– 233.
Lamont, G. T. 1947. The geology of part of the Van Rhynsdorp Division,
Cape Province. Unpublished PhD thesis, University of Cape Town,
Rondebosch.
Partridge, T. C. & Maud, R. R. 1987. Geomorphic evolution of southern
Africa since the Mesozoic. South African Journal of Geology, 90,
179– 208.
Rogers, A. W. 1911. Report on the geological survey of parts of the div-
isions of Van Rhyn’s Dorp, and Namaqualand. Annual Report of the
Geological Commission, Cape of Good Hope, 1911, 7– 84.
Rogers, A. W. & Schwarz, E. H. L. 1904. Geological survey of the north-
western part of Van Rhyn’s Dorp. Annual Report of the Geological
Commission, Cape of Good Hope, 1904, 9– 46.
Von Veh, M. W. 1993. The stratigraphy and structural evolution of the
Late Proterozoic Gariep Belt in the Sendelingsdrif-Annisfontein
area, northwestern Cape Province. Precambrian Research Unit,
University of Cape Town, Bulletin, 38, 1– 174.
 Chapter 19

The Ayn Formation of the Mirbat Group, Dhofar, Oman

PHILIP A. ALLEN1*, RUBEN RIEU2, JAMES L. ETIENNE3, ALBERT MATTER4 & ANDREA COZZI5
1
Department of Earth Science and Engineering, Imperial College London, South Kensington Campus, London SW7 2AZ, UK
2
TOTAL E&P Nederland, Bordewijklaan 18, Den Haag, The Netherlands
3
Neftex Petroleum Consultants Ltd, 97 Milton Park, Abingdon, Oxfordshire OX14 4RY, UK
4
Institute of Geological Sciences, University of Bern, Baltzerstrasse 1, CH-3012 Bern, Switzerland
5
ENI Angola, Rua Nicola Gomes Spencer, 140, PO Box 1289, Luanda, Angola
*Corresponding author (e-mail: philip.allen@imperial.ac.uk)

Abstract: Glacial deposits are found in the Ayn Formation and Shareef Formation of the Mirbat Group close to Mirbat in Dhofar,
southern Oman. The Mirbat Group is most likely a correlative of the Abu Mahara Group of the Huqf Supergroup of northern Oman.
The Ayn Formation, the main subject of this chapter, comprises ,400 m of mainly coarse-grained glaciogenic deposits, ponded in 2-
to .8-km-wide N- to NW-oriented palaeovalleys eroded into crystalline basement, with few or no deposits preserved on intervening
palaeohighs. The Shareef Formation occurs as thin, lenticular, erosional remnants beneath the unconformably overlying Cretaceous.
The Ayn Formation is overlain by a thin (,3 m), discontinuous cap carbonate that passes from carbonate-cemented talus on the
basin margin to stromatolitic carbonate on palaeohighs and resedimented gravity flows on palaeovalley flanks.
The Ayn Formation is younger than its youngest detrital zircons and the youngest late plutons in crystalline basement, constraining it
to ,c. 720 Ma, but its exact age is unknown. The detrital zircon population comprises exclusively Neoproterozoic sources, suggesting
derivation from the juvenile Neoproterozoic crust of the Arabian area.
The composition of fine-grained matrix in glaciogenic diamictite units and of non-glacial mudstones, plotted using the chemical index
of alteration (CIA), suggests strong variations in the intensity of palaeoweathering on contemporary land surfaces between the mechan-
ical weathering-dominated Ayn Formation, and the chemical weathering-dominated overlying Arkahawl Formation, which supports the
notion of major glaciation followed by rapid climatic transit as basin margins were flooded and buried with sediment during post-glacial
transgression. The carbon isotopic ratio (d13C) of the post-glacial carbonate is strongly variable from 23.5‰ to þ5.8‰, whereas
carbonate fissures in the underlying basement range between þ4.1‰ and þ5.7‰.
Two independent palaeomagnetic studies have yielded low palaeomagnetic latitudes for the Mirbat Group.

Precambrian sedimentary rocks have been known in Oman since
early surveys of the Dhofar area of southern Oman (Carter 1852;
Lees 1928) when the Mirbat Sandstone was recognized. Subse-
quently, a fourfold division of the sedimentary rocks in the Huqf
area of east-central Oman was devised (Henson & Elliot 1958;
Morton 1959; Beydoun 1960, 1964), and the Neoproterozoic –
Early Cambrian succession referred to as the Huqf Group (Glennie
1977; Gorin et al. 1982). The Huqf Group was later raised in status
to a Supergroup, comprising Abu Mahara, Nafun and Ara Groups
(Glennie et al. 1974; Gorin et al. 1982; Hughes-Clarke 1988;
Wright et al. 1990), the former containing glaciomarine deposits
in the Jabal Akhdar region of northern Oman. Rieu et al. (2006,
2007a) and Allen (2007) recommended the use of the term
Mirbat Group, to include the Ayn, Arkahawl, Marsham and
Shareef Formations, in Dhofar. The Mirbat Group is unfossilifer-
ous and comprises ,400-m-thick glacially influenced deposits
of the Ayn Formation, overlain by c. 1 km of non-glacial marine
deposits of the Arkahawl and Marsham Formations, followed by
a ,40-m-thick unit of diamictite of the Shareef Formation.
The precise correlation of the Mirbat Group with the Huqf
Supergroup of east-central Oman and the Jabal Akhdar of northern
Oman has been a subject of ongoing debate, largely because of the
inadequacy of the geochronological database. Recent publications
(Rieu et al. 2006, 2007a; Allen 2007) correlate the Mirbat Group of
Dhofar with the Abu Mahara Group of northern Oman, whereas
Kilner et al. (2005) correlated it with the Nafun Group.
The glaciogenic origin of the Ayn Formation was first recog-
nized by Qidwai et al. (1988) and the central part of its outcrop
area was studied sedimentologically by Kellerhals (1993) and
Kellerhals & Matter (2003), who reported a wide range of facies
including subglacial, glaciomarine, fluvial, deltaic, turbiditic and
lacustrine deposits. Rieu et al. (2006) focused on a number of pre-
viously undescribed outcrops and interpreted the Ayn Formation in
the context of the whole outcrop area north and west of Mirbat.
Many Neoproterozoic sedimentary successions are preserved in
marine basins strongly influenced by subaqueous gravity flows,
leading to an active debate on the glacial v. non-glacial origin of
the poorly sorted deposits labelled diamictites (Schermerhorn
1974; Hambrey & Harland 1981; Young & Gostin 1991; Young
1992; Ross et al. 1995; Arnaud & Eyles 2002b; Allen et al.
2004; Eyles & Januszczak 2004, 2007). In contrast, the Ayn For-
mation records a terrestrial to marginal marine succession, and
therefore provides a valuable example of a glaciated basin margin
where glaciofluvial and glaciodeltaic deposits are well developed.
The Ayn Formation is overlain by a discontinuous cap carbonate
that exhibits large and abrupt lateral changes in sedimentary facies.
The presence of glaciogenic diamictites in thin (,40 m) iso-
lated remnants beneath the erosional sub-Cretaceous unconfor-
mity in Jabal Samhan was recognized by Kellerhals (1993) and
confirmed by Rieu (2006), who termed them the Shareef
Formation. Their presence demonstrates a second glacial epoch
preserved in the Mirbat Group, but the paucity of outcrop limits
further studies.
Structural framework and basement geology
The Mirbat Group crops out superbly along a 20 km NE –SW strik-
ing escarpment near the town of Mirbat (Fig. 19.1). It comprises up
to 1.5 km of weakly deformed, almost entirely siliciclastic rocks
that generally dip c. 108 to the NW. In the northeastern part of
the outcrop belt, a sub-vertical fault juxtaposes the lower part of

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 239– 249. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.19
240 P. A. ALLEN ET AL.

Fig. 19.1. (a) Outcrop areas of Neoproterozoic basement and sedimentary rocks in Oman. (b) Geological map of the Mirbat area showing the location of palaeovalleys
discussed in the text. (c) Summary stratigraphic column of the Mirbat Group. Based on Rieu et al. (2006) and Rieu et al. (2007a, b).

the Mirbat Group and basement with the sub-Cretaceous unconfor-
mity (see map in fig. 3 of Rieu & Allen 2008), and in the southern
extremity of the outcrop belt small thrust faults cutting the Mirbat
granodiorite fold the overstepping carbonate and terminate
upwards in the Arkahawl Formation. In general, however, the
Mirbat area escaped the deformation associated with the
‘Western Deformation Front’ to the NW (Allen 2007).
It has been suggested that the Mirbat Group was formed on the
southeastern flank of a c. 1000-km-long NE –SW striking rift basin
imaged from the subsurface (Loosveld et al. 1996). The general
sediment transport direction in the Mirbat Group derived from
palaeocurrent data was to the NW (Kellerhals 1993; Kellerhals
& Matter 2003; Rieu et al. 2006, 2007b; Rieu & Allen 2008).
This direction would therefore correspond to a transverse supply
from a margin situated to the SE.
The Mirbat Group unconformably overlies a complex crys-
talline and metamorphic basement (Platel et al. 1992a, b; Hauser
& Zurbriggen 1992, 1994; Würsten 1994; Worthing 2005; Mer-
colli et al. 2006). Four major units are recognized in the basement
at Mirbat (Mercolli et al. 2006; Bowring et al. 2007): (i) the Juffa
 THE AYN FORMATION
Group, composed of metasedimentary mica-gneisses and amphi-
bolites with a .1300 Ma (Mesoproterozoic) siliciclastic sedimen-
tary protolith; (ii) the Sadh Group, comprising highly deformed
and metamorphosed banded gneisses intruded by slightly
deformed diorites and tonalites (Mahall Complex); (iii) the Tona-
lite Group, composed of three large calc-alkaline plutons dated
780– 800 Ma; and (iv) the Granite Group, comprising different
types of dykes and small intrusive bodies including the Pegmatite
Complex (770 –750 Ma), and the Mirbat Granodiorite
(706 + 40 Ma), Leger Granite (726 + 0.4 Ma) and Shaat Dyke
Swarm (c. 750 –700 Ma). A phase of accretion of juvenile Neopro-
terozoic crust (.900 Ma) was followed by extensive calc-alkaline
magmatism, amphibolite facies metamorphism and intense defor-
mation culminating at c. 800 Ma. Worthing (2005) believed the
NW –SE orientated calc-alkaline Shaat Dyke Swarm to originate
from subduction-modified lithospheric mantle, mobilized by
crustal extension. The dykes appear to be similar in orientation
and geochemistry to a dyke swarm (c. 700 Ma) in the Al Bayda
island-arc terrane of Yemen (Windley et al. 1996). Late stocks
such as the Mirbat Granodiorite (and probably the Leger
Granite) and the Shaat dykes were most likely exhumed by the
time of deposition of the oldest sedimentary rocks of the Mirbat
Group (Qidwai et al. 1988; Platel et al. 1992a, b; Kellerhals &
Matter 2003; Rieu et al. 2006, 2007a).
Stratigraphy
The ,1.5-km-thick Mirbat Group comprises a lowermost glacial
interval (,400-m-thick Ayn Formation) occupying deep palaeo-
valleys incised in crystalline basement (Fig. 19.1). The overlying
Arkahawl Formation records at its base a major post-glacial trans-
gression over the previous basin margin, followed by a 300–
400-m-thick turbidite complex consisting of 1–5-km-wide lobes
embedded in fine-grained distal fan deposits (Rieu & Allen 2008).
The turbidite complex gradationally passes up into c. 500 m of
distal marine mudstone and siltstone. The overlying c. 100-
m-thick Marsham Formation records the progradation of shallow
marine and fluviatile deposits, before a second glaciation rep-
resented by the Shareef Formation, barely preserved as thin rem-
nants beneath the sub-Cretaceous unconformity.
Glaciogenic deposits and associated strata
In the Jabal Akhdar of the Oman Mountains, Neoproterozoic gla-
ciomarine deposits are found in the Abu Mahara Group (Ghubrah
and Fiq Formations), occupying the tectonically deformed core
of a major east –west oriented anticline, with no basement rocks
exposed (Allen et al. 2011). In contrast, the Abu Mahara-
equivalent glaciogenic sedimentary rocks of the Ayn and Shareef
Formations of the Mirbat Group of Dhofar are relatively unde-
formed and occur above an irregular crystalline basement (Keller-
hals & Matter 2003; Rieu et al. 2006). Rieu et al. (2007a) proposed
that the Arkahawl and Marsham Formations represent the missing
stratigraphy between the Ghubrah and Fiq Formations in the Jabal
Akhdar.
Ayn Formation
The Ayn Formation comprises ,400 m of mainly coarse-grained
glaciogenic deposits, ponded in 2- to .8-km-wide N- to NW-
oriented palaeovalleys eroded into basement, with few or no
deposits preserved on intervening palaeohighs. Sedimento-
logical analysis of near-continuous sections through the three
westernmost palaeovalley-fills of the Ayn Formation (Fig. 19.2)
allow four facies associations to be distinguished: (i) subaerial,
241
(ii) fluviodeltaic, (iii) proximal glaciomarine and (iv) distal glacio-
marine. Details can be found in Rieu et al. (2006).
Subaerial facies association. Units of the subaerial facies associ-
ation, generally ,1 m thick, unconformably overlie crystalline
and metamorphic basement at many localities, and comprise
in situ brecciated bedrock interpreted as regolith, and poorly
sorted, clast-supported breccia and crudely stratified diamictite
thought to represent slope deposits. Clast type commonly reflects
the directly underlying bedrock lithology. There is no evidence
such as striated clasts or an underlying striated pavement that sup-
ports a glacial origin of the diamictites in this facies association.
Fluviodeltaic facies association. The fluviodeltaic facies association
comprises mainly coarse-grained sandstones and conglomerates
that make up the lower part of the stratigraphy of palaeovalleys
1, 2 and 3. In palaeovalley 3, this facies association is also found
higher up in the stratigraphy in two 20–30-m-thick units, alter-
nated with glaciomarine deposits. Characteristic architectural
elements include ,25-m-wide and ,2-m-thick channel fills,
and 10– 15-m-thick, stacked tabular foreset beds with a c. 258
apparent dip.
Massive, graded and cross-stratified sandstones and conglomer-
ates are thought to represent low-sinuosity, bedload-dominated
fluvial deposits, whereas the stacked tabular units are Gilbert-type
deltas that identify the delta front. Interbedded siltstones were
deposited by hemipelagic fallout in pro-delta bottomsets, and
dropstone-bearing silty laminites demonstrate rainout from float-
ing ice (Ovenshine 1970; Smith 2000) in a background glaciomar-
ine or glaciolacustrine setting characterized by dilute underflows
or interflows (e.g. Mackiewicz et al. 1984; Powell & Molnia 1989).
In the central outcrop belt (palaeovalley 4), Kellerhals & Matter
(2003) describe glacial striations preserved within the fluviodeltaic
facies association, demonstrating dynamic glacier advance over
fluvioglacial environments.
Proximal glaciomarine facies association. Coarse-grained proximal
glaciomarine deposits constitute the bulk of the upper Ayn strati-
graphy throughout the area. Deposits of this facies association
overstep palaeovalley margins and may comprise single units
that reach up to c. 150 m in thickness in palaeovalley 4 (Kellerhals
& Matter 2003). Massive and weakly stratified diamictites with
abundant striated clasts are the most common lithology in this
facies association. Stratified diamictites contain common outsized
clasts (,1 m) deflecting underlying lamination, and many striated
and/or faceted clasts. Interbedded sandstones and conglomerates
locally fill channels or occur as pebble lags.
The presence of many large outsized, striated and faceted clasts
is suggestive of rainout from floating ice (e.g. Ovenshine 1970;
Syvitski et al. 1996; Smith 2000) of debris carried by subglacial
transport prior to deposition (Boulton 1978; Benn & Evans
1998). Conglomerates within stratified diamictite units represent
either local concentrations of clasts derived from iceberg overturn
(Ovenshine 1970), lag deposits resulting from intensive wave or
current reworking of ice-rafted diamictite in relatively shallow
marine conditions (Eyles et al. 1985; Eyles 1988; Moncrieff &
Hambrey 1990) or debris flows of remobilized sediment (Wright
& Anderson 1982). Channel bodies comprising graded and cross-
stratified sandstones ‘embedded’ in diamictites may have been
deposited by currents issuing from meltwater tunnel outlets
(Rust & Romanelli 1975; Banerjee & MacDonald 1975; Powell
1981). The entire facies association suggests a proximal glacio-
marine environment, close to the grounded ice-margin (Eyles
et al. 1985; Moncrieff & Hambrey 1990; Syvitski et al. 1996),
possibly in a grounding-line fan (Powell 1990; Lønne 1995).
Distal glaciomarine facies association. The distal glaciomarine
facies association consists mainly of laminated mudstones with
subordinate sandstone laminae. Outsized, commonly striated, clasts
 242
P. A. ALLEN ET AL.
Fig. 19.2. Summary sedimentological logs of the Ayn Formation from palaeovalleys 1 to 4 and interpretive correlation panel (after Rieu et al. 2006). (a) Logged sections through palaeovalleys and possible correlation (section
PV-4 modified after Kellerhals & Matter, 2003). Logs; m, mudstone; s, sandstone; g, gravel/conglomerate. Facies notation follows Rieu et al. (2006). (b) Lateral correlation of palaeovalley fills and elevation of basement, based
on logged sections and field mapping. Field map in Rieu (2006) and Rieu & Allen (2008).
 THE AYN FORMATION
deflecting laminae occur throughout this facies association, but are
much less common (c. 1%) than in the stratified diamictites of the
proximal glaciomarine facies association. Distal glaciomarine
facies also include graded sandstones and rare gravel and thin
pebble conglomerates. The predominance of fine-grained lami-
nated deposits points to a relatively quiet depositional environment
dominated by hemipelagic sedimentation. Sediment gravity flow
deposits are more sheet-like and thinner than those found in the
proximal glaciomarine facies association, which together with the
low abundance of outsized clasts, interpreted as ice-rafted debris,
supports a relatively distal glaciomarine depositional environment
(e.g. Mackiewicz et al. 1984; Eyles et al. 1985; Syvitski et al. 1996).
In the central outcrop area (palaeovalley 4), Kellerhals & Matter
(2003, p. 60) describe graded, turbiditic sandstones with drop-
stones at the tops of beds, interleaved with diamictite units, demon-
strating contemporaneous sediment gravity flow processes and
rainout from melting icebergs.
Post-glacial carbonate above Ayn Formation
The post-glacial Arkahawl Formation, in contrast to the Ayn For-
mation, is laterally extensive and covers all remaining basement
relief left after deposition of the Ayn Formation (Rieu 2006). At
the base of the Arkahawl Formation is a ,3-m-thick transgressive
cap carbonate that sharply overlies the underlying glaciomarine or
fluviodeltaic deposits of the Ayn Formation or, where these are
absent, directly overlies crystalline basement (Rieu et al. 2006).
This carbonate is locally overlain by marine sandstones but more
commonly by a 30 –40-m-thick succession of distal marine
shales, indicating rapid deepening of water depths.
Carbonate facies. The carbonate is laterally discontinuous and
varies considerably in thickness and sedimentary facies. Carbon-
ates are locally well developed on palaeohighs, whereas they are
absent or represented by discontinuous, mixed carbonate – silici-
clastic mass flow deposits in the deeper parts of the basin (i.e. in
palaeovalleys). The carbonate comprises shallow-water, mass-
flow and fissure-filling facies.
† Shallow-water facies association. In the western part of the
outcrop area, onlapping onto crystalline basement or thin
breccia, the carbonate is mainly a pink, recrystallized micritic
dolomite, passing upward into wave-rippled calcareous sand-
stones and then a .1.5-m-thick, generally massive dolomite.
In the centre of the outcrop area a locally well-developed car-
bonate with a basal conglomeratic lag overlies fluviodeltaic
deposits. The carbonate is a dark grey, locally organic-rich,
recrystallized micritic limestone showing mainly millimetre-
thin, undulating laminations, small-scale symmetrical wave
ripples (eastern part of palaeovalley 3) and ,1-m-high,
elongate domal stromatolites (centre of palaeovalley 3),
which together indicate shallow-water deposition.
† Carbonate mass-flow facies association. In palaeovalleys 1, 2
and 4, the carbonate is either absent or represented by
carbonate-rich mass flow deposits containing a significant
(,c. 75%) amount of siliciclastic material. Mass-flow deposits
include centimetre-thick graded beds, interpreted as turbidites,
decimetre-thick beds with soft-sediment deformation, inter-
preted as slump deposits, and poorly sorted conglomerates inter-
preted as debris flows. A ,15-m-thick erosive unit of boulder
and cobble conglomerates overlies basement and glacial depos-
its in palaeovalley 1, containing a matrix of siliciclastic sand and
detrital carbonate. The presence of bedding-parallel veneers
of carbonate beneath and above this unit indicates that the con-
glomerates, which contain rare carbonate clasts, were deposited
while carbonate was being produced in the basin.
† Carbonate-filled fissures. In the western part of the area, a
slightly foliated granodiorite (Mirbat Granodiorite) is host to
243
several populations of cross-cutting, millimetre-to-centimetre-
wide and ,10 m deep, fissures filled with detrital carbonate
(dolomite) mixed with minor amounts of sand-sized siliciclas-
tic particles, and void-filling cements. The fills of the fissures
originate from the overlying carbonate (dolomite), which is
further supported by their similarity in C-isotopic values (see
below). In one location a small-scale (,2 cm high) set of fore-
sets of a ‘mini delta’ developed over a centimetre-high negative
step in the wall of the fracture indicates pumping of sediment
and fluids upwards through a connected fracture system (Rieu
et al. 2006). Fissures were evidently opened up at the same
time as carbonate deposition during post-glacial transgression.
Boundary relations with overlying and underlying
glacial strata
The Mirbat Group is sandwiched between a basal unconformity
represented by a series of deep palaeovalleys cut into crystalline
basement, exposed in the bevelled coastal plain close to Mirbat
town, and a regional sub-Cretaceous planar unconformity
exposed high in the cliffs below Jabal Samhan. A map of the
Group and its boundary relations is found in Rieu & Allen
(2008, fig. 3).
Chemostratigraphy
Carbon-isotopic composition of the post-glacial carbonate
Rieu et al. (2006) measured the carbon isotopic composition of
inorganic carbonate (d13C) in the carbonate at five different sites
across the area and in the carbonate-filled fissures in basement
(Fig. 19.3). Tests for diagenetic effects are discussed in Rieu
et al. (2006). The most negative d13C values of the carbonate are
found in the central part of the field area where they increase
from 23.5‰ at the base to þ5.1‰ at the top in laminated and stro-
matolitic carbonate (sections C4 and C5). Further to the west, lami-
nated carbonate overlying basement shows an increasing trend from
þ0.9‰ at the base to þ2.7‰ at the top (section C3) and nearby car-
bonate associated with conglomerate shows upward-increasing
values from þ3.5‰ to þ4.3‰ (section C3). The most positive
values are found in the westernmost part of the area in carbonate-
filled fissures in crystalline basement (between þ4.1‰ and
þ5.7‰) and in the overlying laminated and wave-rippled carbon-
ate, which shows an upward decreasing trend from þ5.8‰ at the
base to þ2.0‰ at the top (sections C1 and C2).
The lateral variation in d13C in the carbonate is interpreted to
represent a secular variation of the C-isotopic composition of
ocean water. Owing to progressive flooding of the palaeotopogra-
phically highest parts of the basin (i.e. in the west), carbonate depo-
sition in these locations would have been delayed with respect to
the basin centre. The resulting d13C signal of the carbonate on
the basement high is therefore ‘base-truncated’. This interpretation
allows the construction of a composite isotope profile through the
carbonate, showing an increasing trend from 23‰ to þ5.8‰ and
then decreasing to þ2‰ at the top (Fig. 19.3). In the central area
(sections C4, C5; palaeovalley 3) values reach a maximum of
þ5.1‰, but the decreasing arm is missing (‘top-truncated’),
which is in accord with the observation that carbonate deposition
was restricted mainly to shallower environments.
Neoproterozoic weathering: Chemical Index of Alteration
Rieu et al. (2007b) studied the compositional and mineralogical
changes in the Mirbat Group in order to test the possible effect
of severe climate swings in the Neoproterozoic on the intensity
of chemical weathering in source areas (Nesbitt & Young 1982;
244 P. A. ALLEN ET AL.

Fig. 19.3. Summary logs, d13C data and major element data of the post-glacial carbonate (after Rieu et al. 2006). See Figure 19.2 for location of C1 –C5. (a) Isotopic
profiles and summary logs are shown in their relative vertical positions, high in the SW where the cap carbonate onlaps basement, and low in the centre of the outcrop area
where the cap overlies fluvial deposits. (b) Composite isotopic profile, with sections C1 and C4 compressed in order to account for different sedimentation rates and to
match sections C2 and C5, respectively. Dashed arrows indicate the relation between cap carbonate and carbonate fissures. (c) Element concentrations for selected
samples from composite section.
 THE AYN FORMATION
McLennan et al. 1993; Fedo et al. 1995; Nesbitt et al. 1996; Schef-
fler et al. 2003; Bahlburg & Dobrzinski 2011). A similar study of
Neoproterozoic sedimentary rocks using the chemical index of
alteration (CIA) is found in Dobrzinski et al. (2004). To ensure a
well-mixed provenance and to minimize the effects of hydrodyn-
amic sorting, Rieu et al.’s (2007b, c) study was limited to mud-
stone beds and the mudstone matrices of diamictites.
Throughout the Mirbat Group there are significant compo-
sitional and mineralogical variations (Fig. 19.4). When plotted in
A –CN –K and Qtz – Pl– Kfs space, the data define trends parallel
to A – CN and Qtz –Pl boundaries, thus suggesting that compo-
sitional variations result from variability in the extent of chemical
weathering in the regolith in source areas. Glacial deposits of the
Ayn Formation (AY2 –AY8) plot close to the feldspar join and
close to the composition of unaltered granodiorite, indicating
that their composition has been little affected by chemical altera-
tion, as reflected by their low average CIA value of 54. Pre-glacial
deposits of the Ayn and Marsham Formation, and non-glacial
deposits of the Arkahawl Formation in particular, are enriched in
Al2O3. This enrichment suggests the incorporation of weathered
material, in agreement with the relatively high abundance of
clay minerals, and is reflected by average CIA values of 60 and
74 for the pre-glacial and non-glacial intervals, respectively. In
Qtz – Pl– Kfs space, a similar trend is revealed, with samples
from the glacial, pre-glacial and non-glacial intervals, respect-
ively, being progressively enriched in quartz with respect to
plagioclase, resulting in progressively higher average MIA (miner-
alogical index of alteration ¼ Qtz/(Qtz þ Kfs þ Pl)) values (18,
36 and 46, respectively). Rieu et al. (2007b) believed that these
CIA
unconformity 50 60 70 0 20
glacial epoch II
Mhm Shf
1200
Non-glacial interlude
or ‘interglacial’
Arkahawl Fm
800
cap carbonate
400
Ayn Fm
glacial epoch I
0m
50 60 70 0 20
CIA
245
trends were best explained by variations in the intensity of palaeo-
weathering of contemporary land surfaces.
Palaeolatitude and palaeogeography
Kilner et al. (2005) reported palaeomagnetic data that passed fold
and reversal tests from Oman, which gave the Muscat region a
palaeolatitude of 138 in the late Neoproterozoic. They recognized
a remnant magnetization that was acquired between deposition and
a period of deformation in the Cambrian – Ordovician (Blendinger
et al. 1990; Gass et al. 1990), and argued (p. 415) that the magne-
tization was acquired at deposition since a similar pattern of rever-
sals occurs at the same stratigraphic level in widely separated
locations in the north, centre and south of Oman. Subsequent
work, however, points strongly to different ages for these strati-
graphic sections (Allen & Leather 2006; Rieu et al. 2006; Rieu
et al. 2007a). The difference between the Oman palaeopoles and
those of Gondwanan terranes (Meert 2003) suggested to Kilner
et al. (2005) that Oman did not accrete with the ‘African’ terranes
comprising greater Gondwana until after c. 600 Ma, the approxi-
mate age of the youngest stratigraphy sampled for palaeomagnet-
ism. Kempf et al. (2000) recovered a Precambrian pole position
from the south of Oman (implying a palaeolatitude of 98) but
noted a similarity with published poles from Gondwana (Congo,
India, Australia) at c. 550 Ma, which was thought to be the deposi-
tional age. The rocks in south Oman providing the low palaeolati-
tude are c. 700 Ma in age, not 550 Ma (Rieu et al. 2006, 2007a),
which suggests that either the south of Oman was in low
MIA
40 60 80
decreasing
temperatures
rapidly
increasing
temperatures
Fig. 19.4. Variations in weathering indices
(CIA and MIA) in the Mirbat Group,
displayed as a function of their stratigraphic
height (from Rieu et al. 2007b). Low CIA
and MIA values are associated with glacial
and pre-glacial conditions inferred from
sedimentological facies, and relatively high
CIA and MIA values are associated with
40 60 80
non-glacial marine deposits. Errors on CIA
MIA are ,2%.
246 P. A. ALLEN ET AL.

palaeolatitudes at c. 700 Ma and experienced similar palaeolati-

tudes at 550 Ma, or that the magnetization was acquired during a
later event close to 550 Ma, when Oman and the adjacent parts
of Gondwana were indeed situated in the tropics on the basis of
their common evaporitic sedimentary facies (Allen 2007). Geo- 870
logical evidence summarized in Allen (2007) indicates an earlier
accretion of Oman into the megasuture forming the eastern
border (in present-day coordinates) of Gondwana. Currently,
therefore, the palaeomagnetic data described by Kempf et al.
(2000) and Kilner et al. (2005) lack a rigorous stratigraphic and
geochronological context. 810
Shareef Fm
(glaciation)
Geochronological constraints
n = 177 (69)
The age of the Mirbat Group is uncertain. The maximum age must
be younger than the age of the youngest intrusions in the basement
complex truncated by the basal unconformity. The carbonate
stratigraphically above the Ayn Formation is therefore younger
than the age of the underlying Mirbat Granodiorite (706 +
40 Ma, Rb –Sr date, Gass et al. 1990). The Leger Granite
(726 + 0.4 Ma U –Pb zircon date, Bowring et al. 2007) cannot 810
be proven to pre-date the Mirbat Group because it is truncated 840

Mirbat Group
by the sub-Cretaceous unconformity to the NE of the outcrop
belt of the Mirbat Group, but it is nevertheless highly likely (see
below). These late intrusions were most likely exhumed by the
time of deposition of the oldest sedimentary rocks of the Mirbat
Group (Qidwai et al. 1988; Platel et al. 1992a, b; Kellerhals & Arkahawl +
Matter 2003; Rieu et al. 2006, 2007a). The dykes at Mirbat do Marsham Fms
not cut the unconformably overlying sedimentary Mirbat Group,
as recognized by Qidwai et al. (1988), so the Mirbat Group sedi-
mentary rocks must be younger than the 696 –744 Ma age of the
dykes derived from Sm– Nd and Rb – Sr dating (Worthing 2005). n = 92 (59)
In order to further constrain the age of the Mirbat Group and the
contributing source areas, use has been made of detrital zircon geo-
chronology. A total of 1057 new U– Pb detrital zircon ages were
produced by laser ablation ICP-MS from the Huqf Supergroup
(Rieu et al. 2007a; Allen 2007). The sampled stratigraphic inter- 840
vals in the Mirbat Group include the Ayn, Arkahawl, Marsham
810
and Shareef Formations. Samples of the Ayn Formation come
from three different diamictite units (D1, D2 and D3 of Kellerhals
& Matter 2003) in Wadi Autunt. The sample of the Arkahawl For-
mation was obtained from turbiditic sandstones in Wadi Hinuna,
c. 75 m above the top of the Ayn Formation. From the Marsham
Formation two samples were collected near Jabal Shareef from
shoreface sandstones. Diamictites of the Shareef Formation were Ayn Fm
sampled in the same section, as well as in a section 3 km to the NE. (glaciation)
The stratigraphic units of the Mirbat Group contain remarkably 722
similar detrital zircon populations, comprising zircons almost n = 180 (111)
exclusively of Neoproterozoic age (Rieu et al. 2007a) (Fig. 19.5).
When only the most concordant data and those with the lowest
analytical errors are considered, main peaks in the Ayn Formation
are revealed at c. 722, 810 and 840 Ma. In the Arkahawl and 0.5 1.0 1.5
Marsham formations, only the c. 810 Ma and c. 840 Ma peaks
are present, whereas in the Shareef Formation the zircon popu-
lation is dominated by c. 870 Ma ages with only a small ‘shoulder’ Age (x109 years)
at c. 810 Ma. The maximum age constraint for the depositional age
Fig. 19.5. Frequency distribution of detrital zircons from the Mirbat Group
of the Ayn Formation and the overlying part of the Mirbat Group is (modified from Rieu et al. 2007a). Grey areas are data produced by filter 1
provided by the c. 722 Ma subpopulation in the Ayn Formation, (,25% discordant; 2s errors ,20% for 235U– 207Pb and ,10% for
represented by four grains with a mean age of 722 +19 Ma. 238
U – 206Pb). Black lines are data produced by filter 2 (,10% discordant; 2s
This age is in agreement with the constraints provided by the errors ,10% for 235U – 207Pb and 238U– 206Pb). Number of analyses (n)
700–750 Ma age of the underlying basement based on Rb – Sr, produced with filters 1 and 2 are displayed outside and inside parentheses
K –Ar and Sm– Nd whole-rock and mineral analyses (Gass et al. respectively. Numbers for peaks are in Ma.
1990; Worthing 2005; Mercolli et al. 2006), but is considered to
be more robust.
The detrital zircon populations are dominated by Neoprotero- the Leger Granite (726 + 0.4 Ma), the 722 +19 Ma subpopu-
zoic ages that are in good agreement with the age range (0.7 – lation of zircons in the Ayn Formation may have been derived
1.0 Ga) of local basement rocks in the Mirbat area. Although from this or associated intrusions, suggesting that the Leger
there is no stratigraphic contact between the Mirbat Group and Granite intruded prior to deposition of the Ayn Formation and
 THE AYN FORMATION
was exhumed shortly after. This subpopulation is similar in age to
the zircons found in tuffaceous ashes of the Ghubrah Formation in
northern Oman (Brasier et al. 2000).
Discussion: the nature of Neoproterozoic glaciation
The Ayn stratigraphy suggests that in basin-marginal settings,
periods of glaciomarine deposition alternated with periods of
fluvial and deltaic deposition during glacial recession. Thus, terres-
trial sediment routing systems functioned strongly during these
periods of recession, by which plentiful sediment could be trans-
ported by high-energy fluvial systems and delivered to more
distal, hemipelagic-dominated areas. CIA analysis suggests a
dominance of mechanical weathering at this time. Although the
simplest interpretation is that the entire glacial epoch was charac-
terized by dynamic glacier advance and retreat cycles, it is possible
that preservation of stratigraphy took place preferentially during a
long, pulsed deglacial phase. The stratigraphy of the Ayn For-
mation is most easily reconcilable with ‘soft snowball’ models
allowing for ice margin fluctuations and the presence of equatorial
oceans (Hyde et al. 2000; Crowley et al. 2001; Peltier et al. 2007).
A central tenet of the Snowball Earth hypothesis is the occur-
rence of laterally extensive cap carbonates, which sharply
overlie most Neoproterozoic glacial deposits, with typically nega-
tive C-isotopic ratios that have been used for global correlation
(Kaufman & Knoll 1995; McKirdy et al. 2001; Halverson et al.
2005). Cap carbonates have long been considered paradoxical
because they suggest an abrupt change from glacial to tropical con-
ditions (Hoffman & Schrag 2002, and references therein). Models
requiring less extreme environmental changes have been proposed
to explain the presence of cap carbonates (Grotzinger & Knoll
1995; Kennedy 1996; Hoffman et al. 1998; Kennedy et al. 2001;
Jiang et al. 2003). Eyles & Januszczak (2004) doubted that cap
carbonates had any precise palaeoenvironmental significance,
and claimed that many of them are made up of detrital carbonate
derived by erosion rather than representing a distinct oceano-
graphic condition.
It is important to stress first that the Mirbat cap carbonate is
clearly associated with a significant palaeoclimatic change,
because it marks the termination of a glacial epoch. Second, a
detrital origin is unlikely, because it occurs within a .1-km-thick
succession devoid of any other primary or detrital carbonate.
However, the cap carbonate described here contrasts with many
other cap carbonates worldwide, as it is discontinuous and shows
considerable variations in facies, thickness and C-isotopic signal
over very short lateral distances (hundreds of metres). Rieu
et al.’s (2006) data suggest that carbonate was produced in the shal-
lower parts of the basin and redistributed downslope from these
palaeohighs, rather than as a blanket of carbonate precipitated
from an oversaturated ocean (Hoffman et al. 1998; Hoffman &
Schrag 2002; Shields 2005). Carbonate production therefore
appears to have been determined by the same depth-dependent
effects as we see in the photosynthesis-mediated processes of
Phanerozoic carbonate production. If carbonate precipitation took
place beyond the shallow waters of the palaeohighs, steep submarine
relief in the palaeovalleys may have prevented the primary depos-
ited carbonate from being preserved in situ.
Conclusions
The Ayn Formation of the Mirbat Group is a c. 400-m-thick suc-
cession of glaciofluvial, glaciodeltaic and glaciomarine sediments
deposited in several deeply incised palaeovalleys cutting the meta-
morphic and igneous basement in the Mirbat area of Dhofar. The
Formation is overlain by a discontinuous, thin (,3 m) and dia-
chronous cap carbonate that marks an abrupt overstepping of the
basin margin and deepening of palaeoenvironments. The Shareef
247
Formation is a second, little-known glaciogenic unit preserved
patchily beneath the sub-Cretaceous unconformity.
The sedimentary evolution of the Mirbat Group is paralleled
by variations in the CIA derived from mudstones and the fine-
grained matrices of diamictites. The CIA suggests conditions
with minimal chemical weathering during the deposition of the
Ayn Formation, followed by intensive weathering in the overlying
Arkahawl Formation above the cap carbonate. This geochemical
trend strongly supports the Ayn Formation as glaciogenic, fol-
lowed by climatic transit as basin margins were flooded during
post-glacial transgression.
The age of the Ayn Formation is constrained by (i) the age
of the youngest detrital zircons (four zircons with a mean of
722 + 19 Ma), (ii) the age of the Shaat dyke swarm cutting
basement and which does not penetrate the Ayn Formation
(696 –744 Ma), (iii) the age of late plutons such as the Mirbat
Granodiorite (706 + 40 Ma), which is definitely erosionally over-
lain by cap carbonate and the Leger Granite (726 + 0.4 Ma), which
is most likely but not definitely older than the Ayn Formation. It is
therefore highly likely that the Ayn Formation is younger than
c. 720 Ma. The age of the Shareef Formation is unknown.
The Mirbat Group contains detrital zircons of exclusively Neo-
proterozoic age, closely matching the range of ages of meta-
morphic and igneous rocks in the underlying basement, which
formed during a period of major crustal growth associated with
subduction at the southeastern periphery of the Arabian-Nubian
Shield.
The senior author is grateful for the financial and logistical support of Petroleum
Development Oman (PDO) and Enterprise Ireland, supplemented in later years by
the Research Committee of ETH-Zürich. We are grateful for the constructive
input on field visits of J. Grotzinger, S. Bowring, P. Hoffman, G. Halverson,
PDO staff (particularly J. Amthor) and the participants of the field trip associated
with the IAS Regional Meeting in Muscat in 2005. I am grateful for the reviews
of N. Dobrinski, I. Fairchild and E. Arnaud. This represents a contribution of
the IUGS- and UNESCO-funded IGCP (International Geoscience Programme)
Project #512.
References
Allen, P. A. 2007. The Huqf Supergroup of Oman: basin development
and context for Neoproterozoic glaciation. Earth Science Reviews,
84, 139– 185.
Allen, P. A. & Leather, J. 2006. Siliciclastic marine sedimentation
in the aftermath of a Marinoan glacial epoch: the Masirah Bay
Formation, Huqf Supergroup, of Oman. Precambrian Research,
144, 167–198, doi: 10.1016/j.precamres.2005.10.006.
Allen, P. A., Leather, J. & Brasier, M. D. 2004. The Neoproterozoic
Fiq glaciation and its aftermath, Huqf Supergroup of Oman. Basin
Research, 16, 507– 534, doi: 10.1111/j.1365-2117.2004.00249.x.
Allen, P. A., Leather, J. et al. 2011. The Abu Mahara Group (Ghubrah
and Fiq Formations), Jabal Akhdar, Oman. In: Arnaud, E.,
Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record
of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 251–262.
Arnaud, E. & Eyles, C. H. 2002. Glacial influence on Neoproterozoic
sedimentation: the Smalfjord Formation, northern Norway. Sedimen-
tology, 49, 765– 788.
Bahlburg, H. & Dobrzinski, N. 2011. A review of the Chemical Index
of Alteration (CIA) and its application to the study of Neoproterozoic
glacial deposits and climate transitions. In: Arnaud, E., Halverson,
G. P. & Shields-Zhou, G. (eds) The Geological Record of Neo-
proterozoic Glaciations. Geological Society, London, Memoirs, 36,
81– 92.
Banerjee, I. & MacDonald, B. C. 1975. Nature of esker sedimentation.
In: Jopling, A. V. & McDonald, B. C. (eds) Glaciofluvial and Gla-
ciolacustrine Sedimentation. Society of Economic Paleontologists
and Mineralogists, Special Publication, 23, 132– 154.
Benn, D. I. & Evans, D. J. A. 1998. Glaciers and Glaciation. Arnold,
London.
248 P. A. ALLEN ET AL.
Beydoun, Z. R. 1960. Synopsis of the geology of East Aden Protectorate.
Report of the 21st International Geological Congress, Copenhagen,
Part 21, 131– 149.
Beydoun, Z. R . 1964. The stratigraphy and structure of the East Aden
Protectorate. Overseas geology and mineral resources supplement
series. Bulletin supplement, 5, 1– 107.
Blendinger, W., van Vliet, A. & Hughes Clarke, M. W. 1990.
Updoming, rifting and continental margin development during the
Late Palaeozoic in northern Oman. In: Robertson, A. H. F.,
Searle, M. P. & Ries, A. C. (eds) The Geology and Tectonics of
the Oman Region. Geological Society, London, Special Publication,
49, 27 –37.
Boulton, G. S. 1978. Boulder shapes and grain-size distribution of
debris as indicators of transport paths through a glacier and till
genesis. Sedimentology, 25, 773– 799.
Bowring, S. A., Grotzinger, J. P., Condon, D. J., Ramezani, J.,
Newall, M. J. & Allen, P. A. 2007. Geochronologic constraints
on the chronostratigraphic framework of the Neoproterozoic Huqf
Supergroup, Sultanate of Oman. American Journal Science, 307,
1097– 1145.
Brasier, M. D., McCarron, G., Tucker, R., Leather, J., Allen, P. A.
& Shields, G. 2000. New U– Pb zircon dates for the Neopro-
terozoic Ghubrah glaciation and for the age of the top of the Huqf
Supergroup, Oman. Geology, 28, 175–178, doi: 10.1130/0091-
7613(2000)028,0175:NUPZDF.2.2.CO;2.
Carter, H. J. 1852. Memoir on the geology of the south east coast of Arabia.
Journal of the Royal Asiatic Society, Bombay Br., 4, 21–96.
Crowley, T. H., Hyde, W. T. & Peltier, W. R. 2001. CO2 levels required
for glaciation of a ‘near-snowball’ Earth. Geophysical Research
Letters, 28, 283–286, doi: 10.1029/2000GL011836.
Dobrzinski, N., Bahlburg, H., Strauss, H. & Zhang, Q. 2004.
Geochemical climate proxies applied to the Neoproterozoic glacial
succession on the Yangtze Platform, South China. In: Jenkins,
G. S., McMenamin, M. A. S., McKay, C. P. & Sohl, L. (eds) The
Extreme Proterozoic: Geology, Geochemistry and Climate.
American Geophysical Union Monograph, 146, 13– 32.
Eyles, C. H. 1988. Glacially and tidally influenced shallow marine sedi-
mentation of the Late Precambrian Port Askaig Formation. Palaeo-
geography, Palaeoclimatology, Palaeoecology, 68, 1 –25.
Eyles, N. & Januszczak, N. 2004. Zipper-rift: a tectonic model for
Neoproterozoic glaciations during the break-up of Rodinia after
750 Ma. Earth Science Reviews, 65, 1 – 73, doi: 10.1016/
S0012-8252(03)00080-1.
Eyles, N. & Januszczak, N. 2007. Syntectonic subaqueous mass flows
of the Neoproterozoic Otavi Group, Namibia: where is the evidence
for global glaciation? Basin Research, 19, 179–198, doi: 10.1111/
j.1365-2117.2007.00319.x.
Eyles, C. H., Eyles, N. & Miall, A. D. 1985. Models of glacimarine
sedimentation and their application to the interpretation of ancient
glacial sequences. Palaeogeography, Palaeoclimatology, Palaeo-
ecology, 51, 15 –84.
Fedo, C. M., Nesbitt, H. W. & Young, G. M. 1995. Unravelling
the effects of potassium metasomatism in sedimentary rocks and
paleosols, with implications for paleoweathering conditions
and provenance. Geology, 23, 921– 924, doi: 10.1130/0091-
7613(1995)023,0921:UTEOPM.2.3.CO;2.
Gass, I. G., Ries, A. C., Shackleton, R. M. & Smewing, J. D. 1990.
Tectonics, geochronology and geochemistry of the Precambrian
rocks of Oman. In: Robertson, A. H. F., Searle, M. P. & Ries,
A. C. (eds) The Geology and Tectonics of the Oman Region. Geologi-
cal Society, London, Special Publications, 49, 585– 599.
Glennie, K. W. 1977. Outline of the geology of Oman. Mémoires de la
Societé Géologique de France, hors série, 8, 25– 31.
Glennie, K. W., Boeuf, M. G. A., Hughes Clark, M. W., Moody-
Stuart, M., Pilaar, W. F. H. & Reinhardt, B. M. 1974. Geology
of the Oman Mountains. KSEPL, Rijswijk, The Netherlands.
Gorin, G. E., Raacz, L. G. & Walter, M. R. 1982. Late Precambrian –
Cambrian sediments of Huqf Group, Sultanate of Oman. American
Association Petroleum Geologists Bulletin, 66, 2609– 2627.
Grotzinger, J. P. & Knoll, A. H. 1995. Anomalous carbonate pre-
cipitates: is the Precambrian the key to the Permian? Palaios, 10,
578– 596.
Halverson, G. P., Hoffman, P. F., Schrag, D. P., Maloof, A. C. &
Rice, A. H. N. 2005. Toward a Neoproterozoic composite carbon-
isotope record. Geological Society of America Bulletin, 117,
1181– 1207.
Hambrey, M. J. & Harland, W. B. 1981. Earth’s Pre-Pleistocene
Glacial Record. Cambridge University Press, Cambridge.
Hauser, A. & Zurbriggen, R. 1992. Geology of the Crystalline Basement
of the Hadbin area (Salalah area, Dhofar, Sultanate of Oman).
Unpublished MSc thesis, University of Berne, Switzerland, 6– 221.
Hauser, A. & Zurbriggen, R. 1994. Geology of the crystalline basement
of the Hadbin area (Salalah area, Dhofar, Sultanate of Oman).
Schweizerische Mineralogisch Petrographische Mitteilungen, 74,
213– 226.
Henson, F. R. S. & Elliot, G. F. 1958. Exhibition of specimens of
‘Collenia’ from pre-Permian sediments in south Arabia. Proceedings
Geological Society London, 1561, 89– 90.
Hoffman, P. F. & Schrag, D. P. 2002. The snowball Earth hypothesis:
testing the limits of global change. Terra Nova, 14, 129–155, doi:
10.1046/j.1365-3121.2002.00408.x.
Hoffman, P. F., Kaufmann, A. J., Halverson, G. P. & Schrag, D. P.
1998. A Neoproterozoic snowball Earth. Science, 281, 1342–1346,
doi: 10.1126/science.281.5381.1342.
Hughes Clarke, M. W. 1988. Stratigraphy and rock nomenclature
in the oil-producing area of interior Oman. Journal Petroleum
Geology, 11, 5 –60.
Hyde, W. T, Crowley, T. J., Baum, S. K. & Peltier, W. R. 2000. Neo-
proterozoic ‘snowball Earth’ simulations with a coupled climate/
ice-sheet model. Nature, 405, 425– 429, doi: 10.1038/35013005.
Jiang, G., Kennedy, M. J. & Christie-Blick, N. 2003. Stable isotopic
evidence for methane seeps in Neoproterozoic post-glacial cap car-
bonates. Nature, 426, 822– 826.
Kaufman, A. J. & Knoll, A. H. 1995. Neoproterozoic variations in the
C-isotopic composition of seawater: stratigraphic and biogeochem-
ical implications. Precambrian Research, 73, 27– 49.
Kellerhals, P. 1993. Igneous Petrology of the Mirbat Complex and
Facies Analysis of the Mirbat Sandstone Formation (Middle and
Upper Member). Unpublished MSc thesis, University of Berne,
Switzerland.
Kellerhals, P. & Matter, A. 2003. Facies analysis of a glaciomarine
sequence, the Neoproterozoic Mirbat Sandstone Formation, Sultanate
of Oman. Eclogae Geologicae Helvetiae, 96, 49 – 70.
Kennedy, M. J. 1996. Stratigraphy, sedimentology and isotope geo-
chemistry of Australian Neoproterozoic post-glacial cap dolostones:
deglaciation, d13C excursions and carbonate precipitation. Journal
Sedimentary Research, 66, 1050– 1064.
Kennedy, M. J., Christie-Blick, N. & Sohl, L. E. 2001. Are Proterozoic
cap carbonates and isotopic excursions a record of gas hydrate desta-
bilization following Earth’s coldest intervals? Geology, 29, 443– 446.
Kempf, O., Kellerhals, P., Lowrie, W. & Matter, A. 2000. Paleomag-
netic directions in late Precambrian glaciomarine sediments of the
Mirbat Sandstone Formation, Oman. Earth & Planetary Science
Letters, 175, 181– 190, doi: 10.1016/S0012-821X(99)00307-6.
Kilner, B., MacNiocaill, C. & Brasier, M. D. 2005. Low-latitude
glaciation in the Neoproterozoic of Oman. Geology, 33, 413– 416,
doi: 10.1130/G21227.1.
Lees, G. M. 1928. The geology and tectonics of Oman and parts of
southeastern Arabia. Quarterly Journal Geological Society London,
84, 585– 670.
Lønne, I. 1995. Sedimentary facies and depositional architecture of
ice-contact glaciomarine systems. Sedimentary Geology, 98, 13– 43.
Loosveld, R., Bell, A. & Terken, J. 1996. The tectonic evolution of
interior Oman. GeoArabia, 1, 28 –50.
Mackiewicz, N. E., Powell, R. D., Carlson, P. R. & Molnia, B. F.
1984. Interlaminated ice-proximal glacimarine sediments in Muir
Inlet, Alaska. Marine Geology, 57, 113–147.
McKirdy, D. M., Burgess, J. M. et al. 2001. A chemostratigraphic over-
view of the late Cryogenian interglacial sequence in the Adelaide
Fold-Thrust Belt, South Australia. Precambrian Research, 106,
149– 186.
McLennan, S. M., Hemming, S., McDaniel, D. K. & Hanson, G. N.
1993. Geochemical approaches to sedimentation, provenance and
tectonics. In: Johnsson, M. J. & Basu, A. (eds) Processes
 THE AYN FORMATION
Controlling the Composition of Clastic Sediment. Geological Society
of America Special Paper, 284, 21 –40.
Meert, J. G. 2003. A synopsis of events related to the assembly of eastern
Gondwana. Tectonophysics, 362, 1 –40.
Mercolli, I., Briner, A. P., Frei, R., Schönberg, R., Nagler, T. F.,
Kramers, J. & Peters, T. 2006. Lithostratigraphy and geochronol-
ogy of the Neoproterozoic crystalline basement of Salalah, Dhofar,
Sultanate of Oman. Precambrian Research, 145, 182– 206.
Moncrieff, A. C. M. & Hambrey, M. J. 1990. Marginal marine glacial
sedimentation in the late Precambrian succession of east Greenland.
In: Dowdeswell, J. A. & Scourse, J. D. (eds) Glacimarine Environ-
ments: Processes and Sediments. Geological Society, London,
Special Publications, 53, 387– 410.
Morton, D. M. 1959. The Geology of Oman. 5th World Petroleum
Congress, 1959, Section 1, paper 14.
Nesbitt, H. W. & Young, G. M. 1982. Early Proterozoic climates and
plate motions inferred from major element chemistry of lutites.
Nature, 299, 715– 717, doi: 10.1038/299715a0.
Nesbitt, H. W., Young, G. M., McLennan, S. M. & Keays, R. R. 1996.
Effects of chemical weathering and sorting on the petrogenesis of
siliciclastic sediments, with implications for provenance studies.
Journal of Geology, 104, 525– 542.
Ovenshine, A. T. 1970. Observations of iceberg rafting in Glacier
Bay, Alaska, and the identification of ancient ice-rafted deposits.
Geological Society America Bulletin, 81, 891– 894.
Peltier, W. R., Liu, Y. & Crowley, J. W. 2007. Snowball Earth preven-
tion by dissolved organic carbon remineralization. Nature, 450,
813– 818.
Platel, J. P., Roger, J., Peters, T., Mercolli, I., Kramers, J. D. & Le
Métour, J. 1992a. Geological Map of Salalah, Sheet NE 40-09, 1:
250 000, with explanatory notes. Directorate General of Minerals,
Oman Ministry of Petroleum and Minerals.
Platel, J. P., Le Métour, J., Berthiaux, A., Buerrier, M. & Roger, J.
1992b. Geological map of Juzor Al Halaaniyat, sheet NE 40-10, scale
1:250,000. Directorate General of Minerals, Oman Ministry of
Petroleum and Minerals.
Powell, R. D. 1981. A model for sedimentation by tidewater glaciers.
Annals Glaciology, 2, 129–134.
Powell, R. D. 1990. Glacimarine processes at grounding-line fans and
their growth to ice-contact deltas. In: Dowdeswell, J. A. &
Scourse, J. D. (eds) Glacimarine Environments: Processes and
Sediments. Geological Society, London, Special Publications, 53,
53 – 73.
Powell, R. D. & Molnia, B. F. 1989. Glacimarine sedimentary pro-
cesses, facies and morphology of the south-southeast Alaska Shelf
and fjords. Marine Geology, 85, 359– 390.
Qidwai, H. A., Khalifa, M. I. & Ba-Mkhalif, K. A. 1988. Evidence of
Permo-Carboniferous glaciation in the basal Murbat Sandstone
Formation, Southern Region, Sultanate of Oman. Journal Petroleum
Geology, 11, 81 –88.
Rieu, R. 2006. Sedimentology, Stratigraphy and Geochemistry of the
Glacially Influenced Neoproterozoic Mirbat Group, Oman. PhD
thesis, ETH-Zürich.
Rieu, R. & Allen, P. A. 2008. Siliciclastic sedimentation in the inter-
lude between two Neoproterozoic glaciations, Mirbat area, southern
Oman: a missing link in the Huqf Supergroup. GeoArabia, 13,
45 – 72.
Rieu, R., Allen, P. A., Etienne, J. L., Cozzi, A. & Weichert, U. 2006.
A Neoproterozoic glacially influenced basin margin succession and
‘atypical’ cap carbonate associated with bedrock paleovalleys,
Mirbat area, southern Oman. Basin Research, 18, 471– 496.
Rieu, R., Allen, P. A., Cozzi, A., Kosler, J. & Bussy, F. 2007a. A
composite stratigraphy for the Neoproterozoic Huqf Supergroup of
Oman: integrating new litho-, chemo- and chronostratigraphic data
249
of the Mirbat area, southern Oman. Journal of the Geological
Society of London, 164, 997– 1009.
Rieu, R., Allen, P. A., Plötze, M. & Pettke, T. 2007b. Compositional
and mineralogical variations in a Neoproterozoic glacially influenced
succession, Mirbat area, south Oman: implications for paleoweather-
ing conditions. Precambrian Research, 154, 248– 265, doi: 10.1016/
j.precamres.2007.01.003.
Rieu, R., Allen, P. A., Plötze, M. & Pettke, T. 2007c. Climatic cycles
during a Neoproterozoic ‘snowball’ glacial epoch. Geology, 35,
299– 302, doi: 10.1130/G23400A.1.
Ross, G. M., Bloch, J. D. & Krouse, H. R. 1995. Neoproterozoic strata
of the southern Canadian Cordillera and the isotopic evolution of
seawater sulfate. Precambrian Research, 73, 71 –99.
Rust, B. R. & Romanelli, R. 1975. Late Quaternary subaqueous outwash
deposits near Ottawa, Canada. In: Joplin, A. V. & McDonald, B. C.
(eds) Glaciofluvial and Glaciolacustrine Sedimentation. Society of
Economic Paleontologists and Mineralogists, Special Publication,
25, 177– 192.
Scheffler, K., Hoernes, S. & Schwark, L. 2003. Global changes
during Carboniferous-Permian glaciation of Gondwana: linking
polar and equatorial climate evolution by geochemical proxies.
Geology, 31, 605–608, doi: 10.1130/0091-7613(2003)031,
0605:GCDCGO.2.0.CO;2.
Schermerhorn, L. J. G. 1974. Late Precambrian mixtites: glacial and/or
nonglacial? American Journal Science, 274, 673– 824.
Shields, G. A. 2005. Neoproterozoic cap carbonates: a critical appraisal
of existing models and the plumeworld hypothesis. Terra Nova, 4,
299– 310.
Smith, I. 2000. Diamictic sediments within high Arctic lake sediment
cores: evidence for lake ice rafting along the lateral glacial margin.
Sedimentology, 47, 1157– 1179.
Syvitski, J. P. M., Andrews, J. T. & Dowdeswell, J. A. 1996. Sediment
deposition in an iceberg-dominated glacimarine environment, East
Greenland: basin-fill implications. In: Solheim, A., Riis, F., Elver-
hoi, A., Faleide, J. I., Jensen, L. N. & Cloetingh, S. (eds) Impact
of Glaciations on Basin Evolution: Data and Models from the
Norwegian Margin and Adjacent Areas. Global Planetary Change,
12, 251– 270.
Windley, B. F., Whitehouse, M. J. & Ba-Bttat, M. A. O. 1996. Early
Precambrian gneiss terranes and Pan-African island arcs in Yemen:
crustal accretion of the eastern Arabian Shield. Geology, 24, 131–134.
Worthing, M. A. 2005. Petrology and geochronology of a Neoprotero-
zoic dyke swarm from Marbat, south Oman. Journal African Earth
Sciences, 41, 248–265.
Wright, R. & Anderson, J. B. 1982. The importance of sediment gravity
flow sediment transport and sorting in a glacial marine environment:
eastern Weddell Sea, Antarctica. Geological Society America
Bulletin, 93, 951–963.
Wright, V. P., Ries, A. C. & Munn, S. G. 1990. Intraplatformal basin-fill
from the Infracambrian Huqf Group, east-central Oman. In:
Robertson, A. H. F., Seale, M. P. & Ries, A. C. (eds) The
Geology and Tectonics of the Oman Region. Geological Society,
London, Special Publications, 49, 601–616.
Würsten, F. 1994. The Precambrian crystalline basement of Salalah.
Unpublished PhD thesis, University of Berne, Switzerland.
Young, G. M. 1992. Neoproterozoic glaciation in the Broken Hill area,
New South Wales, Australia. Geological Society America Bulletin,
104, 840–850.
Young, G. M. & Gostin, V. A. 1991. Late Proterozoic (Sturtian) succes-
sion of the North Flinders Basin, South Australia: an example of
temperate glaciation in an active rift setting. In: Anderson, J. B. &
Ashley, G. M. (eds) Glacial Marine Sedimentation: Paleoclimatic
Significance. Geological Society America Special Paper, 261,
207– 223.
 Chapter 20

The Abu Mahara Group (Ghubrah and Fiq formations), Jabal Akhdar, Oman

PHILIP A. ALLEN1*, JONATHAN LEATHER2, MARTIN D. BRASIER3, RUBEN RIEU4, MARGARET MCCARRON3,
ERWAN LE GUERROUÉ5, JAMES L. ETIENNE6 & ANDREA COZZI7
1
Department of Earth Science and Engineering, Imperial College London, South Kensington Campus, London SW7 2AZ, UK
2
Tullow Oil, 5th Floor, Block C, Central Park Leopardstown, Dublin 18, Ireland
3
Department of Earth Sciences, Oxford University, Parks Road, Oxford OX1 3PR, UK
4
TOTAL E&P Nederland, Bordewijklaan 18, Den Haag, The Netherlands
5
Beicip-Franlab, 232, Avenue Napoléon Bonaparte, 92502 Rueil-Malmaison, PO Box 213, France
6
Neftex Petroleum Consultants Ltd, 97 Milton Park, Abingdon, Oxfordshire OX14 4RY, UK
7
ENI Angola, Rua Nicola Gomes Spencer, 140, PO Box 1289, Luanda, Angola
*Corresponding author (e-mail: philip.allen@imperial.ac.uk)

Abstract: The Abu Mahara Group (c. 725–,645 Ma) of the Huqf Supergroup in the Jabal Akhdar of northern Oman hosts two glacial
successions in the Ghubrah and Fiq formations, separated by the ,50-m-thick volcanogenic Saqlah Member. The .400-m-thick
Ghubrah Formation is dominated by distal glaciogenic rainout diamictites, laminites and turbiditic siltstones, whereas the ,1.5-km-
thick Fiq Formation exhibits a cyclical stratigraphy of proximal and distal marine glaciogenic facies, and non-glacial sediment
gravity flow and shallow marine facies. The Fiq Formation is overlain by a transgressive, isotopically light carbonate known as the
Hadash Formation. A tuffaceous ash interbedded with glacial diamictites of the Ghubrah Formation in Wadi Mistal has yielded a
U –Pb zircon age of 713.7 + 0.5 Ma. The Fiq Formation contains detrital zircons as young as 645 Ma. The use of the CIA (Chemical
Index of Alteration) shows the Fiq Formation to be climatically cyclic, with alternations of high and low chemical weathering of con-
temporary land surfaces driven by phases of glaciation and deglaciation. The transgression into the post-glacial Masirah Bay Formation is
marked by a major increase in chemical weathering.

The Huqf Supergroup crops out in northern Oman (Jabal Akhdar),
east-central Oman (Huqf area) and southern Oman (Mirbat area of
Dhofar) (Fig. 20.1) (Allen et al. 2011) and is penetrated by many
boreholes in the salt basins of the Oman interior. The Huqf Group
recognized by Glennie (1977) in the Huqf area of east-central
Oman was subdivided into the Abu Mahara, Khufai, Shuram and
Buah Formations by Gorin et al. (1982), but no glaciogenic depos-
its are found in this region. In the Jabal Akhdar of northern Oman,
Kapp & Llewellyn (1965) defined the Mistal, Hajir, Mu’aydin and
Kharus Formations, which were correlated (Tschopp 1967) with
the Huqf succession of east-central Oman. Kapp & Llewellyn
(1965), Glennie et al. (1974), and geologists of the Bureau de
Recherches Géologiques et Minières (Beurrier et al. 1986; Rabu
et al. 1986) recognized and mapped the Mistal Formation (syn.
Mistal Conglomerate Formation) as the oldest exposed Neoproter-
ozoic sedimentary rocks in the Jabal Akhdar. Rabu et al. (1986)
divided the Mistal Formation into four members, in ascending
order, the Ghubrah, Saqla, Fiq and Amq Members. In an attempt
to streamline stratigraphic nomenclature across the surface out-
crops and subsurface well penetrations of the salt basins, PDO
geoscientists (PDO internal reports, Loosveld et al. 1996) rede-
fined the Abu Mahara Group as comprising a lower diamictite-rich
Ghadir Manqil Formation and an upper Masirah Bay Formation,
and raised the status of the succession to the Huqf Supergroup,
comprising Abu Mahara, Nafun and Ara Groups (Glennie et al.
1974; Gorin et al. 1982; Hughes-Clarke 1988; Wright et al.
1990). McCarron (2000) referred to the distinctive carbonate
capping the diamictites of the Fiq Member as the Hadash For-
mation, based on the name of the village close to its type locality
above Wadi Mistal in the Jabal Akhdar. When the basin evolution
of the Huqf Supergroup was better understood, the term Abu
Mahara Group was restricted to the stratigraphy below the trans-
gressive Hadash Formation carbonate (McCarron 2000; Leather
et al. 2002; Allen et al. 2004; Allen & Leather 2006), and the
Masirah Bay Formation assigned to the Nafun Group. The strati-
graphic nomenclature currently in use divides the Abu Mahara
Group into a lower Ghubrah Formation and an upper Fiq For-
mation, separated locally by a volcanogenic unit. This unit is tran-
sitional upwards into Fiq Formation, so is included in the Fiq as the
Saqlah Member.
Neoproterozoic glacial deposits are found in the Abu Mahara
Group in the Jabal Akhdar of the Oman Mountains, occupying
the core of a major east – west oriented anticline (Fig. 20.2).
Partly coeval rocks are found in Dhofar (southern Oman) (Allen
et al. 2011), but the Abu Mahara Group is absent from the Huqf
area of east-central Oman. The Group has been penetrated by a
limited number of boreholes in the salt basins of Oman, and has
been imaged on seismic reflection profiles, which show it to be
confined to large north – south oriented, c. 50-km-wide basins
that have been interpreted as rifts (Loosveld et al. 1996). This
rift interpretation is supported by sedimentological and strati-
graphic evidence derived from field studies (Rabu 1988; Rabu
et al. 1993). In the Jabal Akhdar, the sedimentology of the glacio-
genic Fiq Formation indicates sediment derivation from two
opposing margins (Leather et al. 2002; Allen et al. 2004). The
intervening basin has the same scale and orientation as the structu-
rally confined basins imaged on seismic profiles.
The phase of sedimentation responsible for the Fiq Formation
was preceded by a period of extrusive basaltic volcanism,
represented by the Saqlah Member (Rabu 1988; Rabu et al.
1993; Le Guerroué et al. 2005). Although the Saqlah volcanic
and volcaniclastic rocks and the Fiq glaciogenic and non-glacial
sedimentary rocks can be attributed to a continental rift basin,
the Ghubrah stratigraphy below the Saqlah Member is far less
well constrained.
A transgressive, 13C-depleted (down to –5‰) carbonate
(Hadash Formation) oversteps basement-cored Abu Mahara
basin margins, as seen in the Huqf area, where the carbonate

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 251– 262. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.20
252 P. A. ALLEN ET AL.

Thickness (m)
(a) (b) Oman
composite
Nomenclature
stratigraphy

CAMBRIAN
Dhahaban
IRAN A6
Al Noor

ARA GROUP
Musandam

Subsurface
A5

Ara cycles
0
26 N
Athel A4
ARABIAN GULF A3

Birba
Fara A2
Makran Fold Belt 4000 A1
A0
Buah

Jabal Akhdar and Huqf

GULF OF OMAN

NAFUN GROUP
EDIACARAN
Shuram
Qatar Arch

UNITED ARAB EMIRATES

Muscat
Jabal Akhdar
Khufai
Fig. 2 A Saih Hatat 23O N
Oman Mountains Masirah Bay 3200
Qalhat
Hadash
Southern Gulf F7
Salt Basin Jebel Ja'alan
Fahud
Basin

xis
2800

fa
q
Hu
gh

Ghaba F6
Hi

SAUDI ARABIA Basin

rem
ka

Al Jobah

Jabal Akhdar
Ma

F5

HUQF SUPERGROUP
Western Rub Al-Khali B
Salt Basin Huqf area Fiq/Shareef
F3
Masirah Island
O
20 N
2000
South Oman

CRYOGENIAN

ABU MAHARA GROUP

Salt Basin
OMAN n k
si n
gh dB
a
Fla
Hi a r n
F1

sfa
h Wa ste 1600
a Ea Saqlah
Kh
u n-
ud
Gh ARABIAN SEA
Marsham
?
sin
Ba

1200
ld
ha
ag

Ain Sarit Mirbat

W

YEMEN

Mirbat
C 100 km
Al Hota 17o N
Arkahawl 800
52o E 55oE 58o E
Major transgressive
Evaporites of Ara Sandstones and conglomerates surface

Ayn/Ghubrah 400
Carbonate rocks Shales and siltstones
northern Huqf
Mirbat and

Basement
Extrusives in basin-fill/
Glacial diamictites crystalline basement/late plutons

Fig. 20.1. (a) Neoproterozoic outcrops (black) and salt basins (grey) of Oman. Main outcrop areas are Jabal Akhdar of northern Oman (A), Huqf area of east-central
Oman (B) and Mirbat area of Dhofar (C), as well as the metamorphic window of Saih Hatat. (After Allen et al. 2004.) (b) Composite stratigraphy of the Huqf Supergroup
in Oman. (After Allen 2007 and Rieu et al. 2007a.) Arrows indicate main transgressive surfaces, and F1 to F7 are Fiq units of Leather (2001).

rests via an unconformity lined with silcrete upon felsic ignim- of the Oman Mountains, and in the intensely deformed and meta-
brites and tuffs dated 802 Ma and granitic basement dated 822 morphosed domal culmination of the Saih Hatat (Glennie et al.
and 825 Ma (Allen & Leather 2006; Allen 2007; Bowring et al. 1974; Le Métour et al. 1986; Rabu et al. 1993; Gray et al.
2007). Over basement highs, therefore, the unconformity 2005). The Saih Hatat structural deformation was generated
beneath the Nafun Group stretches from c. 800 Ma to the onset during a top-to-the-NE shear, producing a major, refolded,
of deposition of the Nafun Group, whereas in the Abu Mahara NE-facing, recumbent anticlinal fold-nappe within pre-Ordovician
basins, as in the Jabal Akhdar, the stratigraphic succession is sedimentary rocks, including lateral equivalents of the Huqf
more complete and conformable. Supergroup. No detailed stratigraphic or sedimentological work
has been carried out on account of this deformation and meta-
morphism, although the stromatolitic Khufai Formation of the
Structural framework Jabal Akhdar has been correlated lithologically with the Hiyam
Formation of the Saih Hatat (Rabu et al. 1986).
The Huqf Supergroup is exposed in the core of a 700-km-long, The Huqf Supergroup of the Jabal Akhdar is less intensely
30–130-km-wide, east –west-trending antiform in the Jabal Akhdar deformed than in the Saih Hatat and is exposed in a series of
 THE ABU MAHARA GROUP
erosional windows or ‘bowls’, chiefly (from west to east) the
Sahtan, Kharus and Mistal bowls. Huqf Supergroup sedimentary
rocks were tilted (block-faulting and large-amplitude folding
most likely associated with the Hercynian orogeny, Rabu et al.
1986) prior to the deep erosion represented by the pre-Permian
unconformity, and underwent further deformation during an
‘Alpine’ (c. 90 Ma) event associated with the emplacement of
the Samail ophiolite and thrust sheets involving Mesozoic
Tethyan carbonate-dominated rocks (Mann & Hanna 1990).
Alpine deformation imparted a strong cleavage, particularly in
the Ghubrah Formation, where schistosity produces a near-
horizontal pencil cleavage and a roughly north –south stretching
lineation measured from the long axes of clasts in diamictite
units. The gradient in intensity of deformation and metamorphism
towards the east and NE of the Oman Mountains (Le Métour et al.
1986), reaching a peak in the Saih Hatat, indicates that the same
tectonic events affected both the Jabal Akhdar and Saih Hatat
regions. The fact that the penetrative cleavage that affects fine-
grained units of the Huqf Supergroup extends into the overlying
Permian –Mesozoic succession implies that it is Alpine in age,
with an increasing shear deformation downwards. The structural
geology of the Jabal Akhdar illustrated in the work and map of
Rabu et al. (1986), Rabu (1988) and Rabu et al. (1993) and the
separation of the pre-Permian from Alpine deformation, awaits a
further detailed analysis.
Stratigraphy
The Jabal Akhdar region contains excellent exposures of rocks
belonging to the Abu Mahara and Nafun Groups of the Huqf
Supergroup (Fig. 20.1; see Allen 2007, fig. 4, p. 145, for a
summary of stratigraphic nomenclature, and fig. 8, p. 149, for
the composite stratigraphic column). Whereas the base of the
Huqf Supergroup is not seen, the top is an erosional, angular
unconformity beneath the carbonate rocks of the Permian Saiq,
which cuts down variably. In some locations (e.g. Wadi Mistal),
the Saiq cuts down deeply into the Abu Mahara Group, whereas
in others a full succession of Nafun Group is preserved. In Wadi
Bani Awf the Nafun Group passes up into the Fara Formation, a
surface equivalent of part of the Ara Group, which is truncated
by the base– Saiq unconformity. The Abu Mahara Group com-
prises a lower diamictite-rich Ghubrah Formation, separated
from an upper diamictite-rich Fiq Formation by an intervening
253
Fig. 20.2. Geological map of the Jabal
Akhdar showing log localities, main towns
and villages. Log localities: WS1 and WS2,
Wadi Sahtan; WH1 and WH2, Wadi Hajir;
WBK1, Wadi Bani Kharus; WM1, WM2,
WM3 and WM4, Wadi Mistal; MU1 Wadi
Mu’aydin; WBJ1, Wadi Bani Jabir. Arrows
indicate main entrances to wadis. (Modified
from Allen et al. 2004.)
laterally discontinuous volcaniclastic and volcanic (pillowed and
vesicular basalts) Saqlah Member (Le Guerroué et al. 2005). The
Fiq Formation is overlain by a well-developed carbonate named
the Hadash Formation (McCarron 2000), which is the lowermost
unit of the Nafun Group.
Glaciogenic deposits and associated strata
Glaciogenic deposits are found in the Abu Mahara Group in the
Jabal Akhdar at two different stratigraphic levels. The younger
Cryogenian glaciogenic sedimentary rocks are found in the cycli-
cal Fiq Formation (Leather et al. 2002; Allen et al. 2004). The
older Cryogenian glaciogenic deposits are found in the much
less well-understood Ghubrah Formation (Rabu et al. 1986;
Brasier et al. 2000; Leather 2001; Le Guerroué et al. 2005).
The Ghubrah Formation
Very little is known about the Ghubrah Formation. It was initially
recognized by Kapp & Llewellyn (1965) as the lower part of the
Mistal Conglomerate, and Rabu et al. (1986) referred to it as the
Ghubrah Member of the Mistal Formation. The base is not seen.
The Ghubrah Formation crops out in Wadi Mistal (.200 m
thick) and Wadi Mu’aydin in the east, and in Wadi Sahtan
(.400 m thick) in the west of the Jabal Akhdar. Rabu et al.
(1986) contrasted it with the Fiq unit as comprising large thick-
nesses of poorly stratified diamictic conglomerate, compared
with the better-bedded Fiq. Rabu et al. (1986), however, warned
that locally, in particular in Wadi Mu’aydin, the sandstones of
the Fiq interfinger with conglomerate similar to that of the
Ghubrah, rendering the two virtually indistinguishable. Leather
(2001) stated that the facies within the Ghubrah and Fiq For-
mations were ‘notably different’, with the Ghubrah Formation
lacking bedding for much of its thickness.
McCarron (2000), Brasier et al. (2000) and Leather (2001) all
followed Rabu et al. (1986) in believing the Ghubrah to be an
older stratigraphic unit separated from the Fiq Formation by the
volcanogenic Saqlah Member. Le Guerroué et al. (2005) identified
an angular relationship between the two formations, proposing an
unconformity at the base of the Saqlah. However, the relationship
of the Ghubrah to the Fiq, and the sedimentology and structural
geology of the Ghubrah, have never been given the attention
254 P. A. ALLEN ET AL.

they deserve. The U –Pb zircon date reported by Brasier et al.
(2000), and more accurate U –Pb dates on similar material reported
in Bowring et al. (2007), need to be viewed in this context (see
below).
The facies of the Ghubrah Formation are described briefly by
Leather (2001). The Formation comprises a distal glaciomarine
facies association made of massive diamictites, massive and
graded siltstones, dropstone laminites, very minor carbonate
rocks and tuffaceous sandstones.
The massive diamictites are up to hundreds of metres thick, with
gradational upper and lower boundaries into massive and graded
siltstones. The diamictites appear structureless, with some striated
and faceted clasts. These massive diamictites may be thick amal-
gamated debris flows that reworked glacially transported clasts,
but the association with dropstone laminites and the large lateral
extent (.40 km) suggest a glaciomarine origin by fallout from
suspension and rainout of ice-rafted debris, possibly in an outer
shelf/slope setting (Anderson et al. 1984; Deynoux 1985; Eyles
1988; Eyles & Lagoe 1990; Moncrieff & Hambrey 1990; Brodzi-
kowski & van Loon 1991). The associated siltstones indicate quiet
marine sedimentation when ice rafting was reduced, either because
of glacial retreat (Eyles & Lagoe 1990) or movement of iceberg
transport lanes (Moncrieff & Hambrey 1990; Miller 1996). The
dropstone laminites found in Wadi Mu’aydin indicate a clear
ice-rafted component (Powell & Domack 1995). Carbonate
occurs as c. 10-cm-thick boudinaged beds in Wadi Mistal consist-
ing of quartz grains and lithic clasts in a dolomite matrix of
,50 mm rhombs. The tuffaceous and ashy sediments that have
provided the U – Pb date reported by Brasier et al. (2000) and
Bowring et al. (2007) indicate nearby volcanism during deposition
of the Ghubrah Formation.
Sedimentology of the Fiq Formation
The Fiq Formation is well exposed in a number of wadis in the
Jabal Akhdar of northern Oman, particularly in Wadi Sahtan,
Wadi Hajir and Wadi Mistal (Figs 20.2 & 20.3). Although the stra-
tigraphy, first described by PDO geologists in 1965, was mapped
and interpreted as partly glaciomarine (Glennie et al. 1974; Beur-
rier et al. 1986; Rabu et al. 1986; Hughes Clark 1988), the first
detailed sedimentological analysis is in the doctoral thesis of
Leather (2001).
Facies associations. The facies present in the Fiq Formation can be
assigned to four different facies associations: (i) distal glaciomar-
ine, (ii) proximal glaciomarine, (iii) non-glacial sediment gravity
flow and (iv) non-glacial shallow marine. The lithofacies associ-
ations used have been adapted in part from Eyles et al. (1985),
Moncrieff & Hambrey (1990), Brodzikowski & van Loon (1991)

Fig. 20.3. Summary sedimentological logs through the Fiq Formation from 10 localities (see Fig. 20.2), hung from the level of the Hadash Formation. Diamictites
are highlighted to illustrate their lateral continuity and vertical superimposition. F1 to F7 are stratigraphic units of Leather (2001), and T1 to T6 are transgressive surfaces.
Diamictites in F5 and F7 are basin-wide deposits, whereas older diamictites, such as that in F1, are locally developed. (After Allen et al. 2004.)
 THE ABU MAHARA GROUP
and Miller (1996). Lithofacies names have been partly adapted
from Eyles et al. (1983), Moncrieff (1989) and Moncrieff &
Hambrey (1990).
Distal glaciomarine facies association. The distal glaciomarine
facies association represents distal glaciomarine environments
where ice rafting and non-glacial processes dominate, and is
found in Wadi Sahtan, Wadi Bani Awf and Wadi Hajir. It is domi-
nated by massive, sheet-like diamictites (,30 m thick) laterally
and vertically transitional to deep-water mudstones and siltstones.
Clasts include striated and faceted examples indicating subglacial
transport (Wentworth 1936; Boulton 1978; Dowdeswell et al.
1985). An ice-rafted origin is inferred on the basis of the glacially
transported clasts, common transitional lower boundaries and
lack of internal stratification (Anderson et al. 1984; Eyles 1988;
Kellerhals & Matter 2003). The clast-poor nature of many of the
diamictites, their sheet-like geometry and their association with
deep-water deposits suggests that they formed as a result of a com-
bination of settling of suspended sediment and ice rafting, possibly
in an outer shelf/slope setting relatively distant from active ice
margins (Deynoux 1985; Eyles & Lagoe 1990; Moncrieff &
Hambrey 1990; Brodzikowski & van Loon 1991).
Associated dropstone laminites (,50 m thick), found in Wadi
Hajir, comprise grey shales and siltstones with sharp-based turbi-
ditic interbeds and rare outsized clasts, indicating a distal shelf
to slope setting affected by ice rafting, suspended sediment
fallout and dilute turbiditic underflows. Underflows may have
been released by subglacial river output (Eyles 1988; Moncrieff
& Hambrey 1990) or brine-rich currents, as occurs on the
Barents Shelf today (Elverhoi et al. 1989).
Proximal glaciomarine facies association. The proximal glaciomar-
ine facies association formed close to the grounded ice margin
(Moncrieff & Hambrey 1990; Miller 1996) where rainout of
coarse ice-rafted debris and sediment gravity flows interact. Sedi-
ment gravity flows are also found in distal glaciomarine and non-
glacial environments. In summary, the proximal glaciomarine
facies association comprises (i) massive diamictites (from a few
metres thick to 100 m thick) containing striated, faceted and
‘flat-iron’-shaped clasts, indicating subglacial transport prior to
ice rafting; (ii) subordinate but widespread stratified diamictites,
formed by a combination of winnowing of ice-rafted material
(Eyles et al. 1985; Moncrieff & Hambrey 1990), or re-deposition
of massive diamictites by sediment gravity flows in ice-proximal
settings; (iii) clast-supported, laterally discontinuous conglomer-
ates, representing local concentrations of clasts derived from
iceberg overturn (Ovenshine 1970), lag deposits indicating inten-
sive current, wave or tide reworking of ice-rafted diamictite depos-
its in relatively shallow marine conditions (Eyles et al. 1985; Eyles
1988; Moncrieff & Hambrey 1990), and debris flows of remobi-
lized ice-rafted material where the fine sediment has been expelled
into an overlying turbulent suspension (Wright & Anderson 1982);
(iv) massive or graded, rippled, sharp-based sandstones and silt-
stones, with lonestones/dropstones at bed boundaries, are turbidite
deposits formed in a glaciomarine stetting extending some distance
from the grounding line; (v) minor dropstone laminites, due to
minor ice rafting in a quiet distal glaciomarine environment; (vi)
massive and laminated mudstones and siltstones lacking drop-
stones, derived from dilute turbidity currents or overflow plumes
(Mackiewicz et al. 1984; Powell & Molnia 1989). Taken together,
the facies association suggests an ice-proximal environment
subject to outflows from subglacial streams, debris flows, sandy
turbidites, fallout from suspension from turbid plumes, winnowing
and reworking, and ice rafting.
Non-glacial sediment gravity flow facies association. The non-glacial
sediment gravity flow facies association, which may form units
,250 m thick, comprises (i) erosive-based, commonly lenticular
conglomerates and pebbly sandstones, with rounded clasts and
255
large intraformational rafts, representing debris flows or hypercon-
centrated flows (Nardin et al. 1979; Costa 1988); (ii) common
sharp-based, massive and graded, sheet-like, commonly amalga-
mated sandstones, locally with rippled tops and fluted bases,
which are turbidites; (iii) current-rippled sandstones and siltstones
with ubiquitous water-escape structures representing deposition
from more dilute turbidity currents, with local wave-generated
ripples at the top of shallowing-up cycles; (iv) massive and lami-
nated mudstones and convoluted siltstones, which settled slowly
from turbiditic plumes/tails or river plumes/underflows (Stow
et al. 1996); (v) thin, brown carbonate on a transgressive surface
overlying glacial diamictites (Wadi Sahtan, log WS1, unit F6)
suggesting carbonate precipitation at a time of deglacial sea-
level rise.
Non-glacial shallow marine facies association. The non-glacial
shallow marine facies association forms a minor part of the non-
glacial sediments of the Fiq Formation. It comprises (i) laterally
persistent conglomerates composed of well-rounded pebbles and
cobbles, representing transgressive lag (ravinement) deposits win-
nowed of their finer material during flooding; (ii) extensively
wave-rippled sandstones, particularly common in Wadi Sahtan,
probably representing shoreface deposition; (iii) rippled wave-
generated heterolithics, passing up into sharp-based, massive sand-
stones probably representing flows generated by storms.
Depositional history of the Fiq Formation
The Fiq Formation is spatially and temporally highly variable
(Fig. 20.3). This variability results from the fundamental bathy-
metric effects produced by the rifted crustal structure. However,
distinct genetic units reflecting relative sea-level change can be
traced across the entire Jabal Akhdar area. Glacially influenced
sedimentary rocks occur at a number of different levels in the
Fiq Formation, commonly followed abruptly by a facies change
indicating deepening. The facies evolution through time and the
relative sea level recorded by these facies can therefore be evalu-
ated (Leather et al. 2002; Allen et al. 2004; Allen & Etienne 2008).
Unit F1. Unit F1 is a transitional unit between the underlying vol-
canics of the Saqlah Member and the Fiq Formation. Volcaniclas-
tic detritus was derived from a source to the east, probably in the
Saih Hatat area (Le Métour et al. 1986; Villey et al. 1986), and
was deposited in an already-deep marine basin. The occurrence
of diamictites (F1b in Wadi Mistal) containing volcanic clasts indi-
cates glaciation in F1, though its effects appear to be restricted to
the east of the outcrop area.
Units F2 and F3. In the west of the Jabal Akhdar, quiet, deep-water
conditions became established in F2 times, as shown by increasing
interbedding of turbidites derived from the west, followed by shal-
lowing into a wave-rippled shoreface. The overlying F3 proximal
diamictite indicates a resumption of glaciation. In the east of Jabal
Akhdar, the F1b diamictite was abruptly flooded and succeeded by
gravity-flow deposits derived from non-volcanic sources in the
east. An overlying diamictite correlates well with F3 in the west
of Jabal Akhdar, indicating that the entire outcrop area was influ-
enced by glacial processes at this time.
Unit F4. Evidence for an important flooding event following depo-
sition of the F3 diamictite is found across the whole Jabal Akhdar
outcrop area, and there is no record of any glacial processes in the
basin. Subsequently, sediment gravity flows derived from the east
filled the eastern sector of the basin, whereas shallow shoreface
conditions existed in the west, indicating asymmetry in subsidence
and/or sediment supply. In both the east and west, sediment
gravity flows from multiple sources then resumed and dominated
sedimentation across the entire basin. A shutdown in sediment
256 P. A. ALLEN ET AL.
supply in the west resulted in the deposition of sediments during
transgressive phase T3.
Unit F5. Renewed glacial activity across the whole outcrop area
is recorded by proximal diamictites and sediment gravity flows
in the east and distal glaciomarine deposition in the west.
Unit F6. The end of F5 glaciation is indicated by a major flooding
event in the western and central parts of the outcrop area, whereas
sediment gravity flows initially dominated the eastern sector of the
basin. In the western and central parts of the basin, deposition of
unit F6 continued as turbidites derived primarily from western
sources, and shallow-water wave-rippled sandstones were depos-
ited in the east. A locally developed diamictite (F6a) indicates a
weak glacial advance before significant basin-wide flooding (T5
transgression). The overlying lag conglomerate and thin carbonate
rocks and mudstones indicate transgression and establishment of
deep-marine conditions before a gradual shallowing took effect
towards the top of F6.
Unit F7. Coarse-grained gravity-flow deposits are overlain by
proximal glaciomarine diamictites and debris flows and, locally,
by subglacial stream outwash, representing the last of the glacial
phases in the Fiq Formation. The overlying Hadash Formation is
a classic, transgressive postglacial carbonate. This T6 transgres-
sion covered the rift margins and flooded other basement highs
in the Oman region.
The Hadash Carbonate
The ,1.5-km-thick Fiq Formation is overlain by a transgressive,
13
C-depleted dolostone (Hadash Formation) that deepens up into
the marine shales and siltstones of the Masirah Bay Formation
(Allen & Leather 2006) (Fig. 20.4). The Hadash Formation is
dominated by dolostones, and despite being ,15 m thick is later-
ally extensive over at least 80 km throughout the Jabal Akhdar.
Subsurface well penetrations (PDO internal reports) suggest that
the Hadash Formation is found throughout the Oman area.
Surface outcrop studies also demonstrate that a transgressive car-
bonate unit overlies basement crystalline or volcanic rocks in the
Huqf region (Loosveld et al. 1996; McCarron 2000; Leather
2001; Allen & Leather 2006; Allen 2007). Directly overlying
glacial diamictites of the Fiq Formation and with a negative
C-isotopic signature, the 8-m-thick carbonate-dominated part of
the Hadash Formation shares much in common with cap dolo-
stones worldwide (Williams 1979; Fairchild & Hambrey 1984;
Tucker 1986; Narbonne et al. 1994; Kennedy 1996; Hoffman
et al. 1998a; Myrow & Kaufman 1999; Hoffman & Schrag
2002; Halverson et al. 2004). The facies of the Hadash Formation
are described in Allen et al. (2004). Although the broad deposi-
tional environment of the cap carbonate in the Jabal Akhdar is
deep water, the high proportion of siliciclastics and palaeocurrent
directions in Wadi Bani Jabir, the orientation of roll-up structures
(Simonson & Carney 1999) at Hadash and thickness trends
between the logged sections collectively suggest that the deposi-
tional environment of the carbonate deepened to the centre-west.
The cap carbonate therefore partially ‘shales out’ into the deepest-
water lithofacies deposited in the centre of the Fiq basin. The most
important transition in the basin is between the carbonate-
dominated eastern Jabal Akhdar area (Hadash 1, Hadash 2 and
Wadi Mu’aydin) and the western Jabal Akhdar area, which
exhibits a higher concentration of fine-grained siliciclastics
(Wadi Hajir 1, Wadi Hajir 2, Wadi Bani Awf, Wadi Sahtan).
This was also an important transition during the deposition of
the Fiq Member, with thicker, more shale-prone sections evident
in the west. It is therefore likely that the basement structure
controlled water depth and accommodation across this line.
Chemostratigraphy
Carbon-isotopic data
In total, 271 samples were analysed from seven stratigraphic pro-
files through the Hadash Formation in the Jabal Akhdar (Fig. 20.4).
‘Least altered’ samples were selected for being well away from
areas of intense veining and for having good microscopic fabric
preservation revealed from petrographic and cathodoluminescence
data. In total, 81 samples from five stratigraphic sections were ana-
lysed for major and minor elements (details in Leather 2001).
Elemental and isotopic data were used to screen for the effects
of diagenesis and metamorphism, following techniques described
by Brand & Veizer (1980, 1981), Marshall (1992) and Kaufman &
Knoll (1995). Further details can be found in Leather (2001) and
Allen et al. (2004).
The vertical pattern of C-isotope values (Hadash 1 type section)
shows an initial shift to – 4‰ to –8‰ in the basal metre, followed
by a gradual drift back to –1‰ at the top of the carbonate, although
with considerable scatter. This trend is in common with other Neo-
proterozoic cap dolostones (Kennedy 1996; Hoffman et al. 1998b;
Frimmel et al. 2002; Fölling & Frimmel 2002). However, corre-
lation of chemostratigraphic facies from east to west across the
Jabal Akhdar is problematical. The variation in d13C values
(c. 2‰) between closely spaced stations is as great as the amplitude
of the isotopic excursions used for correlation. Leather (2001)
therefore concluded that high-resolution intrabasinal correlation
using d13C curves is highly uncertain.
Further C-isotopic data from the Nafun Group are found in
Burns & Matter (1993), Cozzi & Al Siyabi (2004), Fike et al.
(2006), Le Guerroué et al. (2006a, b, c) and Le Guerroué &
Cozzi (2010).
Chemical Index of Alteration
The use of the Chemical Index of Alteration (CIA) is described in a
separate paper in this volume (Bahlburg & Dobrzinski 2011,
Chapter 6), and details of the technique are not repeated here.
Throughout the Fiq and Masirah Bay formations there are signifi-
cant compositional and mineralogical variations. When plotted
in Al2O3 – CaO þ N2O – K2O and quartz-plagioclase-K-feldspar
space, the data define trends roughly parallel to the Al2O3 –
CaO þ N2O and plagioclase-quartz boundaries, suggesting varia-
bility in the extent of chemical weathering of the sediment in the
source area (Fedo et al. 1995; Nesbitt et al. 1996). The possible
influence of changes in grain size, hydrodynamic sorting, prove-
nance and diagenetic alteration on the composition of the sedi-
ments is discussed in Rieu et al. (2007b).
Both uncorrected CIA values and those corrected for a
maximum amount of K-metasomatism are plotted as a function
of their stratigraphic height (Fig. 20.5). In both cases, a similar
first-order trend is revealed, comprising three intervals during
which chemical weathering was reduced as indicated by rela-
tively low CIA and Mineralogical Index of Alteration (MIA)
values. Reduced chemical weathering in these intervals is in
agreement with the presence of distinctive sedimentary facies
(diamictites, dropstone-bearing laminites) that suggest a cold
climate. These intervals alternate with units characterized by
relatively high CIA and MIA values and that lack evidence for
any glacial influence during sedimentation, which are interpreted
to represent interglacial periods. Importantly, the end of the
entire glacial epoch corresponds to a major increase in CIA
and MIA values in the lowermost Masirah Bay Formation.
These values are the highest found in the succession
(CIA . 80; MIA, 100). A discussion of these trends, and a justi-
fication of their interpretation as due to temporal changes in the
chemical weathering of contemporary land surfaces, are given by
Rieu et al. (2007b, c).
 N 10 km
WBJ1
Wadi Hedak
Nakhl

Wadi Sahtan
S Wadi Bani
Jabir
Wadi Mistal Legend
Fashah Wadi Bani Awf Al Awabi
Thin dolomite beds in fine-grained siliciclastic rocks (Facies W3)
Amq WS1
Sahtan Bowl Wadi Wadi Bani Ghubrah
Hajir Kharus
WS2 WBA2 WM2 WM1 Sand-grade siliciclastics (Facies E5)
WM3
WBA1 Hajir Ghubrah
WBK1
WH2 Bowl Crystalline limestone and dolospar (Facies E4)
WH1

Permian to Cretaceous cover WM4 Wadi Hajir 2 Dolomite-dominated deposits of Facies W2, E1, E2 and E3
Hadash
Huqf Supergroup Thinly-interbedded dolomite and siltstone (Facies W1)
Al Ayn
Quaternary outside Jabal Akhdar
Log locality and section line MU1 Massive glacial diamictites (Fiq unit 7)
Entrance to wadi
Village or Town
Wadi Mu'aydin
Carbon isotope data: Dolomite Calcite

11 Wadi Sahtan
Wadi Bani Awf Hadash 2 Hadash 1

THE ABU MAHARA GROUP

Height in m above base of Hadash Formation

10 (Wadi Mistal) (Wadi Mistal)

W3 Wadi Bani Jabir
9
Wadi Hajir 1
8
W3
7 W3
Wadi Mu'aydin E4
6 W3
E4 E5
5 W2 E4
4 E3
W2 E3 E3
3 E2
E2
W2 W2 E2
2 W1
W1
1 E1 E1 E1 E1
W1 W1 E1
0 -6 -2 0 -2 0
-4 -4 -2 0 -4 0 -6 -4 0 -4 -2 0 -8 -4 0 -3 -2 -1 0
F7 F7 d13C
F7 F7 F7 F7 F7
d13C d13C d13C F7 d13C d13C d13C

Western Jabal Akhdar Eastern Jabal Akhdar

Fig. 20.4. Summary logs of Hadash Formation from eight localities across the Jabal Akhdar region (see Fig. 20.2), with C-isotopic data shown for seven. Facies labels described in Leather (2001) and Allen et al. (2004). Note the
variable horizontal scale for d13C, which demonstrates the significant variation in C-isotopic ratios between sections. Isotopic and geochemical data are in the supplementary material of Allen et al. (2004).

257
258 P. A. ALLEN ET AL.
Boundary relations with overlying and underlying
non-glacial units
The non-sedimentary basement beneath the Huqf Supergroup is
poorly exposed in Oman. No basement rocks are found in the
Jabal Akhdar. Crystalline basement is exposed in south Oman in
the Mirbat region (Platel et al. 1992a) and on the Al Hallaniyah
Islands (Platel et al. 1992b), and in small outcrops in the north
of the Huqf area of east-central Oman (at Al Jobah, Allen &
Leather 2006; Allen 2007; Bowring et al. 2007) and near Sur in
the Jabal Ja’alan (Roger et al. 1992) further to the north. Most
knowledge on the pre-Huqf Supergroup geological history of
Oman is derived from the Mirbat area, recently summarized by
Mercolli et al. (2006) and Allen (2007) (see Allen et al. 2011).
The rocks overlying the Abu Mahara Group, termed the Nafun
Group, have been correlated confidently between the Jabal
Akhdar, the Huqf area of east-central Oman and the subsurface
of the Oman salt basins (Cozzi & Al-Siyabi 2004; Allen &
Leather 2006; Allen 2007; Le Guerroué & Cozzi 2010), in terms
of major lithological trends and carbon isotopic profiles. The
base of the overlying Nafun Group is marked by a major strati-
graphic overstep due to transgression of basin margins, but the
age of the Hadash Formation and the bulk of the Nafun Group is
not well known. There is no evidence of glaciogenic sedimentary
rocks in the Nafun Group, although it contains a very large nega-
tive C-isotopic excursion known as the Shuram anomaly (Fike
et al. 2006; Le Guerroué et al. 2006b; Bristow & Kennedy 2008;
Le Guerroué & Cozzi 2010).
Palaeolatitude and palaeogeography
Kilner et al. (2005) sampled from the Jabal Akhdar, Huqf and
Dhofar regions. Six sites yielding 20 samples were taken from
the Fiq Formation in the Jabal Akhdar, and the Hadash Formation
was sampled at 21 sites giving 60 samples. Kilner et al. (2005)
found a component of magnetization (Component B) that passed
the fold test and was therefore imparted prior to tectonic defor-
mation in the Cambro-Ordovician (Blendinger et al. 1990; Gass
et al. 1990). After correction for tectonic tilt, the declination and
inclination for a sample close to the top of the Fiq Formation in
Wadi Sahtan (inclination of –218) and for a sample 300 m lower
Fig. 20.5. Variations in chemical and
mineralogical indices of alteration (CIA,
Al2O3/(Al2O3 þ CaO þ Na2O þ K2O);
MIA, Qtz/(Qtz þ Kfs þ Pl)) with
stratigraphic height for section in Wadi
Sahtan (a) and critical section across the
glacial –post-glacial transition at Hadash,
Wadi Mistal (b). Numbering of diamictites
and flooding surfaces as in Figure 20.3.
Errors in CIA due to uncertainties in major
element concentrations are ,1.5% (,1 CIA
unit). 2s error bars are indicated for MIA
values. MB Fm., Masirah Bay Formation.
CIA values are corrected to show the
maximum possible effect of potassium
metasomatism during burial diagenesis, as
explained by Rieu et al. (2007c, p. 301).
(After Rieu et al. 2007c.)
in the stratigraphy in Wadi Mistal (inclination of þ498) indicate
a magnetic reversal during Fiq times. For palaeolatitude determi-
nation, a set of samples from the Hadash Formation provide
more data (12 sites), with tilt-corrected inclinations between
–7.68 and –39.38 for reversed polarities, and between þ218 and
þ298 for normal polarities (Kilner et al. 2005, Data Repository
Table 1). The average palaeopole and palaeolatitude based on 25
sites throughout Oman quoted by Kilner et al. (2005) is unfortu-
nately a statistical average of samples of ages that are spread
from c. 800 Ma to c. 640 Ma (Allen 2007; Allen et al. 2011). Elim-
inating all but the Fiq and Hadash samples, it is evident that the
variation in the B component is large, and the average palaeo-
latitude of 138 for the Muscat region is therefore open to doubt.
Geochronological constraints
Geochronological data (Brasier et al. 2000; Amthor et al. 2003;
Allen & Leather 2006; Rieu 2006; Bowring et al. 2007; Rieu
et al. 2007a) were integrated to obtain the best available con-
straints on the ages of stratigraphic units within the Huqf Super-
group by Allen (2007) (Fig. 20.6).
Brasier et al. (2000) provided a U –Pb date on zircons recovered
from a tuffaceous ash interbedded with thin (centimetre-scale)
sandstones and thick, cleaved diamictites from the Ghubrah For-
mation in Wadi Mistal. The sample yielded an intercept age of
723(þ16/– 10) Ma. Resampling and analysis (Bowring et al.
2007) revealed a zircon population of uniform colour and size,
and good crystal forms lacking any obvious evidence of reworking.
Furthermore, of the 25 crystals analysed, 22 defined a discordia
with an upper intercept age of 713.7 + 0.5 Ma, whereas the final
three grains are variably discordant, perhaps due to a small xeno-
crystic component. The lack of an admixture of older zircons, com-
pared to their abundance in nearby detrital beds (sandstone clast
and diamictite matrix, see below), suggests that 714 Ma is a
good estimate of the depositional age of the tuffaceous ash. This
is a rare and valuable constraint on the age of a Neoproterozoic
glaciation because the tuffaceous ash is interbedded with glacial
diamictites.
A number of detrital zircons have been analysed from the
same Ghubrah Formation locality in Wadi Mistal (Bowring et al.
2007). The sandstone clast yielded five grains in the age range
 THE ABU MAHARA GROUP
ABU MAHARA GROUP, JABAL AKHDAR
Neoproterozoic Mesoproterozoic
c.664 (1 grain) 675 810-860 (10 grains
c.920 (1 grain)
645 715-718 840-860
790-810
750
700-720 1050
750-826 (detrital grains in ash)
c.750 (3 grains) 800-830 (4 detrital grains)
860
810 910
735
755 (matrix)
713.7±0.5 (n=22)
812-826 (5 detrital grains)
710
810
0.5 1.0 1.5
Bowring et al. (2007) data (Lahan core) Age (Ga)
Detrital grains (diamictites, turbidites)
Detrital grains (tuffs, ashes)
Syn-magmatic (Ghubrah date)
812– 826 Ma, and a sample of sandy diamictite matrix yielded
three grains dated as 755 Ma.
Rieu et al. (2007a) sampled a graded siltstone situated 1 m
above the thin tuffaceous ash yielding the 713.7 + 0.5 Ma date.
Of the 79 detrital zircon ages (derived by LA-ICPMS) compiled
(selected as ,25% discordant and 2s errors ,10% for
235
U – 207Pb and 238U – 206Pb), 18 passed a stricter filter (,10%
discordant and 2s errors ,20% for 235U – 207Pb and ,10% for
238
U – 206Pb). These most concordant data yield a youngest sub-
population with a peak at 710 Ma, which is in excellent agreement
with the age of the underlying tuffaceous ash. An older subpopu-
lation, which makes up 10% of all analysed zircons, ranges from
750 to 910 Ma, but only two grains pass the stricter filter, with
ages at 810 Ma. It is clearly hazardous to infer the detrital geochro-
nology of the entire Ghubrah Formation from samples collected
from several lithologies at just one locality, and further sampling
and analysis is required. With the data currently available, we
infer an exclusively Neoproterozoic juvenile source for the
Ghubrah zircons, with a syndepositional crystallization age of
714 Ma and a reworked contribution from older sources aged
755– 826 Ma. The c. 755 Ma crystals may represent magmatism
associated with the beginning of continental extension responsible
for the Ghubrah basins, or derivation from c. 760 Ma intrusives
associated with the welding together of arc and gneiss terranes
in the Arabian – Nubian Shield to the west. The older zircons
(c. 810–825 Ma) have identical ages to basement rocks exposed
at Al Jobah (Allen & Leather 2006; Allen 2007) at the northern
end of the Huqf region and to the ages of peak metamorphism
and magmatism in the gneisses of Mirbat (Mercolli et al. 2006;
Allen et al. 2011).
A number of attempts have been made to date the volcanic and
volcaniclastic Saqlah unit at the base of the Fiq Formation.
Bowring et al. (2007) analysed a sample from the Saqlah in
Wadi Mistal, recovering seven zircons, three of which yielded an
age of c. 750 Ma and four of which yielded ages in the range
c. 800 –830 Ma. Further samples of ashy material yielded dates
between 750 and 826 Ma, indicating reworked zircons from
sources indistinguishable from those of the Ghubrah. The most
259
Palaeoproterozoic
Fiq n=307 (167)
Fig. 20.6. Zircon geochronology, Abu
Saqlah n=95 (28) Mahara Group, Jabal Akhdar (after Allen
2007, supplemented by Bowring et al.
2007). Histograms of zircon ages derived by
LA-ICPMS, reported in Rieu (2006) and
Rieu et al. (2007a). Grey shaded area shows
zircons passing first filter, solid line is
zircons passing stricter second filter (see
Ghubrah n=79 (18)
text). Sample number n represents zircons
contributing to the histogram; bracketed
term is the number passing the first filter but
failing the stricter filter. Stars indicate
detrital zircon data in Bowring et al. (2007).
Solid bars beneath the time scale show the
2.0 2.5 main age ranges of detrital zircons. Note the
progressively older average age of the
detrital zircon distribution in the younger
stratigraphy and the appearance of Meso-
and Palaeoproterozoic sources in the Fiq
Formation.
concordant of Rieu et al.’s (2007a) dataset (28 dates) give a
main peak at 860 Ma, and subpopulation peaks at 730, 815 and
910 Ma. The youngest zircons may have been derived from
igneous rocks associated with the accretion of island arc and gneis-
sic terranes in the Arabian –Nubian megasuture (c. 760 Ma), late
post-tectonic plutons such as those of the Mirbat area (c. 700–
750 Ma), or the start of continental extension at c. 720 Ma. The
older zircons reveal the same sources of plutons formed during a
major period of subduction-related crustal growth at .800 Ma.
Although disappointing in not providing a syndepositional age
for the Saqlah, the results confirm older syn-Ghubrah and juvenile
Neoproterozoic basement sources.
Bowring et al. (2007) analysed a turbiditic sandstone from the
Fiq Formation collected from the northern part of Wadi Mistal.
Of 12 single grains analysed, one grain was c. 920 Ma, ten
grains were in the range 810– 860 Ma and one grain yielded an
age of c. 664 Ma. A further sample of an ashy sediment inter-
bedded with dropstone-bearing laminites and diamictites from
core recovered from the Lahan-1 well, thought to be equivalent
to a position very near the top of the Fiq, yielded zircons with
ages in the range c. 645– 718 Ma, with significant scatter. Only
five young grains grouped around 645 Ma were present. Conse-
quently, it is safe to merely conclude that the top of the Fiq
Formation must be younger than 645 Ma. Rieu et al.’s (2007a)
larger data set of Fiq samples from Wadi Sahtan and Wadi
Mistal (307 analyses, 167 of which passed the stricter filter)
reveals important Mesoproterozoic (1%), Palaeoproterozoic
(5%) and even Archaean (,1%) components. Main peaks of the
most concordant data are at c. 790 –810 Ma and c. 840– 860 Ma,
with secondary peaks or shoulders at c. 700– 720 Ma and
c. 750 Ma.
Combining all data, we conclude that the depositional age of the
glaciogenic Fiq Formation must be younger than 645 Ma, but that
the formation mostly contains detrital zircons derived from the
same Neoproterozoic sources as seen in older parts of the Abu
Mahara Group, with two additional components: (i) first cycle
derivation of very old sources, suggesting progressively deep
exhumation/unroofing of accreted terranes in the catchment
260 P. A. ALLEN ET AL.
areas of the Fiq basins, or inheritance of older zircons derived from
Neoproterozoic terranes (Hargrove et al. 2006); (ii) minor mag-
matic sources (.645 Ma) that are most likely contemporaneous
with Fiq basin development.
Discussion and uncertainties
The Fiq Formation has been used as an example of repeated glacial
advance and retreat before post-glacial transgression, signifying
dynamic glaciation and vigorously acting sediment routing
systems throughout the glacial epoch (Allen & Etienne 2008).
Glacially influenced strata, principally proximal and distal
rainout diamictites bearing dropstones, lonestones and faceted
and striated clasts, are embedded in non-glacial stratigraphy domi-
nated by hemipelagics, sandstone turbidites and debrites, but
including wave-rippled sandstones requiring an ice-free sea
surface. Further details of the sedimentology of the Fiq Formation
and its use as a test for the Snowball Earth hypothesis can be found
in Leather (2001), Leather et al. (2002), Allen et al. (2004) and
Allen & Etienne (2008).
Although the glacial intervals in the Sultanate of Oman have
received some prominence in the literature, there remain areas of
uncertainty that would benefit from further work. Chief among
these are the following:
† The age of the Fiq Formation is only constrained to terminate at
younger than 645 Ma based on the presence of a small number
of detrital zircons in core. A more systematic study of detrital
zircon ages in the Fiq Formation is warranted.
† The relationship between the Fiq and Ghubrah formations is
poorly known and requires structural mapping in the key
wadis, especially Wadi Mistal. The absence of a major strati-
graphic gap would imply glaciation took place intermittently
throughout the period from c. 720 to 640 Ma. The presence of
a significant stratigraphic gap and/or angular unconformity
would support the idea of discrete glacial epochs in the older
and younger Cryogenian.
† The Ghubrah Formation is almost entirely unknown sedimento-
logically. Notwithstanding its commonly highly cleaved nature,
there is scope for a sedimentological study of the Formation
across the Jabal Akhdar region, including Wadi Mu’aydin.
Conclusions
The .400-m-thick Ghubrah Formation is dominated by thick,
mostly unstratified marine rainout diamictites, with an interbedded
tuffaceous ash that has yielded a U –Pb zircon age of 713.7 +
0.5 Ma. The overlying ,1.5-km-thick Fiq Formation commences
locally with the volcanogenic Saqlah unit of submarine pillow
basalts, ashes and volcaniclastic rocks, which passes up into the
well-bedded cycles of the Fiq Formation, comprising a wide
range of glaciogenic and non-glacial sediment gravity flow and
shallow marine facies. Detrital zircons from ashy beds in the
upper part of the Fiq Formation are as young as 645 Ma. The
exact relationship between the Ghubrah Formation and the Fiq
Formation is not fully resolved and requires further study.
Calculation of CIA from mudstones and the muddy matrix of
diamictites reveals large excursions attributable to secular vari-
ations in the intensity of chemical weathering on contemporary
land surfaces. Such variations were most likely driven by pulses
of glaciation separated by interglacial periods. The deposition of
the transgressive cap carbonate coincides with a strong increase
in CIA, indicative of climatic transit following glaciation.
Carbon-isotopic ratios in the Hadash carbonate range from –1 to
–8‰, but variability between sample stations is large, making
intrabasinal correlation problematic and extrabasinal correlation
hazardous.
We are grateful for the financial and logistical support of Petroleum Develop-
ment Oman over many years and specifically the help of H. al Siyabi, M.
Newall, J. Amthor and J. Schreurs. We are also grateful for the additional
financial support of Enterprise Ireland and the Research Committee of ETH-
Zürich. A large number of individuals and field parties have provided input
on the Abu Mahara Group of the Jabal Akhdar for which we are grateful:
B. Levell, A. Heward, P. Hoffman, G. Halverson, A. Matter, and the partici-
pants of the field trip associated with the 24th IAS Meeting of Sedimentology
in Muscat, January 2005. We are grateful for the reviews of N. Eyles and I.
Fairchild, and the editing of E. Arnaud. This represents a contribution of the
IUGS- and UNESCO-funded IGCP (International Geoscience Programme)
Project #512.
References
Allen, P. A. 2007. The Huqf Supergroup of Oman: Basin development
and context for Neoproterozoic glaciation. Earth Science Reviews,
84, 139– 185, doi: 10.1016/j.earscirev.2007.06.005.
Allen, P. A. & Leather, J. 2006. Siliciclastic marine sedimentation in
the aftermath of a Marinoan glacial epoch: The Masirah Bay For-
mation, Huqf Supergroup, of Oman. Precambrian Research, 144,
167– 198, doi: 10.1016/j.precamres.2005.10.006.
Allen, P. A. & Etienne, J. L. 2008. Sedimentary challenge to snowball
Earth. Nature Geoscience, 1, doi: 10.1038/ngeo355.
Allen, P. A., Leather, J. & Brasier, M. D. 2004. The Neoproterozoic
Fiq glaciation and its aftermath, Huqf Supergroup of Oman. Basin
Research, 16, 507– 534, doi: 10.1111/j.1365-2117.2004.00249.x.
Allen, P. A., Rieu, R., Etienne, J. L., Matter, A. & Cozzi, A. 2011.
The Ayn Formation of the Mirbat Group, Dhofar, Oman. In:
Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geo-
logical Record of Neoproterozoic Glaciations. Geological Society,
London, Memoirs, 36, 239–249.
Amthor, J. E., Grotzinger, J. P., Schröder, S., Bowring, S. A.,
Ramezani, J., Martin, M. W. & Matter, A. 2003. Extinction of
Cloudina and Namacalathus at the Precambrian – Cambrian boundary
in Oman. Geology, 31, 431–434.
Anderson, J. B., Brake, C. F. & Myers, N. C. 1984. Sedimentation on
the Ross Sea continental shelf, Antarctic. Marine Geology, 57,
295– 334.
Bahlburg, H. & Dobrzinski, N. 2011. A review of the Chemical Index
of Alteration (CIA) and its application to the study of Neoproterozoic
glacial deposits and climate transitions. In: Arnaud, E., Halverson,
G. P. & Shields-Zhou, G. (eds) The Geological Record of Neopro-
terozoic Glaciations. Geological Society, London, Memoirs, 36,
81 – 92.
Beurrier, M., Bechennec, F., Rabu, R. & Hutin, G. 1986. Geological
Map of Rustaq, Sheet NF 40-3D, scale 1:100,000. Directorate General
of Minerals, Oman Ministry of Petroleum and Minerals.
Boulton, G. S. 1978. Boulder shapes and grain-size distribution of debris
as indicators of transport paths through a glacier and till genesis. Sedi-
mentology, 25, 773– 799.
Blendinger, W., van Vliet, A. & Hughes Clarke, M. W. 1990.
Updoming, rifting and continental margin development during the
Late Palaeozoic in northern Oman. In: Robertson, A. H. F.,
Searle, M. P. & Ries, A. C. (eds) The Geology and Tectonics of
the Oman Region. Geological Society, London, Special Publications,
49, 27 – 37.
Bowring, S. A., Grotzinger, J. P., Condon, D. J., Ramezani, J.,
Newall, M. J. & Allen, P. A. 2007. Geochronological constraints
on the chronostratigraphic framework of the Neoproterozoic Huqf
Supergroup, Sultanate of Oman. American Journal of Science, 307,
1097– 1145, doi: 10.2475/10.2007.01.
Brand, U. & Veizer, J. 1980. Chemical diagenesis of a multicomponent
carbonate system 1: Trace elements. Journal of Sedimentary Petrol-
ogy, 50, 1219– 1236.
Brand, U. & Veizer, J. 1981. Chemical diagenesis of a multicomponent
carbonate system 2: Stable isotopes. Journal of Sedimentary Petrol-
ogy, 51, 987– 997.
Brasier, M. D., McCarron, G., Tucker, R., Leather, J., Allen, P. A.
& Shields, G. 2000. New U– Pb zircon dates for the Neoproterozoic
Ghubrah glaciation and for the age of the top of the Huqf Supergroup,
 THE ABU MAHARA GROUP
Oman. Geology, 28, 175–178, doi: 10.1130/0091-7613(2000)028
,0175:NUPZDF.2.2.CO;2.
Bristow, T. F. & Kennedy, M. J. 2008. Carbon isotope excursions and
the oxidant budget of the Ediacaran atmosphere and ocean.
Geology, 36, 863–866, doi: 10.1130/G24968A.1.
Brodzikowski, K. & van Loon, A. J. 1991. Glaciogenic Sediments.
Developments in Sedimentology, 49, Elsevier, Amsterdam, 674.
Burns, S. J. & Matter, A. 1993. Carbon isotopic record of the
latest Proterozoic from Oman. Eclogae Geologicae Helvetiae, 86,
595– 607.
Costa, J. E. 1988. Rheologic, geomorphic, and sedimentologic differen-
tiation of water floods, hyperconcentrated flows, and debris flows.
In: Baker, V. R., Kochel, R. C. & Patton, P. C. (eds) Flood Geo-
morphology. Wiley, New York, 113–122.
Cozzi, A. & Al Siyabi, H. 2004. Sedimentology and play potential of
the late Neoproterozoic Buah carbonates of Oman. GeoArabia, 9,
11 – 36.
Deynoux, M. 1985. Terrestrial or waterlain diamictites? Three case
studies from the Late Precambrian and Late Ordovician glacial
drifts in West Africa. Palaeogeography, Palaeoclimatology,
Palaeoecology, 51, 97 –142.
Dowdeswell, J. A., Hambrey, M. J. & Wu, R. 1985. A comparison of
clast fabric and shape in Late Precambrian and modern glaciogenic
sediments. Journal of Sedimentary Petrology, 55, 691– 704.
Elverhoi, A., Pfirman, S. L., Solheim, A. & Larssen, B. B. 1989.
Glaciomarine sedimentation in epicontinental seas exemplified by
the northern Barents Sea. Marine Geology, 85, 225– 250.
Eyles, C. H. 1988. Glacially and tidally influenced shallow marine sedi-
mentation of the Late Precambrian Port Askaig Formation. Palaeo-
geography, Palaeoclimatology, Palaeoecology, 68, 1– 25.
Eyles, C. H. & Lagoe, M. B. 1990. Sedimentation patterns and facies geo-
metries on a temperate glacially-influenced continental shelf: the
Yakataga Formation, Middleton Island, Alaska. In: Dowdeswell,
J. A. & Scourse, J. D. (eds) Glaciomarine Environments: Processes
and Sediments. Geological Society, London, Special Publications,
53, 363– 386.
Eyles, N., Eyles, C. H. & Miall, A. D. 1983. Lithofacies types and ver-
tical facies models: an alternative approach to the description and
environmental interpretation of glacial diamict and diamict
sequences. Sedimentology, 30, 393– 410.
Eyles, C. H., Eyles, N. & Miall, A. D. 1985. Models of glaciomarine
sedimentation and their application to the interpretation of ancient
glacial sequences. Palaeogeography, Palaeoclimatology, Palaeoe-
cology, 51, 15 –84.
Fairchild, I. J. & Hambrey, M. J. 1984. The Vendian succession of
northeastern Spitzbergen: petrogenesis of a dolomite-tillite associ-
ation. Precambrian Research, 26, 111– 167.
Fedo, C. M., Nesbitt, H. W. & Young, G. M. 1995. Unravelling the
effects of potassium metasomatism in sedimentary rocks and
paleosols, with implications for paleoweathering conditions and pro-
venance. Geology, 23, 921–924, doi: 10.1130/0091-7613(1995)023
,0921:UTEOPM.2.3.CO;2.
Fike, D. A., Grotzinger, J. P., Pratt, L. M. & Summons, R. E. 2006.
Oxidation of the Ediacaran Ocean. Nature, 444, 744– 747, doi:
10.1038/nature05345.
Fölling, P. G. & Frimmel, H. E. 2002. Chemostratigraphic correlation of
carbonate successions in the Gariep and Saldania Belts, Namibia and
South Africa. Basin Research, 14, 69– 88.
Frimmel, H. E., Fölling, P. G. & Eriksson, P. G. 2002. Neoproterozoic
tectonic and climatic evolution recorded in the Gariep Belt, Namibia
and South Africa. Basin Research, 14, 55 –68.
Gass, I. G., Ries, A. C., Shackleton, R. M. & Smewing, J. D. 1990.
Tectonics, geochronology and geochemistry of the Precambrian
rocks of Oman. In: Robertson, A. H. F., Searle, M. P. & Ries,
A. C. (eds.) The Geology and Tectonics of the Oman Region. Geologi-
cal Society, London, Special Publications, 49, 585– 599.
Glennie, K. W. 1977. Outline of the geology of Oman. Mémoires de la
Societé Géologique de France, hors series, 8, 25 – 31.
Glennie, K. W., Boeuf, M. G. A., Hughes Clark, M. W.,
Moody-Stuart, M., Pilaar, W. F. H. & Reinhardt, B. M. 1974.
Geology of the Oman Mountains. KSEPL, Rijswijk, The Netherlands,
423.
261
Gorin, G. E., Raacz, L. G. & Walter, M. R. 1982. Late
Precambrian –Cambrian sediments of Huqf Group, Sultanate of
Oman. American Association Petroleum Geologists Bulletin, 66,
2609–2627.
Gray, D. R., Miller, J. M. & Gregory, R. T. 2005. Strain state and
kinematic evolution of a fold-nappe beneath the Samail Opiolite,
Oman. Journal Structural Geology, 27, 1986– 2007.
Halverson, G. P., Maloof, A. C. & Hoffman, P. F. 2004. The Marinoan
glaciation (Neoproterozoic) in northeast Svalbard. Basin Research,
16, 297– 324.
Hargrove, U. S., Stern, R. J., Kimura, J.-I., Manton, W. I. & Johnson,
P. R. 2006. How juvenile is the Arabian– Nubian Shield? Evidence
from Nd isotopes of pre-Neoproterozoic inherited zircon in the Bi’r
Umq suture zone, Saudi Arabia. Earth & Planetary Science Letters,
252, 308–326.
Hoffman, P. F. & Schrag, D. P. 2002. The snowball Earth hypothesis:
testing the limits of global change. Terra Nova, 14, 129– 155,
doi: 10.1046/j.1365-3121.2002.00408.x.
Hoffman, P. F., Kaufmann, A. J., Halverson, G. P. & Schrag, D. P.
1998a. A Neoproterozoic snowball Earth. Science, 281, 1342– 1346,
doi: 10.1126/science.281.5381.1342.
Hoffman, P. F., Kaufman, A. J. & Halverson, G. P. 1998b. Comings
and goings of global glaciation on a Neoproterozoic tropical platform
in Namibia. GSA Today, 8, 1– 9.
Hughes Clarke, M. W. 1988. Stratigraphy and rock nomenclature in the
oil-producing area of interior Oman. Journal of Petroleum Geology,
11, 5– 60.
Kapp, H. E. & Llewellyn, P. G. 1965. The geology of the Central Oman
Mountains. Report S00005-9, Geological Group, Petroleum Develop-
ment Oman.
Kaufman, A. J. & Knoll, A. H. 1995. Neoproterozoic variations in the
C-isotopic composition of seawater: stratigraphic and biogeochem-
ical implications. Precambrian Research, 73, 27– 49.
Kellerhals, P. & Matter, A. 2003. Facies analysis of a glaciomarine
sequence, the Neoproterozoic Mirbat Sandstone Formation, Sultanate
of Oman. Eclogae Geologicae Helvetiae, 96, 49– 70.
Kennedy, M. J. 1996. Stratigraphy, sedimentology and isotope geochem-
istry of Australian Neoproterozoic postglacial cap dolostones: degla-
ciation, d13C excursions, and carbonate precipitation. Journal of
Sedimentary Research, 66, 1050–1064.
Kilner, B., MacNiocaill, C. & Brasier, M. D. 2005. Low-latitude gla-
ciation in the Neoproterozoic of Oman. Geology, 33, 413– 416,
doi: 10.1130/G21227.1.
Leather, J. 2001. Sedimentology, chemostratigraphy and geochronology
of the lower Huqf Supergroup, Oman. PhD thesis, Trinity College
Dublin, vols 1 and 2.
Leather, J., Allen, P. A., Brasier, M. D. & Cozzi, A. 2002. Neoproter-
ozoic snowball Earth under scrutiny: evidence from the Fiq glaciation
of Oman. Geology, 30, 891– 894.
Le Guerroué, E. 2010. Duration and synchroneity of the largest negative
carbon isotope excursion on Earth: The Shuram/Wonoka
anomaly. Comptes Rendus Geosciences. doi: 10.1016/j.crte.2009.
12.008.
Le Guerroué, E. & Cozzi, A. 2010. Veracity of Neoproterozoic
negative C isotope values: The termination of the Shuram
negative excursion. Gondwana Research, 17, 653–661, doi:
10.1016/j.gr.2009.11.002.
Le Guerroué, E., Allen, P. A. & Cozzi, A. 2005. Two distinct
glacial successions in the Neoproterozoic of Oman. GeoArabia, 10,
17– 34.
Le Guerroué, E., Allen, P. A. & Cozzi, A. 2006a. Chemostratigraphic
and sedimentological framework of the largest negative carbon isoto-
pic excursion in Earth history: the Neoproterozoic Shuram Formation
(Nafun Group, Oman). Precambrian Research, 146, 68 – 92, doi:
10.1016/j.precamres.2006.01.007.
Le Guerroué, E., Allen, P. A., Cozzi, A., Etienne, J. L. & Fanning,
M. 2006b. 50 million year duration negative carbon isotopic
excursion in the Ediacaran ocean. Terra Nova, 18, 147–153, doi:
10.1111/j.1365-3121.2006.00674.x.
Le Guerroué, E., Allen, P. A. & Cozzi, A. 2006c. Parasequence devel-
opment in the Ediacaran Shuram Formation (Nafun Group, Oman):
high resolution stratigraphic test for primary origin of negative carbon
262 P. A. ALLEN ET AL.
isotopic ratios. Basin Research, 18, 205– 220, doi: 10.1111/
j.1365-2117.2006.00292.x.
Le Métour, J., Villey, M. & de Gramont, X. 1986. Geological map of
Quryat, Sheet NF 40-4D, scale 1:100,000. Directorate of Minerals,
Oman Ministry of Petroleum and Minerals.
Loosveld, R., Bell, A. & Terken, J. 1996. The tectonic evolution of
interior Oman. GeoArabia, 1, 28 –50.
Mackiewicz, N. E., Powell, R. D., Carlson, P. R. & Molnia, B. F.
1984. Interlaminated ice-proximal glaciomarine sediments in Muir
Inlet, Alaska. Marine Geology, 57, 113– 147.
Mann, A. & Hanna, S. S. 1990. The tectonic evolution of pre-Permian
rocks, Central and Southeastern Oman Mountains. In: Robertson,
A. H. F., Searle, M. P. & Ries, A. C. (eds) The Geology and
Tectonics of the Oman Region. Geological Society, London,
Special Publications, 49, 307– 325.
Marshall, J. D. 1992. Climatic and oceanographic isotopic signals from
the carbonate rock record and their preservation. Geological Maga-
zine, 129, 143– 160.
McCarron, G. M. E. 2000. The Sedimentology and Chemostratigraphy
of the Nafun Group, Huqf Supergroup, Oman. PhD thesis, University
of Oxford.
Mercolli, I., Briner, A. P., Frei, R., Schönberg, R., Nagler, T. F.,
Kramers, J. & Peters, T. 2006. Lithostratigraphy and
geochronology of the Neoproterozoic crystalline basement of
Salalah, Dhofar, Sultanate of Oman. Precambrian Research, 145,
182– 206.
Miller, J. M. G. 1996. Glacial sediments. In: Reading, H. G. (ed.) Sedi-
mentary Environments: Processes, Facies and Stratigraphy. Black-
well Science Ltd, Oxford, 454– 484.
Moncrieff, A. C. M. 1989. Classification of poorly sorted sedimentary
rocks. Sedimentary Geology, 65, 191–194.
Moncrieff, A. C. M. & Hambrey, M. J. 1990. Marginal marine glacial
sedimentation in the late Precambrian succession of east Greenland.
In: Dowdeswell, J. A. & Scourse, J. D. (eds) Glaciomarine
Environments: Processes and Sediments. Geological Society,
London, Special Publications, 53, 387– 410.
Myrow, P. M. & Kaufman, A. J. 1999. A newly discovered cap carbonate
above Varanger-age glacial deposits in Newfoundland. Canadian
Journal of Sedimentary Research, 69, 784– 793.
Narbonne, G. M., Kaufman, A. J. & Knoll, A. H. 1994. Integrated che-
mostratigraphy and biostratigraphy of the Windermere Supergroup,
northwestern Canada: implications for Neoproterozoic correlations
and the early evolution of animals. Geological Society of America
Bulletin, 106, 1281–1292.
Nardin, T. R., Hein, F. J., Gorsline, D. S. & Edwards, B. D. 1979. A
review of mass movement processes, sediment and acoustic charac-
teristics, and contrasts in slope and base-of-slope systems versus
canyon-fan-basin systems. In: Doyle, L. J. & Pilkey, O. H. (eds)
Geology of Continental Slopes. Society of Economic Paleontologists
and Mineralogists, Special Publication. 27, 61 –73.
Nesbitt, H. W., Young, G. M., McLennan, S. M. & Keays, R. R. 1996.
Effects of chemical weathering and sorting on the petrogenesis of
siliciclastic sediments, with implications for provenance studies.
Journal of Geology, 104, 525– 542.
Ovenshine, A. T. 1970. Observations of iceberg rafting in Glacier Bay,
Alaska, and the identification of ancient ice-rafted deposits. Bulletin
Geological Society America, 81, 891– 894.
Platel, J. P., Roger, J., Peters, T., Mercolli, I., Kramers, J. D. & Le
Métour, J. 1992a. Geological Map of Salalah, Sheet NE 40-09, 1:
250000, with explanatory notes. Directorate General of Minerals,
Oman Ministry of Petroleum and Minerals.
Platel, J. P., Le Métour, J., Berthiaux, A., Buerrier, M. & Roger, J.
1992b. Geological map of Juzor Al Halaaniyat, sheet NE 40-10, scale
1:250,000. Directorate General of Minerals, Oman Ministry of
Petroleum and Minerals.
Powell, R. D. & Molnia, B. F. 1989. Glaciomarine sedimentary pro-
cesses, facies and morphology of the south –southeast Alaska Shelf
and fjords. Marine Geology, 85, 359– 390.
Powell, R. & Domack, E. 1995. Glaciomarine processes and sediments.
In: Menzies, J. (ed.) Modern Glacial Environments. Butterworth-
Heinemann, Oxford.
Rabu, D. 1988. Géologie de l’authochthon des montagnes d’Oman, la
fenetre du Jabal Akhdar. PhD thesis, Université Pierre et Marie
Curie, Paris 6, and documents BRGM 130.
Rabu, D., Bechennec, F., Beurrier, M. & Hutin, M. 1986. Geological
map of Nakhl. Sheet NF-40-3E, scale 1:100,000. Oman Ministry of
Petroleum and Minerals, Directorate General of Minerals.
Rabu, D., Nehlig, P. et al. 1993. Stratigraphy and structure of the Oman
Mountains. Bureau de Récherches Géologiques et Minières, 221.
Rieu, R. 2006. Sedimentology, Stratigraphy and Geochemistry of the
Glacially Influenced Neoproterozoic Mirbat Group, Oman. PhD
thesis, ETH-Zürich.
Rieu, R., Allen, P. A., Cozzi, A., Kosler, J. & Bussy, F. 2007a. A com-
posite stratigraphy for the Neoproterozoic Huqf Supergroup of Oman:
integrating new litho-, chemo- and chronostratigraphic data of the
Mirbat area, south Oman. Geological Society London Journal, 164,
997– 1009.
Rieu, R., Allen, P. A., Plötze, M. & Pettke, T. 2007b. Compositional
and mineralogical variations in a Neoproterozoic glacially influenced
succession, Mirbat area, south Oman: Implications for paleoweather-
ing conditions. Precambrian Research, 154, 248– 265, doi: 10.1016/
j.precamres.2007.01.003.
Rieu, R., Allen, P. A., Plötze, M. & Pettke, T. 2007c. Climatic cycles
during a Neoproterozoic ‘snowball’ glacial epoch. Geology, 35,
299– 302, doi: 10.1130/G23400A.1.
Roger, J., Béchennec, F., Janjou, D., Le Métour, J., Wyns, R. &
Buerrier, M. 1992. Explanatory Notes to the Geological Map of
Ja-alan. 1:100,000 Sheet NF 40-8E. Directorate General of Minerals,
Ministry of Petroleum and Minerals.
Simonson, B. M. & Carney, K. E. 1999. Roll-up structures: Evidence of
in situ microbial mats in Late Archaean deep shelf environments. In:
Hagadorn, J. W., Pflueger, F. & Bottjer, D. J. (eds) Unexplored
Microbial Worlds. Palaios. 14, 13 –24.
Stow, D. A. V., Reading, H. G. & Collinson, J. D. 1996. Deep
seas. In: Reading, H. G. (ed.) Sedimentary Environments: Pro-
cesses, Facies and Stratigraphy. Blackwell Science Ltd, Oxford,
395– 453.
Tucker, M. E. 1986. Formerly aragonitic limestones associated with
tillites in the Late Proterozoic of Death Valley, California. Journal
of Sedimentary Petrology, 56, 818– 830.
Tschopp, R. H. 1967. The general geology of Oman. Proceedings of the
7th World Petroleum Congress, Mexico, 2, 231.
Villey, M., Le Métour, J. & Gramont, X. 1986. Geological map of
Fanjah, sheet NF 40-3F, scale 1:100,000. Oman Ministry of Pet-
roleum and Minerals, Directorate General of Minerals.
Wentworth, C. K. 1936. An analysis of the shape of glacial cobbles.
Journal of Sedimentary Petrology, 6, 85 –96.
Williams, G. E. 1979. Sedimentology, stable isotope geochemistry and
palaeoenvironment of dolostones capping late Precambrian glacial
sequences in Australia. Journal of the Geological Society of Austra-
lia, 26, 377–386.
Wright, R. & Anderson, J. B. 1982. The importance of sediment gravity
flow sediment transport and sorting in a glacial marine environment:
eastern Weddell Sea, Antarctica. Bulletin Geological Society
America, 93, 951–963.
Wright, V. P., Ries, A. C. & Munn, S. G. 1990. Intraplatformal
basin-fill from the Infracambrian Huqf Group, east-central Oman.
In: Robertson, A. H. F., Searle, M. P. & Ries, A. C. (eds) The
Geology and Tectonics of the Oman Region. Geological Society,
London, Special Publications, 49, 601–616.
 Chapter 21

The Tambien Group, Northern Ethiopia (Tigre)

NATHAN R. MILLER1*, DOV AVIGAD2, ROBERT J. STERN3 & MICHAEL BEYTH4

1
Department of Geological Sciences, Jackson School of Geosciences, University of Texas at Austin, Austin, TX 78712-0254, USA
2
Institute of Earth Sciences, Hebrew University of Jerusalem, Jerusalem 91904, Israel
3
Geosciences Department, University of Texas at Dallas, Richardson, TX 75083-0688, USA
4
Geological Survey of Israel, Jerusalem, Jerusalem 95501, Israel
*Corresponding author (e-mail: nrmiller@mail.utexas.edu)

Abstract: The Tambien Group of northern Ethiopia (Tigre), with probable correlatives in Eritrea, is a 2 –3-km-thick siliciclastic–
carbonate succession that was deposited in an intra-oceanic arc platform setting within the southern Arabian– Nubian Shield (ANS)
area (southern extension of the Nakfa Terrane) of the Mozambique Ocean. Its deposition occurred prior to ocean closure between con-
verging fragments of East and West Gondwana and concomitant structural emergence of the East African Orogen (EAO). The Tambien
Group is well exposed and best studied in the Mai Kenetal and Negash synclinoria, where litho- and chemostratigraphy (including
d13Ccarb, 87Sr/86Sr) provide the basis for a composite reference section. Two glaciogenic intervals have been suggested from exposures
within the Didikama and Matheos Formation in the Negash Synclinorium. No reliable palaeomagnetic data exist to constrain the palaeo-
latitude of Tambien Group deposition and the southern ANS, but palaeogeographic reconstructions and evaporite pseudomorphs in lower
carbonate units (Didikama Formation) imply low to intermediate latitudes (,458). Integration of available geochronological information
(regional magmatism and detrital zircon) suggests c. 775– 660 Ma as a plausible window constraining deposition of the prospective
glacial intervals.
The Tambien Group appears to preserve a coherent chemostratigraphic framework that can be effectively subdivided according to
shifts in d13Ccarb polarity [polarity intervals A (þ), B (–), C (þ), D (–)]. Slates underlying and interstratified with polarity interval A
carbonate preserve evidence of extreme chemical weathering that lessened prior to deposition of polarity interval B carbonate.
Tambien Group carbonate units have sedimentological characteristics consistent with both shallow and deeper marine depositional set-
tings. The lower prospective glacial interval lacks diagnostic sedimentological evidence of synglacial deposition, but is overlain by nega-
tive d13C carbonate (polarity interval B) with sedimentological characteristics consistent with well-documented cap-carbonate
successions. The upper prospective glacial interval in the Negash Synclinorium (Matheos Diamictite) best exhibits characteristics con-
sistent with glaciogenic deposition (matrix-supported polymictic clasts, possible dropstones, possible bullet-nosed and striated clasts). In
contrast to pericratonic rift margin settings that are common for Cryogenian glaciogenic deposits, palaeogeographic reconstructions for
the 775 –660 Ma timeframe place northern Ethiopia within an intra-oceanic setting that was likely far removed from cratonic hinterlands.
More work on Tambien Group sedimentology, geochronology and palaeogeography is required to better evaluate the extent and timing of
glacial conditions associated with the prospective glaciogenic intervals.

Supplementary material: Supplementary Table 21.1 of Tambien Group geochronological age constraints is available at http://www.
geolsoc.org.uk/SUP18462.

The Tambien Group is exposed (Fig. 21.1) throughout portions of
northern Ethiopia (Tigre Province) and Eritrea (NNE extensions
from Figure 21.1; Bizen domain and Adobha Abi terrane of
Beyth et al. (2003) and De Souza Filho & Drury (1998)), in
greenschist-grade terranes comprising the southern portion of the
Arabian –Nubian Shield (ANS). It may be equivalent to similar
carbonate-rich units in NE Sudan (Bailateb Group; Stern et al.
1994), SW Saudi Arabia (Hali Group; Greenwood et al. 1976),
and possibly western Yemen (inferred from its Red Sea conjugate
position). Its deposition marks mainly marine siliciclastic and car-
bonate sedimentation within the Mozambique Ocean, an extinct
Neoproterozoic ocean basin destroyed during the late Neoprotero-
zoic consolidation of Greater Gondwana with the emergence of the
East African Orogen (EAO) (Stern 1994). The Tambien Group is
best studied in northern Ethiopia from the Mai Kenetal (western
flank near 138550 N, 388500 E) and Negash (southern extent near
138500 N, 398370 E) synclinoria, where much of the Tambien
Group is continuously exposed, and these localities provide the
substantial basis for regional litho- and chemostratigraphy.
Two possible glaciogenic intervals have been suggested within
the Tambien Group (Beyth et al. 2003; Miller et al. 2003, 2009).
The lower interval occurs as a greywacke-conglomerate interval
within slate of the lower Didikama Formation in the Negash
Synclinorium (Fig. 21.2a, column D), and the top of this interval
may correlate with the base of the Assem Limestone in the Mai
Kenetal Synclinorium (see discussion). The upper prospective
glacial interval tops the Tambien Group as diamictite of the
Matheos Formation in the core of the Negash Synclinorium
(Fig. 21.2a), and this interval may be equivalent to arkosic sand-
stone and conglomerate of the Dugub Formation that similarly
tops the Tambien Group in the western Shiraro area (Fig. 21.1,
Avigad et al. 2007). The upper diamictite unit at Negash was orig-
inally defined as the ‘Pebbly slate’ (Beyth 1972). It was
subsequently assigned within the Matheos Formation (Garland
1980) and informally described as ‘Pebbly Slate (diamictite)’ in
Miller et al. (2003), ‘Slate/Pebbly slate (Diamictite Slate)’ in
Alene et al. (2006) and ‘Negash Diamictite’ in Avigad
et al. (2007). Miller et al. (2009) subdivide the Matheos Formation
into three members, the youngest corresponding to the diamictite
facies. In consideration of a possible lower glaciogenic interval
in the Negash Synclinorium, the informal term ‘Matheos Diamic-
tite’ is suggested for the upper prospective glacial interval. In
addition to these northern Ethiopian localities, Eritrea hosts poss-
ible glaciogenic units that have not been systematically studied.
The southern ANS is still very much a frontier region in need of
systematic sedimentological, geochemical, and geochronological
studies of Neoproterozoic units. Much of what is known about the
Cryogenian Period for the region has been learned in only the past
decade. The earliest suggestion of a Late Proterozoic glaciation
was by Bibolini (1920), who described faceted clasts (facce piane)
within an unnamed pebbly mudstone-conglomerate unit (conglom-
erati poligenici) in northern Eritrea. The basic Neoproterozoic
stratigraphic framework for northern Ethiopia was established in
regional mapping by Beyth (1972), including description of the

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 263– 276. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.21
264 N. R. MILLER ET AL.
units now posited to have glacial associations. Motivated by the
Snowball Earth hypothesis, a number of reconnaissance studies
have since explored the depositional context of the Tambien
Group (Beyth et al. 2003; Miller et al. 2003; Alene et al. 2006; see
also Stern et al. 2006). More comprehensive regional investigations,
involving U–Pb zircon geochronology, chemical weathering
indices, and higher resolution sampling for integrated C- and
Sr-isotope stratigraphy, have further refined the age and range of
litho- and chemostratigraphic variations in the Tambien Group
(Sifeta et al. 2005; Avigad et al. 2007; Miller et al. 2009).
Structural framework
(See additional region-specific structural information in the
‘Glaciogenic deposits and associated strata’ section.) The greater
Fig. 21.1. Location of key Tambien Group
exposures (unshaded units) (A, Shiraro area; B,
Mai Kenetal Synclinorium; C, Negash
Synclinorium; D, Samre area) within northern
Ethiopia (Tigre Province) modified from Miller
et al. (2009). Metavolcano–sedimentary block
boundaries (Tadesse et al. 1999) occur only
within the Tsaliet Group, and are inferred to
represent accreted arc terranes or slivers in a
supra-subduction zone setting. Stars show
geochronological localities for syn-tectonic
intrusives (white), post-orogenic intrusives
(black) and detrital zircons (white stars with black
dots) discussed in the text and shown in
Figure 21.5 (also in Supplementary Table 21.1).
Prospectively equivalent upper Tambien Group
strata of the Gulgula Group occur in western
Eritrea (arrow) just west of the Shiraro Area, as
well as in the Adobha Abi terrane and Bizen
Domain (arrows) to the north in Eritrea. Dashed
box east of Mai Kenetal outlines the Werii study
area of Sifeta et al. (2005). M. Alem marks the
western limb locality (Madahne Alem) within the
Negash Synclinorium bearing 774.7 + 4.8 Ma
zircons in slate c. 16 m below lowest Tambien
Group carbonate beds (Avigad et al. 2007). The
Negash Synclinorium is structurally bounded to
the east by the Atsbi Horst (AH ). Inset map
(below) shows the location of an unnamed pebbly
mudstone unit in northeastern Eritrea (Cecioni
1981) that could correlate to the Tambien Group.
ANS consists of a patchwork of Neoproterozoic tectonostrati-
graphic terranes, now bisected by the Oligocene and younger
Red Sea rift. ANS terranes include significant volumes of
juvenile Neoproterozoic crust generated within spreading centers,
arc and back-arc settings of the Mozambique Ocean (Stern 1994).
Sutures between terranes, many with ophiolites and dated meta-
morphic assemblages, document collisional deformation and help
constrain the timing of terrane amalgamation. Many sutures have
appreciable strike – slip offsets, and these may broadly relate to
c. 600 Ma escape tectonics (Burke & Sengör 1986), during which
ANS terranes were progressively sandwiched by, and offset
between, obliquely converging Gondwana cratonic blocks (de
Souza Filho & Drury 1998, and references therein). Tambien
Group exposures in Tigre occur within the presumed southern
extension of the Nakfa terrane of Eritrea. The Nakfa terrane is
one of several suture-bounded low-grade volcano-sedimentary
 THE TAMBIEN GROUP, NORTHERN ETHIOPIA 265

(a)
Relation A. B. C. D.
to syn- Shiraro Plains Mai Kenetal Negash Negash
& post- Region Synclinorium Synclinorium Synclinorium
tectonic
Tadesse Beyth 1972 Miller et al. 2009
intrusives 1999 Beyth 1972 (Tadesse 1999) Beyth 1972
(Garland 1980)
Pz & Mz
Adigrat Ss Adigrat Ss Enticho Ss Enticho Ss Enticho Ss
Seds

Dugub Fm - - - Mentebtab Fm
Algal* Ls Mai Kenetal Ls (200m) Pebbly slate (200m)
Diamictite Mbr
(Tselim Imni Ls) ?

Matheos Fm
---- Tambien Group ----
Shiraro Group ?
Black detrital ls
Enda Zebi Fm Transitional Mbr
Shiraro Cgl Tsedia Slate (500m) (300m)
? (subdivided into
Post-orogenic granitoids

Black Ls Mbr
Arkosic ss upper Bilato and Mica dolomite &
lower Logmiti Slates) slate (700m)
Mentebtab Ss
Assem LS (300m) Slate & dolomite
Dolomite > Slate

Didikama Fm
Bedded ls (Filafil Ls) (200m)
Didikama Fm

Mica dolomite
Slate > Dolomite
Purple slate w/grn
Purple slate w/grn Werii Slate (1000m) reduction spots; GCI ?
reduction spots (Segali Slate) greywacke (300m) Lower Slate
? (Atsbi Horst)

Syn- Undiff. Tsaliet Tsaliet Tsaliet

Tsaliet Group
Metavolc. Metavolcanics Metavolcanics Metavolcanics
(b)
d13C carb Polarity

Mai Kenetal Synform Negash Synform Afro-Arabian Peneplain

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Mai Kenetal LS C (++) Matheos Fm D (-) Diamictite Mbr ? ? ?
Transition Mbr 3
2 Tsedia Slate C (+) C (++) Black LS Mbr
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ unconformity?
Assem LS B (-) Didikama Fm B (-) Upper (dol>slate)
~~~~~~~~~~~~~~~ GCI ? ? ? 1
? ? ? paraconformity?
Werii Slate A (+?) A (+) Lower (slate >dol)
Lower Slate (Atsbi Horst)

Tsaliet Metavolcanics Tsaliet Metavolcanics

Fig. 21.2. (a) Lithostratigraphic subdivision of the Tambien Group in previous work. Prospective glacial intervals in columns A, C and D are superimposed from Beyth
et al. (2003) and Miller et al. (2009). The upper Algal Limestone unit in the Shiraro plains region (column A), originally termed the ‘Algal, stromatoporoidea (?)
limestone’ by Beyth (1972) was not mapped in the Axum Sheet (Tadesse 1999) and may occur in the Badme region (north of Shiraro, Fig. 21.1). (b) Chemo- and
lithostratigraphic correlations between the Mai Kenetal and Negash synclinoria used to create a composite Tambien Group stratigraphic section (see discussion).
Stratigraphic intervals in each section are alphabetized (A –D) according to the succession of associated d13Ccarb polarity (þ or – ) intervals. See Figure 21.3 for associated
chemostratigraphic characteristics (circled data sets: d13CTOC, 87Sr/86Sr, [Sr] and chemical weathering index) that support the correlations. Numbers 1–3 correspond to
the different segments used in constructing the composite stratigraphy of Figure 21.6. Abbreviations (A&B): Fm, formation; GCI, Greywacke –Conglomerate Interval,
the lower prospective glacial interval in the Negash Synclinorium; grn, green; Ls, limestone; Mbr, member; Metavolc, metavolcanics; Mz, Mesozoic; Pz, Palaeozoic;
Ss, sandstone; Undiff, undifferentiated. (Modified from Miller et al. 2009.)

terranes comprising the greater Nubian Tokar superterrane et al. 2005). An unconformable basal contact also was postulated
(Kröner et al. 1991). from considerable lateral variations observed among basal
The Nakfa terrane supracrustal assemblage records a lower Tambien metasediments (Arkin et al. 1971; Beyth 1972; Tadesse
sequence of igneous rocks associated with arc magmatism in and 1999). Tambien Group deposition, including the two prospective
around the southern ANS sector of the Mozambique Ocean, as glacial intervals, is thought to have occurred in an intra-oceanic
exemplified by calc-alkaline plutons and associated metavolcanic platform setting (above a substratum of consolidated arc terranes)
rocks of the Tsaliet Group. In western Tigre, the Tsaliet Group may following the main phase of Tsaliet arc magmatism in the region
be preserved as a series of accreted volcanic arc terranes within a (c. 780 + 30 Ma; Avigad et al. 2007). The extent to which
suprasubduction setting (Fig. 21.1; volcano-sedimentary blocks of Tsaliet arc subterranes were fully accreted prior to Tambien
Tadesse et al. 2000). Cessation of Tsaliet arc volcanism was fol- Group deposition is uncertain, and some syndepositional relief
lowed by deposition of weathered arc detritus (Sifeta et al. 2005) differentiation related to ongoing shortening and/or extension is
and in turn increasing proportions of carbonate sediments, as possible (Miller et al. 2009).
exemplified by the Tambien Group. The contact between the The entire Nakfa basement complex was deformed after depo-
Tsaliet and Tambien Group is poorly understood, ranging from sition of the Tambien Group in conjunction with closure of the
seemingly conformable and gradational to fault-bounded (Sifeta Mozambique Ocean between fragments of east and west Gondwana,
266 N. R. MILLER ET AL.
and the concomitant structural emergence of the EAO, beginning
c. 630 Ma. As a consequence, NNE-trending upright/overturned
folds and faults dominate the regional structural grain, with Tambien
Group exposures best preserved in synclinoria and grabens. Inter-
nal minor or secondary folds and faults complicate measured sec-
tions in the Mai Kenetal and Negash synclinoria. Restoring
shortening (200%) associated with these collisional structures
greatly extends the minimal aerial extent of the Tambien basin or
basins. The deformed Tsaliet-through-Tambien Group succession
was subsequently intruded by c. 610 Ma (‘Mareb’) granitoids
associated with crustal thickening and differentiation of the matur-
ing EAO. Whether or not, and for how long, Tambien Group depo-
sition continued prior to emplacement of the ‘Mareb’ post-orogenic
granitoids is unknown. Despite these post-depositional orogenic
processes, temperatures and pressures did not exceed low-
greenschist grade metamorphism and well-preserved primary sedi-
mentary structures are common in many Tambien Group carbonate
successions (Alene et al. 2006; Miller et al. 2009).
Extensive erosion of the EAO resulted in cutting of a widespread
regional unconformity prior to Cambro-Ordovician time (Avigad
et al. 2005), herein termed the Afro-Arabian Peneplain (AAP).
Uppermost Tambien Group exposures in the core of the Negash
Synclinorium (Matheos Diamictite) and possibly equivalent silici-
clastic deposits of the Dugub Formation in the western Shiraro
area (Fig. 21.1) are thought to be northern Ethiopia’s youngest
preserved Neoproterozoic metasediments below the AAP.
Stratigraphy
Beyth (1972) documented Tambien Group outcrops in the Shiraro
area, in several en echelon synclinoria to the east (i.e. Mai Kenetal,
Tsedia, Chemit and Negash), and the Escarpment area east of
Adigrat (Fig. 21.1, study areas A– C). Apparent regional facies
variations within the internal stratigraphy of the Tambien Group
were described, and a preliminary correlation was proposed for
three areas: Shiraro (west), Mai Kenetal (centre) and Negash
(east). Subsequent workers have modified the stratigraphy in
various ways as described below and shown in Figure 21.2a.
Figure 21.2b shows a recent correlation scheme proposed for the
Tambien Group based on integrated litho- and chemostratigraphy
of the Mai Kenetal and Negash synclinoria (Miller et al. 2009),
which is further elaborated in the discussion section.
Above the Tsaliet Metavolcanics, Beyth (1972) assigned four
formational designations within the Tambien Group based on the
Mai Kenetal type section (in stratigraphic order): Werii Slate,
Assem Limestone, Tsedia Slate and Mai Kenetal Limestone. In
the Negash Synclinorium, Beyth defined informal facies (in strati-
graphic order: greywacke and purple slate with green reduction
spots, slate and dolomite, mica dolomite and slate, black detrital
limestone, and pebbly slate), which he suggested might correlate
with the Mai Kenetal section (Fig 21.2a, columns C v. B).
Dolomite-slate successions similar to those of the lower Tambien
Group in the Negash Synclinorium were subsequently documented
in the Escarpment east of Adigrat, as well as in synclines in the
Shiraro area. Garland (1980) formalized Beyth’s Tsaliet Metavolca-
nics as the Tsaliet Group, and subdivided the Negash Tambien
Group succession (Fig. 21.2a, column D) into the lower Didikama
Formation (after similar dolomitic rocks in the Shiraro Area) and
overlying Matheos Formation (equivalent to Beyth’s informal
black detrital limestone and pebbly slate facies). In mapping of
the Mai Kenetal synclinorium, Tadesse (1999) used local names
to address the possibility that some of Beyth’s original type areas
(i.e. Werii Slate, Tsedia Slate) might reside in different volcano-
sedimentary arc terranes (Fig. 21.1). The main difference between
the two lithostratigraphic schemes is differentiation of Beyth’s
Tsedia Slate into a lower Logmiti Slate and overlying Bilato Lime-
stone and Slate (Fig. 21.2a, column B). Tadesse (1999) also formal-
ized a new Neoproterozoic stratigraphic scheme for the Shiraro
area, splitting Beyth’s informal Tambien Group facies into the
basal Didikama Formation and overlying Shiraro Group
(Fig. 21.2a, column A). Designated within the latter were three
metasiliciclastic formations (Enda Zebi, Dugub and Mentebtab
formations), each with respective lateral facies variations.
In the eastern limb of the Negash Synclinorium, Beyth (1972)
mapped the base of the Tambien Group as an interval of arkosic
sandstone and conglomerate, above a possible tectonic contact
with the Astbi Horst (Fig. 21.1, Fig. 21.2a column C). This
coarse metasiliciclastic interval corresponds to the lower prospec-
tive glacial interval, the Greywacke-Conglomerate interval (GCI)
of the Didikama Formation. The basal contact of the GCI has not
been studied in detail, but the underlying depositional sequence
involves a minimum of several hundred metres of conformably
bedded fine-grained marine metasedimentary units (mainly slate
with episodic thin carbonate interbeds up to a few centimetres
thick). This lower slate succession contrasts from coarse volcanic
agglomerate and conglomerate of the Tsaliet Group exposed at the
top of the Atsbi Horst (Fig. 21.1), but the contact between these
units has not been systematically mapped or studied. Miller
et al. (2003, 2009) include this lower slate interval (below the
GCI) within the Tambien Group, as the informal Lower Slate
member of the Didikama Formation (Fig. 21.2a, column D).
Above the GCI, the Didikama Formation transitions into varie-
gated slate that includes the first prominent dolomite beds (infor-
mal Slate . Dolomite member) and thicker bedded dolomite at
the top of the formation (informal Dolomite . Slate member).
The GCI and overlying purple slate (with green reduction spots;
the ‘Purple Egg Slate’ of Beyth 1972), with estimated combined
thickness of 300 m, forms a distinctive dark (grey-purple)
marker unit that can be traced laterally throughout much of the
c. 30-km-long exposed eastern synclinorial limb.
The Didikama Formation is overlain sharply, and very likely
unconformably (Beyth 1972, p. 74; Garland 1980, p. 14), by dis-
tinctive black limestone of the Matheos Formation. Miller et al.
(2009) subdivide the Matheos Formation into three members.
The basal Black Limestone Member is well laminated and partly
detrital (intraclasts), and transitions upward into non-calcareous
slate (Transitional Member) and eventually pebbly slate (Diamic-
tite Member).
Prospective glaciogenic deposits and associated strata
Lower prospective glaciogenic unit
Greywacke –Conglomerate Interval, lower Didikama Formation (Lower
Tambien Group). Beyth (1972) described the GCI within the
eastern flank of the Negash Synclinorium as poorly sorted, light-
grey to black greywacke, containing mostly medium grained to
pebble-sized, subrounded quartz, in a mud (slate) supported
matrix (Fig. 21.2a, column C). Thin conglomeratic layers and
ripple marks (exposed in bedding planes) also occur within the
unit. The unit fines upward into purple slate with distinctive
ovate greenish-yellow patches, interpreted as reduction spots.
This lower interval (c. 300 m thick, Beyth 1972) is overlain by var-
iegated slate (orange and green), which in turn grades upward
(with poor outcrop) into the lowest exposed dolomite beds of the
Didikama Formation. Similar purple spotted slate is reported
below the Didikama Formation in the Shiraro area (Beyth 1972;
Fig. 21.2a, column A). The GCI is underlain by well-layered
slate (with occasional sub-decimetre-thick carbonate interbeds
and minor foliation) that crops out continuously down-section
(eastward) for at least several hundred metres in the hanging
wall flank of the Atsbi Horst (Fig. 21.1).
Upper Werri Slate (Lower Tambien Group). In the NW flank of the
Mai Kenetal Synclinorium, the Assem Limestone lies sharply
and conformably above the Werii Slate and shares
 THE TAMBIEN GROUP, NORTHERN ETHIOPIA
sedimentological and isotopic characteristics of well-studied cap-
carbonate sequences (i.e. abrupt basal lithological transition to
high-energy carbonate facies, basal dolomite bed (c. 1 m thick)
transitioning upward into limestone, consistently negative
d13Ccarb, primary and/or early diagenetic (pre-compaction) pris-
matic and radial fibrous cements (Miller et al. 2009, figs S2D –
G; Hoffman et al. 2007)). The observation that many Cryogenian
carbonate successions with negative d13C compositions are
affiliated with glaciogenic intervals (e.g. Yoshioka et al. 2003;
Halverson et al. 2005; Corsetti et al. 2007 and other arguments,
see discussion) raises the possibility that the pre-Assem Limestone
depositional sequence could have a glacial affiliation. The under-
lying Werri Slate is mainly finely laminated, well-foliated, non-
calcareous slate that does not exhibit obvious glaciogenic sedimen-
tological characteristics. However, the nature of the upper Werri
Slate and its contact with the Assem Limestone has not been
studied regionally and poorly sorted metasiliciclastic intervals
are documented, at least locally, in the Werri Slate. For example,
Beyth (1972) reported occurrences of greywacke within the
Werii Slate regionally, and encountered poorly sorted rhyolitic
agglomerate with well-rounded metre-scale quartzitic fragments
at the southern end of Mai Kenetal syncline (around 138470 N
and 388510 E). Greywacke is also observed near its gradational
contact with Tsaliet Group (Alene et al. 2006).
Upper prospective glaciogenic unit
Diamictite Member, Matheos Formation (Upper Tambien Group). The
Matheos Formation, comprising the interior of the Negash Syncli-
norium, records a conformable transition from black limestone to
diamictite. Ranging from faintly layered (massive) to well-bedded,
with near vertical dips, the c. 250-m-thick basal Black Limestone
Member (Fig. 21.2a, column D) forms a distinctive ridge that
rims the interior of the Negash Synclinorium. Individual beds
range in thickness from decimetre- to metre-scale and are typically
finely laminated in thin section. Intraclastic grainstone intervals are
common. Towards the synclinorial axis, limestone beds thin
(becoming sub-decimetre thick) and interstratify with increasing
proportions of light to dark grey phyllitic slate. The overlying c.
100 m Transitional Member includes non-calcareous phyllitic
slate that passes upward into pebbly slate, without interbedded
carbonate. The primary depositional fabric involves mainly fine
(,1 to 3 cm scale) horizontal beds. Initial disparate sedimentary
clasts are sub-centimetre scale (compositions have not been
studied). The overlying c. 200-m-thick Diamictite Member
occupies the tightly folded core of the Negash Synclinorium and
is distinguished by an overall upward increase in clast abundance
and size (typically ,10 cm, but up to 20 cm in diameter). Clasts
are matrix-supported and the finer matrix retains horizontal layer-
ing, with some lateral pinching and swelling, throughout the
member. Syn-orogenic compression of the synclinal core has
deformed the diamictite to varying extents. More pelitic parts of
the diamictite display foliation and associated clasts are often some-
what elongated with pressure shadows. Despite these superimposed
features, clasts that appear to preferentially deform underlying
matrix laminae are relatively common. Diamictite clasts have sub-
rounded, elongate and angular shapes (including some bullet-nosed
clasts), and some planar clast surfaces may be striated (Miller et al.
2003). Clast lithologies are polymictic, including felsic volcanic
rocks, fine-grained black limestone and dolomite (including well-
rounded clasts that retain primary sedimentary structures; e.g.
oolite), low-grade semipelitic sediments, and rare volcanic con-
glomerate consistent with the upper Tsaliet Group.
Dugub Formation, Shiraro Group (Upper Tambien Group). Conglo-
merate and arkosic metasediments comprise youngest Tambien
Group exposures in the lowland Shiraro area of western Tigre
(Shiraro Group of Tadesse 1999; Fig. 21.2a, column A). These
267
rocks overlie carbonate units assigned to the Didikama For-
mation (Tadesse 1999) and crop out within the Shiraro Block
(Fig. 21.1), a graben-like depression that is fault-bounded to
the east against the Adi Hageray Block and extends westward
beyond the studied area into Eritrea. The areally most extensive
unit is the Dugub Formation, which we summarize following
Tadesse (1999) and our own studies 4.5 km ENE of the town
of Shiraro (Fig. 21.1).
The Dugub Formation consists of weakly metamorphosed con-
glomerate, arkosic sandstone and siltstone (Tadesse 1999), which
are regionally intercalated at different scales. Well-preserved
primary sedimentary structures are common as graded bedding,
cross-lamination, ripple marks, and flame and slump structures.
Elliptical (rounded to subrounded) and well-sorted metaconglome-
rate clasts (8 cm) comprise up to 40% of the rock volume. Clast
compositions include low-grade (Tsaliet-like) volcanic rocks
(chlorite schist, tuffaceous metasediments), phyllite, granite,
quartz pegmatite, quartzite and chert. Fine- to medium-grained
chlorite, muscovite, feldspar and quartz are notable components
of the metaconglomerate groundmass. Carbonate is notably
absent, but is reported (as scarce marble layers a few to tens of
metres thick) in the underlying Enda Zebi Formation (Tadesse
1999). Dugub metasandstone contains quartz pebbles (up to a
few millimetres across) and grades laterally and vertically into
non-pebbly sandstone and siltstone. Cross bedding exhibits set
heights ranging between 0.1 and 0.5 m; coset heights range up to
1.5 m. Orientations of asymmetrical ripple marks and slump
structures suggest west-southwestward transport and palaeoslope
directions (Tadesse 1999; Avigad et al. 2007).
Prospective glaciogenic deposits in Eritrea
The high-grade Arag terrane in NE Eritrea (c. 400 km NNW of
Negash) has a widespread unnamed pebbly mudstone unit, consist-
ing of silicic clasts within an argillaceous-siliceous matrix (Verri
1909; Bibolini 1920, 1921, 1922). The unit is reported as widely
exposed between 178N and 178450 N along its north-northwesterly
trend (c. 90 km) and penetrated by post-orogenic granitoids (acid
aplitic intrusions, Cecioni 1981), which have not been dated.
Cecioni (1981) described the pebbly mudstone matrix as a hard,
violet, clay-quartz cement, and the clasts as pebbles and boulders
of quartzite and silicified schist. Some of the small pebbles
(,1 cm) are pitted and knotted; others are angular. Bibolini
(1920, 1921) noted the occurrence of faceted clasts and postulated
a glaciogenic origin. To our knowledge this is the earliest suggestion
of a possible Late Proterozoic (‘algonkiano’) glaciation within the
ANS. Cecioni (1981) preferred a gravity flow origin for this unit,
but recognized a possible glaciogenic association due to the
occurrence of faceted pebbles. This unit trends more or less along
strike with Tambien Group exposures in northern Ethiopia and
prospective Bizen Domain equivalents in southern Eritrea, but has
apparently not been further studied in relation to a glacial
association.
The Neoproterozoic supracrustal succession of western Eritrea
(just west of the Shiraro area in northern Ethiopia, Fig. 21.1) con-
sists of volcano-sedimentary assemblages (Augaro Group) uncon-
formably overlain by basin-fill metasediments of the Gulgula
Group (Teklay et al. 2003; Teklay 2006). The Gulgula succession
involves greenschist-grade polymictic conglomerate, shale, phyllite
with interstratified marble lenses, in addition to carbonaceous sand-
stone, quartz arenite and allodapic carbonate, which, like the
Tambien Group, are deformed as upright, recumbent, and isoclinal
folds. Teklay (2006) describes the polymictic conglomerates as
clasts, both matrix- and clast-supported, ranging from a few centi-
metres to a half-metre of granite, slate, phyllite and epidotite. The
Gulgula Group merges southwards into the Shiraro block in north-
ern Ethiopia, suggesting a composite Gulgula-Shiraro depositional
basin and a lithostratigraphic affiliation with the Shiraro Group
268 N. R. MILLER ET AL.
(upper Tambien Group). The origin of Gulgula Group polymict
conglomerates, in addition to those described above in the
Shiraro Group (Dugub Formation) require further study to clarify
whether or not glacial processes were involved.
Boundary relations with overlying and underlying
non-glacial units
Lower prospective glaciogenic unit
Greywacke-Conglomerate Interval (GCI) of the Didikama Formation,
Negash Synclinorium. Where examined in the eastern limb of the
Negash synclinorium, the GCI occurs within the Didikama For-
mation (Fig. 21.2a, columns C and D) above thick slate (Lower
Slate Member) and below variegated (purple/green/orange)
slate that, in turn, grades upward into lowest dolomite beds of
the Didikama Formation (Slate . Dol Member). The continuity
of the transition from the Lower Slate member to the GCI is uncer-
tain due to minor folding and possible faulting associated with the
eastward structural transition to the Atsbi Horst (Fig. 21.1),
whereas the transition above the GCI appears to be continuous
and conformable.
Upper Werri Slate, Mai Kenetal Synclinorium. Miller et al. (2009)
make chemostratigraphic arguments that the equivalent of the
GCI may occur in the Mai Kenetal Synclinorium below the
Assem Limestone, either within the upper Werii Slate or
between the Werii Slate and Assem Limestone as a paraconfor-
mity. The abrupt Werii Slate-Assem Limestone contact can be
followed laterally in the field as well as traced (.13 km) in
satellite imagery.
Upper prospective glaciogenic unit
Matheos Formation Diamictite Member, Negash Synclinorium. The
Matheos Formation Diamictite Member is mapped only within
the folded core of the Negash Synclinorium. Within the Matheos
Formation, it overlies a distinctive but gradational transition
from black limestone (Black Limestone Member) to non-
calcareous slate (Transitional Member). Although a direct
contact has not been mapped, the oldest overlying sediments are
Ordovician Enticho Sandstone, the base of which marks the AAP.
Dugub Formation, Shiraro Area. The Dugub Formation in the
western Shiraro Area (Fig. 21.1) occupies a grossly similar
stratigraphic position within the Tambien Group that could be
equivalent to the Matheos Diamictite, but boundary relations for
this low-lying unit are poorly known.
Chemostratigraphy
Chemostratigraphic data for Tambien Group exposures in Ethiopia
(Alene et al. 1999, 2006; Miller et al. 2003; Sifeta et al. 2005) and
likely equivalents in Eritrea (Beyth et al. 2003) have become
available in only the last decade (Fig. 21.3). For the most part, che-
mostratigraphic surveys have been limited in terms of types of ana-
lyses performed and stratigraphic sampling frequency within a
given lithological unit, with fewer than 30 sample intervals
considered in any one study (Fig. 21.3b). Miller et al. (2009) sub-
stantially expanded the regional chemostratigraphic database,
assessing 100 carbonate and 30 slate intervals from the Shiraro,
Samre, Mai Kenetal and Negash regions. The integrated Tambien
Group record now constitutes a significant chemostratigraphic data
set for evaluating Cryogenian marine secular variations (including
71 stratigraphic intervals with 87Sr/86Sr measurements on
least-altered units), with the Mai Kenetal and Negash synclinoria
data comprising most (c. 80%) analyses (Fig. 21.3). This compi-
lation reveals significant chemostratigraphic trends, from which
regional correlations are proposed (Fig. 21.2b, discussion).
Carbonate d13C compositions for the Tambien Group range
between –8 and þ8‰ and can be characterized as fluctuating
repeatedly upsection between positive and negative polarity
(Fig. 21.3). Low-grade Bizen Domain stromatolitic carbonates
in SE Eritrea, considered to be Tambien Group equivalents
(Beyth et al. 2003), are among the most negative d13Ccarb values
reported (range, – 7.2 to –4.8‰; 3 cal, 5 dol) but sample preser-
vation was not critically assessed. Highest d13Ccarb compositions,
with maxima near þ7‰, derive from the Mai Kenetal Limestone
and Matheos Formation Black Limestone Member. The negative
d13Ccarb compositions of the Assem Limestone (c. –1 to – 4‰)
are similar to those comprising the lower negative d13Ccarb interval
in the Negash Synclinorium (c. – 1 to – 3‰) (despite their contrast-
ing mineralogies), as well as negative d13Ccarb intervals mapped as
Assem Limestone in the Chemit ( –4.4 to –3.0‰) and Tsedia
(– 4.5 to –0.7‰) synclinoria (Alene et al. 1999, 2006; Miller
et al. 2009). Bizen Domain stromatolitic dolomites with negative
d13Ccarb (Beyth et al. 2003) may correlate with the negative
d13Ccarb interval of the Didikama Formation. The Negash
carbonate sequence is so far unique in recording two negative
d13Ccarb excursions.
Carbonate d18O compositions range between 0 and –16‰, with
dolomite typically enriched (by 2– 5‰) relative to stratigraphi-
cally proximal (or prospectively equivalent) limestone. Although
substantial scatter is apparent, median compositions are mainly
between –4 and –10‰, similar to other Cryogenian datasets
(Fig. 21.3; Jacobsen & Kaufman 1999; Robb et al. 2004; Halver-
son et al. 2005, supplementary information). Values more negative
than –11‰ may be below the modal range of Cryogenian samples
(Kaufman et al. 1993), and could be particularly altered. The Mai
Kenetal Limestone and Matheos Formation Black Limestone
(units considered best preserved for d13Ccarb and 87Sr/86Sr, see
discussion) have median compositions between –7.6 and
–3.5‰, with Matheos Black Limestones underlying the transition
to diamictite deposition most enriched.
Carbon-isotopic compositions of organic matter (d13CTOC) in
the Tambien Group, principally from the Mai Kenetal and
Negash synclinoria, are mainly in the range of –20 to –30‰,
with extremely light values (e.g. , –40‰) suggesting contri-
butions from methanogenic biomass (Fig. 21.3). Both Mai
Kenetal and Negash successions show similar stratigraphic enrich-
ment trends in d13CTOC with pre-diamictite values in uppermost
Matheos Formation limestones about 3‰ heavier than uppermost
Mai Kenetal Limestone exposures.
The Tambien Group has a large compilation of 87Sr/86Sr com-
positions in studies by Miller et al. (2003, n ¼ 9; 2009, n ¼ 71)
and Alene et al. (2006, n ¼ 5). Results obtained from the same
units are complementary among the studies despite contrasts
in sample processing approaches. Least-altered 87Sr/86Sr compo-
sitions (normalized to SRM 987 ¼ 0.710240) are mainly
between 0.7055 and 0.7068 (Fig. 21.3). The Mai Kenetal succes-
sion shows a stratigraphic enrichment trend that compares with a
more extensive enrichment trend in the Negash succession.
The lower negative d13Ccarb interval at Negash has 87Sr/86Sr
compositions (avg: 0.706178 + 40, n ¼ 2) within error of those
for Assem Limestone (avg: 0.706175 + 14, n ¼ 14)
(Miller et al. 2009). Negative d13Ccarb compositions in the
Assem Limestone may initiate with 87Sr/86Sr compositions near
0.70597 (Miller et al. 2009).
Tambien Group carbonate units have Sr compositions ranging
up to about 4000 ppm, with pronounced stratigraphic enrichment
evident in both the Mai Kenetal and Negash synclinoria. Sr com-
positions less than 1000 ppm typify limestone and dolostone in
the Assem Limestone and Didikama Formation (also Didikama
Formation in Shiraro and Samre areas), whereas much higher con-
centrations averaging 2000– 3000 ppm occur in the Mai Kenetal
 (a)

THE TAMBIEN GROUP, NORTHERN ETHIOPIA

(b)

Fig. 21.3. Summary of chemostratigraphic data sets for the Tambien Group. (a) Statistical summaries of chemostratigraphic (d13Ccarb, d18Ocarb, d13CTOC, 87Sr/86Sr, Sr concentration and chemical weathering index) data
for Tambien Group lithostratigraphic units by locality and in stratigraphic order. Letters A –D mark stratigraphic units of alternating d13Ccarb polarity; corresponding circled data sets indicate possible correlations between the
Mai Kenetal and Negash synclinoria (black vertical bars) and other Tambien Group localities (see discussion); line types of circled data sets clarify vertical correlations. Black triangles indicate prospective glaciogenic intervals.

269
Data compiled from Miller et al. (2009) and Alene et al. (2006, vein calcite samples excluded) are restricted to samples considered least altered for 87Sr/86Sr. Chemical Index of Weathering (CIA or PIA) data exclude
samples containing lithic fragments or carbonate minerals. (b) Distribution of chemostratigraphic analyses for each study used to compile the upper graph statistical summaries. The histogram (top-to-bottom) bar order
corresponds to analysis type, as listed from left to right under the ‘Total Analyses/Least Altered Analyses’ compilation.
270 N. R. MILLER ET AL.

Limestone and Matheos Formation Black Limestone Member. derivations from comparable juvenile crustal sources. Associated
Intermediate Sr compositions (avg: 1198 + 91 ppm, n ¼ 2, chemical weathering indices (Fig. 21.3), which proxy the degree
Miller et al. 2009) occur in the Tsedia Slate in the Mai Kenetal Syn- of weathering from fresh rock (50) to full clay conversion (100),
clinorium. Similar transitional compositions are missing in the are generally between 70 and 90 for the Chemical Index of Altera-
Negash Synclinorium and instead there is a sudden stratigraphic tion (CIA, Nesbitt & Young 1982) and .75 for the Plagioclase
jump in Sr concentrations between the Didikama Formation dolo- Index of Alteration (PIA, Fedo et al. 1995).
mite and directly overlying Matheos Formation Black Limestone.
A range of Sr concentrations have been measured for matrix car-
bonate (980 + 712 ppm, n ¼ 11; Miller et al. 2003) and carbonate Palaeolatitude and palaeogeography
clasts in the Matheos Formation Diamictite Member (512 ppm,
oolite cobble; Miller et al. 2009). There are no reliable palaeomagnetic constraints for Tambien
The petrological and chemical transition from largely metavol- Group deposition within the southern ANS. Strike –slip displace-
canic and metasiliciclastic (slate) units in the Tsaliet Group and ment and structural shortening associated with closure of the
lower Tambien Group (Werri Slate) to predominant carbonate Mozambique Ocean and formation of the EAO constitute
deposition in the higher Tambien Group (Assem Limestone and significant challenges for early Cryogenian palaeogeographic
Didikama Formation) was investigated by Sifeta et al. (2005) in reconstructions within the ANS.
the Werii area east of the Mai Kenetal Synclinorium (Fig. 21.1). Palaeogeographic reconstructions spanning the 775–660 Ma
Metavolcanic rocks are sub-alkaline, with chemical fingerprints interval (Fig. 21.4a–c) generally place the ANS (as inferred
that are compatible with island arc (primitive or evolved) and/ oceanic arcs) within the greater Mozambique Ocean between sub-
or MORB settings. Lower Werri Slate compositions indicate sequently flanking Gondwana fragments (e.g. between India and

(a) (b) (c)

Au
Au -
Continents with In
c. 750 Ma
Pac
Ka Pa

-M
palaeomagnetic ific In Ka cif

Ma
e ic
Au data iqu
mb

a
In SC S Mo
za
M
EA Az

r
e

r
mbiqu

aste

aste
Moza
L Az

m
Ada

Ada
EAO
Mozambique Co SF RP RP

La
Ocean Co
Adola

La
Ka o SF o
n
Co RP ilian Sah ilia
SF
Am Br
az Br
az
Brasiliano
WA
Am

Ocean

Am
Postulated Schematic
WA subduction WA continental
zone collision

800 Ma 750 Ma 630 Ma

After Meert 2003; Meert & Torsvik 2003 Collins & Pisarevsky 2005
(d)
West Gondwana Tambien Group ( )
Mz-Cz Orogen
?
Pz-Mz Orogen c. 550 Ma
+
+
+

Pz Orogen Saharan
+

AAP
+

Neoproterozoic Orogen West Meta-

+
+

ANS
African craton
+
+

Cration
AAP
+

East Gondwana
+

AO
+

+
+
+

Congo
+
+
+

Indian
tic ~ E

SF
+

Craton
+

+
+
+

Mad Shield
+

Amazonian Australia
+
+

+
+

Craton
+

SL
~ Antarc
+

+ +
+
+

W.
+
+

+
+

+
+

Kalahari East
+

RP Craton N
+

Antarctic
+

S
+

Shield
+
+

+
+

Subduction Zone
Southern limit of Early Palaeozoic sandstone
marking African-Arabian peneplain (AAP)
1000 km
Modified: Meert & Lieberman 2008

Fig. 21.4. Global palaeogeographic reconstructions for (a) 800 Ma, (b) 750 Ma, (c) 630 Ma and (d) 550 Ma, showing the inferred palaeogeographic location and
structural-tectonic setting of the ANS/Tambien Group (black star) during the Cryogenian Period. Prospective glaciogenic intervals in the Tambien Group were likely
deposited during the earlier Cryogenian (.630 Ma) prior to the emergence of the East African Orogen (EAO). (a, b) Rifting and break-up of Rodinia (c. 900–750 Ma)
was associated with sea-floor spreading, arc and back-arc basin formation, and terrane accretion in the Mozambique Ocean. (c) Accommodation space in the southern
ANS basin likely inverted by 630 Ma in response to closure of the Mozambique Ocean between converging elements of West and East Gondwana and emergence of the
EAO. Occurrence of undeformed post-orogenic intrusives of this age or older that puncture the deformed Neoproterozoic supracrustal sequence in Ethiopia and Eritrea,
suggests that the Tambien Group at 630 Ma was likely deformed and uplifted within the EAO. (d) Location of the Tambien Group at c. 550 Ma within the context of
Gondwana amalgamation and emergence of the Antarctic-EAO. The modern southern limit of Early Palaeozoic sandstone (superimposed from Avigad et al. 2005)
documents the minimal extent of the Afro-Arabian Peneplain (AAP) and extent of regional uplift and Cambro-Ordovician erosion associated with the emergent EAO
(Avigad et al. 2005). Reconstructions modified from (a) Meert 2003; Meert & Torskvik 2003; (b, c) Collins & Pisarevsky 2005; (d) Meert & Lieberman 2008; Grey et al.
2008. Abbreviations for cratons and continents: Am, Amazonia; AuMa, Australia/Mawson Block; Az, Azania; B, Baltica; Co, Congo/Tanzania/Bangweulu Block; EA,
East Antarctica; Ka, Kalahari Block; La/L, Laurentia; In, India; Mad, Madigascar; RP, Rio de la Plata; Sah, Saharan Metacraton; SC, South China; SF, Sao Francisco;
S, Siberia; WA, West Africa; Adola, Adamastor, Braziliano, Mozambique, Pacific, oceanic basins; Mz, Mesozoic, Cz, Cenozoic; Pz, Palaeozoic.
 THE TAMBIEN GROUP, NORTHERN ETHIOPIA 271

500 PHZ Enticho Ss (Ord)

Ediacaran C detrital zircon
550 2 2
Post-orogenic plutons Fig. 21.5. Radiometric age constraints
age distribution 2 Mai
Mareb R. Kenetal bearing on the age of the Tambien Group
Eritrea Sibta Shire 6 7
600 1 1 6 within the magmatic evolution of the
8
9 Hawzien Negash southern ANS and EAO. Tsaliet arc
650 E. Sudan Youngest detrital zircons in upper magmatic products and ‘Mareb’
Age (Ma)

NEOPROTEROZOIC

prospective glacial intervals Matheos Tambien post-orogenic pluton ages arranged

700
Cryogenian

Dugub Fm Diamictite Group according to their relative west-to-east

(Shiraro) Zircons below 1st dolomite beds (Negash) Carbonate positions in Tigre. Magmatic equivalents
750 1 2 2 1 1 Deposition for localities in Eritrea and Sudan (open
1
800 Azeho Deset 2 Madahne BSS squares) are shown to the left. The Enticho
5 4 Eritrea
2
Hawzien Alem Sandstone detrital zircon age spectrum
9 3 Chila (Negash)
850 10 Rama (Avigad et al. 2007) shown at the left of the
Ton.

E. Sudan
900
STUDY
1. Avigad et al. (2007)
2. Tadesse et al. (2000)
3. Tadesse et al. (1999)
4. Teklay (1997)
5. Teklay et al. (2002)
Tsaliet Group syn-tectonic granites and metavolcanics figure derives from the Enticho area as
shown in Figure 21.1. The shaded
Relative W-to-E position (Tigre) horizontal bar shows the age range (827+6
to 770 + 7 Ma) for the Bitter Springs stage
DATE TYPE/METHOD
(BSS) negative d13Ccarb excursion of
Australia (Halverson et al. 2005, 2007; and
6. Miller et al. (2003) Pb-Pb zircon U-Pb zircon
7. Asrat et al. (2004)
references therein). Note that this range
Rb-Sr Zircon evaporation
8. Teklay et al. (2001) overlaps with the main phase of Tsaliet
Sm-Nd Youngest detrital zircon
9. Kröner et al. (1991) syn-tectonic arc magmatism; see text for
CHIME zircon (SHRIMP U-Pb)
10. Teklay et al. (2003) discussion.

Congo cratons). For example, Collins & Pisarevsky (2005) place Geochronological constraints
ophiolite-bearing strata of the Adola Belt (southern Ethiopia) out-
board of the Congo Craton (Fig. 21.4b). As the Adola Belt is gener- The age of Tambien Group deposition is bracketed by magmatism
ally considered to mark the southern extent of the EAO, it is reasonable associated with formation of the underlying Nakfa basement
to infer a similar outboard location for the Tambien Group. complex (Tsaliet Group) and orogenic thickening and differen-
The earliest reliable regional palaeomagnetic data derive from tiation related to maturation of the EAO (‘Mareb’ granitoids).
the time of Gondwana amalgamation (Fig. 21.4d); late Cryogenian Affiliated magmatism is widely dated throughout the ANS (e.g.
(593 + 15 Ma) Dokhan volcanics of Egypt (Davies et al. 1980; Meert 2003; Johnson & Kattan 2007, and references therein).
Wilde & Youssef 2000), interpreted to have a subtropical palaeo- Our geochronological review is restricted to southern ANS
latitude (P-lat: 20.6 + 5.08, A95: 10.08, Trindade & Macouin studies in eastern Sudan (Kröner et al. 1991), Eritrea (Teklay
2007; but see Nairn et al. 1987). The Dokhan palaeopole has 1997; Teklay et al. 2001, 2003; Andersson et al. 2006), northern
been widely used in subsequent palaeogeographic reconstructions Ethiopia (Tadesse et al. 1997, 2000; Miller et al. 2003; Asrat
(e.g. Meert 2003; Meert & Torsvik 2003; Macouin et al. 2004; et al. 2004; Avigad et al. 2007) and western Ethiopia (Ayalew
Trindade & Macouin 2007). et al. 1990) closest to Tambien Group exposures and its prospec-
Low subtropical palaeolatitudes (9 –138) were also reported for tive equivalents (Figs 21.1, 21.5, Supplementary Table 21.1).
Huqf Supergroup units in Oman thought to represent deposition Detrital zircon age spectra from mature Ordovician sand (Enticho
during and after an upper Cryogenian (Fiq Formation) glaciation Sandstone) capping the AAP in Tigre are interpreted by Avigad et al.
before 544 Ma (Kempf et al. 2000; Kilner et al. 2005), when (2007) to reflect the magmatic history of the regional Neoprotero-
Oman was likely amalgamating with the ANS (detrital zircon zoic–Ordovician basement complex, as it was uplifted and eroded
data in Rieu et al. 2007). Subsequent work (Rieu et al. 2006; within the EAO. The Enticho Sandstone zircon age distribution
Allen 2007, and references therein) demonstrates that Mirbat (Fig. 21.5) reinforces that Neoproterozoic magmatism occurred in
Group localities in both palaeomagnetic studies correspond to an two main episodes: (i) arc-related calc-alkaline magmatism
older Cryogenian (Ghubrah-Ayn Formations) glaciation, con- c. 850–740 Ma (c. 780 Ma peak), largely related to petrogenesis
strained radiometrically to be ,722 Ma and ongoing at of the Tsaliet Group, and (ii) post-orogenic magmatism c. 660–
711.8 + 1.6 Ma. The fact that palaeolatitudes for Oman units 580 Ma (c. 630 Ma peak) after Tambien Group deposition.
associated with both glacial intervals (between c. 722 and Additional geochronological constraints elaborated below and
544 Ma) are similarly low, and also close to published c. 550 Ma shown in Figure 21.5 further suggest that the lower and upper
palaeopoles from Gondwana, has yet to be fully reconciled with prospective glacial intervals within the Tambien Group were depos-
the available geologic data (Allen 2007). However, even if low ited during the lull between these two magmatic episodes, with 775
palaeolatitudes are confirmed, Oman had yet to accrete with the to .660 Ma as a plausible depositional window (see Discussion).
ANS at the time of the older Cryogenian (Ghubrah-Ayn For-
mations) glaciation, and therefore these constraints could not be
precisely extended to the Tambien Group. Pre-Tambien Group magmatism
To the extent that the ANS occupied a gross latitudinal range
similar to the Congo-Sào Francisco and/or East-Sahara cratons Deformed (syn-tectonic) plutons and metavolcanic rocks of the
during the early Cryogenian (c. 750 Ma), as implied by various Tsaliet Group (lower calc-alkaline magmatic components of the
palaeogeographic reconstructions (e.g. Fig. 21.4a,b; Trindade & Nakfa terrane) have been dated by a number of techniques (e.g.
Macouin 2007, fig. 3a), the Tambien Group may have occupied Pb/Pb zircon, Rb –Sr, Sm –Nd, CHIME zircon, U –Pb zircon,
low to intermediate latitudes (,458). Occurrence of evaporite zircon evaporation) with varying associated uncertainties (Sup-
pseudomorphs in lower Didikama Formation dolomite (interpreted plementary Table 21.1 – Geochronological Data). These ages
as ,774.7 + 4.8 Ma, Miller et al. 2009) may support a subtropical provide an indirect lower age limit for Tambien Group carbonate
palaeolatitude, as palaeomagnetically constrained evaporite basins deposition. The oldest rocks in this region include an
as old as 2.25 Ga have statistically significant volume-weighted 862 + 6 Ma granite clast within the Gulgula Group (Fig. 21.1,
concentrations in the palaeo-subtropics (Evans 2006). Teklay et al. 2003), 854 + 3 Ma deformed volcanic rocks and
272 N. R. MILLER ET AL.
811 + 11 Ma granite from Eritrea (Teklay 1997). In the Axum
area (western Tigre), syn-tectonic pluton ages range between
806 + 21 Ma and 756 + 33 Ma (Tadesse et al. 2000). A
784 + 14 Ma syn-tectonic granitoid occurs SSW of Hauzien
(Fig. 21.1; Avigad et al. 2007). A conformable felsic interval
shortly (c. 17 m) below initial dolomite beds of the Didikama For-
mation, in the western limb of the Negash Synclinorium (Madahne
Alem), produced a zircon U – Pb date of 774.7 + 4.8 Ma (Avigad
et al. 2007; Figs 21.4 and 21.5).
Post-Tambien Group magmatism
Post-orogenic ‘Mareb’ granitoid plutons (Fig. 21.1) penetrate the
older deformed Neoproterozoic complex (including the Tambien
Group) throughout northern Ethiopia and Eritrea. In Tigre, ages
for these commonly rounded and undeformed intrusives range
from 613.4 + 0.9 Ma (Mai Kenetal Granite, Avigad et al.
2007) to 545 + 24 Ma (Mareb Granite, Tadesse 1997). The
Negash Pluton, just west of the Negash syncline, is c. 606 Ma
(606.0 + 0.9 Ma, Miller et al. 2003; 607 + 7 Ma, Asrat et al.
2004). A small granitoid body puncturing the western Negash
limb (lower Didikama Formation) is interpreted to have a compar-
able age, but did not render suitable zircons for geochronology
(Miller et al. 2009). Ages of post-orogenic intrusives are somewhat
older in Eritrea (628 + 4 Ma, 622 + 1 Ma; Teklay et al. 2001) and
Sudan (SE of Tokar: 652 + 14 Ma, Kröner et al. 1991). The good
agreement between individually dated syn-tectonic and post-
orogenic magmatic products with corresponding modal peaks in
the Enticho Sandstone zircon age distribution (Fig. 21.5), and
the lack of detrital zircons younger than c. 739.2 + 6.3 Ma in
either of the prospective glaciogenic units capping the Tambien
Group (Dugub Fm, Matheos Diamictite) are strong evidence that
the Tambien Group is older than EAO post-tectonic granitic
rocks (Avigad et al. 2007). In the vicinity of Mai Kenetal and
Negash, this constraint is c. 610 Ma, but its upper age is likely
much older considering that regional deformation and crustal
thickening must have preceded the ‘Mareb’ intrusives. The
oldest dated post-tectonic pluton in the region (652 + 14 Ma;
Kröner et al. 1991) and Enticho sandstone detrital zircon record
(initiation of the second magmatic phase at c. 660 Ma; Avigad
et al. 2007) suggest that the Tambien Group may have been depos-
ited prior to c. 660 Ma (Fig. 21.5).
Discussion
Regional chemostratigraphic context of the prospective
glacial intervals
The observation that many Tambien Group exposures in Tigre
exhibit a similar succession of lithofacies that have comparable che-
mostratigraphic characteristics suggests that the Tambien Group pre-
serves a regionally coherent chemostratigraphic framework (Beyth
1972; Alene et al. 2006; Miller et al. 2009). Tambien Group deposi-
tional history, including the depositional context of the prospective
glaciogenic intervals, can be effectively considered relative to
stratigraphic changes in the polarity of associated d13Ccarb. These
relationships are best demonstrated in the Mai Kenetal and Negash
synclinoria (Figs 21.2b, 21.3) but are consistent with the stratigraphic
succession in other areas (e.g. Shiraro and Samre).
The carbonate successions in the Mai Kenetal and Negash syn-
clinoria both initiate above thick marine slate sequences; however,
the nature of this transition to carbonate deposition differs in each
locality. In Mai Kenetal the transition from Werii Slate to the
Assem Limestone is a sharp conformable or paraconformable
contact, the Assem Limestone beginning with and maintaining
negative d13Ccarb compositions. In Negash, faulting in the lower
portion of each synclinorial limb interrupts the stratigraphic
continuity of the slate-to-carbonate transition, leading to negative
d13Ccarb compositions above the lower prospective glacial interval
(GCI) in the eastern limb and positive d13Ccarb compositions
above variegated slate in the western limb (see Miller et al. 2009
for details). The available regional data suggest that Tambien
Group carbonate deposition began with positive d13Ccarb compo-
sitions (interval A). For Negash this interval (c. 250 m thick) in
the western limb may crop out within an east-verging thrust
block, whereas the interval in the eastern limb may be faulted
out or yet unrecognized below the GCI. The nature and continuity
of the transition from Tsaliet Group agglomerates to lower Didi-
kama Formation slate and dolomite remains poorly understood
and a systematic stratigraphic transition from lower positive (inter-
val A) to higher negative (interval B) d13Ccarb compositions has yet
to be documented in a conformable sequence. Based on similar
lithostratigraphic position and character (slate with negligible car-
bonate content and advanced chemical weathering indices;
Fig. 21.3), the informal lower slate member of the Didikama
Formation could be equivalent to the Werii Slate and these units
are tentatively included in polarity interval A.
The lower prospective glacial interval in both sequences
underlies a lower carbonate interval with negative d13Ccarb (interval
B) that is in turn overlain by a distinctive black limestone succession
with the most enriched d13Ccarb compositions of the entire Tambien
Group (interval C; Fig. 21.2b). Polarity interval B and C units in
both localities have highly complementary d13CTOC, 87Sr/86Sr and
Sr compositions that support their regional correlation (Fig. 21.3).
The lithological and chemostratigraphic transition from interval B
to C appears to be continuous in Mai Kenetal, whereas this transition
in Negash is abrupt and likely associated with an unconformity
(Miller et al. 2009). Above the black limestone interval in the
Negash Synclinorium is the second ‘upper’ negative d13Ccarb excur-
sion (interval D) associated with a conformable transition to the
upper prospective glacial unit (Matheos Diamictite).
Composite chemostratigraphic reference section
Figure 21.6 presents a composite chemostratigraphic reference
section for the Tambien Group based on the correlations between
the Mai Kenetal and Negash synclinoria shown in Figure 21.2b
and the reasoning above. The Mai Kenetal section above the Werii
Slate (Assem Limestone, Tsedia Slate and Mai Kenetal Limestone)
essentially replaces the Negash sequence above the basal positive
d13Ccarb interval of the lower Didikama Formation and below
(or within) the lower portion of the Matheos Formation Black
Limestone Member. This composite effectively replaces dolomite
lithologies below a probable unconformity (Negash) with a continu-
ous and conformable succession of limestone lithologies; these seem
to show gradational chemostratigraphic trends into highly enriched
values that correlate well with the lower Matheos Formation Black
Limestone Member (Miller et al. 2009). The abrupt transition
from positive (polarity interval A) to negative (polarity interval
B) d13Ccarb compositions may be a structural contact, but we
suggest that the relative stratigraphic order of d13Ccarb polarity is
maintained in the composite. The 87Sr/86Sr evolution of Cryogen-
ian seawater (albeit poorly delineated) involves a general trend of
increasing 87Sr/86Sr (Halverson et al. 2007). That least altered
samples from polarity interval A dolomite beds have lowest
87
Sr/86Sr compositions is consistent with early Cryogenian depo-
sition within the Tambien Group carbonate platform. The evolution
of Tambien Group depositional environments is now considered
based on the composite reference section (Fig. 21.6).
Palaeoenvironmental changes
The Tambien Group involves mainly a transition from weathered
Tsaliet arc detritus (lower slate intervals; Sifeta et al. 2005) to
 THE TAMBIEN GROUP, NORTHERN ETHIOPIA 273

d13Ccarb Sr (ppm) TOC

Enticho
Sandstone -8 -4 0 4 8 0 2000 4000 0.0 0.1 1.0
2000

Tr Diam.

matrix
oolite
(d)
Interval devoid of bedded carbonate
Matheos Fm
cobble
1800
Black LS

1600
3
MK
LS

1400
(c)
Stratigraphic Height (m)

Bilato
Tambien Group

1200
Tsedia Slate
Logmiti

1000

800
Assem

(b)
LS

MTS
600
CC ?
2 ?
Slate
>Dol

400
(a)

774.7±4.7 Ma
Lower

200
Slate

Extreme weathering indices (PIA: 92-99)

0
1
Tsaliet Gp 0.705 0.706 0.707 -15 -10 -5 0 -45 -35 -25
(Meta-
volcanics) 87Sr/86Sr d Ocarb
18
d CTOC
13

MTS CC
Sheet crack cements Evaporite pseudomorphs Stromatolites Internal slump Molar tooth structures Probable dropstones Cap carbonate-like textures

Fig. 21.6. Composite chemostratigraphic reference sequence for the Tambien Group based on the Mai Kenetal and Negash synclinoria. Segments 1 –3 denote the
basis for constructing this composite sequence (cf. Fig. 21.2b) and the black triangles denote the prospective glacial intervals. The composite sequence is differentiable by
four changes (a –d, separated by horizontal dashed lines) in the polarity of associated d13Ccarb and consists of limestone units except for segment 1 (dolomite comprising
the lower portion of the Didikama Formation in the western limb of Negash). The lower horizontal shaded interval indicates the stratigraphic range of slates having
extreme chemical weathering indices (PIA, plagioclase index of alteration of Fedo et al. 1995). The upper horizontal shaded interval corresponds to the upper portion of
the Matheos Formation lacking bedded carbonate; the darker grey d13Ccarb field in the Matheos Diamictite shows the range for diamictite matrix carbonate (Miller et al.
2003), whereas the uppermost data point corresponds to an oolite cobble (Miller et al. 2009). Abbreviations: Ls, Limestone; Dol, dolomite; Tr, transitional,
Diam-Matheos diamictite; MK, Mai Kenetal.

predominant carbonate deposition in a marine arc-accretion plat- associated with polarity interval A are interpreted to have been
form setting. The lack of obvious ash beds, volcaniclastic intervals deposited in high alkalinity tidal flat and intertidal settings. Mod-
and syn-tectonic intrusive rocks in the main carbonate succession erate energy shallow intertidal or subtidal settings are interpreted
(polarity interval B and higher) suggests deposition during a mag- for the Assem Limestone and upper Didikama Formation
matically quiet interval (Avigad et al. 2007), and the upward (polarity interval B) on the basis of prominent domal and interdi-
waning of slate deposition in favour of carbonate may represent gitate stromatolites, coarse rip-ups, and cross-bedded grainstones
a phasing out of arc magmatic activity in the region. (Miller et al. 2009, figs S2, S3). The base of the Assem Lime-
Chemical weathering indices of slates in polarity interval A stone occurs abruptly above the Werii Slate as a metre-thick
suggest an overall stratigraphic trend of increasing chemical weath- dolomite bed, with sedimentary features similar to those
ering of source areas to extreme levels, which lessened somewhat described for well-studied transgressive cap-carbonate sequences.
prior to deposition of polarity interval B (Fig. 21.6). The association Lower carbonate units of the Tambien Group (polarity intervals
of extreme weathering indices with increasing 87Sr/86Sr is consist- A –B) have low TOC contents with highly variable d13CTOC,
ent with an interval of rapid chemical weathering of arc terranes in including light compositions, which together with the sedimen-
the southern ANS portion of the Mozambique Ocean. The fine tary characteristics are consistent with nearshore environments
grained nature of slates and absence of shallow water indicators (Miller et al. 2009, figs 7B,C and 9).
suggests a moderately deep-water depositional setting. Deeper subtidal environments with lower (but still variable)
Above the lower slate intervals, the Tambien Group carbonate energy levels and ubiquitous micrite are interpreted for upper
pile is broadly differentiable between lower carbonate units carbonate units of the Tambien Group (polarity interval C).
(polarity intervals A –B) suggestive of shallow marine deposi- The transition appears to begin at the base of the Tsedia Slate in
tional settings and upper carbonate units (polarity interval C) sug- Mai Kenetal (equivalent to the proposed unconformity between
gestive of deeper marine environments. Based on the occurrence the Didikama and Matheos Formations in Negash), and this bound-
of evaporite pseudomorphs, sheetcrack cements, microbialami- ary could mark the base of a transgressive system following
nates and stromatolites (Miller et al. 2009, fig. S5), carbonates regression. Both the Mai Kenetal Limestone and the Matheos
274 N. R. MILLER ET AL.
Formation Black Limestone are distinctive dark grey to black lime-
stone units characterized by fine horizontal layering and high lateral
continuity (Miller et al. 2009, figs S3 and S6). The lack of stroma-
tolites and high-energy sedimentary structures (coarsely graded
beds, cross-bedding) but occurrence of dark intraclastic intervals
and intraformational slumping may indicate slope deposition
below storm wave base. These upper carbonate units are relatively
enriched in TOC that has less variable d13CTOC, consistent with
more open marine environments (Miller et al. 2009, fig. 7B,C).
Upper carbonate units are also most enriched in d13Ccarb and
d13CTOC, particularly in the upper part of the Matheos Formation
Black Limestone Member. This mutual stratigraphic enrichment
pattern, together with distinct stratigraphic increases in 87Sr/86Sr
and Sr concentrations, is consistent with high rates of organic
matter burial and related fractionation of the upper (photic)
marine d13CDIC pool in the lead up to Matheos Diamictite depo-
sition. Although d18Ocarb is the marine proxy most likely to
undergo post-depositional alteration, it is notable that the intervals
with most enriched compositions (consistent with cryogenic
sequestering of d16O) occur in association with the two prospective
glacial intervals (Fig. 21.6). The possibility that d18O records fluc-
tuations in ice volume is more likely for the upper prospective
glacial interval because the lower prospective glacial interval is
underlain by dolomite, which is typically enriched by 2 –4‰
over coeval limestone (Jaffrés et al. 2007, and references therein).
Timing of the prospective glacial intervals
Integration of available geochronological information suggest c.
775–660 Ma as a plausible window for Tambien Group deposition
bracketing the two prospective glacial intervals. This depositional
window is not definitive, but corresponds to the lull between
Tsaliet arc magmatism and later magmatism associated with matu-
ration of the EAO, on the basis of regional stratigraphy and cross-
cutting relationships. Direct dates on magmatic products from each
magmatic episode vary locally within Tigre but concur with the
detrital zircon age distribution from the immediately overlying
Ordovician Enticho Sandstone (Fig. 21.5). Somewhat older arc
magmatism is indicated in Eritrea and eastern Sudan.
The 774.7 + 4.8 Ma date obtained from zircons recovered from
variegated (tuffaceous?) slate, c. 17 m below the first positive
d13Ccarb (polarity interval A) dolomite beds of the Didikama For-
mation, is a possible maximum age constraint for significant
(bedded) carbonate deposition in the Tambien Group (Fig. 21.5).
The interval was originally interpreted as either a volcaniclastic
interval or sill injected into a volcaniclastic interval. Subsequent
petrographic analysis of this interval revealed probable sedimen-
tary textures, including rounded quartz grains and a large (3 cm)
rounded quartzose clast. If this date holds, the lower prospective
glacial interval occurring at least c. 250 m higher is likely to be
appreciably younger. The upper prospective glacial interval is
older than deformation and later magmatism associated with the
EAO. The oldest indication of EAO magmatic activity in the
southern ANS is from a 652 + 14 Ma post-orogenic pluton in
eastern Sudan (southern Red Sea Hills, c. 500 km NNW of
Negash; Kröner et al. 1991), whereas locally dated ‘Mareb’ gran-
itoids in Tigre are c. 610 Ma (avg: 609.8 + 3.8 Ma, Mai Kenetal,
Negash, Hauzien plutons).
The dated Red Sea Hills locality in eastern Sudan occurs
c. 50 km NW of the area of prospective glacial deposits (pebbly
mudstones with flattened clasts) in northern Eritrea (Fig. 21.1)
(Cecioni 1981). If the pebbly mudstones are glaciogenic and
correlative with the upper prospective glacial interval of the
Tambien Group, the Red Sea Hills age offers an important
minimum age constraint for the Tambien Group. The most
precise age estimates on older (pre-pebbly mudstone) associated
arc metavolcanic and plutonic rocks in the Red Sea Hills area,
obtained by single zircon 207Pb/206Pb methods (c. 870–
827 Ma), are substantially older than ages determined by Rb –Sr
whole rock methods (c. 770– 670 Ma). The younger ages have
been interpreted to reflect isotopic disturbance due to regional
deformation of the arc complex, possibly as it accreted with the
African continent (Kröner et al. 1991; Saharan Metacraton?).
Structural deformation of the orogen may thus have begun
before 670 Ma. As this deformation was post-depositional and pre-
ceded the first post-orogenic intrusives (652 + 14 Ma) by tens of
millions of years, the Tambien Group minimum age, and the age
of the upper prospective glacial interval, could be appreciably
older than 652 + 14 Ma.
Evidence for glacial influence on sedimentation
The lower prospective glacial interval of the Tambien Group,
as the GCI in Negash and a possibly equivalent interval or paracon-
formity at the top of the Werri Slate is most controversial.
Direct sedimentologic evidence of syn-glacial deposition (i.e.
definitive dropstones, striated clasts) is so far lacking, with the
possible exception of matrix supported horizons in the GCI
(Beyth 1972). On the other hand, both of these intervals are over-
lain by negative d13Ccarb carbonates with similar chemostrati-
graphic characteristics. Where the base of this carbonate interval
is well exposed (Assem Limestone, Mai Kenetal) it exhibits fea-
tures consistent with well-documented Cryogenian post-glacial
cap carbonates.
Alene et al. (2006) discounted a glacial association for the
Assem Limestone, but did not document or sample the base of
this unit above its abrupt contact with Werri Slate. They suggested
the Assem Limestone may correspond to a non-glacial negative
d13Ccarb interval possibly equivalent to the Bitter Springs stage
(BSS) of Australia (as suggested by Halverson et al. (2005,
2007) for similar negative d13Ccarb intervals within early Cryogen-
ian sections from NW Canada (Little Dal Group) and Svalbard
(Akademikerbreen Group)). The age for the BSS in Australia
requires interbasinal correlations that are difficult to confirm, and
the age range has a corresponding large degree of uncertainty
between 827 + 6 Ma (Gairdner Dyke Swarm) and 777 + 7 Ma
(Boucat Volcanics) among various studies (Halverson et al.
2005, 2007). Although a correlation with the BSS is possible, the
c. 827 –770 Ma age range would correspond to a time of active
arc magmatism in the southern ANS (Fig. 21.5). The Assem Lime-
stone and higher Tambien Group succession lacks evidence of
active magmatism (ash beds/slate intervals/syn-tectonic intru-
sives) suggesting it was deposited after the c. 780 Ma acme of
arc magmatism in the region. The lower Negash negative
d13Ccarb interval occurs several hundred metres above the varie-
gated slate interval dated at 774.7 + 4.8 Ma (zircon). This
suggests that the lower negative d13Ccarb interval could be appreci-
ably younger than c. 775 Ma. As there are definitively younger
post-glacial cap carbonates (e.g. above the c. 755 Ma Kaigas For-
mation, southern Namibia, Gariep Belt, Hoffmann et al. 2006)
we suggest that a lower prospective glacial association within
the Tambien Group remains a viable interpretation. It is conceiva-
ble that the glaciogenic association could be indirect. For example,
the cap-carbonate features and negative d13Ccarb excursion could
be products of enhanced upwelling associated with global
cooling and/or vigorous circulation associated with post-glacial
climate change in a setting that did not earlier accumulate glacially
transported clastic sediments. Both of these intervals in the Mai
Kenetal and Negash synclinoria warrant additional study to
evaluate glacial associations.
Of the proposed upper glaciogenic intervals for the Tambien
Group, the Matheos Formation Diamictite Member best exhibits
sedimentological characteristics consistent with glacially influ-
enced deposition. In addition to matrix supported polymict
clasts, clasts that appear to preferentially deform underlying
bedding are interpreted as ice-rafted debris. Some bullet-nosed
 THE TAMBIEN GROUP, NORTHERN ETHIOPIA
and possibly striated clasts are consistent with mechanical abrasion
during glacial entrainment (Kuhn et al. 1993). This interval
is further significant because of its gradational contact with the
underlying Matheos Black Limestone Member, which records a
decline in d13Ccarb and possibly also 87Sr/86Sr, prior to diamictite
deposition. If the latter is not the product of diagenetic alteration,
the declining Sr-isotope compositions could signal a strong
decrease in continental weathering input relative to oceanic
hydrothermal input, as consistent with increasing ice cover.
The possibly equivalent Dugub Formation in the Shiraro area
occupies a gross stratigraphic position similar to the Matheos
Diamictite as well as similar detrital zircon age distributions
(Avigad et al. 2007), but the underlying stratigraphy is poorly
known. Palaeocurrent proxies suggest a north-northeastern
source area for clasts, which considering the prospective glacial
intervals in northern and eastern Eritrea could be consistent with
a regional phase of glaciation.
Tectonic and palaeogeographic setting
The intra-oceanic setting of the Tambien Group contrasts distinctly
from pericratonic rift margin settings that are often associated with
Cryogenian glaciogenic units. Later Cryogenian intervals are unli-
kely to be directly recorded throughout much of the ANS because
the emergence of the EAO (by 660 Ma) would likely have either
destroyed accommodation space or uplifted related strata to
levels planed by subsequent erosion. The EAO could have hosted
alpine/continental glaciation, however, perhaps even instigating
initial downcutting of the vast AAP (Fig. 21.4d; Stern et al. 2006).
The US– Israel Binational Science Foundation (grant no. 2002337) supported this
study. We thank K. Mehari, D. Küster and T. Tadesse for field expertise and
assistance, A. Abraham (Chief Geologist, Geological Survey of Ethiopia) and
Mekele University for logistical support during field seasons, and M. Abdel-
Salem for helpful feedback on geological aspects of the ANS. B. Schilman and
A. Ayalon of the Geological Survey of Israel provided stable isotope expertise.
The manuscript was improved by thoughtful reviews from P. Johnson, M. Pope
and E. Arnaud. This represents a contribution of the IUGS- and UNESCO-
funded IGCP (International Geoscience Programme) Project #512.
References
Alene, M., Conti, A., Sacchi, R. & Zuppi, G. 1999. Stable isotope com-
position (13C and 18O) of Neoproterozoic limestones and dolomites
from Tigre, North Ethiopia. Bollettino della Società Geologica
Italiana, 118, 611– 615.
Alene, M., Jenkin, G. R. T., Leng, M. J. & Darbyshire, D. P. 2006. The
Tambien Group, Ethiopa: an early Cryogenian (ca. 800– 735 Ma)
Neoproterozoic sequence in the Arabian– Nubian Shield. Precam-
brian Research, 149, 79 –89.
Allen, P. A. 2007. The Huqf Supergroup of Oman: basin development
and context for Neoproterozoic glaciation. Earth Science Reviews,
84, 139– 185.
Andersson, U. B., Ghebreab, W. & Teklay, W. 2006. Crustal evolution
and metamorphism in east-central Eritrea, south-east Arabian-Nubian
Shield. Journal of African Earth Sciences, 44, 45– 65.
Arkin, Y., Beyth, M., Dow, D. B., Levitte, D., Haile, T. & Hailu, T.
1971. Geological map of Mekele Sheet area ND 37-11 Tigre Pro-
vince. Imperial Ethiopian Government, Ministry of Mines, Geologi-
cal Survey of Ethiopia, scale 1:250,000.
Asrat, A., Barbey, P., Ludden, J. N., Reisberg, L., Gleizes, G. &
Ayalew, D. 2004. Petrology and isotope geochemistry of the
Pan-African Negash Pluton, northern Ethiopia: mafic-felsic magma
interactions during the construction of shallow-level calc-alkaline
plutons. Journal of Petrology, 45, 1147–1179
Avigad, D., Sandler, A., Kolodner, K., Stern, R. J., McWilliams,
M. O., Miller, N. & Beyth, M. 2005. Mass-production of Cambrian
quartz-rich sandstone as a consequence of chemical weathering of
275
Pan-African orogens: implications for global environment. Earth
and Planetary Science Letters, 240, 818–826.
Avigad, D., Stern, R. J., Beyth, M., Miller, N. & McWilliams, M.
2007. Detrital zircon U– Pb geochronology of Cryogenian diamictites
and Lower Palaeozoic sandstone in Ethiopia (Tigrai): age constraints
on Neoproterozoic glaciation and crustal evolution of the southern
Arabian– Nubian Shield. Precambrian Research, 154, 88– 106.
Ayalew, T., Bell, K., Moore, J. M. & Parrish, R. R. 1990. U-Pb and
Rb-Sr geochronology of the Western Ethiopian Shields. Geological
Society of America Bulletin, 102, 1309– 1316.
Beyth, M. 1972. The geology of central western Tigre, Ethiopia. PhD
thesis, Bonn, University of Bonn.
Beyth, M., Avigad, D., Wetzel, H. U., Matthews, A. & Berhe, S. M.
2003. Crustal exhumation and indications for snowball Earth in the
East African Orogen: North Ethiopia and East Eritrea. Precambrian
Research, 123, 187–201.
Bibolini, A. 1920. Risultali preliminary delle osservazioni faite nel
Nord-est della Colona Eritrea. Asmara.
Bibolini, A. 1921. Sui conglomerati di Rore Babla e dei Monti Haggar
in Colonia Eritrea. Bollettino della Società Geologica Italiana, 40,
169– 176.
Bibolini, A. 1922. Contributions a l’étude de la géologi de l’Afrique
orientale Italienne. 13th International Geological Congress (1922,
Brussels, Belgium), Title Comptes rendus de la XIIIe session, en
Belgique, parts 1 –3, 797– 814.
Burke, K. & Sengör, C. 1986. Tectonic escape in the evolution of
continental crust. In: Reflection seismology – The continental crust
American Geophysical Union, Geodynamic series, 14, 41 –53.
Cecioni, G. 1981. Precambrian pebbly mudstones in Eritrea, northeastern
Ethiopia. In: Hambrey, M. J. & Harland, W. B. (eds) Earth’s Pre-
Pleistocene Glacial Record. Cambridge University Press, A24, 150.
Collins, A. S. & Pisarevsky, S. A. 2005. Amalgamating eastern Gond-
wana: the evolution of the Circum-Indian Orogens. Earth-Science
Reviews, 71, 229–270.
Corsetti, F. A., Stewart, J. H. & Hagadorn, J. W. 2007. Neoprotero-
zoic diamictite-cap carbonate succession and d13C chemostratigraphy
from eastern Sonora, Mexico. Chemical Geology, 237, 129–142.
Davies, J., Nairn, A. E. M. & Ressetar, R. 1980. The palaeomagnetism
of certain late Precambrian and early Palaeozoic rocks from the Red
Sea Hills, eastern desert, Egypt. Journal of Geophysical Research,
85, 3699–3710.
De Souza Filho, C. R. & Drury, S. A. 1998. A Neoproterozoic supra-
subduction terrane in northern Eritrea, NE Africa. Journal of the
Geological Society, London, 155, 551– 566.
Evans, D. 2006. Proterozoic low orbital obliquity and axial-dipolar geo-
magnetic field from evaporite palaeolatitudes. Nature, 444, 51 – 55.
Fedo, C. M., Nesbitt, H. W. & Young, G. M. 1995. Unraveling the
effects of potassium metasomatism in sedimentary rocks and palaeo-
sols, with implication for palaeoweathering conditions and prove-
nance. Geology, 23, 921– 924.
Garland, C. R. 1980. Geology of the Adigrat area. Ministry of Mines
Memoir No. 1, 51, Addis Ababa, Ethiopia. 1:250,000 map.
Gray, D. R., Foster, D. A., Meert, J. G., Goscombe, B. D., Armstrong,
R., Truow, R. A. J. & Passchier, C. W. 2008. A Damaran Perspec-
tive on the Assembly of Southwestern Gondwana. Geological Society,
London, Special Publications, 294, 257–278.
Greenwood, W. R., Hadley, D. G., Anderson, R. E., Fleck, R. J. &
Schmidt, D. L. 1976. Late Proterozoic cratonization in southwestern
Saudi Arabia. Philosophical Transactions of the Royal Society of
London, 280, 517– 527.
Halverson, G. P., Hoffman, P. F., Schrag, D. P., Maloof, A. C. &
Rice, A. H. N. 2005. Toward a Neoproterozoic composite carbon-
isotope record. Geological Society of America Bulletin, 117,
1181–1207.
Halverson, G. P., Dudás, F. Ö., Maloof, A. C. & Bowring, S. A. 2007.
Evolution of the 87Sr/86Sr composition of Neoproterozoic seawater.
Palaeogeography, Palaeoclimatology, Palaeoecology, 256,
103– 129.
Hoffman, P. F., Halverson, G. P., Domack, E. W., Husson, J. M.,
Higgins, J. A. & Schrag, D. P. 2007. Are basal Ediacaran
(635 Ma) post-glacial ‘cap dolostones’ diachronous? Earth and
Planetary Science Letters, 258, 114– 131.
276 N. R. MILLER ET AL.
Hoffmann, K.-H., Condon, D. J., Bowring, S. A., Prave, A. R. &
Fallick, A. 2006. Lithostratigraphic, carbon (d13C) isotope and
U –Pb zircon age constraints on early Neoproterozoic (ca. 755 Ma)
glaciation in the Gariep Belt, southern Namibia, Snowball Earth
Conference, 16 – 21 July 2006, Ascona, Switzerland [abstract], 51.
Jacobsen, S. B. & Kaufman, A. J. 1999. The Sr, C and O isotopic evol-
ution of Neoproterozoic seawater. Chemical Geology, 161, 37 –57.
Jaffrés, J. B. D., Shields, G. A. & Wallmann, K. 2007. The oxygen
isotope evolution of seawater: a critical review of a long-standing
controversy and an improved geological water cycle model for the
past 3.4 billion years. Earth-Science Reviews, 83, 83– 122.
Johnson, P. R. & Kattan, F. H. 2007. Geochronologic dataset for
Precambrian rocks in the Arabian peninsula. Saudi Geological
Survey Open-File Report SGS-OF-2007-3, 21, tables.
Kaufman, A. J., Jacobsen, S. B. & Knoll, A. H. 1993. The Vendian
record of C- and Sr-isotopic variations: implications for tectonics
and paleoclimate. Earth and Planetary Science Letters, 120,
409– 430.
Kaufman, A. J., Knoll, A. H. & Narbonne, G. M. 1997. Isotopes,
ice ages, and terminal Proterozoic earth history. Proceedings of the
National Academy of Sciences USA, 94, 6600– 6605.
Kempf, O., Kellerhals, P., Lowrie, W. & Matter, A. 2000. Palaeo-
magnetic directions in late Precambrian glaciomarine sediments of
the Mirbat Sandstone Formation, Oman. Earth and Planetary
Science Letters, 175, 181– 190.
Kilner, B., Conall, M. N. & Brasier, M. 2005. Low-latitude glaciation
in the Neoproterozoic of Oman. Geology, 33, 413–416.
Kröner, A., Linnebacher, P., Stern, R. J., Reischmann, T., Manton,
W. & Hussein, I. M. 1991. Evolution of the Pan-African island arc
assemblages in the southern Red Sea Hills, Sudan, and in south-
western Arabia as exemplified by geochemistry and geochronology.
Precambrian Research, 53, 99 –118.
Kuhn, G., Melles, M., Ehrmann, W. U., Hambrey, M. J. & Schmiedl,
G. 1993. Character of clasts in glaciomarine sediments as an indicator
of transport and depositional processes, Weddell and Lazarev Seas,
Antarctica. Journal of Sedimentary Petrology, 63, 477– 487.
Macouin, M., Besse, J., Ader, M., Gilder, S., Yang, Z., Sun, Z. &
Agrinier, P. 2004. Combined palaeomagnetic and isotopic data
from the Doushantuo carbonates, South China: implications for the
‘snowball Earth’ hypothesis. Earth and Planetary Science Letters,
224, 387– 398
Meert, J. G. 2003. A synopsis of events related to the assembly of eastern
Gondwana. Tectonophysics, 362, 1 –40.
Meert, J. G. & Lieberman, B. S. 2008. The Neoproterozoic Assembly
of Gondwana and its relationship to the Ediacaran – Cambrian
radiation. Gondwana Research, 14, 5– 21.
Meert, J. G. & Torsvik, T. H. 2003. The making and unmaking of a
supercontinent: Rodinia revisited. Tectonophysics, 375, 261–288.
Miller, N. R., Alene, M., Sacchi, R., Stern, R., Conti, A., Kröner, A.
& Zuppi, G. 2003. Significance of the Tambien Group (Tigre,
N. Ethiopia) for snowball Earth events in the Arabian –Nubian
Shield. Precambrian Research, 121, 263– 283.
Miller, N. R., Stern, R. J., Avigad, D., Beyth, M. & Schilman, B.
2009. Neoproterozoic carbonate-slate sequences of the Tambien
Group, N. Ethiopia (I): pre-‘Sturtian’ chemostratigraphy and regional
correlation. Precambrian Research, 170, 129– 156.
Nairn, A. E. M., Perry, T. A., Ressetar, R. & Rogers, S. 1987. A
palaeomagnetic study of the Dokhan volcanic formation and younger
granites, eastern desert of Egypt. Journal of African Earth Sciences,
6, 353–365.
Nesbitt, H. W. & Young, G. M. 1982. Early Proterozoic climates and
plate motions inferred from major element geochemistry of lutites.
Nature, 299, 715–717.
Rieu, R., Allen, P. A., Etienne, J. L., Cozzi, A. & Wiechert, U. 2006.
A Neoproterozoic glacially influenced basin margin succession and
‘atypical’ cap carbonate associated with bedrock palaeovalleys,
Mirbat area, southern Oman. Basin Research, 18, 471– 496.
Rieu, R., Allen, P. A., Cozzi, A., Kosler, J. & Bussy, F. 2007. A
composite stratigraphy for the Neoproterozoic Huqf Supergroup of
Oman: integrating new litho-, chemo- and chronolstratigraphic data
of the Mirbat area, southern Oman. Journal of the Geological
Society, London, 164, 997–1009.
Robb, L. J., Knoll, A. H., Plumb, K. A., Shields, G. A., Strauss, H. &
Veizer, J. 2004. The Precambrian: the Archean and Proterozoic
Eons. In: Gradstein, F., Ogg, J. & Smith, A. G. (eds) A
Geologic Time Scale 2004. Cambridge University Press, Cambridge,
129– 140.
Sifeta, K., Roser, B. P. & Kimura, J. I. 2005. Geochemistry, provenance,
and tectonic setting of Neoproterozoic metavolcanic and metasedi-
mentary units, Werri area, Northern Ethiopia. Journal of African
Earth Sciences, 41, 212–234.
Stern, R. J. 1994. Arc assembly and continental collision in the Neopro-
terozoic East African Orogen: implications for the consolidation of
Gondwanaland. Annual Reviews of Earth and Planetary Sciences,
22, 319– 351.
Stern, R. J., Kröner, A., Bender, R., Reischmann, T. & Dawoud,
A. S. 1994. Precambrian basement around Wadi Halfa, Sudan: a
new perspective on the evolution of the East Saharan Craton. Geolo-
gische Rundschau, 83, 564–577.
Stern, R. J., Avigad, D., Miller, N. R. & Beyth, M. 2006.
Geological Society of Africa Presidential Review: Evidence for
the Snowball Earth Hypothesis in the Arabian –Nubian Shield and
the East African Orogen. Journal of African Earth Sciences, 44,
1 –20.
Tadesse, T. 1997. The Geology of Axum Area (ND 37-6). Ethiopian Insti-
tute of Geological Surveys, Addis Ababa (Memoir No. 9).
Tadesse, T. 1999. Axum sheet geological map. Geological Survey of
Ethiopia, Addis Ababa, Ethiopia. 1:250,000 map.
Tadesse, T., Hoshino, M. & Sawada, Y. 1999. Geochemistry of low-
grade metavolcanic rocks from the Pan African of the Axum area,
northern Ethiopia. Precambrian Research, 99, 101– 124.
Tadesse, T., Hoshino, M., Suzuki, K. & Izumi, S. 2000. Sm– Nd, Rb– Sr
and U– Pb zircon ages of syn and post-tectonic grantoids from the
Axum area of northern Ethiopia. Journal of African Earth Sciences,
30, 313– 327.
Teklay, M. 1997. Petrology, geochemistry and geochronology of Neopro-
terozoic magmatic arc rocks from Eritrea: implications for crustal
evolution in the southern Nubian Shield. Department of Mines,
Asmara, Eritrea, Memoir 1, 125.
Teklay, M. 2006. Neoproterozoic arc –back-arc system analog to
modern arc –back-arc systems: evidence from tholeiite – boninite
association, serpentinite mudflows and across-arc geochemical
trends in Eritrea, southern Arabian –Nubian shield. Precambrian
Research, 15, 81 – 92.
Teklay, M., Kröner, A. & Metzger, K. 2001. Geochemistry, geochro-
nology and isotope geology of Nakfa intrusive rocks, northern
Eritrea: products of a tectonically thickened Neoproterozoic arc
crust. Journal of African Earth Sciences, 33, 283–301.
Teklay, M., Kroner, A. & Mezger, K. 2002. Enrichment from plume
interaction in the generation of Neoproterozoic arc rocks in northern
Eritrea: implications for crustal accretion in the southern Arabian-
Nubian Shield. Chemical Geology, 184, 167– 184.
Teklay, M., Haile, T., Kröner, A., Asmerom, Y. & Watson, J. 2003. A
back-arc palaeotectonic setting for the Augaro Neoproterozoic mag-
matic rocks of western Eritrea. Gondwana Research, 6, 629–640.
Trindade, R. I. F. & Macouin, M. 2007. Palaeolatitude of glacial depos-
its and palaeogeography of Neoproterozoic ice ages. Comptes Rendus
Geoscience, 339, 200–211.
Verri, P. 1909. Contributo allo studio geografio della Colonia Eritrea. Bol-
lettino della Società Geografica Italiana, 10, 251–320. Carta Geolo-
gica 1:1,500,000.
Wilde, S. A. & Youssef, K. 2000. Significance of SHRIMP dating of the
Imperial porphyry and associated Dokhan volcanics, Gebel Dokhan,
northeastern desert, Egypt. Journal of African Earth Science, 31,
403– 413.
Yoshioka, H., Asahara, Y., Tojo, B. & Kawakami, S. 2003.
Systematic variations in C, O, and Sr isotopes and elemental
concentrations in Neoproterozoic carbonates in Namibia: impli-
cations for a glacial to interglacial transition. Precambrian Research,
124, 69– 85.
 Chapter 22

Evidence for Early and Mid-Cryogenian glaciation in the Northern Arabian –Nubian Shield
(Egypt, Sudan, and western Arabia)

ROBERT J. STERN1*, PETER R. JOHNSON2, KAMAL A. ALI1,3 & SUMIT K. MUKHERJEE1,4

1
Geosciences Department, U Texas at Dallas, Richardson TX 75080-3021, USA
2
Johnson and Vranas Associates, Ltd., Geological Consulting, 6016 SW Haines Street, Portland, Oregon 97219-7046, USA
3
Present address: Faculty of Earth Sciences, King Abdulaziz University, Jeddah 21589, Saudi Arabia
4
Present address: BP America Exploration & Production Company, Houston, TX, USA
*Corresponding author (e-mail: rjstern@utdallas.edu)

Abstract: Evidence of Early- to Mid-Cryogenian (c. 780 Ma and c. 740 Ma) glacial activity is summarized for the northern
Arabian–Nubian Shield (ANS), including structural framework, stratigraphy, lithological descriptions and relationships with
younger and older units, banded iron formation chemostratigraphy, other characteristics, geochronological constraints, and discus-
sion. The ANS is a broad tract of juvenile continental crust, formed from accreted arc-backarc basin terranes developed around
the margins of the Mozambique Ocean. As a result, these successions formed in marine environments at some distance from con-
tinental margins. Deposits include banded iron formation (BIF) and possibly glacial diamictite scattered over broad regions of the
Central Eastern Desert of Egypt, NW Arabia and possible correlative units in NE Sudan. The older (c. 780 Ma) examples (Meritri
group, NE Sudan; basal Mahd group, Arabia) occur in the central ANS, on the southern flank of an important lithospheric boundary,
an ophiolite-decorated suture zone. Mahd group diamictite is thin (1– 5 m thick) and rests above the earliest (Cryogenian) ANS
unconformity. The Meritri group interval near Port Sudan is much thicker and part of a deformed passive margin. Both Mahd
and Meritri group deposits need further study before they are accepted as glaciogenic; confirmation of this interpretation would indi-
cate that Neoproterozoic glacial activity began at least as early as 780 Ma ago. The younger (c. 740 Ma) glacial deposits include
diamictite and BIF: the Atud diamictite and BIFs of the Central Eastern Desert of Egypt and the correlative Nuwaybah diamictite
and BIF of NW Arabia. Northern ANS-BIF is a well-layered chemical sediment of interlaminated hematite-magnetite and jasper. A
glacial origin for the Atud-Nuwaybah diamictites is inferred because large clasts and matrix zircons have ages (Palaeoproterozoic
and Neoarchean) and compositions (especially quartzite, arkose, and microdiamictite) that require transport from outside the ANS
Cryogenian basin. Northern ANS-BIF may also reveal glacial influence, having been deposited in response to reoxygenation of a
suboxic ocean. The 740 Ma diamictite and/or BIF may correlate with Tambien Group diamictites in Ethiopia (Miller et al.
2011). Northern ANS diamictite and BIF were deposited in an oceanic basin of unknown size, as indicated by association with
abundant ophiolites; they are strongly deformed, obscuring many primary features.
There is no strong evidence for or against Ediacaran glaciation in the ANS, largely because the region was uplifted at this time.
The c. 600 Ma ANS peneplain may have been partly cut by Ediacaran glaciation. Some of the post-accretionary basins of Arabia
could preserve glaciogenic deposits of Ediacaran age, but assessing this possibility requires further investigation.

The Arabian –Nubian Shield (ANS) consists of mostly Neoproter-
ozoic outcrops around the Red Sea in NE Africa and West Arabia,
exposed by Oligocene and younger uplift and erosion. The ANS is
one of the largest tracts of juvenile continental crust of Neoproter-
ozoic age on Earth; its evolution accompanied a supercontinent
cycle that defined Neoproterozoic tectonics, beginning with the
break-up of the end-Mesoproterozoic supercontinent Rodinia in
early Neoproterozoic time (Stern 2008). ANS juvenile crust
(intra-oceanic arcs and oceanic plateaux) was generated around
and within the Mozambique Ocean and coalesced as this ocean
closed (Stern 1994). The tectonic cycle culminated in a protracted
collision beginning c. 630 Ma, forming the East African Orogen
(EAO), an important weld in the end-Neoproterozoic supercontinent
of ‘Greater Gondwana’ (Stern 1994) or ‘Pannotia’ (Dalziel 1997). In
reconstructed Gondwana, the EAO extends from the Mediterranean
(Tethys) southward along the eastern margin of Africa and across
East Antarctica. Sedimentary evidence of early Cryogenian glacial
episodes is likely to be preserved in ANS sequences, because
these episodes occurred when ANS crustal components were
mostly below sea level (Stern et al. 2006). In contrast, evidence of
late Cryogenian to Ediacaran glacial episodes may be absent
because this was a time of collision and uplift in the ANS. Some evi-
dence for Ediacaran climate may be preserved in post-accretionary
basins of Ediacaran age in Arabia (Johnson 2003).
In this chapter we describe possible glaciogenic units from both
flanks of the Red Sea in the northern ANS; the interested reader
should see the chapter by Miller et al. (2011) for an overview of
possible glacial deposits from the southern ANS as well as Allen
et al. (2011a, b) for chapters covering successions in Oman, to
the east of the ANS. The units summarized here include (i)
Meritri group metaconglomerate (E. Sudan) and basal Mahd
group diamictite (central Arabian Shield); (ii) Atud diamictite
(E. Egypt) and Za’am group Nuwaybah diamictite (NW Arabian
Shield); and (iii) BIF from E. Egypt, NW Saudi Arabia
(Sawawin deposit) and NE Sudan (Fodikwan). Locations of
these units are shown in Figure 22.1 and co-ordinates are listed
in Table 22.1. We know less about the Meritri group and basal
Mahd group and are less confident that they are glaciogenic than
we would like to be, but if confirmed, they would be the oldest
known Neoproterozoic glaciogenic deposits. These are discussed
here to encourage further detailed studies.
The deposits of E. Sudan and the central Arabian Shield have not
been the subject of focused sedimentological study. The Meritri
group conglomerate of Sudan was first identified and briefly
described by Abdelsalam & Stern (1993). The basal Mahd group
diamictite was first mentioned by Johnson et al. (2003) and then
discussed in somewhat greater detail by Stern et al. (2006).
The diamictite and banded iron formations to the north in Egypt
and NW Arabia (Fig. 22.1) have been studied for some time. The
Atud Formation was first described as conglomerate (El-Essawy
1964). Since its first recognition, the Atud Formation has only
been recognized in eastern Egypt between 258N and 268N.
At the type locality near Gebel Atud, it includes schist, metamud-
stone, and metagreywacke in addition to diamictite, all of which

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 277– 284. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.22
278 R. J. STERN ET AL.

Cairo Jordan suffered greenschist –facies metamorphism (El-Essawy 1964).

The term ‘conglomerate’ is not appropriate for the poorly sorted,
29°
matrix-supported deposits of the Atud Formation. These better
fit the description of diamictite by Flint et al. (1960) as poorly
Area of Figure 3 sorted, heterolithic, and very coarse terrigenous sediments.
Saudi Arabia
Accordingly, in this chapter we refer to the Atud diamictite
(Stern et al. 2006).
Area of Figure 3

Al Muwaylih
Hurghada
27°
Safajah
Nil
e

Quseir Al Wajh
Structural framework

The Meritri group conglomerate (Sudan) is preserved within the B’ir

25° Marsa Alam Umq-Nakasib Suture Zone, whereas the Mahd group diamictite
Egypt (Saudi Arabia) is preserved just south of this suture zone. This is
Al Madinah
one of the longest and best-defined ophiolite-decorated suture
zones in the ANS (Johnson et al. 2003) and extends (with the Red
Sea closed) ENE–WSW over 600 km from the Nile in Sudan into
23°
Red Sea the central Arabian Shield. The suture zone itself consists of rocks
that originated in a variety of juvenile oceanic environments and
Biír Umq suture Mahd gp. include strongly deformed ophiolite nappes, and metavolcanic,
Sudan F Jiddah
diamictite metasedimentary, and intrusive rocks. Dating of the ophiolites, vol-
canic rocks, and pre- and syntectonic plutons indicates that oceanic
21°
magmatism in the region was active c. 870–830 Ma, whereas sutur-
ing occurred c. 780–760 Ma (Hargrove III et al. 2006). Structural
utu
re Port Sudan complexities (folding, faulting, shearing, etc.) are especially severe
sibs for the Meritri group, which is part of a southwards-directed nappe
N aka
19° stack (Fig. 22.2a), discussed in the following section (Abdelsalam
200 km Meritri gp. & Stern 1993). The Mahd group diamictite is less folded and
diamictite faulted because it lies south of the suture zone.
31° 33° 35° 37° 39° 41° The Atud and Nuwaybah diamictites and associated BIF in
Suture Egypt and NW Saudi Arabia are also folded and faulted. BIF is
Phanerozoic Shear zone
Exposed Neoproterozoic crust
strongly deformed although much of this deformation may have
Diamictite
Ophiolitic complexes Banded Iron Formation occurred as a result of slumping of dense, weak sediments in a tec-
tonically unstable basin. BIF and surrounding sediments are meta-
Fig. 22.1. Locality map of the northern ANS showing the approximate morphosed to greenschist facies. Tectonic deformation began as a
locations of geological units described in this chapter. The dark, dashed square result of collision between arc terranes prior to c. 680 Ma (Ries
designates the area shown in Figure 22.3. Early Cryogenian (c. 780 Ma) et al. 1983). Other deformation resulted from pervasive left-lateral
diamictite sequences (Meritri and Mahd) are confined to the Nakasib and Bir strike-slip shearing along the Najd fault system, which was active
Umq suture areas in the south, whereas Atud and Nuwaybah diamictite and BIF during Ediacaran time (Sultan et al. 1988). Najd deformation was a
sequences are confined to Egypt and NW Saudi Arabia. far-field manifestation of collision between fragments of east and
west Gondwana (Abdelsalam & Stern 1997). Najd faulting
imparted a penetrative shear fabric to most Cryogenian supracrus-
tal units in Egypt and NW Arabia.
Table 22.1. Locations of Neoproterozoic deposits of possible glacial origin in
the northern ANS
Stratigraphy
Nation Location and lithology Age (Ma) Latitude (N) Longitude (E)
Cryogenian supracrustal sequences in the northern ANS are
Egypt dominated by variably deformed immature clastic metasediments
Wadi Abu Marwat BIF c. 750 268310 338380 (greywackes) and metavolcanic rocks; rare sedimentary carbon-
Wadi Kareim BIF c. 750 258560 500 348020 ates also occur. Cryogenian stratigraphic reconstructions are
Wadi Kareim diamictite c. 750 258560 500 348020 complicated because of strong deformation as well as the pres-
Wadi Muweilha diamictite c. 750 258520 348540 3000 ence of several accreted terranes (Johnson & Woldehaimanot
Wadi El Dabbah BIF c. 750 258490 348090 2003). Cryogenian metasedimentary successions are also litho-
Um Gerifat BIF c. 750 258400 348200 logically monotonous. Distinctive units such as sedimentary car-
Um Ghamis BIF c. 750 258370 348190 bonates are uncommon, although these become increasingly
Wadi Sitra BIF c. 750 258310 348140 important in the ANS farther south in Sudan, Eritrea, Ethiopia,
Wadi Mubarak diamictite c. 750 258260 348340 and SW Arabia. As a result, ANS Cryogenian supracrustal
El Hadid BIF c. 750 258210 348080 units often lack a useful formal stratigraphic framework. This
El Imra BIF c. 750 258170 348270 is especially true for Egypt, where Stern (1981) informally
Um Nar BIF c. 750 258160 348160 divided the Cryogenian supracrustal succession of the Central
Gebel Atud diamictite c. 750 258010 348270 Eastern Desert into a basal ‘older (ophiolitic) metavolcanics’,
Saudi Arabia ‘metasediments’, and ‘younger (arc-like) metavolcanics’. Egyp-
Wadi Sawawin BIF c. 750 278540 358460
tian Cryogenian successions are unconformably overlain by
Nuwaybah diamictite c. 750 268270 368240
Ediacaran clastic sediments of the Hammamat Group and the
Mahd diamictite c. 780 238240 2200 408460 5200
Dokhan Volcanics. A similar situation exists for Sudan. In con-
Sudan
Meritri Gp. conglomerate c. 780 198380 368420
trast, there are a plethora of stratigraphic names for Cryogenian
Fodikwan BIF – 218440 368420
supracrustal successions of Saudi Arabia, largely based on the
results of quadrangle mapping.
 EARLY AND MID-CRYOGENIAN GLACIATION IN THE NORTHERN ARABIAN –NUBIAN SHIELD 279

Fig. 22.2. Stratigraphic summaries of possibly glaciogenic Cryogenian units described in this chapter, locations presented in Table 22.1. (a) Stratigraphic chart for
possibly glaciogenic conglomerates of the Meritri group, Nakasib suture, NE Sudan. Column A is a measured section for the lower part of the Arba’at volcanic rocks
along Khor Arba’at. Column B is a representative section for the sedimentary part of the Salatib group along Khor Salatib. Column C is a representative section for
diamictite of the Meritri group along Khor Meritri. Column D summarizes the stacking order observed in the nappes of the Nakasib suture (from Abdelsalam & Stern
1993). (b) Simplified stratigraphic column for Mahd group rocks showing stratigraphic position of basal diamictites (after Lowther 1994). (c) Simplified stratigraphic
column for Atud diamictite and associated BIF in Wadi Kareim, Central Eastern Desert of Egypt, from Ali et al. (2010). (d) Simplified stratigraphic column for
Nuwaybah diamictite in NW Saudi Arabia (after Davies 1985).
280 R. J. STERN ET AL.

0 50 100 km east of Gebal Atud (Table 22.1, Fig. 22.3). The Wadi Kareim
33˚E Banded Iron Formation
occurrence is especially significant (Figs 22.2c & 22.3),
28˚N Diamictite because this is the only locality where the Atud diamictite is
Ophiolitic melange found in clear stratigraphic relationship with BIF. The diamictite
Egypt is part of a supracrustal succession in which metavolcanic rocks
Ophiolites
HU
are conformably overlain by immature metasedimentary rocks
Major city (c. 100 m thick of wackestone-sandstone, siltstone and diamic-
Local Road tite) and BIF. Metavolcanic rocks at the base of the section
27˚N Highway are c. 100 m thick and are truncated below by a thrust fault.
36˚E
Egyptian diamictite and BIF are found in regional association
SA with Cryogenian ophiolites, a tripartite association that indicates
37˚E
Abu Marawat that both sedimentary units were deposited in an oceanic basin.
Sawawin Deformation obscures small-scale lateral variations of the Atud
28˚ Diamictite; the BIF is restricted to the Central Eastern Desert
ALM
QU
26˚N Qift - Quseir Ro
ad
Kareim of Egypt, with nine occurrences between 258150 N and 268350 N
Muweilih Saudi Arabia (Fig. 22.3, Table 22.1).
Dabbah In NW Saudi Arabia, the diamictite and BIF are geographically
Um Gerifat separate parts of the Za’am group, the oldest known unit in the
Um Ghamis Mobarak Midyan terrane. The diamictite (Nuwaybah Formation) is in the
Gebal EL Hadid El Imra 27˚ upper part of the Zaam group, which is an assemblage of basaltic,
MA Nuwaybah
andesitic, and subordinate rhyolitic flows and tuffs, abundant
25˚N Idfu - Marsa Alam Road
Um Nar volcaniclastic sandstone and siltstone, and volcaniclastic conglom-
Atud erate (Fig. 22.2d). The Zaam group was deformed and metamor-
AL
phosed (greenschist facies) prior to c. 660 Ma, and neither
HU: Hurghada MA: Marsa Alam
SA: Safajah AL: Al Wajh
original top nor bottom is exposed. It is overlain unconformably
QU: Quseir ALM: Al Muwaylih
by the younger Thalbah group. The significance of the basal
26˚
Thalbah group unconformity is unknown but could correspond
to Cryogenian glacial erosion. The lateral extent of the Nuwaybah
Fig. 22.3. Location of the Atud and Nuwaybah diamictite and associated BIF, diamictite is not yet known.
as well as ophiolitic rocks in Egypt and Saudi Arabia, northern ANS, with the
Further north, in Wadi Sawawin, the BIF-bearing succession
Red Sea closed. Diamictites in this area are thought to be early Cryogenian in
consists of Ghawjah Formation metavolcanic rocks overlain by
age. See Figure 22.1 for regional location (modified after Ali et al. 2010)
and Table 22.1 for geographical coordinates of sites.
Silasia Formation metasedimentary rocks. Ghawjah metavolcanic
rocks are correlated with the ‘younger metavolcanic’ rocks of
Egypt. The Silasia Formation consists of c. 1-km-thick immature
clastic metasedimentary rocks with BIF-bearing units near the
Meritri group metaconglomerate of possible glacial origin in top; the sequence is intruded by diabase sills. The stratigraphic
E. Sudan is associated with five informal groups, separated by and age relationships shown by the successions at Wadi Kareim
thrusts so that original depositional relationships are unresolved and Wadi Sawawin are remarkably similar except that the diamic-
(Fig. 22.2a). The c. 2-km-thick Arbaat (metavolcanic) group lies tite is absent from the latter succession.
at the base of the succession and appears to be autochthonous.
Subsequent units are thrust southwards on top of the Arbaat
group and one another and include from the base: the Salatib Possible glaciogenic deposits and associated strata
group, a 1.2-km-thick succession of clastic metasediments, car-
bonates, conglomerate and felsic tuffs, the Meritri group composed (A1) Meritri group metaconglomerate, E. Sudan
of coarse conglomerate, lithic greywacke, limestone, red sandstone
and felsic metavolcanics, the Nakasib ophiolite and at the top, the Abdelsalam & Stern (1993) inferred that the original thickness for
arc-like Shalhout group (Fig. 22.2a; Abdelsalam & Stern 1993). the Meritri group in Khor Meritri was c. 2 km. They subdivided the
Lateral variations within the Meritri group show changes from group into four formations comprising, from oldest to youngest:
granitic-clast dominated in the east to volcanic-clast dominated conglomerate (possibly glaciogenic); lithic greywacke; inter-
in the west. calated limestone, red sandstone and felsic tuff; and felsic volcanic
Diamictite at the base of the c. 780 Ma Mahd group (Fig. 22.2b) rocks. The coarsest clastic beds are made up of abundant polymict
unconformably overlies diorite and tonalite of the c. 810 Ma conglomerate intercalated with minor lithic greywacke and lime-
Dhukhr batholith, indicating an episode of possibly glacial stone. The polymict conglomerate is matrix supported, with the
erosion at c. 780 –810 Ma (Johnson et al. 2003). Continuity of matrix made up of lithic greywacke and minor carbonate. Clasts
unit to the north is truncated by terrane boundary; possible conti- include granite, granodiorite, diorite, rhyolite, ignimbrite and car-
nuity to the south and east is unknown. The basal Mahd group bonate as well as subordinate clastic metasediments. Intermediate
unconformity is one of the oldest unconformities known in the to felsic volcanic clasts become more abundant to the west. In the
ANS. Another early Cryogenian unconformity, which might be NE, along Khor Meritri, plutonic clasts are most abundant, com-
related to the basal Mahd group unconformity, is found farther prising c. 50% of total clasts, whereas volcanic clasts are c. 35%
south in the Arabian Shield, between plutonic rocks of the and clasts of metasediments make up c. 15%. Abdelsalam (pers.
c. 800 Ma An Nimas batholith and metamorphosed conglomerate, comm. 2007) measured 76 of these clasts, finding a maximum
limestone, and sandstone of the 780 –795 Ma Hali group (Cooper size of 70  40  35 cm3 (clasts are deformed and stretched);
et al. 1979). This encourages speculation that heretofore unrecog- clasts .40 cm are common. The lithic greywacke unit is made
nized Cryogenian glacial deposits might exist in the southern up of lithic greywacke intercalated with minor felsic volcanic
Arabian Shield. layers and limestones. Locally the lithic greywacke grades into
The Atud diamictite is a distinctive lithology found in the conglomerate with smaller clasts of mostly felsic volcanic rocks.
Egyptian metasedimentary succession. This has been reported Sedimentary structures include graded bedding in the sandstone,
from four areas in the Central Eastern Desert of Egypt: Wadi cross-bedding and channels. Channel structures and cross-bedding
Kareim, Wadi Mobarak, Wadi Muweilih, and the type locality indicate that the palaeocurrent direction was from SE to NW.
 EARLY AND MID-CRYOGENIAN GLACIATION IN THE NORTHERN ARABIAN –NUBIAN SHIELD
(A2) Basal Mahd group diamictite, central Arabian Shield
Although dominated by volcanic rocks, the Mahd group has a 1–
5-m-thick diamictite resting on the unconformity. The diamictite is
matrix-supported, with a dark-grey, immature, arkosic matrix con-
taining abundant, angular to sub-angular clasts (up to 30 cm
across) of granitic and felsic volcanic rocks.
(B1) Atud diamictite, E. Egypt
The Atud diamictite consists of massive poorly sorted and rounded
clasts, from gravel to boulder, in a sheared grey matrix. Atud clasts
include quartzite, highly altered granitoid, and a distinctive arkosic
breccia (microdiamictite) (Ali et al. 2010). Clasts and matrix
zircons in the Atud diamictite are dominated by c. 750 Ma granitic
rocks and microdiamictite but with a significant amount of
Palaeoproterozoic and Neoarchean granitic rocks and quartzite
(Ali et al. 2010).
We have also examined Atud diamicite from the eastern half of
the Wadi Mobarak belt in Egypt. Basement exposures around
Wadi Mobarak are dominated by highly deformed ophiolitic frag-
ments of serpentinites, metagabbro, and greenschist-facies mafic
metavolcanic rocks, whereas the sedimentary sequence includes
tuff, shale, schist and Atud diamictite (Akaad et al. 1995). Diamic-
tite clasts are similar to those observed in Wadi Kareim.
(B2) Nuwaybah Formation (Za’am group) diamictite,
NW Saudi Arabia
The Nuwaybah locality is located within the Al Wajh quadrangle
in the northwestern part of the Arabian Shield. The diamictite
locality we studied is beautifully exposed in a roadcut along
the Red Sea highway (Table 22.1). Clasts and matrix zircons in
the Nuwaybah diamictite are dominated by c. 750 Ma granitic
rocks and distinctive arkose/micro-diamictite but with a signifi-
cant amount of Palaeoproterozoic and Neoarchaean granitic
rocks and quartzite (Ali et al. 2010), similar to lithologies and
ages of the Atud diamictite in Egypt.
(C1) Banded iron formation (BIF), E. Egypt
BIF of early Cryogenian age is found in the Central Eastern Desert
of Egypt, with nine occurrences between latitudes 258150 and
268350 N. The Fodikwan BIF in NE Sudan may be related
(Table 22.1, Fig. 22.1), but more work is needed to test this possi-
bility. BIF occurs as fairly regular bands interbedded with meta-
sediments and metavolcanics in a zone that originally had a
stratigraphic thickness of 100– 200 m, within which the aggregate
BIF thickness is about 10 –20 m (Sims & James 1984). Egyptian
BIF is mostly an oxide facies, consisting of interlaminated hema-
tite and jasper, and containing 40– 46% Fe (Sims & James 1984).
In several locales, BIF-bearing metasediments are intruded by
metadiabase sills.
(C2) Silasia Formation BIF, NW Saudi Arabia
BIFs in NW Saudi Arabia occupy a more restricted region than do
their Egyptian counterparts. Arabian BIF occurs within the Silasia
Formation, which, like the Egyptian section, consists of volcano-
genic greywackes that appear to rest conformably on metavolcanic
rocks (Ghawjah Formation). The exposed thickness of the Silasia
Formation is estimated to be c. 1160 m in the reference area of
Wadi Sawawin (Goldring 1990). Arabian BIF is mostly present
as an oxide facies, consisting of interbedded hematite and jasper,
and containing 40–46% Fe (Goldring 1990). Similar to the Egyp-
tian section, the Silasia Formation is intruded by metadiabase sills
up to 100 m thick.
281
Boundary relations with overlying and underlying
non-glacial units
(A1) Meritri Group metaconglomerate, E. Sudan
Contacts with underlying and overlying units are faulted.
(A2) Basal Mahd Group diamictite, central Arabian Shield
Contacts with underlying units are unconformable. Contacts with
overlying units are conformable.
(B1) Atud diamictite, E. Egypt
Contacts with underlying units are only clear at Wadi Kareim,
where they appear conformable.
(B2) Nuwaybah Formation (Za’am group) diamictite,
NW Saudi Arabia
The basal contact of the Nuwaybah Formation is thought to be con-
formable on the underlying Umm Ashsh Formation (part of the
Za’am group). The top of the Formation is obscured by granite
intrusions and by an angular unconformity with younger sedimen-
tary rocks (Thalbah group).
(C1) BIF, E. Egypt
This BIF is found in conformable stratigraphic relationship with
underlying immature, possibly tuffaceous greywackes. It is associ-
ated with metavolcanic rocks and intrusive diabase.
(C2) Silasia Formation BIF, NW Saudi Arabia
The BIF-bearing sediments are conformable above metavolcanic
rocks of the Ghawjah Formation. The top of the formation is not
exposed at the Sawawin locality because it is intruded by metadiorite.
Chemostratigraphy
This section is applicable only for the BIF. There are no significant
carbonate sediments in the region, although these become increas-
ingly important to the south (in Sudan, S. Saudi Arabia, Eritrea,
and Ethiopia, see Allen et al. 2011a, b; Miller et al. 2011). NW
Saudi Arabia (Sawawin) BIF has been sampled systematically
and stratigraphically, in order to identify the nature of its sources
(Mukherjee 2008). Samples of Egyptian BIF were also studied in
order to understand regional variations. Rare earth elements
(REE) are particularly useful for studying the BIF because these
record the composition of equilibrium seawater that the BIF
precipitated from.
REE patterns for the Sawawin BIFs are similar to a mixture of
modern shallow suboxic seawater (German et al. 1991; Webb &
Kamber 2000) with low-T hydrothermal vent fluid solution (Bau
& Dulski 1999) suggesting a dominant hydrothermal input of
REEs and by analogy Fe into the BIFs (Mukherjee 2008). BIF
iron input sources are isotopically dominated by hydrothermal
vent fluids but continental runoff is also significant as revealed
by Ce/Ce* and Eu/Eu* (Mukherjee 2008).
Other characteristics
Gold mineralization may be associated with BIF at Abu
Marawat (Botros 2002). BIF here occurs as sharply defined hor-
izons within a volcanic-sedimentary succession, which is region-
ally metamorphosed into greenschist facies. Gold concentrations
282 R. J. STERN ET AL.
of up to 2.15 ppm occur in the BIF, either enclosed in the flaky
hematite crystals of hematite-rich layers or as fine inclusions in
magnetite-rich bands.
El-Habaak & Mahmoud (1995) identified spherical bodies as
Eosphaera tyleri in jasper-rich layers from Wadi Kareim BIF,
Egypt. Those occur as more or less clear spherules of quartz sur-
rounded by thin veneers of very fine hematite granules. They
recognized two distinct varieties of E. tyleri, one of which is
c. 15 mm in diameter and the other c. 60 mm; El-Habaak &
Mahmoud (1995) suggested that biological activity played a role
in spherule formation and BIF deposition.
The Mahd group basal diamictite is overlain by hypabyssal intru-
sive to volcaniclastic rocks (caldera complex) that hosts a major
epithermal gold deposit (Mahd adh Dhahab Mine). The age of gold
mineralization is not known with certainty, but is likely to have
been syngenetic with Mahd Group igneous activity (c. 760–
c. 780 Ma; Hargrove III et al. 2006). The deposit consists of quartz
veins and stockwork that contains copper, zinc, iron, and lead sul-
phides and very fine-grained gold and silver (averaging 5–30 mm),
mostly as tellurides, associated with the sulphides (Moore 1979).
Palaeolatitude and palaeogeography
Extensive deformation and metamorphism makes palaeomagnetic
determinations unreliable, and for this reason few have been
carried out (Reischmann et al. 1992).
Geochronological constraints
(A1) Meritri Group metaconglomerate, E. Sudan
The age of the Meritri group is constrained by the age of underlying
metavolcanics and intrusive plutonic rocks, reported by Stern &
Abdelsalam (1998). Conventional multigrain analyses of two size
fractions of zircons separated from a metarhyolite lava of the appar-
ently underlying Arba’at Formation yielded a nearly concordant
age of 790 + 2 Ma. The supracrustal sequence, including the
Meritri group, was intruded by the Arba’at quartz diorite, which
yielded a single fraction, nearly concordant U–Pb zircon age of
779 + 3 Ma (Stern & Abdelsalam 1998). Faulting leads to uncer-
tainties about the exact relationship between the intrusion and
Meritri group, but these ages suggest that deposition of the Meritri
diamictite may be constrained between 779 + 3 and 790 + 2 Ma.
(A2) Basal Mahd Group diamictite, central Arabian Shield
The age of this diamictite is constrained by the age of subjacent and
superjacent igneous rocks. The Dhukhr complex, which uncon-
formably underlies the diamictite, has robust conventional multi-
grain U –Pb zircon crystallization ages of 811 + 4 Ma (Stoeser
& Stacey 1988) and 816 + 4 Ma (Calvez & Kemp 1982). The
Hufayriyah batholith, which also unconformably underlies the
Mahd group c. 60 km north –NE of the diamictite locality, has a
U –Pb zircon SHRIMP age of 785 + 6 Ma (Hargrove III et al.
2006). Rhyolite in the Mahd group upsection from the basal dia-
mictite has been dated by U –Pb zircon SHRIMP techniques at
777 + 5 Ma (Hargrove III et al. 2006). In addition to the U –Pb
zircon SHRIMP age, the Hufayriyah batholith also yields a con-
ventional multigrain U –Pb zircon age of 760 + 10 Ma (Calvez
& Kemp 1982); this result may represent a younger pulse of
Hufayriyah intrusion and is not necessarily a robust constraint on
the age of rocks at the base of the Mahd group.
(B1) Atud diamictite, E. Egypt
The age of this unit is constrained to be younger than both the age
of the underlying metavolcanics (Fig. 22.2c) and the age of the
youngest clast within the diamictite. Ali et al. (2009) report a
SHRIMP U –Pb zircon ion probe age of c. 750 Ma for the meta-
volcanics that lie beneath the Atud diamictite at Wadi Kareim.
One metavolcanic sample yielded a weighted mean 206Pb– 238U
age of 769 + 29 Ma.
As noted above, the Atud diamictite at Wadi Kareim is
dominated by Cryogenian clasts and matrix material but con-
tains abundant clasts of Palaeoproterozoic and Neoarchaean
granitic rocks and pre-Neoproterozoic quartzite. Clasts in the
diamictite are as young as 754 + 15 Ma (Ali et al. 2010).
Because of stratigraphic relationships shown on Figure 22.2c,
the maximum age of c. 750 Ma for the metavolcanics and
754 + 12 Ma for the Atud diamictite at Wadi Kareim also pro-
vides a maximum age for BIF deposition at this locality (Ali
et al. 2009, 2010).
Geochronological data support the inference that clasts in
the Atud diamictite sample much older rocks than are exposed in
the Eastern Desert of Egypt and so must have been transported
some distance. Two granitic cobbles from NW of Marsa Alum
(also referred to as the Wadi Mobarak metasedimentary unit)
yielded highly discordant conventional multigrain U –Pb zircon
upper intercept ages of 1120 and 2060 Ma (Dixon 1981). Dixon
(1979) obtained a discordant U –Pb zircon upper intercept of
2.3 Ga for a granitic cobble from Atud conglomerate outcrops
west of Quesir. SHRIMP geochronological studies by Ali et al.
(2010) confirm and extend Dixon’s conclusions that many Atud
clasts are pre-Neoproterozoic.
(B2) Nuwaybah Formation (Za’am Group) diamictite,
NW Saudi Arabia
Like the Atud diamictite, this unit is dominated by Cryogenian
material but contains abundant clasts of Palaeoproterozoic and
Neoarchaean granitoids and quartzite. The youngest clasts in the
Nuwaybah diamictite yield SHRIMP U –Pb zircon ages of
765 + 22 Ma (Ali et al. 2010). Another clast is an arkose with a
SHRIMP U – Pb zircon age of 766 + 5 Ma, which we interpret
as the age of the basement supplying the arkose, which was
rapidly eroded and deposited, lithified, and re-eroded, all appar-
ently within a very few millions of years.
The formation has a minimum age of deposition constrained by
the age of intrusions in the Zaam Group near the diamictite. These
include the Buwaydah complex, dated by the conventional U –Pb
zircon method at 725 + 4 Ma (Hedge 1984) and the Imdan
complex, dated by the SHRIMP U –Pb zircon method at
676 + 6 Ma (15 data points (Kennedy et al. 2011) and convention-
al U – Pb zircon method at 660 + 4 Ma (Hedge 1984). Zaam Group
felsic tuff 175 km SE of the diamictite locality yields U– Pb zircon
SHRIMP ages of 711 + 10 Ma and 708 + 4 Ma (Kennedy et al.
2004, 2005). However, the stratigraphic relationship between the
diamictite and these tuffs is poorly understood, thus the signifi-
cance of these data for constraining the depositional age of the dia-
mictite is not clear.
(C1) BIF, E. Egypt
BIF age is best known from Wadi Kareim, where ages of conform-
ably underlying metavolcanics and diamictite indicate that it is
younger than c. 750 Ma (Ali et al. 2009, 2010), but perhaps not
much younger.
(C2) Silasia Formation BIF, NW Saudi Arabia
The Silasia Formation, which hosts this BIF, is intruded by pluto-
nic rocks of the Muwalylih suite, dated by U –Pb zircon techniques
at 710–725 Ma (Hedge 1984). Unfortunately, Hedge (1984) did
not provide analytical details, the number of zircon populations
or statistics about the results (uncertainty and mean square of the
weighted deviates). Ali et al. (2011) report a U –Pb zircon ion
 EARLY AND MID-CRYOGENIAN GLACIATION IN THE NORTHERN ARABIAN –NUBIAN SHIELD
probe weighted mean 206Pb/238U age of 763 + 25 Ma for
Ghawjah metavolcanics beneath the BIF. Silasia Formation was
deformed and then intruded by Sawawin complex diorite, which
yields a U – Pb zircon concordia age of 661.5 + 2.3 Ma (Ali
et al. 2011). Silasia Formation BIF is thus , 763 + 25 Ma, prob-
ably c. 750 Ma, similar in age to Egypt BIF.
Discussion
The foregoing summary indicates that there is compelling evi-
dence of a c. 740 Ma age glaciation and plausible evidence of a
c. 780 Ma glaciation in the northern ANS. Indistinguishable brack-
eting ages of the Meritri group (790 + 2 Ma to 779 + 3 Ma) and
Mahd group (785 + 6 Ma to 777 + 5 Ma) indicate that deposition
of these units may have occurred about the same time. A glacio-
genic interpretation for c. 780 Ma Mahd-Meritri groups might
help explain the enigmatic Bitter Springs d13C excursion of
approximately the same age, thought to be associated with glacia-
tion but for which no definitive evidence of glaciation has been
identified to date (Halverson et al. 2005, 2007). The association
of the Mahd diamicite with the oldest unconformity known from
the Arabian Shield is also consistent with a glacial interpretation
and is tentative evidence of a c. 780 Ma local continental
glaciation, probably sourced to the south (present coordinates).
Certainly, more work is needed to map, study sedimentology and
stratigraphy, date clasts and matrix, and examine intervening
sedimentary and possible igneous environments of Mahd and
Meritri groups deposits.
Evidence for a c. 740 Ma glaciation is found in the ophiolite-
Atud diamictite-BIF basin and includes far travelled exotic clasts
(based on age and composition) and association of diamictite
with BIF. By virtue of their age, these deposits may correlate
with early Cryogenian glaciogenic successions elsewhere (Fair-
child & Kennedy 2007). The Atud-Nuwaybah diamictite may be
strictly correlative, with similar ages of deposition, clast lithology
and distribution of Cryogenian, Palaeoproterozoic and Neoarch-
aean clast ages (Ali et al. 2010). The association of diamictite
with BIF is consistent with the genetic coupling suggested by
Snowball Earth concepts (Hoffman 2005), although alternative
interpretations exist. Based on similar age constraints (the young-
est detrital zircons in the Ethiopian diamictites are also c. 0.75 Ga),
Atud diamictite may also be broadly synchronous with Negash and
Shiraro diamictites in northern Ethiopia (Avigad et al. 2007; Miller
et al. 2003, 2011). Neoproterozoic diamictites from the ANS (this
chapter and Miller et al. 2011) do not seem to correlate directly
with diamicitites of the Abu Mahara Group of the Huqf Super-
group, Oman, which are younger in age (700– 735 Ma) and are
dominated by c. 860 Ma zircons that are not common in the
Atud/Nuwaybah diamictites (Rieu et al. 2007).
Because pre-Neoproterozoic basement is unknown in Egypt east
of the Nile, Dixon (1981) concluded that these clasts were derived
from older crust to the west or south, perhaps from the Saharan
Metacraton (Abdelsalam et al. 2002). Dixon (1979) suggested
that Atud clasts were transported such great distances by ice
rafting, but other possible explanations (e.g. meteorite impact
ejecta blanket, far-travelled submarine debris flow) are possible.
There is controversy regarding how the Egyptian BIFs formed,
although ideas published in the geological literature were mostly
developed prior to the Snowball Earth hypothesis (Fairchild &
Kennedy 2007). Sims & James (1984) suggested that BIF
formed as chemical precipitates during lulls in dominantly subaqu-
eous, calc-alkaline volcanism, apparently within an intraoceanic
island-arc environment. The close association in time and space
between volcanic activity and deposition of the BIF suggests a
genetic relation (Sims & James 1984).
Finally, it should be noted that there is no strong evidence for or
against an Ediacaran glaciation in the ANS. The ANS peneplain
may have been partly cut during this time, but regional uplift
resulted in generally poor preservation (Stern et al. 2006). Some
283
of the post-accretionary basins of Arabia may preserve glaciogenic
deposits of Ediacaran age, but focused investigations are needed to
establish or refute this possibility (Johnson 2003).
We are grateful for support by many individuals and agencies over the years
including the Sudan Geologic Survey, Saudi Geological Survey, Nuclear Materials
Agency of Egypt, NASA, NSF and the US–Egypt Joint Technical Program. We
also thank M. Abdelsalam, D. Avigad, M. Beyth, N. Miller, F. Kattan, V. Pease,
M. Whitehouse and A. Kröner for discussions on this topic, A. Collins, V. Pease
and D. Avigad for thoughtful reviews of the manuscript, and E. Arnaud for
careful final editing. This represents a contribution of the IUGS- and UNESCO-
funded IGCP (International Geoscience Programme) Project #512.
References
Abdelsalam, M. G. & Stern, R. J. 1993. Tectonic evolution of the
Nakasib suture, Red Sea Hills, Sudan: evidence for a late Precambrian
Wilson cycle. Journal of the Geological Society, London, 150,
393– 404.
Abdelsalam, M. G. & Stern, R. J. 1997. Sutures and shear zones in the
Arabian-Nubian Shield. Journal of African Earth Sciences, 23,
289– 310.
Abdelsalam, M. G., Liégeois, J. P. & Stern, R. J. 2002. The Saharan
Metacraton. Journal of African Earth Sciences, 34, 119–136.
Akaad, M. K., Noweir, A. M. & Abu-El-Ela, A. M. 1995. The volcano-
sedimentary association and ophiolites of Wadi Mubarak, eastern
Desert, Egypt. Proc. Inter. Conf. 30 Years Cooper., 69. Geological
Survey of Egypt Special Publication, 231–248.
Ali, K. A., Stern, R. J., Manton, W. I., Kimura, J.-I. & Khamis, H. A.
2009. Geochemistry, Nd isotopes, and U –Pb SHRIMP zircon dating
of Neoproterozoic volcanic rocks from the Central Eastern Desert of
Egypt: new insights into the c. 750 Ma crust-forming event. Precam-
brian Research, 171, 1 –22.
Ali, K. A., Stern, R. J., Manton, W. I., Johnson, P. R. & Mukherjee,
S. K. 2010. Neoproterozoic diamictite in the Eastern Desert of Egypt
and Northern Saudi Arabia: evidence of c. 750 Ma glaciation in the
Arabian– Nubian Shield. International Journal of Earth Sciences,
99, 705– 726.
Ali, K. A., Stern, R. J. et al. 2011. Geochemical, U– Pb zircon and Nd
isotopic investigations of the Ghawjah metavolcanics of northwestern
Saudi Arabia: the 750 Ma crust-forming event correlated across the
Red Sea. Lithos, 120, 379– 392.
Allen, P. A., Rieu, R., Etienne, J. L., Matter, A. & Cozzi, A.
2011a. The Ayn Formation of the Mirbat Group, Dhofar, Oman.
In: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds)
The Geological Record of Neoproterozoic Glaciations. Geological
Society, London, Memoirs, 36, 239–249.
Allen, P. A., Leather, J. et al. 2011b. The Abu Mahara Group
(Ghubrah, and Fiq Formations), Jabal Akhdar, Oman. In: Arnaud,
E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological
Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 251–262.
Avigad, D., Stern, R. J., Beyth, M., Miller, N. & McWilliams, M. O.
2007. Detrital zircon U– Pb geochronology of Cryogenian
diamictites and Lower Paleozoic sandstone in Ethiopia (Tigrai):
age constraints on Neoproterozoic glaciation and crustal evolution
of the southern Arabian– Nubian Shield. Precambrian Research,
154, 88– 106.
Bau, M. & Dulski, P. 1996. Distribution of yttrium and rare-earth
elements in the Penge and Kuruman iron-formations, Transvaal
Supergroup, South Africa. Precambrian Research, 79, 37 –55.
Bau, M. & Dulski, P. 1999. Comparing yttrium and rare earths in hydro-
thermal fluids from the Mid-Alantic Ridge: Implications for Y and
REE behaviour during near-vent mixing and for the Y/Ho ratio of
Proterozoic seawater. Chemical Geology, 155, 77 – 90.
Botros, N. S. 2002. Metallogeny of gold in relation to the evolution of the
Nubian Shield in Egypt. Ore Geology Reviews, 19, 137– 162.
Calvez, J.-Y. & Kemp, J. 1982. Geochronological investigations in the
Madh adh Dhahab quadrangle, Central Arabian Shield. BRGM-TR-
02-5, Saudi Arabian Deputy Ministry for Mineral Resources.
Cooper, J. A., Stacey, J. S., Stoeser, D. G. & Fleck, R. J. 1979. An
evaluation of the zircon method of isotopic dating in the Southern
284 R. J. STERN ET AL.
Arabian Craton. Contributions to Mineralogy and Petrology, 68,
429– 439.
Dalziel, I. W. D. 1997. Neoproterozoic –Paleozoic geography and tec-
tonics: review, hypothesis, environmental speculation. Geological
Society of America Bulletin, 109, 16 –42.
Davies, F. B. 1985. Geologic map of the Al Wajh quandrangle, sheet 26B,
Kingdom of Saudi Arabia. Saudi Arabian Deputy Ministry for
Mineral Resouces, Geoscience Map 83A, Jeddah.
Dixon, T. H. 1979. The evolution of continental crust in the Late Precam-
brian Egyptian Shield. Ph.D. Thesis, UC San Diego.
Dixon, T. H. 1981. Age and chemical characteristics of some pre-Pan-
African rocks in the Egyptian Shield. Precambrian Research, 14,
119– 133.
El-Essawy, M. A. 1964. Geology of the area east of Gabal Atud,
Eastern Desert, Egypt. MSc thesis, Assiut University, Assiut,
Egypt, 235.
El-Habaak, G. H. & Mahmoud, M. S. 1995. Carbonaceous bodies of
debatable organic provenance in the banded iron formation of the
Wadi Kareim area, Eastern Desert, Egypt. Journal of African Earth
Sciences, 19, 125– 133.
Fairchild, I. J. & Kennedy, M. J. 2007. Neoproterozoic glaciation in the
Earth System. Journal of the Geological Society, London, 164,
895– 921.
Flint, R. F., Sanders, J. E. & Rodgers, J. 1960. Diamictite, a substitute
term for symmictite. Geological Society of America Bulletin, 71,
1809– 1810.
German, C. R., Holiday, B. P. & Elderfield, H. 1991. Redox cycling of
rare earth elements in the suboxic zone of the Black Sea. Geochimica
et Cosmochimica Acta, 55, 3553–3558.
Goldring, D. C. 1990. Banded iron formation of Wadi Sawawin district,
Kingdom of Saudi Arabia. Transactions of the Institution of Mining
and Metallurgy (Sect. B: Applications for Earth Science), 99,
B1 –B14.
Halverson, G. P., Hoffman, P. F., Schrag, D. P., Maloof, A. C. &
Rice, A. H. N. 2005. Toward a Neoproterozoic composite carbon-
isotope record. Geological Society of America Bulletin, 117,
1181– 1207.
Halverson, G. P., Dudás, F. Ö., Maloof, A. C. & Bowring, S. A. 2007.
Evolution of the 87Sr/86Sr composition of Neoproterozoic seawater.
Palaeogeography, Palaeoclimatology, Palaeoecology, 256, 103–129.
Hargrove III, U. S., Stern, R. J., Griffin, W. R., Johnson, P. R. &
Abdelsalam, M. G. 2006. From Island Arc to Craton: Timescales
of Crustal Formation along the Neoproterozoic Bi’r Umq Suture
Zone, Kingdom of Saudi Arabia, Saudi Geological Survey.
Hedge, C. E. 1984. Precambrian Geochronology of part of northwestern
Saudi Arabia, Kingdom of Saudi Arabia, U.S. Geological Survey.
Hoffman, P. F. 2005. On Cryogenian (Neoproterozoic) ice-sheet
dynamics and the limitations of the glacial sedimentary record.
28th DeBeers Alex. Du Toit Memorial Lecture. South African
Journal of Geology, 108, 557– 576.
Johnson, P. R. 2003. Post-amalgamation basins of the NE Arabian shield
and implications for Neoproterozoic III tectonism in the northern
East African orogen. Precambrian Research, 123, 321– 338.
Johnson, P. R. & Woldehaimanot, B. 2003. Development of the
Arabian –Nubian Shield: perspectives on accretion and deformation
in the East African Orogen and the assembly of Gondwana. In:
Yoshida, M., Windley, B. F. & Dasgupta, S. (eds) Proterozoic
East Gondwana: Supercontinent Assembly and Breakup. Geological
Society, London, Special Publications, 206, 289–325.
Johnson, P. R., Abdelsalam, M. G. & Stern, R. J. 2003. The Bi’r Umq-
Nakasib suture zone in the Arabian-Nubian shield: a key to under-
standing crustal growth in the East African Orogen. Gondwana
Research, 6, 523–530.
Kennedy, A., Johnson, P. R. & Kattan, F. H. 2004. SHRIMP geochro-
nology in the northern Arabian Shield. Part I: Data acquisition. Saudi
Geological Survey Open File Report, SGS-OF-2004-11.
Kennedy, A., Johnson, P. R. & Kattan, F. H. 2005. SHRIMP geochro-
nology in the northern Arabian Shield. Part II: Data acquisition 2004.
Saudi Geological Survey Open File Report, SGS-OF-2005-10.
Kennedy, A., Kozdroj, W., Johnson, P. R. & Kattan, F. H. 2011.
SHRIMP geochronology in the northern Arabian Shield. Part III.
Data Acquisition, 2006. Saudi Geological Survey, Open-File
Report SGS-OF-2007-9.
Lowther, J. M. 1994. Mahd adh Dhahab gold deposit. In: Collenette,
P. & Grainger, D. J. (eds) Mineral Resources of Saudi Arabia. Saudi
Arabian Directorate General of Mineral Resources, Special Publi-
cation SP-2, Jeddah, 105– 111.
Ludwig, K. R. 2000. SQUID 1.00. A User’s Manual, Berkeley Geochro-
nology Center, Berkeley.
Miller, N. R., Alene, M., Sacchi, R., Stern, R., Conti, A., Kröner, A.
& Zuppi, G. 2003. Significance of the Tambien group (Tigrai,
N. Ethiopia) for Snowball Earth events in the Arabian– Nubian
Shield. Precambrian Research, 121, 263– 276.
Miller, N. R., Avigad, D., Stern, R. J. & Beyth, M. 2011. The Tambien
Group, Northern Ethiopia (Tigre). In: Arnaud, E., Halverson, G. P.
& Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic
Glaciations. Geological Society, London, Memoirs, 36, 263–276.
Mukherjee, S. K. 2008. Petrography, age (uranium-lead zircon), geo-
chemical and isotopic studies of the Sawawin Banded Iron Formation
(BIF), Northwestern Saudi Arabia: Implications for understanding
Neoproterozoic climate change. PhD thesis, The University of
Texas at Dallas, 147.
Reischmann, T., Bachtadse, V., Kröner, A. & Layer, P. 1992. Geo-
chronology and palaeomagnetism of a Late Proterozoic island arc
terrane from the Red Sea Hills, northeast Sudan. Earth and Planetary
Science Letters, 114, 1– 15.
Ries, A. C., Shackleton, R. M., Graham, R. H. & Fitches, W. R. 1983.
Pan-African structures, ophiolites and mélange in the Eastern Desert
of Egypt: a traverse at 268N. Journal of the Geological Society of
London, 140, 75 –95.
Rieu, R., Allen, P. A., Cozzi, A., Kosler, J. & Bussy, F. 2007. A com-
posite stratigraphy for the Neoproterozoic Huqf Supergroup of
Oman: integrating new litho-, chemo-, and chronostratigraphic data
of the Mirbat area, southern Oman. Journal of the Geological
Society, London, 164, 997–1009.
Sims, P. K. & James, H. L. 1984. Banded iron-formations of late Protero-
zoic age in the central Eastern Desert, Egypt; geology and tectonic
setting. Economic Geology, 79, 1777– 1784.
Stern, R. J. 1981. Petrogenesis and tectonic setting of late Precambrian
ensimatic volcanic rocks, Central Eastern Desert of Egypt. Precam-
brian Research, 16, 195–230.
Stern, R. J. 1994. Arc assembly and continental collision in the Neopro-
terozoic East African orogen: implications for the consolidation of
Gondwanaland. Annual Reviews of Earth and Planetary Sciences,
22, 319– 351.
Stern, R. J. 2008. Neoproterozoic crustal growth: the solid Earth system
during a critical time of Earth history. Gondwana Research, 14,
33 – 50.
Stern, R. J. & Abdelsalam, M. G. 1998. Formation of continental
crust in the Arabian –Nubian shield: evidence from granitic rocks
of the Nakasib suture, NE Sudan. Geologische Rundschau, 87,
150– 160.
Stern, R. J., Avigad, D., Miller, N. R. & Beyth, M. 2006. Evidence
for the snowball Earth hypothesis in the Arabian –Nubian Shield
and the East African Orogen. Journal of African Earth Sciences,
44, 1 – 20.
Stoeser, D. B. & Stacey, J. S. 1988. Evolution, U– Pb geochronology,
and isotope geology of the Pan-African Nabitah orogenic belt of
the Saudi Arabian Shield. In: El-Gaby, S. & Greiling, R. O. (eds)
The Pan-African Belt of NE Africa and Adjacent Areas. Friedr.
Vieweg & Sohn, Braunschweig, 227– 289.
Sultan, M., Arvidson, R. E., Duncan, I. J., Stern, R. J. & Kaliouby,
B. E. 1988. Extension of the Najd shear system from Saudi Arabia to
the central Eastern Desert of Egypt based on integrated field and
Landsat observations. Tectonics, 7, 1291–1306.
Webb, G. E. & Kamber, B. S. 2000. Rare earth elements in Holocene
reefal microbialites; a new shallow seawater proxy. Geochimica et
Cosmochimica Acta, 64, 1557– 1565.
 Chapter 23

Glacial deposits of the Bokson Group, East Sayan Mountains, Buryatian Republic,
Russian Federation

NICKOLAY M. CHUMAKOV
Geological Institute of the Russian Academy of Sciences, Pyzhevsky per. 7, Moscow 119017, Russia (e-mail: chumakov@ginras.ru)

Abstract: The Bokson Group (Gr.) forms the platform cover in northern part of the Tuva-Mongolian Massif, south of the East Sayan
Mountains. Widespread diamictites occur in the lower part of the Bokson Gr., in the Zabit Formation (Fm.). These contain erratic, faceted
and striated stones along with dropstones. They have disconformable or gradual lower boundaries and conformable or transitional upper
boundaries with overlying deposits. The isotopic d13Ccrb curve records some maximum and minimum values as a background of insig-
nificantly varying values (from – 3 to 2‰). Maximum values (up to 5.5‰), are recorded in deposits underlying diamictites, whereas the
overlying beds of the Zabit Formation (Fm.) yielded three peaks or minima down to 24‰. A fourth negative anomaly derived from the
middle part of the Bokson Gr. in the lower part (close to base) of the Tabinzurt Fm. is as low as – 5.7‰. The deposits underlying the
diamictites contain Ediacaran (Vendian) microfossils. Cloudina sp. was discovered in the granule of diamictites, whereas beds just
above include small shelly fossils (SSF) of the Nemakit-Daldynian type (Cambrotubulus decurvatus Miss., Anabarites trisulcatus
Miss.), which is overlain by a member with abundant fossils of the Tommotian assemblage. The diamictites represent glaciomarine
deposits formed in the uppermost Ediacaran or lowermost Cambrian stage (Upper Vendian).

The Bokson Gr. occurs in the Oka River basin, in the southern part
of the East Sayan Mountains, Buryatian Republic of the Russian
Federation. Intensive investigation of the region began in the mid-
twentieth century in search of bauxite (Il’ina 1958) and phosphor-
ite (Volkov et al. 1972), with detailed stratigraphic work published
later (e.g. Khomentovskiy et al. 1985; Terleev & Zadorozhnyi
1996; and others). Reconstruction of sedimentation history
(Kheraskova & Samygin 1992) with specific emphasis on the
Sarkhoy Gr.
nature of the diamictites (Osokin & Tyzhinov 1998) was made
later. Initially the Bokson Gr. was defined as a formation but
further studies resulted in dividing it into five, sometimes six, for-
mations. Diamictites are usually described from the lower part of
the Zabit Fm. (Kheraskova & Samygin 1992; Pokrovskiy et al.
1999; and others) or from the basal, mainly terrigenous, Khushatay
Fm. (Osokin & Tyzhinov 1998; Kuzmichev 2004). The type
section of the diamictic member of the Zabit Fm. occurs in the
Bokson River basin (Fig. 23.1, I) and that of the Khushatay Fm.
in the Sarkhoy River basin (Fig. 23.1, II).
Structural framework
The Bokson Gr. forms a platform cover of the Tuva-Mongolian
Massif, unconformably lying above the Tonian-Cryogenian?
(Upper Riphean) folded basement composed of deposits shed in
an island arc and active continental margin settings. During the
early Palaeozoic Caledonian folding, the Bokson Gr. experienced
significant deformation, which makes correlation of units between
different sites difficult: deformation includes folding, disjunctive
faulting and overthrusting (Terleev & Zadorozhnyi 1996; Fedo-
tova & Khain 2002). In the north, the deposits were metamor-
phosed to chlorite-sericite phase, less frequently than quartz-
biotite shales (Kheraskova & Samygin 1992; Kuzmichev 2004).
The Bokson Gr. fills the Bokson-Sarkhoy Synclinorium in the
northern part of the Tuva-Mongolian Massif (Fig. 23.1). The
lower part of the group is of flyschoid type and the upper part of
the group consists of typical carbonate platform sediments.
Stratigraphy
Diamictites form the lower part of the Bokson Gr. and uncon-
formably overlie Tonian-Cryogenian? (Upper Riphean) ophiolites,
diabases, turbidites and conglomeratic breccias of the Dzhund-
zhugur Fm. or the sedimentary-volcanogenic Sarkhoy Gr. whose
volcanic rocks have yielded Rb – Sr ages of 718 + 30 Ma (ero-
chrona, Buyakayte et al. 1989). Both thickness and composition
of the lower part of Bokson Gr. vary noticeably laterally. In the
type section the lowermost Bokson Gr. is the Zabit Fm. A compo-
site section of the lower part of the Bokson Gr. at the type section
consists of the following from the base upwards (Kheraskova &
Samygin 1992; Osokin & Tyzhinov 1998; Kuzmichev 2004):
Dzhundzhugur Fm., Sarkhoy Gr.
(1) Deformed basic volcanogenic rocks, dark-coloured clayey and
siliceous shales, diabases.
(2) Weathering crust (a few metres).
Angular unconformity
The Boxon.
The Zabit Fm.
(3) Boulder and large pebbled conglomerates composed mainly
of fragments of underlying deposits, less frequently granites
(3 –40 m).
(4) Diamictites with a mudstone matrix with scattered stones of
different size, composition and roundness (0 –100 m).
(5) Grey and dark grey dolomites, often clastic and oncolitic with
abundant sedimentary structures indicative of subaqueous
slumping (200 –225 m).
(6) Breccia consisting of both angular and rounded fragments of
dolomite and carbonate matrix; frequent structures indicative
of underwater slumping (5– 15 m).
(7) Massive, locally sandy, grey and dark grey dolomites (15 m).
(8) Yellow-grey dolomitic phosphorites and phosphatic dolomites
(30 –35 m).
(9) Dark grey dolomites, partly brecciated and stromatolitic
(140 –200 m).
Hiatus and palaeokarst
The Tubinzurta Fm.
(10) Siltstone member containing a bauxite bed.
(11) Variegated and black siltstones, breccia-like dolomites.
(12) Grey and black silicified dolomites.
Hiatus and palaeokarst
The Khuzhirtai Fm.
(13) Dolomites passing into dark grey limestones with archae-
otheans, trilobites and brachiopods characteristic of the Atda-
banian Stage of the Lower Cambrian (.500 m).

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 285– 288. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.23
286
Irkutsk
52o
Baikal L.
0 100 km
Khubsugul L.
(a)
100 o o Ediacaran (Upper Vendian) to Middle
104
1 5
2 6
3 7
4 8 0 10
(b)
In the western part of the Bokson-Sarkhoy Synclinorium, along the
Sarkhoy River (Fig. 23.1, II), the Zabit Fm. is separated from the
Cryogenian basement by a thick (up to 500 m) red clastic
sequence, the Khushatay Fm. of the Bokson Gr. Some researchers
(Osokin & Tyzhinov 1998; Kuzmichev 2004) reported the pres-
ence of diamictites at the base and top of Khushatay Fm. In the
middle part of the Khushatay Fm., black clayey limestones and
shales up to 150 m thick occur (Kheraskova & Samygin 1992).
Sections of the Khushatay Fm change significantly over short dis-
tances. It is likely, that the basal diamictite member of the Zabit
Fm. is coeval to the upper part of the Khushatay Fm.
Glaciogenic deposits and associated strata
The diamictites of the Zabit and the Khushatay Fm. have a wide
lateral distribution (more than 130  360 km2, taking into
account the presence of Bokson Gr. equivalents in the northern
Mongolian Khubsugul area). Lithology of the diamictite
members of the Bokson Gr. was briefly described by Osokin &
Tyzhinov (1998) and Kheraskov & Samygin (1992). These
diamictites are massive, but sometimes crudely bedded. Thickness
of the massive beds varies from 20 to 55 m. These are composed of
greenish and cherry-coloured shale or carbonate shale matrix with
scattered pebbles, and small and large boulders derived from rocks
underlying the Bokson Gr.: sandstones, quartzites, jaspers, diabases
and gabbro, as well as pebbles and boulders of sandstones and dolo-
mites containing stromatolites or fossils of the algae Renalcis.
Lenses of dolomitic breccias occur locally. Dolomitic boulders
and breccias are confined to the sections where the Zabit Fm. is
separated from the Cryogenian basement by the Khushatay Fm.,
which includes dolomites. The diamictites contain erratic stones
similar to the granitoids of pre-Mesoproterozoic crystalline base-
ment that crop out on the Gargan Massif to the east. Semi-angular
and poorly rounded stones dominate over rounded ones. Some clast
are of flat-iron shape, with flattened and striated faces. Stones are
dispersed irregularly and sometimes form ‘stone nests’. The dia-
mictites include rare conglomerate interbeds and sandstone lenses.
The massive diamictites are associated with lenses and sub-
ordinate beds of thin and rhythmically bedded shales and mud-
stones containing scattered dropstones, which press through or
break underlying beds and are themselves in turn draped by over-
lying beds. Dolomites, which overlie the diamictites, are grey or
black, massive or brecciated, and sometimes stromatolitic.
Boundary relations with overlying and underlying units
In most sections, the diamictites and basal conglomerates uncon-
formably overlie the Sarkhoy Gr. (Osokin & Tyzhinov 1998). In
the Bokson River basin their contact is marked by a weathering
N. M. CHUMAKOV
o
100
Fig. 23.1. Geographic position of the
Bokson Gr. and its main sections. (a) Map
showing general location of study area
(box). (b) Geological map of described area.
1, granitoides; 2, Middle Cambrian–
Ordovician; 3, the Bokson Group, latest
52 o Cambrian; 4, Tonian-Cryogenian? (Upper
Riphean); 5, pre-Mesoproterozoic (pre-
Riphean) crystalline basement; 6, main
thrusts; 7, main faults; 8, boundary of the
Russian Federation and Mongolia; I,
20 km Bokson River Basin; II, Sarkhoy River
Basin.
horizon (Kuzmichev 2004). In the Sorok River basin and else-
where, the boundary between the underlying deposits and the dia-
mictites is gradational (Osokin & Tyzhinov 1998). The upper
boundary of the diamictite member with overlying dolomites is
sharp but conformable and locally this boundary is gradational
(Osokin & Tyzhinov 1998; Kuzmichev 2004).
Chemostratigraphy
Isotopic analyses (oxygen and carbon) of the Bokson Gr. carbonate
rocks were carried out by B.G. Pokrovskiy and coauthors
(Pokrovskiy et al. 1999). Results obtained are illustrated in
Figure 23.2. Carbonate samples were dissolved in H2PO4 at
25 8C for 1.5 h in the case of the calcitic fraction; dissolution
was made at 100 8C for 1 h for the dolomitic fraction. A correction
to isotopic oxygen values was introduced on different fractionation
of calcite and dolomite (þ1.1‰ for the latter).
The d13C range determined for calcite varied from 24.4 to
24.9‰ in beds that underlie the diamictites; from 22.9 to 1.8‰
for the Zabit Fm. units that overlie the diamictites; from – 4 to
1.5‰ for the Tabinzurt Fm.; from –2.1 to 0.3‰ for the Khuzhirtay
Fm.; and from –2.1 to 0.7‰ for the lower part of the Nyurga Fm.
The isotopic d13C curve demonstrates distinct positive values cor-
responding to the beds under the diamictites: three negative peaks
for the lower carbonate part of the Zabit Fm.; an interval of mainly
positive low values for the upper Zabit – lower Tabinzurt Fm. beds,
a sharp drop in value for the Tabinzurt beds that overlie the bauxite
member; change with some oscillations to almost zero for the
uppermost Tabinzurt Fm.; and low negative values for the
Khuzhirtay Fm.
Other characteristics
The member of dolomites with interbeds of microcrystalline dolo-
mitic phosphorite occurs in the upper part of the Zabit Fm. in the
Bokson area. The thickness of this member varies from 35 to
50 m. Total thickness of phosphorite beds is on average 5.5 m.
Average contents of P2O5 in this phosphorite is 8.3% (Yanshin
& Zharkov 1986).
The basal bauxite member of the Tabinzurt Fm. overlies a karst
surface of the Zabit Fm. in the Bokson area (Kuzmichev 2004).
This member is 4– 23 m thick and consists of massive, brecciated
and laminated reddish bauxite consisting of diaspore, boehmite
and a higher proportion of Fe2O3 and SiO2 (Il’ina 1958).
Palaeolatitude and palaeogeography
No valid results of palaeomagnetic investigations have been
obtained and published for the Bokson Gr. and adjacent
 BOKSON GROUP, EAST SAYAN MOUNTAINS
deposits. Following regional palaeotectonic reconstructions for
the late Cryogenian and Ediacaran (Early and Late Vendian),
the Tuva-Mongolian Massif and its constituents seems to be
in low latitudes of around 08 (Fedotova & Khain 2002) and
2308 (Smith 2001) respectively.
Geochronological constraints
The Rb –Sr age of volcanic rocks in the Sorkhoy Gr. unconform-
ably underlying the Bokson Gr. has been estimated to be
718 + 30 Ma (erochrona, Buyakayte et al. 1989). The basal
Khushatay Fm. of the Bokson Gr. has yielded the acritarch Gran-
omarginatasphaeria judomica Pjat. and multiserial sheaths of
organic-walled microfossils Polytrichoides lineatus Hermann,
suggestive of an Ediacaran (Vendian) age (Veis & Vorob’eva
1993). Cloudina sp. (attributed by A. Y. Zhuravlev) was recovered
from granule of the diamictites (Kheraskova & Samygin 1992),
and the SSFs Cambrotubulus decurvatus Miss. and Anabarites tri-
sulcatus Miss., characteristic of the Nemakit-Daldynian (Many-
kaian) Stage of Siberia, were recovered 30 m above the
diamictitic member (Terleev & Zadorozhnyi 1996; Postnikov &
Terleev 2004). Higher up the section, the phosphoritic member
of the Zabit Fm. contains abundant assemblages of SSF (Cambro-
tubulus decurvatus Miss., Anabarites trisulcatus Miss., Tisitheca
sp., Igorella sp., Fomitchella infundibuliformis Miss.), silicified
algae (Obruchevella magna Golov., O. parva Reitl., Heliconema
sp., Eosynechococcus meorei Hoffm. and other forms), radio-
larians (Palaeocenosphaera parva Nazarov, Entasitinia sp.) and
other fossils characteristic of the Lower Cambrian Tommotian
stage (Postnikov & Terleev 2004).
287
Fig. 23.2. Lithological and
chemostratigraphic succession of the lower
part of Bokson Gr. (from Pokrovskiy et al.
1999, with additions). 1, limestones; 2,
massive dolostones; 3, stromatolitic
dolostones; 4, dark dolostones, calcarenites
and calcirudites; 5, phosphatic calcarenites;
6, limey shales; 7, bauxites; 8, argillites and
siltstones; 9, diamictites; 10, eluvial debris;
11, silicification; 12a, calcite; 12b,
dolomite; ]1, Lower Cambrian; V2,
Ediacaran-Cambrian (Upper Vendian); R3,
Cryogenian (Upper Riphean) (Sorkhoy
Gr.); SS, Nemakit-Daldyn SSF (small
shelly fossils); Cl, Cloudina;
Mf, microfossils.
Discussion
Sedimentary environments
A glacial origin of the Zabit diamictites is suggested by the wide
distribution of massive and bedded diamictites at the same strati-
graphic interval, the presence of erratic and ‘flat iron’ shaped
stones with striated faces and dropstones in thin-bedded rocks
(Osokin & Tyzhinov 1998). Osokin & Tyzhinov (1998) proposed
that the diamictites originated in glacial terrestrial, lacustrine and
shelf environments. The opinion of these and other authors
(Kuzmichev 2004 and others) about the glacial origin for the dia-
mictites seems correct. However, the presence of continental
glacial deposits in the Bokson Gr. is doubtful. There are no signs
of glaciodislocations, zones of assimilation or glacial pavements
at the base of massive diamictites. The preferred orientation of
elongated stones typically found in basal tillites does not occur
in the Bokson diamictites. On the contrary, the stones are oriented
in a random fashion and often form ‘stone nests’. Gradual
transitions between alternating diamictites and thin-bedded sedi-
ments and numerous indicators of underwater slumping suggest
that the Bokson diamictites are glaciomarine deposits, partly
reworked by slumps and debris flows, as noted by Kheraskova &
Samygin (1992).
There are other interpretations of the Bokson Gr. diamictites.
Some researchers suggest that these diamictites represent sedimen-
tary olistostromes formed by reworking of Sarkhoy Gr. deposits
(Khomentovskiy et al. 1985). Other geologists consider the dia-
mictites to be tectonic olistostromes formed in front of the nappe
where the Bokson Gr. overthrusts Palaeozoic deposits (Terleev
& Zadorozhnyi 1996; Fedotova & Khain 2002; and others). This
288 N. M. CHUMAKOV
tectonic interpretation contradicts the fact that the Bokson Group
has a gradational contact with the underlying beds in some areas,
convincingly demonstrated by regional mapping (Khomentovskiy
et al. 1985; Kheraskova & Samygin 1992; Kuzmichev 2004).
Another argument against tectonic olistostromes is the fact that
the carbonate stones in the diamictites have high positive d13C,
similar only to carbonate rocks below the diamictites. Carbonate
rocks of the Bokson Gr. that overlie the diamictites have only
negative d13C values.
Osokin & Tyzhinov (1998) suggested a glacial origin for the
carbonate breccia of the Zabit Fm. (see bed 6 in stratigraphy
section described above). They base this on scattered carbonate
gravel and small pebbles that occur in some lenses of rhythmi-
cally thin-bedded limestones and dolomites related with the brec-
cias. The description of the composite type section presented
above, notes that the breccias and enclosing deposits bear abun-
dant traces of underwater slumping, suggestive that breccias were
most likely products of slides and slumps on slopes of the car-
bonate platform. The rhythmically and thin bedded limestones
and dolomites appear to represent sediments deposited from
related turbidity currents. A non-glacial origin is supported by
the absence of any foreign clasts in the breccias and
thin-bedded rocks.
Age
In terms of the framework proposed by the International Strati-
graphic Committee, the diamictites of Bokson Fm. can be
referred to the uppermost part of the Ediacaran or the lowermost
part of the Lower Cambrian (Nemakit-Daldyn Stage). In terms of
the Russian stratigraphic scale, diamictites of the Bokson Fm.
can be referred to the uppermost part of the Kotlin or the lower-
most part of the Nemakit-Daldyn Stages of the Vendian. The dia-
mictites of the Zabit Fm. occur above the Khushatai Fm., which
contains microfossils, similar to those of the latest Ediacaran
(Late Vendian) (Veis & Vorob’eva 1993). The diamictites contain
Cloudina sp. (Kheraskova & Samygin 1992) and lie below beds
containing Cambrotubulus decurvatus Miss. and Anabarites tri-
sulcatus Miss. (Terleev & Zadorozhnyi 1996; Postnikov &
Terleev 2004), indicative of a Nemakit-Daldynian age. These
data together with the Tommotian fossils present higher in the
Zabit Fm. allow the diamictites to be referred to the latest Edia-
caran (Late Vendian) (Kuzmichev 2004), and more specifically,
either a slightly antedated Nemakit-Daldyn age or concurrent
to its commencement (Chumakov 2009a, b). The Cloudina in
the diamictites was likely redeposited from underlying succes-
sion. The lowermost negative d13C anomaly in the Zabit Fm.
can be correlated to the negative anomaly of d13C at the base
of the Nemakit-Daldyn stage (Pokrovskiy & Missarzhevskiy
1993). On an interregional scale, it is possible to correlate the
Bokson diamictites to the Baykonur Glacial Horizon in Kazakh-
stan and Kyrgyzstan (Chumakov 1981, 2011) and to the Hankal-
chough and Luoguan formations of northern China (Baode et al.
1986), based on their similar stratigraphic position.
Research projects were supported by grants of the Russian Fund of Basic Inves-
tigations No. 11-05-00232 and Program No. 25 of the Presidium of the Russian
Academy of the Sciences. The author would like to acknowledge stimulating
discussions arising from participation in IGCP project 512 ‘Neoproterozoic Ice
Ages’ (http://groups.google.com/IGCP-512?nl¼enIGCP-512@?hl¼en). This
represents a contribution of the IUGS- and UNESCO-funded IGCP (International
Geoscience Programme) Project #512.
References
Buyakayte, M. I., Kuzmichev, A. B. & Sokolov, D. D. 1989. 718
million year – Rb –Sr erochrona Sarkhoy Group of East Sayana.
Doklady Academy of Sciences of USSR, 309, 150–154.
Chumakov, N. M. 1981. Upper Proterozoic glaciogenic rocks and their
stratigraphic significance. Precambrian Research, 15, 373–396.
Chumakov, N. M. 2009a. The Baykonurian Glaciohorizon of the Late
Vendian. Stratigraphy and Geological Correlation, 17, 373–381.
Chumakov, N. M. 2009b. Neoproterozoic Glacial Events in Eurasia. In:
Gaucher, C., Sial, A. N., Halverson, G. P. & Frimmel, H. E.
(eds) Neoproterozoic-Cambrian Tectonics, Global Change and
Evolution: a focus on southwestern Gondwana. Developments in
Precambrian Geology, 16, Elsevier, The Netherlands, 389– 403.
Chumakov, N. M. 2011. Glacial deposits of the Baykonur Formation,
Kazakhstan and Kyrgyzstan. In: Arnaud, E., Halverson, G. P. &
Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic
Glaciations. Geological Society, London, Memoirs, 36, 303–307.
Fedotova, A. A. & Khain, E. V. 2002. Tectonics of the south East Sayan
and its position in the Ural– Mongolian Belt (in Russian). Nauchnyi
Mir, Moscow.
Guan Baode, Wu Ruitang, Hambrey, M. J. & Geng Wu. 1986. Glacial
sediments and erosional pavement near the Cambrian – Precambrian
boundary in western Henan Province, China. Journal of Geological
Society, London, 143, 311–323.
Il’ina, A. V. 1958. Geology and origin of Boxon bauxites in East Sayan.
In: Dolgopolov, N. N. (ed.) Bauxites their Mineralogenesis Origin
(in Russian). Academy of Sciences of USSR Press, Moscow, 267–281.
Kheraskova, T. N. & Samygin, S. G. 1992. Tectonic environments of
formation of Vendian-Middle Cambrian clastic-carbonates sequence
of the East Sayan. Geotectonics, 6, 18 –36.
Khomentovskiy, V. V., Pak, K. L., Postnikov, A. A. & Skopintsev, V. G.
1985. Geology structure of the basin of Ukha-Gol River (in Russian). In:
Khomentovskiy, V. V. (ed.) Stratigraphy of the Late Precambrian and
Lower Palaeozoic of Siberia. Riphean and Vendian. Institute of the
Geology and Geophysics, Novosibirsk, 76–106.
Kuzmichev, A. B. 2004. Tectonic History of the Tuvino-Mongolian
Massif (in Russian). Probel-2000, Moscow.
Osokin, P. V. & Tyzhinov, A. V. 1998. Precambrian tilloids Oka-
Khubsugul phosphorite basin (East Sayan, North-West Mongolia).
Lithology and Mineral Resources, 2, 162–176.
Pokrovskiy, B. G. & Missarzhevskiy, V. V. 1993. Isotopic correlations
boundary sections of Precambrian and Cambrian of Siberian
Platform. Doklady Academii Nayk SSSR, 329, 768– 771.
Pokrovskiy, B. G., Letnikova, E. F. & Samygin, S. G. 1999. Isotopic
stratigraphy of Boxon Group, Vendian-Cambrian of East Sayan.
Stratigraphy and Geological Correlation, 7, 23 –41.
Postnikov, A. A. & Terleev, A. A. 2004. Stratigraphy of Neoproterozoic
of the Altay-Sayan folded region. Geology and Geophysics, 45,
295–309.
Smith, A. G. 2001. Palaeomagnetically and tectonically based global
maps for Vendian to Mid-Ordovician time. In: Zuravlev, A. Yu &
Riding, R. (eds) The Ecology of the Cambrian Radiation. Columbia
University Press, New York, 11– 46.
Terleev, A. A. & Zadorozhnyi, V. M. 1996. Discovery of Palaeozoic
Foraminifera in ‘Precambrian’ of East Sayan (Sarkhoy River).
Doklady Academy of Sciences of Russia, 351, 373–374.
Veis, A. F. & Vorob’eva, N. G. 1993. First finds wall-organic microfossils
in Upper Precambrian of Boxon-Sarkhoy basin (East Sayan). Strati-
graphy and Geological Correlation, 1, 27 –32.
Volkov, R. I., Zaytsev, N. S., Il’in, A. V. & Osokin, P. V. 1972.
Ukhangol phosphorite field of East Sayan. Soviet Geology, 2,
94 – 107.
Yanshin, A. L. & Zharkov, M. A. 1986. Phosphorus and Potassium in
Nature (in Russian). Nauka, Novosibirsk.
 Chapter 24

The Neoproterozoic glacial formations of the North and Middle Urals

NICKOLAY M. CHUMAKOV
Geological Institute Russian Academy of Sciences, Pyzhevskiy per. 7, Moscow 109017, Russia (e-mail: chumakov@ginras.ru)

Abstract: In the North and Middle Urals, four Neoproterozoic diamictite-dominating units are known: the Churochnaya, Tany and
Koyva formations and Lower Starye Pechi Subformation. A glacial origin for the diamictites is indicated by a number of characteristic
features such as erratic stones, striated and faceted clasts, shales with dropstones, preferential orientation of elongated stones, the wide
distribution and confinement to certain stratigraphic levels of the deposits, and the association with distinct post-glacial dolostones. Most
diamictites were deposited in a glaciomarine environment and sourced from an ice sheet on the East European craton. The Churochnaya
Formation also contains subordinate terrestrial and probably seasonal sea ice deposits. This formation accumulated on a glaciated con-
tinental shelf, while the Tany, Koyva and Starye Pechi formations were deposited on the outer shelf and continental slope of the eastern
margin of the East European Craton. The Middle Ural sections contain the White Sea (Ediacaran) Metazoa assemblage, typical late
Neoproterozoic microfossils, stromatolite associations, and radiometric dates obtained from volcanic tuffs, granosyenites, trachytes
and trachyandesites. The combined data suggest that the Lower Starye Pechi Subformation was deposited in the middle of the Ediacaran
Period (Early Vendian). The Churochnaya Formation of the North Urals and the Tany and Koyva formations of the Middle Urals were
likely deposited in the late Cryogenian period (Early Vendian).

Neoproterozoic glacial deposits are known in the western parts of
the North, Middle and South Urals. They are best exposed and
studied along the Pulyudov Ridge and Churochnaya River in the
northern Urals and along tributaries of the Chusovaya River in
the Middle Urals (Fig. 24.1). This chapter focuses on glacial
deposits of the North and Middle Urals; glacial deposits of the
South Urals were recently described and are reviewed elsewhere
(Chumakov 1981a, b, c, 1992, 1998; Maslov 2000).
Until the mid-1960s, the late Precambrian strata of the North and
Middle Urals were regarded as Palaeozoic in age and the occur-
rence of associated ‘tillite-like conglomerates’ (diamictites) was
explained by subaqueous slumping. This view was reflected in
many publications and geological maps of the Urals and USSR
territory. Long regional investigations resulted in the elaboration
of the late Precambrian timescale as reviewed by Ablizin et al.
(1982). Modern stratigraphic investigations of Neoproterozoic
sedimentary sequences in the Urals, including glacial deposits,
began in the second half of the twentieth century during lithologi-
cal investigations (the Polyudov Ridge; Borovko 1967) and
geological mapping (the Middle Urals, Ablizin et al. 1982) that
was verified by Ehlakov & Morozov (2006).
Glacial deposits of the Polyudov Ridge in the North Urals are
confined to the Churochnaya Formation. Three glacial levels
were established in the Middle Urals: the Tany and Koyva
formations of the Serebryanka Group and in the Lower Starye
Pechi Subformation of the overlying Sylvitsa Group (Fig. 24.2).
The Tany and Koyva formations are replaced eastward by the
Wil’va Formation, which contains some diamictite beds. The
possible occurrence of diamictites in the Us’va, Fedotovka and
Kernos formations has not been confirmed by recent investi-
gations. The Us’va and Fedotovka shales locally contain single
lonestones.
Structural framework
The Neoproterozoic deposits of the Polyudov Ridge accumu-
lated in the marginal shelf zone of the Russian platform and
deformed during the Timanian and, more significantly, during
the Hercynian orogeny. Now they are exposed in the cores of
marginal brachyanticlines of the Uralian folded zone, disrupted
by many faults, and variably foliated. In the Middle Urals,
Neoproterozoic sedimentation occurred in the outer shelf and
continental slope environments. Sections of this region are
three times thicker than those of the Polyudov Ridge and
contain a higher percentage of siliciclastics in the form of turbi-
dites and shales (i.e. flysch). The Neoproterozoic deposits of the
middle Urals, along with younger deposits, were locally heavily
deformed during the Hercynian, with isoclinal folds and numer-
ous faults and overthrusts. The Tany, Koyva and Starye Pechi
formations were metamorphosed to lower greenschists facies
at this time. The Wil’va Formation was metamorphosed to
greenschist facies.
Stratigraphy
Neoproterozoic stratigraphy, lithologies and pertinent radiometric
ages of the northern and middle Urals under consideration are
schematically shown in Figure 24.2. The Polyudov Ridge section
begins with the thick (800 –900 m) siliciclastic Rassolny For-
mation (not shown in Fig. 24.2), which is overlain by pink lime-
stones and marls of the Deminskaya Formation. Above is the
Niz’va Formation, comprising mainly calcareous dolomites with
stromatolites similar to those of the late Mesoproterozoic –early
Neoproterozoic (Upper Riphean) Min’yar and Uk formations of
the South Urals (Raaben 2007). The Niz’va Formation is uncon-
formably overlain by the clastic Ust-Churochnaya Formation,
which, in turn, is unconformably overlain by the Churochnaya For-
mation. The latter consists of red sandstones in the lower half and
diamictites with an interval of dolomite and black shales in the
upper part. The black shales are overlain by a thick sandy-shaly
sequence, which is divided by an unconformity into the Il’ya-Vosh
and Kocheshor formations.
The late Precambrian succession of the Middle Urals is more
complicated. Sequences that are approximately coeval with those
of the Polyudov Ridge are considerably thicker, more lithologi-
cally diverse, and divisible into four groups, each separated by a
depositional hiatus. Each group is further subdivided into for-
mations as shown in Figure 24.2.
Glaciogenic deposits and associated strata
The Churochnaya Formation
This formation is 350–500 m thick. Its lower part (c. 150 m)
consists of red sandstones. The upper part of the formation

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 289– 296. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.24
290 N. M. CHUMAKOV

o o
56 58

R.
ra
Po

she
ly
d

Vi
ov
Moscow Perm

Ri
Krasnovishersk

dg
e
Ka
m
aR
.

o
60

Phanerozoic
Solikamck Sylvitsa Gr., Kocheshor,
Il’ay-Vozh Fms
Serebryanka Gr., Churochnzya,
Ust-Churochnaya Fms
Riphean

main disjunctive
dislocations
Main Neoproterozoic sections
.
aR Mezhevaya Utka R.
’v
Kos
. Serebryanka R.
va R
Us’ Sylvitsa R.
Koyva R.
Vil’va R.
Us’va R.
R. Churochnaya R.
ovaya
C hus
58 o
Perm

0 25 50 km Fig. 24.1. Late Mesoproterozoic to

Neoproterozoic (Vendian and Upper
Riphean deposits of the North and Middle
Urals).

(c. 200–350 m) is composed of massive and bedded green-grey
or reddish diamictites containing some beds of sandstones,
shales and conglomerates (Fig. 24.3). The diamictites are over-
lain by the dolomite member (10 m) and above it by black
shales (.20 m). The diamictites have a sand-silty matrix with
dispersed stones (2 –10% of the diamictite volume) of variable
size, roundness and composition. Some diamictites show indis-
tinct bedding, thin lenses and interbeds of sandstones and
shales, and nested clasts. In addition to the predominant sand-
stone and quartzite, there are many erratic stones, including
granites, gneisses and schists derived from the crystalline base-
ment of the Russian plate. The proportion of erratic stones
increases upsection. Between 3 and 7% of stones are striated,
grooved and faceted. Rare diamictites beds contain a large pro-
portion of well-rounded stones of medium size. Some diamic-
tites preserved a preferred orientation among elongated stones.
Rhythmically banded shales with dropstones occur at the base
of the diamictites.
The diamictites are overlain by a dolomite member (up to 10 m).
This dolostone is pale, slightly calcareous (1 –7% CaCO3), distinct
in structure and texture, and commonly massive or brecciated. In
places it is thinly bedded and distorted by subaqueous slumps.
Abundant carbonate, chert and quartz veins cut the dolomite
and encrust fragments of breccia. Thin, intercalated black chert
(2– 3 cm) overlies the dolostone. The chert is followed by
.20 m of black shale with a relatively high phosphorous content
(0.1 –1%). The black shales are overlain by multicolour shales
with sandstone intercalations.
Boundary relations with overlying and underlying non-glacial
units. Conglomerates at the base of the Churochnaya Formation
rest on a probable unconformity. Breccias in the upper part of
the cap dolomite are apparently related to weathering and karstifi-
cation. Nevertheless, the hiatus between the dolomite and over-
lying chert was probably not significant, as a 1.5-m-thick
diamictite bed occurs 12 m above the base of the black
shale member.
Chemostratigraphy. Dolomites and limestones of the Niz’va
Formation, which unconformably underlies the Churochnaya
Formation, are characterized by slightly variable but on
average low, positive d13C values ( –0.8 to þ2.5‰) and stable
d18O values ( –10 to – 6‰). Dolomites overlying the diamictites
of the Churochnaya Formation have stable negative d13C values
(from –3 to – 5‰).
 NEOPROTEROZOIC OF THE NORTH AND MIDDLE URALS 291

Polyudov Ridge Central Ural

O3 O3

. .. .. .. .. .
. .. .. .. .. . Ust’-Sylvitsa Fm. . .. .. .. .. . diamictite
. .. .. .. .. .
350 m .....
..... . .. .. .. .. . . .. .. .. .. .

Sylvitsa Gr.
Kocheshor Fm. Chernyi Kamen’ Fm. ..... ..... sandstone
500 m .. .. .. .. .. 557 ± 13 U-Pb, z 1700 m .. .. .. .. ..
..... shales
545 K-Ar, g Perevolok Fm.
570 K-Ar, g ..... 300 m dropstone
..... .....
..... Starye Pechi Fm. .....
.....
limestone
300 m
.....
Kernos Fm. ..... dolstone
Il’ay-Vozh Fm. 300 m .....
.....
480 m .. .. .. .. .. Buton Fm. volcanic rock
610; 620 K-Ar, g
Serebrayanka Gr.

400 m
cap dolomite
Koyva Fm.
Churochnaya Fm. 700 m
stromatolite
350 m 569 ± 42 Sm; 559 ± 16 Rb
. .. .. .. .. .
..... .....
Garevka Fm. .....
200-700 m .....

Tany Fm. .....

.....

800 m
.....
..... Us’va Fm. .....
Ust’- . .. .. .. .. .
1200 m . . . . .
Churochnaya Fm. .....
.....
..... Fedotovka Fm.
Basegi Gr.

720 m . . . . .
1200 m
635 K-Ar, g .....
. .. .. .. .. .
Shcherogrovitsy Fm.
0-900 m
671+24 Rb; 671+24 U-Pb, z
.....
Oslyanka Fm. .....
300 m .....
Kedrovka Gr.

Niz’va Fm. Klyktan Fm.

950 m Fig. 24.2. Stratigraphic position of Vendian
750 K-Ar, g 1300 m
..... glacials at the North and Middle Urals
Deminskaya Fm. .....
Sinii Gory Fm. ..... (Ablizin et al. 1982; Chumakov 1998;
300 m 1000 m ..... Maslov et al. 2007). Radiometric dating
techniques: g, glauconite; z, zircon.

Tany Formation Boundary relations with overlying and underlying non-glacial

The Tany Formation (up to 800 m thick) consists of two members
of massive dark grey diamictites divided by a sandstone member.
The base of the thick Lower Member (c. 350–450 m) consists of
alternating mafic volcanics, schists, limestones and diamictites
(Ablizin et al. 1982). The lower member of diamictites in some
section is capped by a dolomite bed 8 m thick (Fig. 24.4). The
lithology of diamictites clasts is diverse and includes quartz are-
ntites, granites, gneisses and carbonates. A considerable portion
(up to 45%) of the diamictite clasts are felsic igneous rocks. The
abundance and size of granites and gneiss clasts increase to
the SW (towards the Russian platform) and in the upper part of
the lower member (Ablizin et al. 1982). Some blocks of granite
and gneiss are up to 3.5 m in diameter. Nests of medium-sized
and large stones are common. Some rare shale interbeds contain
dropstones. The Middle Member consists of bedded quartz-
feldspathic sandstones (80– 150 m). The Upper Member includes
massive and bedded diamictites and thin intercalations of lami-
nated shales with dropstones (Chumakov 1992).
units. The lower boundary of the Tany Formation is probably
erosional judging from the abundance of clasts from the underlying
Kedrovka and Basegi groups within the basal diamictites.
The upper boundary of the Tany Formation is sharp and
conformable.
Koyva Formation
The Koyva Formation contains diamictites only to the north of
the Sylvitsa River. In general, diamictites are similar to those of
the Tany Formation but differ in their reddish colour and con-
finement of bedded diamictites and laminated shales with drop-
stones to the upper part of the formation. Father to the north, at
the Vil’va, Us’va and Kos’va rivers, the Koyva Formation con-
tains alkali basalts flows. In the stratotype section at the Koyva
River, diamictites are overlain by a transitional member of thin-
bedded variegated shales with small dropstones and then by the
dolomite member (up to 6 m thick; Fig. 24.5). Thin (1– 2 cm)
292 N. M. CHUMAKOV

Lower Starye Pechi Subformation

massive diamictites
In the Lower Starye Pechi Subformation, diamictites alternate with
sedimentary and quartzites stones
sandstones and are overlain by thin-bedded shales containing scat-
granitoid and gneisses stones tered sand grains and pebble dropstones (Fig. 24.6). The matrix of
the diamictites is dark grey sandy-silt, and the diamictites are
diamictites with well rounded stones bedded in the upper part of the formaton. Clasts in the diamictite
sandstones with pebbles vary in size and shape; they are dominated by fragments of the
underlying rocks, but some erratic pebbles of quartz, quartzites
sandstones and plagio-granites occur as well. Striated and grooved clasts
have been found. Diamictite wedges are found at the base of
shales
some of the diamictite beds.
dolomites
Boundary relations with overlying and underlying non-glacial
chert units. The contact between the Lower Starye Pechi Subformation
black rocks and the underlying Kernos Formation is generally assumed to be an
unconformity (Ablizin et al. 1982). Nevertheless, locally the lower
brown-red rocks Starye Pechi Formation contains interbedded sandstones, which
striated stones
are typical of the Kernos Formation, and thus imply a transitional
contact (Chumakov 1998). The upper boundary of the Starye Pechi
dropstones Subformation is transitional.
stone fabric
ch .. stone nests
.. Geochronological constraints
== vague bedding
rhythmic lamination Polyudov Ridge
==
subaqueous slump structure No precise radiometric data are available for the Neoproterozoic
formations of the Polyudov Ridge. Numerous radiometric ages
higher content of phosphorus
were obtained in the middle of last centenary by K –Ar dating
of glauconites (Garris et al. 1964; Borovko 1967; Keller &
Chumakov 1983). They show a rather wide range of ages. The
average K –Ar ages recalculated with recent 40K decay constants
== (l40Kb- ¼ 4.962  10210 a21; l40Ke- ¼ 0.581  10210 a21) are
30 750 + 25 Ma for the Niz’va Fm., 660 + 30 Ma for the Ust’-
Churochnaya Fm., 610 + 20 Ma for the Il’ay-Vozh Fm., and
== 570 + 17 Ma for the lower and 545 + 15 Ma for the upper part
.. of the Kocheshor Formation, respectively.
The Niz’va Fm. contains abundant stromatolites. Their assem-
blage consists of Poludia mutabilis Raab., Gymnosolen ramsayi
Steinm., Conophyton miloradovichi Raab., Poludia polimorpha
Raab., Minjaria uralica Kryl., Linella ukka Kryl., and other
0 m forms typical of the Cryogenian (uppermost Upper Riphean) of
the Middle and South Urals, the Timans and Svalbard (Raaben
2007). In general, K –Ar ages from the Niv’zva Formation are
Fig. 24.3. Representative stratigraphic section of the upper Ghuroghnaya consistent with the chronology implied by stromatolite
Formation, Ghurochnaya River. biostratigraphy.

interbeds of dolomites occur in rhythmically laminated shales Middle Urals

overlying the dolomite.
Boundary relations with overlying and underlying non-glacial uni-
ts. The lower and upper boundaries of the Koyva Formation are
transitional (Ablizin et al. 1982).
Chemostratigraphy. In the stratotype section at the Koyva River the
upper dolomite member yielded d13C values from –3 to –5‰.
South of the Koyva River, where there are no diamictites or
shales with dropstones within the Kyva Formation, the basal bed
of the upper dolomite member is 2 m thick and has d13C values
from – 4 to –7‰ (A.V. Maslov, pers. comm.).
Other characteristics. At the Us’va River, the lower part of the
Koyva Formation contains alkali basalts, tuffs and banded hemati-
tic rocks and ores, the latter of which vary from 10 to 40 m thick.
Abundant non-skeletal Metazoa (Cyclomedusa davdi Sprigg,
Dickinsonia tenuis Glaessner and Wade, Irridinitus multiradiatus
Fedonkin, Vaizitsinia sophia Sokolov and Fedonkin and other
forms) characteristic of the White Sea and Eriacaran biota occur
in the Chernyi Kamen’ Formation in the upper Sylvitsa Group
(Grazhdankin et al. 2005, 2007; Maslov et al. 2007). A U –Pb
(SHRIMP-II) zircon age of 557 + 13 Ma was obtained on a tuff
within the Chernyi Kamen’ Formation (Maslov et al. 2007). This
age provides a maximum constraint for all of the Neoproterozoic
diamictites of the Middle Urals.
More controversial dates were obtained for the individual
glacial formations of the succession using less reliable radiometric
techniques. The diamictites of the Starey Pechi Subformation seem
to be younger than 569 + 42 or 559 + 16 Ma based on Sm– Nd
(clinopyroxene) and Rb – Sr (whole rock) ages from the Koyva tra-
chyandesites (Petrov et al. 2005). The Sm –Nd age is imprecise,
whereas the Rb – Sr age is similar to an age of 557 + 13 Ma
 NEOPROTEROZOIC OF THE NORTH AND MIDDLE URALS
Mezhevaya Utka R. Serebryanka R. Us’va R.
Garevka Fm.
3
2
Tany Fm.
1
0m
50
100
150
Us’va Fm.
obtained by Maslov et al. from 1.5 km upsection (Sylvitsa
Group). Macrofossil associations lower in the Sylvitsa Group are
similar to the Miaohe biota of the uppermost Doushantou For-
mation of South China (Grazhdankin et al. 2007), implying a
late Ediacaran age.
It was long accepted that grano-syenites of the Troitsk massif
(the Kos’va River) cut the Tany and Garevka formations of the
Serebryanka Group and that they were emplaced between depo-
sition of the Serebryanka and Sylvitsa groups. This hypothesis
implies that the diamictites of the Tany Formation would be
older than 630 + 20 (Pb/Pb zircon) or 621 + 12 Ma (Rb –Sr
whole rock; Petrov et al. 2005). More recent mapping and
dating, however, suggest that the Troitsk grano-syenites cut only
the much older Shechegrovittsy Formation (Basegi Group) and
have an age of 671 + 24 Ma (SHRIMP U –Pb zircon)(Ronkin
et al. 2007). The differences between the ages may be related to
two phases of magmatic activity in the Troitsk massif. In any
case, the diamictites of the Koyva and Tany formations are
constrained to between 557 + 13 and 671 + 24 Ma, which is con-
sistent with biostratigraphic data.
Palaeolatitudes and palaeogeography
Palaeomagnetic investigations of the late Proterozoic rocks of the
South Urals have failed to reveal primary palaeomagnetic direc-
tions. According to palaeogeographic reconstructions by Li et al.
(2008), the eastern margin of the Russian Platform was located
293
dimictites,
crystalline clasts
dimictites,
sedimentary clasts
bedded diamictites
laminated shales
with dropstones
sandstones
shales
limestones
dolomites
striated stones
dropstones
rhythmic lamination
bedding
Fig. 24.4. Stratigraphic sections of the Tany
Formation.
at the latitude of around 608S between 630 and 600 Ma, then
displaced to 208S between 600 and 550 Ma.
Discussion
Sedimentary environments
Massive and bedded diamictites contain shale interbeds with drop-
stones, clasts with typical glacier striae, grooves and facets, erratic
stones (including very large ones), and nests of stones. They are
overlain by carbonates resembling typical Neoproterozoic ‘cap
dolomites’. In combination, these features suggest that most dia-
mictites of the Tany and Koyva formations and the Lower Starye
Pechi Subformation are glaciomarine deposits. The same con-
clusion can be drawn for most diamictites of the Churochnaya
Formation, except for members characterized by till-like fabric
indicated by preferred orientations in elongated stones and the
occurrence of varved shales, both of which suggest a continental
origin. Some other members of the Churochnaya Formation
contain mainly well-rounded clasts that may have been deposited
by seasonal ice.
The Niz’va, Ust-churochnaya and Churochnaya formations
were deposited in inner shelf environments on the margin of the
Russian Plate. The flysch-like nature of sediments lying under,
between and over the diamictites of the Serebryanka Group and
the Starye Pechi Formation show that they were deposited on the
outer shelf to continental slope.
294 N. M. CHUMAKOV

Possible correlations
m
The glaciogenic deposits of Starye Pechi Formation are older
diamictites than those of the Chernyi Kamen’ Formation (which contain
Edacarian Metazoa and a tuff dated as 557 + 13 Ma; Maslov
bedded diamictites et al. 2007). The Lower Starye Pechi Subformation may
therefore be correlated tentatively to middle Ediacaran glacial
sandstones horizons of the East European craton (the Glussk and Mor-
tensnes formations) and the glacial Gaskiers Formation in
70 Newfoundland.
multicolour shales The Koyva Formation shares similar stratigraphy and facies
with the Tany Formation, from which it is separated by a turbidite
black shales sequence (the Garevka Formation) that is noticeably reduced in
thickness in some places and pinches out near the Koyva River.
dolomites Therefore, the Koyva, Garevka and Tany formations were likely
deposited during a single glacial epoch. This composite unit is
very large boulders slightly older than Buton Formation containing Ediacaran Obru-
chevella (Golovenok et al. 1989) and much younger than the
Shchegrovitsy Formation and Troitsk grano-syenites massif
raft of diamictites (671 + 24 Ma). These are tentatively correlated to the lower
glacial horizon of the Lower Vendian (upper Cryogenian) on the
. .. stone nests East European craton (the Blon’ and Smalfjord formations), the
glacial Nantuo Formation in South China, and the Yerelina Sub-
26 dropstones group of South Australia.
The age and correlation of the Churochnaya Formation is
lamination subject to debate. The overlying Il’ya-Vozh and Kocheshor for-
mations are separated by an erosional unconformity (Fig. 24.2),
and it is possible that the Kocheshor Formation is equivalent to
rhythmic lamination the lower Sylvitsa Goup. In this case, the Churochnaya and Il’ya-
Vozh formations together may correlate with the upper Sereb-
6 ryanka Group (Maslov 2004). It follows from this hypothesis
that the Churochnaya Formation may be coeval with the glacial
Koyva Formation (Fig. 24.2). The traditional correlation of the
Niz’va Formation to the middle Neoproterozoic of the Karatavian
(Upper Riphean) stratotype of the South Urals and the Klytkan
20 Formation of the Middle Urals is supported by similar stromato-
lites assemblages (Raaben 2007) and isotopic data (d13C values
from –0.5 to 2.0‰ for the Niz’va Formation).

6 Regional palaeogeography

The lithology of diamictite clasts suggests that crystalline base-

ment and sedimentary cover of the Russian plate were the
main sources for the Churochnaya, Tany and Koyva formations.
Local sources for the diamictite clasts have also been estab-
50 lished, particularly for the Lower Starye Pechi Subformation
(Ablizin et al. 1982) and Vil’va Formation in the eastern
Urals (Suslov & Teterin 1997). The regional palaeogeographical
reconstructions (Chumakov & Sergeev 2004) suggest that an
inland ice sheet existed during Tany time on the eastern part of
the Russian Plate, whereas a marginal belt of shelf glaciers lay
to the east of it (Churochnaya, Tany and Koyva formations).
. .. Further east, sedimentation was dominated by turbidity currents
and ice rafting in an outer shelf –slope environment (Tany,
Koyva and Wil’va formation). The existence of extensive inland
glacier sheets in the northeastern part of East European Craton
300 is also evident in a number of other Early Vendian (late Cryogen-
ian) glacial units on the craton. The most complete late Cryogenian
glaciogenic sequences are located in the marginal parts of the
... East European craton and are further described elsewhere in this
volume. Within the cratonic interior, glacial sedimentation
occurred in aulocogens that developed in the middle-Late Riphean
(late Mesoproterozoic– Neoproterozoic) (Bessonova & Chumakov
1969; Chumakov 1971, 1981b, c, 1992).
Fig. 24.5. Stratigraphic section of the upper Koyva Formation, Koyva River.
Our research was supported by grant 11-05-00232 from the Russian Fund of Basic
Investigations and Program No 25 of the Presidium of the Russian Academy of
 NEOPROTEROZOIC OF THE NORTH AND MIDDLE URALS 295

m diamictites
bedded diamictites

sedimentary and quartzite stones

Sp2 30 granitoid stones

sandstones with small pebbles

calcareous sandstones
massive quartz-feldspathic sandstones
Sp 4 shales
1,5
shales with dropstones
5,5
dropstones
. ..
rhythmic lamination
Sp1 11
vague bedding
. .. till wedges
. .. stone nests
6

10
20 m

6
10
5

Kr
0
15
Fig. 24.6. Stratigraphic section of lower
Starye Pechi Formation, Sylvitsa River. Sp,
Starye Pechi Formation; Kr, Kernos
Formation.

Sciences. The authors would like to acknowledge stimulating discussions arising
from participation in IGCP project 512 ‘Neoproterozoic Ice Ages.’
References
Ablizin, B. D., Klyuzhina, M. L., Kurbatskaya, F. A. & Kurbatskiy,
A. M. 1982. Upper Riphean and Vendian of the West Slope of the
Middle Ural (in Russian). Nayka, Moscow.
Bessonova, V. Ya. & Chumakov, N. M. 1969. Upper Precambrian
glacial deposits of western regions of USSR. Lithology and Mineral
Resources, 2, 73– 89.
Borovko, N. G. 1967. Vendian and Lower Paleozoic of Polyudov Ridge
of the North Urals (in Russian). VSEGEI, Leningrad.
Chumakov, N. M. 1971. Vendian glaciation of the Europe and North
Atlantic. Doklady Academii Nauk SSSR, 198, 419– 422.
Chumakov, N. M. 1981a. Late Precambrian Kurgashglya tilloids,
soutern Urals. In: Hambrey, M. & Harland, W. B. (eds) Earth’s
Pre-Pleistocene Glacial Record. Cambridge University Press,
Cambridge, 674–677.
Chumakov, N. M. 1981b. Late Precambrian glacial deposits of
the Vilchitsy Formation of western regions of the USSR.
In: Hambrey, M. & Harland, W. B. (eds) Earth’s Pre-
Pleistocene Glacial Record. Cambridge University Press, Cam-
bridge, 655– 659.
Chumakov, N. M. 1981c. Late Precambrian glacial deposits of the Blon
Formation, Belorussia, USSR. In: Hambrey, M. & Harland,
W. B. (eds) Earth’s Pre-Pleistocene Glacial Record. Cambridge
University Press, Cambridge, 660–662.
Chumakov, N. M. 1992. The Problems of Old Glaciations (Pre-
Pleistocene Glaciogeology in the USSR). Harwood Academic
Publishers, Pennsylvania.
Chumakov, N. M. 1998. The key section of Vendian glacial deposits in
the South Urals (Kurgashly Formation, Krivoluksky graben) in
Russian. In: Knipper, A. L., Kurenkov, C. A. & Semikhatov,
M. A. (eds) The Urals: Fundamental Problems of Geodynamics
and Stratigraphy. Nauka, Moscow, 138 –153 (in Russian).
Chumakov, N. M. & Sergeev, V. N. 2004. Problems of climatic zonality
of the Late Precambrian. Climate and biotic events (in Russian)
In: Semikhatov, M. A. & Chumakov, N. M. (eds) Climate During
296 N. M. CHUMAKOV
the Epochs of Principal Biosphere Rearrangements. Transactions of
Geological Institute of RAS, 550, 271.
Ehlakov, Yu. A. & Morozov, G. G. 2006. Stratigraphy (in Russian).
In: Kudryashov, A. I. (ed.) Mineral-Stuff Resources of Permian
territory. Knizhnaya Ploshchad’, Perm, 49– 63.
Garris, M. A., Kasakov, G. A. & Keller, B. M. 1964. Geochronological
scale of Upper Proterozoic (Riphean and Vendian). In: Absolute Age
of Geological Formations. Nauka, Moscow, 431– 455 (in Russian).
Golovenok, V. K., Belova, M. Iu. & Kurbatskaia, F. A. 1989. First
find of Obruchevella Reitlinger in Vendian sediments of Middle
Urals. Doklady Academii Nauk SSSR, 309, 701–705.
Grazhdankin, D. V., Maslov, A. V., Mustill, T. M. R. & Krupenin,
M. T. 2005. The Ediacaran White Sea biota in the Central Urals.
Doklady Earth Sciences, 401, 382– 385.
Grazhdankin, D. V., Nagovitsin, K. E. & Maslov, A. V. 2007. Late
Vendian Miaohe-type ecological assemblage of the East European
platform. Doklady Earth Sciences, 417, 1183–1187.
Keller, B. M. & Chumakov, N. M. (eds) 1983. Stratotype of Riphean.
Stratigraphy. Geochronology. Nauka, Moscow (in Russian).
Li, Z. X., Bogdanova, S. V. et al. 2008 Assembly, configuration, and
break-up history of Rodinia: a synthesis. Precambrian Research,
160, 179– 210.
Maslov, A. V. 2000. Some features of sedimentation in Early Vendian
at the South and Middle Ural. Lithology and Mineral Resources,
6, 624–639.
Maslov, A. V. 2004. Riphean and Vendian sedimentary sequences of the
Timanides and Uralides, the eastern periphery of the East European
Craton. In: Gee, D. G. & Pease, V. L. (eds) The Neoproterozoic
Timanide Orogen of Eastern Baltica. Geological Society, London,
Memoirs, 30, 19– 35.
Maslov, A. V., Grazhdankin, D. V. et al. 2007. U –Pb (SRIMP II) age
of the zircons from ash tuffs of the Chernyi Kamen’ Formation,
Sylvitsa Group of Vendian (Middle Ural). Doklady Russian
Academy of Sciences, Geosciences, 411, 354– 359.
Petrov, G. A., Maslov, A. B. & Ronkin, Yu. L. 2005. Pre-Paleozoic
magmatic complexes of the Kvarkush-Kamennogorsk anticlinorium
(the Middle Ural): new data on the geochemistry and the geody-
namics (in Russian). Lithosphere, 4, 42 –69.
Raaben, M. E. 2007. Stromatolitic formations of the Riphean of East-
European plateform. Stratigraphy and Geological Correlation, 1,
35 – 46.
Ronkin, Yu. L., Maslov, A. V., Petrov, G. A., Matukov, D. I. &
Suslov, S. B. 2007. In situ U– Pb (SRIMP II) date of zircons from
granosyenites of Tritsk Massiv (Kvarkush-Kamennogorsk Meganti-
clinorium, Middle Ural). Doklady Russian Academy of Sciences,
Geosciences, 412, 87 –92.
Suslov, S. B. & Teterin, I. P. 1997. On the origin ‘exotic’ stones of
‘tillite like’ conglomerates of Upper Wil’va Subformation. In: Ibla-
minov, R. G. (ed.) Geology and Mineral Resources of the Western
Ural. Permian State University, Perm, 30 – 31 (in Russian).
 Chapter 25

Glacial deposits of the Nichatka Formation, Chara River basin and review
of Upper Precambrian diamictites of Central Siberia

NICKOLAY M. CHUMAKOV
Geological Institute of the Russian Academy of Sciences, Pyzhevsky per. 7, Moscow 119017, Russia (e-mail: chumakov@ginras.ru)

Abstract: Outcrops of the Nichatka Formation are located in the southern centroclinal part of the marginal Berezovskaya basin of the
Siberian craton. Deposits of the gently sloping southeastern limb of the asymmetrical basin are unaltered, whereas those of the steep,
partly overturned western limb exhibit cleavage and are locally slightly metamorphosed. In the Late Precambrian the Berezovskaya
basin was a shallow marine area of the passive margin of the Siberian craton. The Nichatka Formation is the basal unit of the Dal’nyaya
Tayga Group of the Patom Supergroup. On the southeastern limb of the Berezovskaya Basin, the formation is composed of massive and
bedded diamictites, conglomerates, sandstones and thin-bedded rocks with dropstones. The diamictites contain erratic blocks of different
size and roundness. The formation thins out eastward. On the western limb, the formation is represented predominantly by graded-bedded
and less frequently massive diamictites, sandstones and rare thin bedded members with dropstones. The diamictites contain clasts with
glacial facets and striae. The formation unconformably lies on the Ballaganakh Group of the Patom Supergroup or older rocks and grades
into a marker member of thin-bedded red dolomitic marls and dolomites. Overlying carbonate-terrigenous formations of the Patom
Supergroup are characterized by two negative (down to –6 and – 10.5‰) d13C anomalies and one positive (up to 3.9‰) anomaly.
87
Sr/86Sr ratios in the carbonates above the Nichatka Formation increase upsection from 0.70725 in the overlying deposits to
0.70837 in the Nemakit-Daldyn (Fortunian) horizon of the early Cambrian. The Nichatka Formation as well as the host deposits are
assigned to the Lower Vendian (Late Cryogenian) by reliable geological and isotopic correlations to biostratigraphically better
studied Upper Precambrian sequences of the Ura Uplift in the Lena River basin. Stratigraphically similar glacial units were traced
along the margins of the Baikal-Patom Highland up to the southern end of Lake Baikal and further to the NW towards East Sayan.
The Siberian Craton was the main source area for these glacial units, which points to the existence of the great inland glaciers at the
southern part of this craton. Some older Upper Precambrian diamictites of unclear age and genesis are also recorded in Central Siberia.

Diamictites, sandstones and conglomerates, which lie unconform-
ably on Lower Proterozoic–Archaean granites and quartzites in
the Lake Nichatka area, are named the Nichatka Formation
(Zhuravleva et al. 1959). From this area the formation extends to
the NE up to the eastern bank of the Chara River and to the north
up to the Dzhelinda River and its inflows (Fig. 25.1b). Later,
some researchers (Ivanov et al. 1995 and others) incorrectly referred
to the Nichatka Formation along the western limb of the Berezovs-
kaya Basin as the Dzhemkukan Formation, based on long-distance
correlations with deposits of different structural and facies zones
and composition. The stratotype section of the formation is
located on the northwestern bank of Lake Nichatka, 3.5 km to the
NW of the mouth of the Shirik River (578470 3600 N, 1178370 0900 E–
578480 0600 N, 1178380 3500 E). Although a glacial origin for the for-
mation was assumed in the mid-twentieth century, geological
facts confirming this idea have been discovered and published
only recently (Chumakov 1993).
Structural framework
In the eastern area of distribution of the Nichatka Fm. gently
dipping (5 –158) strata compose the southeastern limb of the asym-
metrical marginal Berezovskaya Basin of the Siberian Craton
(Fig. 25.1b). The deposits have not been altered by secondary pro-
cesses. In the steep and locally overturned western limb, the
Nichatka Fm. is exposed in the cores of faulted anticlines. There
the deposits are intensely cleaved and experienced lower green-
schist metamorphism. Deformation of the western limb may
have been caused by the clockwise rotation of the Aldan Shield
during the Middle Palaeozoic (Shatsillo 2006).
The Nichatka Fm. accumulated in the shallow eastern margins
of the Late Precambrian Patom basin, which was a passive
margin of the Siberian Craton (Chumakov et al. 2007). The
terrigenous-carbonate Patom Supergroup was formed in the
basin during the Late Precambrian. Towards the east, all units of
the supergroup, including the Nichatka Formation, are reduced
in thickness, and terrigenous formations are partly substituted by
carbonate platforms. Sedimentation was cyclic. The Nichatka For-
mation is a basal part of the second Dal’naya Taiga cycle. During
sedimentation of the Patom Supergroup, the basin transgressed
over the craton. This resulted in overlapping of older deposits by
younger ones with unconformities at the margins. Because of thin-
ning and probably erosional processes, the Nichatka formation is
missing in sections along the left bank of the Chara River.
Stratigraphy
A sequence of the Patom Supergroup in the southern Berezovskaya
Basin is shown in Figure 25.2. Analogues of the groups and
formations of the Patom Supergroups were established there by
means of successive correlation of sections (Zhuravleva et al.
1959), which was supported by geological mapping (Ivanov
et al. 1995), similarity of stromatolitic and microphytolitic associ-
ations (Dol’nik 2005) and recent data on C and Sr isotopes
(Pokrovsky et al. 2006). The basal Ballaganakh Gr. is exposed in
the western limb of the basin. The group is mainly composed of
sandstones and small-pebbled conglomerates. Its upper part is rep-
resented by massive and stromatolitic dolomites and limestones of
the Mariinsky Fm. An erosional surface of this formation is over-
lapped by the Nichatka Formation, which is a basal part of the
Dal’nyaya Tayga Gr. The Nichatka Fm. is composed of dia-
mictites, thin-bedded clay-rich siltstones with lonestones (drop-
stones), conglomerates and sandstones. The formation decreases
in thickness eastward and thins out or was eroded on the right
bank of the Chara River. The Nichatka Fm. is well correlated
with the Bol’shoy Patom Fm. of the transverse Ura Uplift, which
complicates the northeastern sector of the Patom folded arc.
Both formations are similar in stratigraphic position, lithological
composition and genesis. In other sectors of the arc, both for-
mations are replaced by sands and shales of the Dzhemkukan Fm.
The Nichatka Fm. is conformably overlain by the Kumakh-
Ulakh Fm., which begins with a member of red-brown dolomites
marls and dolomites. The rest of the Kumakh-Ulakh Fm. is com-
posed of black and variegated shales with limestone interbeds.

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 297– 302. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.25
298 N. M. CHUMAKOV
The formation is correlative to the Barakun Fm. of the Patom
Supergroup. The Dal’nyaya Tayga Gr. is crowned by the Sen’
Fm. represented by alternating sandstones and dolomites in the
lower part and by limestones, massive stromatolitic and oncolitic
dolomites in the upper one. Therefore, some geologists divide
the Sen’ Fm. into two formations: Imalyk and Tokko (Petrov
1976). The Sen’ Fm. is correlated to the shaly Valyukhta Fm. of
the Patom folded arc and to the terrigenous Ura and carbonate
Kalancha formations of the Ura Uplift. The Dal’nyaya Tayga Gr.
is overlain by the Zhuya Gr. represented by a single Torgo Fm.,
which comprises a variegated marl member below and aphanitic
algal and oolitic limestones and dolomites in the larger upper
part. Within the Patom folded arc, the Torgp Fm. corresponds to
the Nikol’skoe Fm. of variegated marls and the Chencha Fm.
of limestones.
The Precambrian deposits of the southern Berezovskaya
depression is terminated by the Zherba and Tinnnaya formations,
which are identical in composition to synonymous formations of
the Patom arc and Ura Uplift. The Zherba Fm. is mostly composed
of quartzite-like, glauconite-bearing sandstones, and the Tinnaya
Fm. by bituminous limestones, dolomites and dolomitic breccias.
In the Ura Uplift, the uppermost beds of the latter contain Nemakit-
Daldyn fauna (Khomentovskiy et al. 2004). The overlying Yuedey
Fm. of variegated marls, dolomites and limestones is correlative
to the Lower Cambrian Nokhtuysk Fm. of the Ura Uplift contain-
ing the Tommotian fauna at the base and to the Pestrotsvetnaya
Fm. of the Aldan Shield.
Glaciogenic deposits and associated strata
Sections of the Nichatka Fm. are essentially variable in lithological
composition. The stratotype section on the northwestern bank of
Lake Nichatka (Fig. 25.3, 2) exhibits repeatedly alternating units
of massive, less frequently bedded diamictites (5 –10 m thick)
and thin-bedded clay-rich mudstones and sandstones (1.5 –20 m
thick). The diamictites consist of lilac-grey clayey-silty-sandy
matrix containing 10– 15% of scattered stones of different form
and size (2– 75 cm, singular up to 170  100 cm). Roundness
varies from class 0 to 5 on a five-class scale, but falls into
classes 1 or 2 on average because of the dominance of only slightly
rounded or unrounded stones. Most stones are fragments of grey
granites and reddish quartzites similar to those underlying the
Fig. 25.1. (a) Geographical position of
Nichatka Fm. and coeval glacial units at
Baikal-Patom Highland: Nch, Nichatka
Fm.; Dz, Dzhemkukan Fm.; Bp, Bol’shoy
Patom Fm.; B.P., Bol’shoy Patom R.
(b) Geological scheme for the south
centroclinal part of Berezovskaya Basin: 1,
Mezsozoic syenite-porphyry; 2, Tinnaya,
Zherba and Torgo formations; 3, Sen’ and
Kumakh-Ulakh formations; 4, Nichatskaya
Fm.; 5, more old Proterozoic deposits; 6,
Palaeoproterozoic and Archaean granites;
7, Archaean gnesses and granites; 8, faults
and thrusts. Numbers in circles indicate type
sections of Nichatka Fm.: 1, upper course of
Dzhelinda R.; 2, west side of Nichatka L.; 3,
Mokryi Kumakh-Ulakh R.
diamictites, as well as foreign gneisses and pink granites. Some
stones (a few percent) have striated and grooved surfaces. Longi-
tudinal, variably sized, subparallel striae and grooves are often
confined to newly made abrasion facets. Grooves on granite
pebbles sometimes resemble a series of small spear-shaped chat-
termarks (Chumakov 1993, fig. 3). Stone nests (clast clusters)
and variously rounded and then split stones are characteristic.
Elongated stones are predominantly oriented to the north and
NE. Lower parts of the diamictite are usually bedded, suggesting
a gradual transition from the underlying thinly bedded rocks
with dropstones to massive diamictites. However, at the base of
some massive diamictites there are signs of erosion or disharmonic
folding and brecciation of underlying thinly bedded rock. In this
case, overlying diamictites contain undeformed rafts of underlying
rocks and small lenses and boudins of assimilated sandstones.
Units between diamictites are composed of alternations of lami-
nated red siltstones and mudstones with graded fine-grained
arkosic sandstones. Upper parts of these units are less distinctly
bedded and contain scattered pebbles and boulders (0.5 –150 cm
across). These tear underlying beds and are enveloped by over-
lying ones; such dropstones bear abraded faces with striae and
grooves. At lateral contacts of pebbles and boulders, synsedimen-
tary deformation structures were observed. In general, lithological
changes throughout the section reflect a gradual transition from
thinly bedded members to bedded diamictites.
A few kilometres to the NE, diamictites and thinly bedded rocks
are replaced by conglomerates with subordinate diamictite inter-
beds. The conglomerates include large boulders consisting of
granites, gneisses, quartzites and boulders and plates of finely
crystalline and oolitic dolomites up to 2.3  2  1.8 m in size.
Further to the north, the conglomerates are rapidly substituted by
a sequence of cross-bedded arkosic sandstones and poorly sorted
small-pebbled conglomerates (Chumakov 1993). The Nichatka
Fm. demonstrates similar composition and structure throughout
the southeastern limb of the Berezovskaya Basin.
The character of the formation is different in the western limb
(Figs. 25.3, 1). The most complete and representative section of
the Nichatka Fm. is located at the Creek Opornyi, a left inflow
of the Dzhelinda River (588030 2500 N, 1178350 2600 E – 588030 3200 N,
178360 0300 E). There the formation is mainly represented by
yellowish-grey and grey diamictites with graded bedding, which
frequently constitute single gradational rhythms with sandstones.
Massive and bedded diamictites are subordinate. There are also
 NICHATKA FORMATION, CHARA RIVER BASIN
interbeds of cross-bedded sandstones and members of thinly
bedded shales with dropstones. Beds of conglomerates and
breccia are rare. Massive diamictites contain predominantly
pebbles, and small and medium-sized boulders, but large boulders
and blocks of granites 0.5  1.2 m in size also occur in some
members. The stones have a roundness of 1– 2.5 (according to
the five class scale). Clasts in the diamictites are mostly rep-
resented by local and erratic granites, quartzites and gneisses.
Pebbles and poorly rounded boulders of dolomites similar to
those of the Mariinsky Fm. are frequent. Some stones bear sub-
parallel striae. Massive diamictites enclose small sand lenses
with erosional or gradational lower boundaries as well as signs
of underwater slides in the form of plastic folds and slide rolls.
Boundary relations with overlying and underlying
non-glacial units
On the southeastern limb of the Berezovka Basin, the Nichatka Fm.
overlies with angular disconformity Archaean (Fig. 25.1, 3) and
Early Proterozoic gneisses and granites, dark red quartzites and
299
Fig. 25.2. Stratigraphic position of the
Nichatka Fm.: 1, diamictites; 2,
conglomerates; 3, sandstones; 4, shales; 5,
limestones; 6, dolostones; 7, marls; 8,
dolostone breccia; 9, oolitic carbonates; 10,
stromatolitic carbonates; 11, lowermost
value of 87Sr/86Sr of formations
(chemostratigraphic data by Pokrovsky
et al. 2006).
pyrophyllitic shales of the Purpol Fm. On the western limb of
the basin the Nichatka Fm. lies on an erosional surface above the
Upper Proterozoic Mariinsky or Bugarikhta formations. The
base of the Nichatka Fm. is composed in this region of conglomer-
ates and breccia of massive stromatolitic and oncolitic dolomites
and limestones of the Mariinsky Fm. Locally, dolomite surfaces
show buried relief (the Bogayukhta River, 578500 0200 N,
1178270 5500 E).
The upper boundary of the Nichatka Fm. is more uniform. The
formation is everywhere gradually overlain by a basal member of
the Kumakh-Ulakh Fm., which consists of alternating laminated
red dolomitic marls and pink dolomites. The thickness of this
member is more than 3 m in the southeastern limb and up to
10 m in the western limb of the Berezovskaya Basin. Contacts of
the Nichatka Fm. and this carbonate member are observed in the
middle course of the Sen’ River, 2 km below the Uraga mouth
(578500 0200 N, 1178270 5500 E) and at the Creek Opornyi. At the
first locality the Nichatka diamictites grade into coarse and then
fine-grained clayey sandstones (2 m), which are overlain by
platy, sandy and then by calcareous mudstones (8 m). The latter
are conformably overlain by red-brown thin-bedded dolomitic
300 N. M. CHUMAKOV
marls of the Kumakh-Ulakh Fm. (0.7 m) passing upward into lami-
nated red clayey dolomites (1.5 m apparent thickness). Similar
conformable and gradual transition between the Nichatka and
Kumakh-Ulakh formations is recorded in the Creek Opornyi
section. There the uppermost Nichatka diamictites grade into red-
brown sandstones (.5 m) with scattered pebbles and boulders.
Alternating with shales, the sandstones are replaced by finely
bedded variegated shales (15 m) including thin sandstone inter-
beds and higher up by a basal member of the Kumakh-Ulakh
Fm. (10 m). This member begins with marls and, higher, exhibits
alternating beds (2 –10 cm) of red micritic dolomites and dolomite
marls.
Chemostratigraphy
C, O and Sr isotopic compositions of carbonates were studied by
Pokrovsky et al. (2006) in craton sections of the Patom Supergroup
along the Chara River, which runs across the southeastern limb of
the Berezovskaya Basin. Results for the calcite component are
described below and represented in part in Figure 25.2. Values
Fig. 25.3. Type sections of Nichatka Fm.:
1, conglomerates; 2, massive diamictites; 3,
bedded diamictites; 4, shales with
dropstones; 5, sandstones; 6, shales; 7,
limestones; 8, dolstones; 9, marls; 10,
granites; 11, striated and faceted stones;
12, dropstones; 13, glaciotectonic
deformations; 14, till-like stone fabric; 15,
stone nests; 16, split, semi-rounded stones;
17, rhythmic bedding and lamination; 18,
erosional channels; 19, small sandstone
lenses; 20, slide and slump structures;
21, slide rolls; 22, lamination; 23,
cross-bedding; 24, conglobreccies. PR1,
Palaeoproterozoic; km, Kumakh-Ulakh
Fm.; mr, Mariinsky Fm.; nč, Nichatka Fm.
for the dolomite component are slightly higher but exhibit parallel
trends. The d13C curve shows two negative anomalies. The first
negative anomaly in the Kumakh-Ulakh Fm. ranges from –2.9
to – 6.3‰ Vienna Pee Dee Belemnite (VPDB) standard, reaching
a minimum value in the lower part of the formation and increasing
upsection. d18P values vary considerably from 14.9 to 22‰
Vienna Standard Mean Ocean Water (VSMOW). The second
negative anomaly, showing a decrease to –8.0 to –10.5‰, is
found in the Torgo Fm. d18P varies insignificantly from 21 to
23‰ with some values as low as 19‰. Between these two negative
anomalies there is a moderately positive one (from – 2.7 to
þ3.4‰) with a maximal d13C value of þ3.4 to þ3.9‰ in the
upper part of the Sen’ Fm. d18P values for this formation vary
between 22 and 25‰, reaching 27‰ in its upper part. The
overlying carbonate Tinnaya Fm. is characterized by gradually
increasing d13C values from –0.8 to þ2.1‰ and d18P values
from 20 to 23‰.
The lowermost 87Sr/86Sr ratio for samples from the Kumakh-
Ulakh Fm. with low values of 87Rb – 86Sr (0.001 – 0.005), Mn –Sr
(0.07 –0.12) and Fe/Sr (2.05 –3.30) is 0.70725 (three determi-
nations: Sr 881–2013 ppm; range 0.70725 –0.70734). For the
 NICHATKA FORMATION, CHARA RIVER BASIN
Torgo Fm. the lowermost 87Sr/86Sr ratio is 0.70837 (one deter-
mination: Sr 158 ppm). For the Tinnaya Fm. the lowermost ratio
is 0.70799 (four determinations: Sr 299–812 ppm; range of
0.7079 –0.70832).
Palaeolatitude and palaeogeography
No palaeomagnetic investigations of Upper Precambrian deposits
have been carried out in the Berezovskaya Basin. According to the
majority of recent global reconstructions (Powell et al. 2001;
Smith 2001; Meert & Torsvik 2003; Shatsillo 2006; and others),
the Siberian craton, including this basin, was located at low lati-
tudes at that time.
Geochronological constraints
There are no available radiometric and reliable biostratigraphic
data on the Upper Precambrian deposits of the Berezovskaya
Basin. They can be dated by means of correlation with biostratigra-
phically better studied sections of the Patom Supergroup of the Ura
Uplift (the Lena River). The corrrelations are based on successive
section-by-section tracing of formations, stromatolites and
isotopic data (Chumakov et al. 2011). d13C anomalies can also
provide confident correlations. The lower Kumakh-Ulakh negative
d13C anomaly corresponds to the negative anomaly for the upper
Bol’shoy Patom –lower Barakun interval of the Ura Uplift. The
positive Sen’ anomaly can be correlated with the analogous
Barakun –Valyukht anomaly of that uplift, and the significant
negative Torgo anomaly can be easily correlated with the analo-
gous Zhuya anomaly (Pokrovsky et al. 2006). These correlations
are supported by similar 87Sr/86Sr ratios for both regions. These
data suggest that the Nichatka Fm. is correlative to the glacial
Bol’shoy Patom Fm. and that the lower Sen’ Fm. corresponds to
the Ura Fm. The last formation contains a rich assemblage of
acanthomorphic palynomorphs (Chumakov et al. 2007; Vorob’eva
et al. 2008) very similar to the Ediacaran (lower Vendian) Kel’tma
microbiota of the Russian plate (Veis et al. 2006) and to ‘late Per-
tatataka’ [¼ECAP (Ediacaran Complex Acanthomorph Palyno-
flora), Grey 2005] microbiota of Australia. The upper part of the
Tinnaya Formation of the Ura Uplift contains small-shelly fauna
of Nemakit-Daldyn type (Khomentovskiy et al. 2004).
Discussion
Sedimentary environments
The Nichatka Fm. has a great number of features that are typical of
glacial deposits. It contains many massive and bedded diamictites
with clasts of variable size, roundness and composition. These
stones have, on average, roundness falling into classes 1 or 2
(using the five classes scale). Erratic and abraded stones (with
clearly striated and grooved faces and spear-like chattermarks)
are common. Rounded and subsequently split stones are also fre-
quent. Dropstones and stone nests (clast clusters) occur commonly
in laminated rocks, which alternate with bedded diamictites.
The disharmonic deformation at the base of some massive dia-
mictites resembles glaciotectonic disruption, and the orientation
of elongated stones is similar to that of tills, suggesting that the
Nichatka Fm. on the southeastern limb of the Berezovskaya
Basin contains basal tillites. The prevailing northern and northeast-
ern orientation of elongated stones and the occurrence of erratic
stones of plagiogneisses and micaceous greisses along with
local stones indicate the Aldan Shield as a source for Nichatka
deposition in the SE of the basin. The basal tillites alternate
with thin and rhythmically bedded siltstones and mudstones
with dropstones. Such thin-bedded deposits are likely to be
301
glaciolacustrine. Along strike, the basal tillites are replaced by
boulder-sized conglomerates, and further along strike by cross-
bedded arkosic sandstones. These deposits, which can be inter-
preted as glaciofluvial, form extensive fans penetrating into the
adjacent basin. Distal sediments of the fans occur on the western
limb of the Berezovskaya Basin. They are mostly represented
there by graded bedded and massive diamictites. Massive diamic-
tites contain rare striated stones and frequent small sand lenses, the
tops of which are marked by thin silt interlayers and the bases of
which exhibit a gradual transition into the sandy-clayey-silty
matrix of diamictite.
There are also small sand lenses filling gentle incisions, showing
that local low-energy currents washed out clay and silt from the
diamictons. The supposition is that massive and bedded diamic-
tites were deposited by floating shelf glaciers, whereas graded
bedded diamictites represent glaciogenic deposits reworked by
mud and debris flows. The direction of cross-bedding in the
sandy interlayers and the lithology of the clasts suggest that
material was transported both from the east and SE and partly
from the west to the western end of the basin. Diamictites
include rare units of thin and rhythmically bedded siltstones and
mudstones with dropstones, which may represent distal turbidite
deposits where stones dropped from melting icebergs.
Age
Reliable stratigraphic correlation between the sections of the
Berezovskaya Basin and the Ura Uplift along with biostratigraphic
data existing for Ura Uplift succession (Chumakov et al. 2011)
allow us to suppose that the Tinnaya and probably Zherba
formations can be tentatively assigned to the Nemakit-Daldyn
Stage of the Upper Vendian, the Zhuya Group to the Kotlin and
Redkino horizons of the Middle Vendian, and the Dal’nyaya
Tayga Group (and the Nichatka Fm. correspondingly) to the
lower part of Laplandian Horizon of the Lower Vendian. Accord-
ing to the timescale of the International Commission on Stratigra-
phy, the Tinnaya Fm. must be referred to the pre-Tommotian
(Fortunian) interval of the early Cambrian, the Zhuya Group and
apparently the upper part of the Laplandian Horizon to the Edia-
caran System, and the lower part of the Laplandian Horizon to
the Upper Cryogenian. In the context of distant correlations, the
data presented above suggest that the Nichatka Fm., like the
Bol’shoy Patom Fm., is correlative to the Nantuo Fm. of South
China and probably the Yerelina Subgroup of Australia.
Regional palaeogeography
As described earlier (Chumakov 1993, 2009), glacial deposits
coeval with the Nichatka Fm. are traceable (with some interrup-
tions) along almost the entire margin of the Baikal-Patom folded
zone (Fig. 25.1a). They are known as the Dzemkukan Fm. to the
north of the Lake Nichatka between the Dzhelinda and Chencha
rivers. Farther on, they are exposed in the Ura Uplift along the
Lena River, where they form the Bol’shoy Patom (or Dzemkukan
Fm.). To the SW, glacial deposits are exposed along the western
margin of the Baikal-Patom folded zone, where they are also
called the Dzemkukan Fm. The Dzemkukan Fm. can then be
traced from the Lena –Bol’shoy Patom interfluve up to the
Vitim –Chaya interfluve. Although slightly thinner, the Dzemku-
kan Fm. of this region is similar in composition and structure to
the Bol’shoy Patom Fm. (Chumakov 1993). To the SW of the
Malaya Chuya River head, the Ballaganakh and lower part of the
Dal’nyaya Tayga groups with the glacial deposits thin out. At
the Chaya River, analogues of the upper part of the Patom Super-
group form the Baikal Supergroup extending up to Lake Baikal. At
the southern end of the lake at the base of the supergroup there
appears a sequence of diamictites, conglomerates and sandstones,
which are called the Bugul’deka Member and are correlated to the
302 N. M. CHUMAKOV
Dzemkukan, Bol’shoy Patom and Nichatka formations (Chuma-
kov 1993; Sovetov & Komlev 2005).
These geological correlations have recently been supported by
C- and Sr-isotopic studies, which revealed considerable similarity
between the Baikal Supergroup and the upper part of the Patom
Supergroup (Kuznetsov & Letnikova 2005), and indirectly by
Pb/Pb dating of the Uluntuu limestones (550 + 40 Ma, Kuznetsov
et al. 2006). All of the above-mentioned glacial units were united
into a single Middle Siberian Glacial Horizon (Chumakov 1993).
By means of lithological and sequence stratigraphic data, the
Bugul’deyka Member can now be correlated to glacial deposits
of the Marninsk Fm. of the Sayany region (Sovetov & Komlev
2005). This makes the area of distribution of the Middle Siberian
Glacial Horizon much wider.
As stated above, the main source for the Nichatka deposits
was the Aldan Shield. Adjacent parts of the Siberian Craton also
supplied debris for the Bol’shoy Patom Fm. of the Ura Uplift
(Chumakov & Krasil’nikov 1991) and the Dzemkukan Fm. at
the western margin of the Baikal-Patom folded zone. Glaciers
moved from the craton to the western Baikal area (Sovetov &
Komlev 2005). Sometimes, debris was transported to the Sayany
region from the folded framework of the Siberian craton
(Sovetov 2002; Sovetov & Komlev 2005), and also to the
western part of the Berezovskaya Basin. In general, The Siberian
Craton was the main sourceland for the Middle Siberian Glacial
Horizon. This suggests that the main centres of Middle Siberian
glaciation were located on the Siberian Craton, and at least the
entire southern part of the craton was covered by glaciers,
which, as evinced in the wide distribution of glaciomarine facies,
flowed into the Baikal Ocean and surrounding seas. The very
extensive area of distribution of the glacial deposits (almost
2000 km long) unambiguously indicates the existence of inland
glacial sheets at all southern parts of the Siberian Craton during
the Middle Siberian glaciation.
Research was supported by grants of the Russian Fund of Basic Investigations
(No. 11-05-00232), Program No. 25 of Presidium of Russian Academy of the
Sciences. The author would like to acknowledge stimulating discussions arising
from participation in IGCP project 512 ‘Neoproterozoic Ice Ages’. This rep-
resents a contribution of the IUGS- and UNESCO-funded IGCP (International
Geoscience Programme) project #512.
References
Chumakov, N. M. 1993. Riphean Middle Siberian Glaciohorizon. Strati-
graphy and Geological Correlation, 1, 17 –28.
Chumakov, N. M. 2009. Neoproterozoic glacial events in Eurasia.
In: Gaucher, C., Sial, A. N., Halverso, G. P. & Frimmel, H. R.
(eds) Neoproterozoic– Cambrian Tectonics, Global Change and
Evolution: a Focus on Southwestern Gondwana. Developments in
Precambrian Geology. Elsevier, The Netherlands, 16, 389– 403.
Chumakov, N. M. & Krasil’nikov, S. S. 1991. Lithology of
Riphean tilloids; Ura Uplift. Lithology and Mineral Resourses, 3,
58 –78.
Chumakov, N. M., Pokrovsky, B. G. & Melezhik, V. A. 2007. Geologic
history of Patom Supergroup, Late Precambrian, Middle Siberia.
Doclady Academii Nayk. Geologiya, 413, 379– 383.
Chumakov, N. M., Pokrovsky, V. G. & Melezhik, V. A. 2011. The
glaciogenic Bol’shoy Patom Formation, Lena River, central
Siberia. In: Arnaud, E., Halverson, G. P. & Shields-Zhou, G.
(eds) The Geological Record of Neoproterozoic Glaciations. Geo-
logical Society, London, Memoirs, 36, 309–316.
Dol’nik, T. A. 2005. Stromatolites and Mikrophytolites in Stratigraphy
of Riphean and Vendian of Folded Framework of the South Part of
Siberian Craton (in Russian). Geo, Novosibirsk.
Grey, K. 2005. Ediacaran Palinology of Australia. Memoir 31. Associ-
ation of Australian Palaeontologists, Canberra.
Ivanov, A. I., Lifshits, V. I., Perevalov, O. V., Strakhova, T. M.,
Yablonovskiy, B. V, Gryzer, M. I., Il’inskaya, Kh. G. & Golove-
nok, V. K. 1995. Precambrian of Patom Highland (in Russian).
Nedra, Moscow.
Khomentovskiy, V. V., Postnikov, A. A., Karlova, G. A., Kochnev,
B. B., Yakshin, M. S. & Ponomarchuk, V. A. 2004. Vendian of
Baikal-Patom Upland (Siberia). Geology and Geophysics, 45,
465– 484.
Kuznetsov, A. B. & Letnikova, E. F. 2005. Opening Baikal Branch of
Palaeoasian ocean: Sr and C data (in Russian). In: Koryakin, Y. V.
(ed.) Tectonics of the Earth Crust and Mantle. The Tectonic Regu-
larities of the Occurrences Mineral Recourses. GEOS, Moscow, I,
352– 355.
Kuznetsov, A. B., Ovchinnikova, G. B., Kaurova, O. K. & Letni-
kova, E. F. 2006. Pb/Pb age and Sr chemostratigraphy of carbonate
rocks of the Baikal Group, the south-west Pribaykal’e (in Russian).
Isotopic dating of the ore formation, magmatism, sedimentation and
metamorphism. GEOS, Moscow, 1, 362– 365.
Meert, J. G. & Torsvik, T. H. 2003. The making and unmaking of a
supercontinent: Rodinia revisited. Tectonophysics. 375, 261– 288.
Petrov, A. F. 1976. Precambrian orogenic complexes of west part of
Aldan shield (in Russian). Nayka, Novosibirsk, 120.
Pokrovsky, B. G., Melezhik, B. A. & Buyakayte, M. I. 2006. Geo-
chemistry of isotopes C, O, Sr and S, chemostratigraphy and environ-
ments of sedimentation of Late Precambrian deposits of Patom
trough. Lithology and Mineral Resourses, 5, 505– 530.
Powell, C. McA., Pisarevsky, S. A. & Winwate, M. T. D. 2001. An ani-
mated history of Rodinia. Geological Society Australia Abstracts,
65, 85 – 87.
Shatsillo, A. V. 2006. Palaeomagnetism of the Vendian of the South Part
of the Siberian Craton and Some Aspects of Late Precambrian
Palaeogeodynamics (in Russia). Institutes Physics of the Earth
Russian Academy of Sciences, Moscow.
Smith, A. G. 2001. Palaeomagnetically and tectonically based global
maps for Vendian to Mid-Ordovician time. In: Zuravlev, A. Yu
& Riding, R. (eds) The Ecology of the Cambrian Radiation. Colum-
bia University Press, New York, 11 – 46.
Sovetov, J. K. 2002. Vendian foreland basin of the Siberian cratonic
margin: Palaeopangean accretionary phases. Russian Journal of
Earth Sciences, 4, 363– 387.
Sovetov, Yu. K. & Komlev, D. A. 2005. Tillites at the base of Oselok
Group Prisayan’ya and lower boundary of Vendian in south-west
part of Siberian Craton. Stratigraphy and Geological Correlation.
13, 3 – 34.
Veis, A. F., Vorob’eva, N. G. & Golubkova, E. Y. 2006. First find of
Lower Vendian microfossils at Russian plate: taxonomy composition
and biostratigraphic significance. Stratigraphy and Geological
Correlation, 14, 28– 46.
Vorob’eva, N. G., Sergeev, V. N. & Chumakov, N. M. 2008. New
occurrences of Lower Vendian microfossils in Ura Formation:
problem Patom Supergroup of Middle Siberia. Doclady Academii
Nayk. Geologiya, 419, 782– 787.
Zhuravleva, Z. A., Komar, V. A. & Chumakov, N. M. 1959. Strati-
graphic correlations Patom Complex with sedimentary deposits
west and north slopes of Aldan Shield. Doclady Academii Nayk
USSR, 128, 1026– 1029.
 Chapter 26

Glacial deposits of the Baykonur Formation, Kazakhstan and Kyrgyzstan

NICKOLAY M. CHUMAKOV
Geological Institute of the Russian Academy of Sciences, Pyzhevsky per. 7, Moscow 119017, Russia (e-mail: chumakov@ginras.ru)

Abstract: The Baykonur Formation extends with some interruptions for over 1700 km along the eastern and northern margins of the
Upper Precambrian Syrdarya and Tarim microcontinents. The Baykonur Formation constitutes the upper part of the Ulutau Group,
which lies on an erosional surface of granosyenite with a U–Pb age of 720 + 20 Ma and is overlain by Lower Cambrian vanadium-
bearing carbonaceous-siliceous shales. The Baykonur Formation is mostly composed of diamictite with erratic and striated stones,
and includes thin-bedded shale beds with lonestones (dropstones) and a ‘cap dolomite’ unit at its top. The lithological composition of
this formation implies a glaciomarine origin, while its stratigraphic position suggests a Late Vendian or an Early Cambrian age (the
latest Ediacaran or Early Cambrian according to scale of the International Commission on Stratigraphy).

The Baykonur Formation represents a horizon of Upper Precam-
brian diamictites, which extend with some interruptions as an
extended arc for over 1700 km from the Ulutau Mountains of
Central Kazakhstan (Fig. 26.1, 1) through the Bol’shoy Karatau
Ridge and Dzhebagly Mountains of Southern Kazakhstan
(Fig. 26.1, 3, 4, 5), the Chatkal, Sandalash (Fig. 26.1, 6) and
Kokiirim-Too (Fig. 26.1, 7) ridges of Western Kyrgyzstan up to
the Naryn-Too and Dzhetym-Too ridges of Central Kyrgyzstan
(Fig. 26.1, 8) and the Sarydzhaz Ridge of Eastern Kyrgyzstan
(Fig. 26.1, 9) and further up to the China boundary. The Baykonur
Formation is the upper formation of the group called Ulutau in
Kazakhstan and Dzhetym in Kyrgyzstan. Its name is derived
from the Baykonur River flowing from the southwestern slope of
the Ulutau Mountains where the stratotype section of the formation
is located. The second type section, deposits of which are less
metamorphosed and better studied, is an outcrop along the Rang
River, which flows from the northeastern slope of the Bol’shoy
Karatau Ridge. Probable stratigraphic analogues of the Baykonur
Formation are the diamictite of the lower part of the Kyrshibakty
Formation in the Malyi Karatau Ridge (Korolev & Maksumova
1984) and the upper diamictitic part of the Kopal Formation
exposed in the Mointy River basin (Fig. 26.1, 2).
The Baykonur diamictites were first recovered in 1924 by D. V.
Nalivkin, who referred them as Upper Palaeozoic glacial deposits.
Some time later, the Baykonur Formation was considered to be
Cambrian ‘facies of the foot of rocky shores’ (Borovikov 1955),
proluvium (Makarychev 1967), subsurface sliding deposits
(Volin 1966; Kholodov 1973; and others) or subaqueous sliding
deposits with an admixture of mountain glacial debris (Knipper
1963). In the 1960s it became evident that the Baykonur diamic-
tites belong to the Upper Precambrian (Korolev 1963). A majority
of geologists agreed with their Precambrian age but continued,
after Knipper, to interpret them as polygenic deposits of subaqu-
eous slides, clastic and grain flows or, partly, as proluvium and
mountain glacial deposits (Zaytsev & Kheraskova 1979; Kheras-
kova 1981a, 1986; Korolev & Maksumova 1984; and others). At
the same time, evidence was gradually accumulating to indicate
that the Baykonur diamictites were to a considerable extent
formed by glacial activity (Ankinovich 1961; Zubtsov 1972;
Chumakov 1978; Azerbaev 1988).
Structural framework
The Baykonur Formation is confined to the synclinorium of
the Caledonian basement stretching along the ancient eastern
active margins of the Syrdarya and northern margin of the
Tarim microcontinents (Kiselev 2001). The Baykonur Formation
experienced folding, thrusting and other dislocations during
Palaeozoic ‘Caledonian’ orogeny. The deposits are usually
cleaved (locally significantly) and weakly metamorphosed up to
lower greenschist facies.
Stratigraphy
One of the type sections of the upper part of the Ulutau Group is a
section at the Rang River (the Bol’shoi Karatau Ridge), where the
following members are exposed (from the base upward, Fig. 26.2).
The Aksumbe Formation
(1) Pale limestones alternating with green and black clay-rich
shales (200 m).
The Baykonur Formation
(2) Dark thin-bedded clay-rich shales with scattered stones of
quartz, dolomite, quartz porphyres and shales. The deposits
were disturbed by subaqueous slumps (10 m).
(3) Diamictite with a muddy matrix and scattered stones of differ-
ent size, roundness and composition; bedding is disturbed by
sliding, ‘quivering’ undulatory lamination and rare dolomite
olistoliths (200 m).
(4) Dark grey, brown on surface, bedded dolomite with intrafor-
mational breccias and signs of sliding (7.5 m).
The Kurumsak Formation
(5) Black cherts and vanadium-bearing carbonaceous-siliceous
shales (250 m).
Sections of the Baykonur Formation in the Ulutau Mountains
(Fig. 26.2) and other areas of distribution are similar to the strato-
type section. Thicknesses (from 600 to 100 m or less) and names of
units above and below the Baykonur Formation are variable.
Glaciogenic deposits and associated strata
The Baykonur Formation is mostly composed of diamictite with
subordinate interbeds and members of shale, sandstone, conglom-
erate and breccia. The diamictite retains its main characteristic
features throughout the region, but may vary in appearance and
degree of metamorphism. The diamictite matrix usually comprises
dark grey mudstone with variable portions of sand and dolomite
grains. Increased content of the latter imparts a brownish-grey

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 303– 307. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.26
304 N. M. CHUMAKOV
colour to the matrix. There may also be admixture of chlorite,
depending on the metamorphic grade. Chlorite imparts a
greenish-grey colour to the matrix. Pyrite inclusions are also fre-
quent at times. The matrix contains chaotically scattered clasts
of different sizes, roundness and composition that can amount to
between 1 and 20% of the total rock. Their predominant size
ranges from 1 to 10–15 cm, although more rare larger fragments
up to few metres across are also found. Semi-rounded and angular
stones dominate over rounded and well-rounded stones. Striated
and faceted clasts occur rarely. The clasts are mainly composed of
rocks underlying the diamictite, but there are also erratic rocks that
are foreign to the underlying successions (Azerbaev 1988).
Fig. 26.1. Outcrops of Baykonur
Formation and coeval glacial units in
Middle Asia. a, Outcrops of Baykonur and
Kopal Formations; b, boundary of states.
Kazakh Republic: 1, Ulutau Mountains; 2,
Mointy River; 3, North Bol’shoy Karatau
Ridge; 4, Central Bol’shoy Karatau Ridge;
5, South Bol’shoy Karatau Ridge and
Dzhebagly Mountains. Kirgiz Republic: 6,
Chatkal Mountain Range; 7, Kokirim
Mountain Range; 8, Naryntau and
Dzhetymtau Mountain Ranges; 9,
Sarydzhaz Mountain Range. I, Uzbek
Republic; II, People’s Republic of China.
In the Rang section, the diamictite contains a great number of
small and large (up to a few metres long) dolomite olistoliths,
brecciated dolomite and fragments of clay-rich sandstone beds.
In some places olistoliths are deformed, disrupted and associated
with smaller fragments of the same rocks (Korolev & Maksumova
1984). In addition to the dolomites, there are fragments and blocks
of arkosic sandstone. Clast clusters (‘stone nests’) are common.
Some parts of the diamictite exhibit a massive texture. Varying
concentrations of clasts in the Baykonur diamictite may make
impressions of coarse and lenticular bedding. In the Rang section,
varying dolomite content defines vague bedding exhibiting
gradual boundaries every 0.2– 5 m within the diamictite, which
Fig. 26.2. Type sections of Baykonur
Formation (Ulutau Mauntins: Knipper
1963; Kheraskova 1986; Dzhetymtau
Range: Korolev & Maksumova 1984).
 BAYKONUR FORMATION, KAZAKHSTAN AND KYRGYZSTAN
is often characterized by fine plastic crimping. Large-scale soft-
sedimentary deformation provides evidence of subsurface sliding.
Shales
Greenish-grey thin-bedded shales constitute definable units
(members) 5–40 m thick in the lower part and less commonly in
the middle part of the Baykonur Formation. The shales usually
include scattered small and medium-sized, and more rarely large
pebbles and boulders. The lonestones cut through into the under-
lying beds, while the overlying deposits drape the stones
(Knipper 1963; Azerbaev 1988). The shales bear frequent signs
of subaqueous sliding.
Sandstones
The sandstones are composed of feldspar and quartz and form
thin interbeds, lenses and small members within the diamictite
beds. Locally the sandstones have fine (up to 2 cm) rhythmical
graded bedding. The rhythms begin with coarse-grained sand-
stones or small-pebbled conglomerates or grits, which gradually
pass into fine-grained sandstones. Many rhythms have an
uppermost thin interbed of siltstones with carbonate cement
(Knipper 1963).
Dolomite
In many sections, the Baykonur Formation is terminated by a dolo-
mite bed. Its thickness reaches 7.5 –10 m in the Rang section. The
dolomite is calcareous (up to 10% of CaCO3), slightly sandy (up to
11%), and has a dark grey colour, which is pale yellow or light
brown on weathered surfaces (Chumakov 1992). Its structure is
massive or bedded, frequently breccia-like, with signs of sub-
aqueous slides. Some dolomite interlayers represent typical intra-
clast breccias consisting of numerous flat dolomite clasts.
Conglomerates
Conglomerates occur as separate beds and lenses. They greatly
increase in amount on the western limb of the Baykonur synclinor-
ium (Kheraskova 1986). The clasts in these conglomerates are the
same as those in the diamictites.
Boundary relations with overlying and underlying
non-glacial units
In the Ulutau Mountains, the Baykonur Formation lies conform-
ably on underlying deposits in the axial part of the Baykonur syn-
clinorium and on an erosional surface on the synclinorium limbs
(Kheraskova 1981a, 1986). An unconformity at the base of the
Baykonur Formation can also be observed at the Rang section of
the Bol’shoy Karatau Ridge and at the Dzetymtau Ridge near
the Kalmakashu Pass (Korolev & Maksumova 1984).
The upper boundary of the Baykonur Formation is distinct but
conformable in the Baykonur synclinorium (Kheraskova 1981a)
and at Rang River. Further to the south, in the Tien Shan Moun-
tains, there is a hiatus at the top of the formation locally associated
with a weathering crust (Korolev & Maksumova 1984).
In the Mointy region, the Kopal Formation contains a diamictite
member very similar to the Baykonur Formation (Kheraskova
1981b). It lies conformably on top of dolomites of the lower part
of the Kopal Formation. In some sections, the Basagin Formation
gradually replaces the diamictites. It contains stromatolites of
Cambrian type and is covered without hiatus by the Kyzylzhar
305
Formation, which demonstrates a Middle Cambrian fauna
(Zaytsev & Kheraskova 1979; Kheraskova 1981b).
Chemostratigraphy
No chemostratigraphic investigations of Neoproterozoic depos-
its have been carried out in regions of the distribution of
Baykonur Fm.
Other characteristics
A unit of black, thin and rhythmically bedded vanadium-bearing
carbonaceous chert occurs 5–30 m above the base of the Kurum-
sak Formation of the Bol’shoy Karatau. The shales are 10 –12 m
thick and include phosphorite concretions. The chert contains
75– 90% SiO2, 1.5 –15% Corg, 0.05–1.5% V2O5 and up to 2%
BaO (Kholodov 1973). Under the microscope, fossil radiolarians
and sponge spicules can be observed in the chert (Kholodov 1973).
Palaeolatitude and palaeogeography
No valid results for palaeomagnetic investigations have been pub-
lished for the Baykonur Formation and adjacent deposits. Accord-
ing to general palaeotectonic reconstructions, the Kazakhstan and
Kyrgyzstan microcontinents were located at low latitudes during
the Cryogenian-Ediacaran (Vendian) (Kheraskova et al. 2003).
Geochronological constraints
The Baykonur Formation is the uppermost formation of the Ulutau
(Dzhetym) Group. In the Bol’shoy Karatau Ridge, the group lies
unconformably on deeply eroded granosyenite with a U – Pb
zircon age of 720 + 20 Ma. (Kiselev 2001). The formation has a
conformable contact with overlying carbonaceous-siliceous
shales of the lower Koktal Formation (the Ulutau Mountains) or
its stratigraphic and lithological equivalent, the Kurumsak For-
mation (the Bol’shoy Karatau Ridge). The Koktal carbonaceous
chert (c. 30–50 m higher than the base of the formation) yielded
acritarchs Micrhystridium aff. dissimilare Volk., M. tornatum
Volk., M. lubomlense Kirjan., Cymatiosphaera? membranacea
Kirjan., Leosohaeridia sp. ‘This acritarch assemblage is character-
istic of the upper Lower Cambrian and very similar to the associ-
ation from the Vergol Horizon of the East European platform’
(Krylov et al. 1986). The Koktal Formation lies beneath the Kok-
bulak Formation containing Middle and Late Cambrian trilobites
(Ergaliev 1965).
Discussion
Sedimentary environments
The wide lateral distribution of Baykonur diamictites in the same
stratigraphic position, as well as the occurrence of erratic and
striated stones, suggests a significant role for glaciers in their
genesis. Interbeds and whole units of thin-bedded shales with drop-
stones and frequent subsurface sliding indicate that the diamictites
were formed in glaciomarine environments and subsequently
affected by flows of variable density. A characteristic succession
of post-glacial deposits is worthy of attention: diamictites are
directly overlain by a marker bed of dolomites resembling
cap dolomite and this latter unit is overlapped by carbonaceous
chert containing phosphorite concretions and considerable
306 N. M. CHUMAKOV
concentrations of organic carbon (1.5 –15%). This facies succes-
sion is similar to many Upper Precambrian post-glacial sequences
(Chumakov 1992; Zhu et al. 2007).
Age
The sharp unconformity between the Ulutau Group and the under-
lying granosyenites, which have a U – Pb zircon age of
720 + 20 Ma, provides evidence for a considerably younger age
for the group as a whole. The Baykonur Formation has a conform-
able contact with the overlying Kurumsak Formation or its strati-
graphic analogues, for example the Koktal Formation. The
erosional disconformity at the top of the Baykonur Formation in
the Tien Shan sections is therefore unlikely to reflect a great
hiatus. This suggestion is supported by persistent, but thin (few
metres) cap dolomite of the Baykonur Formation over a long dis-
tance along the Bol’shoy Karatau Ridge and in some other sec-
tions. The conformable overlapping of the Baykonur Formation
by the Lower Cambrian Koktal Formation indicates a relatively
young age for the Baykonur Formation. As mentioned above, the
microfossils characteristic of the upper Lower Cambrian (the
Vergol Horizon correlatable with the Holmia Zone) were found
some tens of metres above the Koktal Formation base. Strata
between those with Early Cambrian microfossils and the Baykonur
cap dolomite were formed in a basin experiencing condensed sedi-
mentation. This suggests that their accumulation, although occur-
ring over a long time, would be unlikely to have taken more than
several million years. From these considerations an inference
can be made that the Baykonur Formation was formed during
the Late Vendian or initial Early Cambrian, that is, during the
latest the Ediacaran or Early Cambrian according to the scale of
the International Commission on Stratigraphy.
Glaciogenic deposits similar to the Baykonur Formation depos-
its are also known in regions adjacent to Kyrgyzstan in China,
namely in the Kuruktag Range (Wang et al. 1981; Chen et al.
1981). The glacial Hankalchough Formation has a very similar
stratigraphic position to the Baykonur Formation, just below
deposits with Lower Cambrian fossils (Zhu et al. 2011). The
basal member of the Lower Cambrian in the Kuruktag Range con-
sists of black carbonaceous chert containing phosphorite, as in
Kyrgyzstan and Kazakhstan. A similar stratigraphic position has
also been reported for the glacial Hongtiegou Formation of
Chaidam Basin (Shen et al. 2010) and Luoquan Formation
of North China (Mu 1981; Guan et al. 1986) and glacial deposits
of the Bokson Formation (Chumakov 2009, 2011). I have com-
bined all enumerated formations as the Baykonurian Glacial
Horizon (Chumakov 1978, 1981). The widespread extent of the
Baykonurian Glacial Horizon points to a significant glacial event
occurring close to the Precambrian –Cambrian boundary that
affected the Kazakhstan, Kyrgyzstan, Tuva-Mongolian and
Tarim microcontinents as well as the North China Craton. It is
possible that the large negative d13C anomaly that occurred close
to the lower boundary of the Nemakit-Daldyn Horizon (Knoll
2000; Zhu et al. 2007; and others) is related to this Baykonurian
Glaciation.
Research was supported by grants from the Russian Fund of Basic Investigations
(Nos 11-05-00232 and 10-05-00294), and Program No. 25 of the Presidium of the
Russian Academy of the Sciences. The author would like to acknowledge stimu-
lating discussions arising from participation in IGCP project 512 ‘Neoproterozoic
Ice Ages’. This represents a contribution of the IUGS- and UNESCO-funded
IGCP (International Geoscience Programme) project #512.
References
Ankinovich, S. G. 1961. Lower Paleozoic vanadium-bearing basin of the
North Tien Shan and West Margin of Central Kazakhstan (in
Russian). Part I, Alma-Ata, 272.
Azerbaev, N. A. 1988. Lithologic features and origin of Vendian con-
glomerates of Bol’shoy Karatau (in Russian). Izvestia of Academy
of Sciences of Kazakh SSR. Geological Series, 2, 53 – 63.
Borovikov, L. I. 1955. Lower Paleozoic of Dzhezkazgan-Ulutau Region
of West Part of Kazakhstan (in Russian). Nedra, Moscow.
Chen, J., Zhang, H., Xing, Y. & Ma, G. 1981. On the Upper Precam-
brian (Sinian Suberathem) in China. Precambrian Research, 15,
207– 228.
Chumakov, N. M. 1978. Precambrian Tillites and Tilloids (in Russian).
Nauka, Moscow.
Chumakov, N. M. 1981. Upper Proterozoic glaciogenic rocks and their
stratigraphic significance. Precambrian Research, 15, 373–396.
Chumakov, N. M. 1992. The Problems of Old Glaciations (Pre-
Pleistocene Glaciogeology in the USSR). Harwood Academic Pub-
lishers, Pennsylvania.
Chumakov, N. M. 2009. The Baykonurian Glaciohorizon of the Late
Vendian. Stratigraphy and Geological Correlation, 17, 373–381.
Chumakov, N. M. 2011. Glacial deposits of the Bokson Group, East
Sayan Mountains, Buryatian Republic, Russian Federation. In:
Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geo-
logical Record of Neoproterozoic Glaciations. Geological Society,
London, Memoirs, 36, 285–288.
Ergaliev, G. Kh. 1965. On stratigraphy of Vendian and Cambrian Bay-
konur Karatau-Dzhebagly zone (in Russian). Izvestia of Academy of
Sciences of Kazakh SSR. Geological series, 6, 31 –43.
Guan, B., Wu, R., Hambrey, M. J. & Geng, W. 1986. Glacial sediments
and erosional pavement near the Cambrian – Precambrian boundary
in western Henan Province, China. Journal of Geological Society,
London, 143, 311–323.
Kheraskova, T. H. 1981a. Late Precambrian tilloid of Baykonur
Formation in Ulutau Mountains, Central Kazakhstan, USSR. In:
Hambrey, M. J. & Harland, W. B. (eds) Earth’s Pre-Pleistocene
Glacial Record. Cambridge University Press, Cambridge,
348– 352.
Kheraskova, T. H. 1981b. Late Precambrian tilloid of Kopal Formation
in the Atasu-Mointy interfluve, Central Kazakhstan, USSR. In:
Hambrey, M. J. & Harland, W. B. (eds) Earth’s Pre-Pleistocene
Glacial Record. Cambridge University Press, Cambridge,
358– 360.
Kheraskova, T. N. 1986. Vendian-Cambrian Rock Associations of
Caledonides of Asia (in Russian). Nayka, Moscow.
Kheraskova, T. N., Didenko, A. N., Bush, V. A. & Volozh, Yu. A.
2003. The Vendian– Early Paleozoic history of continental margin
of Eastern Paleogondwana, Paleoasian Ocean, and Central Asian
Foldbelt. Russian Journal of Earth Sciences, 5, 165–184.
Kholodov, B. N. 1973. Sedimentary Ore Genesis and Metallogeny of
Vanadium (in Russian). Nauka, Moscow.
Kiselev, V. V. 2001. Analogues of the Sinian complex in the central and
northern Tien Shan. Geology and Geophysics, 42, 1453– 1463.
Knipper, A. L. 1963. Tectonics of Baikonur Synclinorium (Central
Kazakhstan) (in Russian). Publishing Office Academy of Sciences
of the USSR, Moscow.
Knoll, A. H. 2000. Learning to tell Neoproterozoic time. Precambrian
Research, 100, 3– 20.
Korolev, V. G. 1963. About Cambrian boundaries in the Middle Asia
(in Russian). Trudy Frunzinskogo politekhnicheskogo institute, Geo-
logiya, Gornoe delo, 10, 16 –21.
Korolev, V. G. & Maksumova, R. A. 1984. Precambrian Tillites and
Tilloids of Tien Shan (in Russian). Ilim, Frunze.
Krylov, N. N., Sergeev, V. N. & Kheraskova, T. N. 1986. Discovery
of Cambrian microfossils in deposits of Baikonur Synclinorium
(in Russian). Izvestiya of Academy of Sciences of the USSR, Seriya
geologicheskaya, 1, 51– 56.
Makarychev, G. I. 1967. Stratigraphy of Protorozoic and Lower Paleo-
zoic deposits of Bol’shoy Karatau (in Russian). Bulleten’ Moskovs-
kogo obshchestva ispytateley prirody, XXXII, 31 –46.
Mu, Y. 1981. Luoquan tillite of the Sinian System in China. In: Hambrey,
M. J. & Harland, W. B. (eds) Earth’s Pre-Pleistocene Glacial
Record. Cambridge University Press, Cambridge, 402–413.
Shen, B., Xiao, S., Zhou, G., Kaufman, A. J. & Yuan, X. 2010. Carbon
and sulphur isotope chemostratigraphy of the Neoproterozoic Quanji
Group of the Chaidam Basin, NW China: Basin stratification in the
 BAYKONUR FORMATION, KAZAKHSTAN AND KYRGYZSTAN
aftermath of an Ediacaran glaciation postdating the Shuram event?
Precambrian Research, 177, 241– 252.
Volin, A. V. 1966. Breccias of sliding and tillites in relation with problem
glaciations and moving of poles (in Russian). In: Materials k sovesh-
chaniyu ‘Obshchie sakonomernosti geologicheskikh yavleniy’.
Leningrad, 1, 31 – 46.
Wang, Y., Lu, S., Gao, Z., Lin, W. & Ma, G. 1981. Sinian tillites of
China. In: Hambrey, M. J. & Harland, W. B. (eds) Earth’s Pre-
Pleistocene Glacial Record. Cambridge University Press, Cam-
bridge, 386– 401.
Zaytsev, Y. A. & Kheraskova, T. N. 1979. Vendian of Central Kazakh-
stan (in Russian). Moscow University Publisher, Moscow.
307
Zhu, M. & Wang, H. 2011. Neoproterozoic glacigenic diamictites in
the Tarim Block, NW China. In: Arnaud, E., Halverson, G. P. &
Shields-Zhou, G. (eds) The Geological Record of Neoprotero-
zoic Glaciations. Geological Society, London, Memoirs, 36,
367– 378.
Zhu, M., Strauss, H. & Shields, G. A. 2007. From snowball earth to the
Cambrian bioradiation: calibration of Ediacaran –Cambrian earth
history in South China. Palaeogeography, Palaeoclimatology,
Palaeoecology, 254, 1 –6.
Zubtsov, E. I. 1972. Precambrian tillites of Tien Shan and their strati-
graphic significance. Bulleten’ Moskovskogo obshchestva ispytateley
prirody, XLVII, 42 – 56.
 Chapter 27

The glaciogenic Bol’shoy Patom Formation, Lena River, central Siberia

NICKOLAY M. CHUMAKOV1*, BORIS G. POKROVSKY1 & VICTOR A. MELEZHIK2

1
Geological Institute Russian Academy of Sciences, Pyzhevskiy 7, Moscow 119017, Russia
2
Geological Survey of Norway, Leiv Eirkssons Vei 39, N-7491 Trondheim, Norway
*Corresponding author (e-mail: chumakov@ginras.ru)

Abstract: The Bol’shoy Patom Formation (Fm.) is part of the Upper Precambrian Patom Supergroup, which comprises a siliciclastic and
carbonate succession divided into (from base to top) the Ballaganakh, Dal’nyaya Tayga and Zhuya Groups (Gr.). The supergroup was
deposited within the bay-like passive margins of the Siberian craton. The Bol’shoy Patom Fm. is the lower unit of the Dal’nyaya Tayga
Gr. Massive and stratified diamictites with subordinate sandstones, mudstones, siltstones, conglomerates and conglo-breccias form the
Bol’shoy Patom Fm. New biostratigraphic (microfossils of Pertatataka type, Ediacaran fossils) and chemostratigraphic data (87Sr/86Sr,
d13C) point to a Vendian (Late Cryogenian and Ediacaran) age for the Dal’nyaya Tayga and Zhuya Groups. Unsorted diamictite matrix, a
very wide range of size and roundness of dispersed erratic clasts, the presence of glacial grooves on boulders, dropstones and till pellets in
laminated mudstones and siltstones are all evidence for intense ice rafting and ice-shelf sedimentation in the northeastern part of the
Patom basin during deposition of the Bol’shoy Patom Fm.

A ‘conglomerate’ (diamictite) unit was first identified in the lower
reaches of the Bol’shoy Patom River and on the southern banks of
the Lena River (10 km upstream of the mouth of the Bol’shoy
Patom River) in 1935 by Z. M. Starostina and formally named in
1941 as the Bol’shoy Patom Fm. by A. A. Predtechensky (Chuma-
kov 1959). This name has been used in many publications up to
today (Keller 1963; Bobrov 1979; Chumakov 1993; Vorob’eva
et al. 2007), although some authors incorrectly refer to it as the
Dzhemkukan Fm., based on correlation with a unit of very differ-
ent lithology (shales and sandstones) in the Zhuya River area
(Ivanov et al. 1995 and others). The best outcrops of the
Bol’shoy Patom Fm. are in canyons of the Bol’shoy Patom and
Ura River. The upper part of the formation is seen along the
Lena River. The type section of the formation is situated on the
eastern limb of the Zheday anticline in a canyon of the Bol’shoy
Patom River.
Lungersgauzen (1963) first suggested a glacial origin for the dia-
mictites of the Bol’shoy Patom Fm. (he called it the Ust-Patom
Fm.). Lungersgauzen described them as lithified basal moraines
and fluvioglacial deposits. Subsequently, evidence for bedding,
graded bedding and subaqueous slumps in the diamictites led
Lungersgauzen’s interpretation to be modified (Chumakov 1965;
Chumakov & Krasil’nikov 1991).
Structural framework
The Bol’shoy Patom Fm. forms the cores of two anticlines that
exhibit a northeastern trend and an en-echelon-like arrangement.
The anticlines form the apex of the transverse Ura Uplift, which is
oriented perpendicular to a huge marginal arch of folds bordering
the Patom fold system. The uplift consists of the simply folded,
thick (.6 km) Neoproterozoic Patom Supergroup (SGr.). The
more southerly Zheday Anticline (Fig. 27.1b) crosses the lowermost
course of the Bol’shoy Patom River and the middle course of the
Lena River south of the settlement Chapaevo. The Ura anticline
(Fig. 27.1a) is located to the east of the Lena River along the Ura
River. Both anticlines have flat crests and moderate limb dips
(30–608). Rocks of the Bol’shoy Patom Fm have been slightly
altered, mostly by early epigenetic processes. Faint cleavage of the
shale is developed only at the southwestern end of the Zheday fold.
The Patom SGr. was deposited within the bay-like passive
margins of the Siberian craton. The supergroup disconformably
overlies older rift structures filled by conglomerates, sandstones
and volcanic rocks (metamorphic andesites and andesito-basalts).
The supergroup succession points to the cyclic progradation of a
sedimentary wedge from craton margins into the basin (Chumakov
et al. 2007).
Stratigraphy
The Patom SGr. comprises a siliciclastic and carbonate succession
divided into three parts (from base to top): the Ballaganakh, Dal’-
nyaya Tayga and the Zhuya Groups (Gr.). The general succession
and lithostratigraphy of the two upper groups are shown in
Figure 27.2 and Table 27.1. The Bol’shoy Patom Fm. is the
lower unit of the Dal’nyaya Tayga Gr. Lateral step-by-step litho-
stratigraphic correlations of the groups and their formations,
using some robust lithological and isotopic markers, allow most
formations to be traced more than 1000 km along the marginal
folded arch of the Patom fold system. Beside the Ura Uplift, dia-
mictite formations also occur in the lower part of the Dal’nyaya
Tayga Gr. in two other uplifted areas of the folded arch: at its
northwestern flank (Bol’shaya Chuya R. – ‘Dzhemkukan’ Fm.)
and its eastern flank (Dzhelinda and Chara Rs. – Nichatka Fm.).
There were three main cycles of progradation of Patom SGr.
sedimentation represented by the Ballaganakh, Dal’nyaya Tayga
and Zhuya Groups, respectively (Chumakov et al. 2007). The
lower Ballaganakh Gr. is up to 6.5 km thick in the inner part of
the Patom arch. It thins to 0.5 km at the craton margin and thins
out completely on the craton. The group consists of conglomerates
at the base, sandstones in the more substantial middle part and car-
bonates in the upper part (Mariinskiy Fm.). The Mariinskiy Fm.
contains limestones and shales in the inner part of the Patom
arch and stromatolitic and oolitic dolomites of carbonate platforms
at the margins of the craton. The Dal’nyaya Tayga Gr. was depos-
ited conformably on the Mariinskiy Fm. in the deep anoxic basin of
the Patom zone and bordering arch. Margin disconformities occur
at the base of the Dal’nyaya Tayga Gr. Near the edge of the craton
the Mariinskiy Fm. was preserved as thin erosional relicts only.
The Dal’nyaya Tayga Gr. consists of black slates and limestones
and some sandstones in the Patom basin and in its margin fold
arch. The Dal’nyaya Tayga Gr. is divided in this region into
three formations (from the bottom): Dzhemkukan, Barakun and
Valyukhta. To the NE, in the Ura uplift, black shales and sand-
stones of the Dzhemkukan Fm. give way to diamictites and associ-
ated rocks of the Bol’shoy Patom Fm. The lower part of the

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 309– 316. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.27
310 N. M. CHUMAKOV ET AL.

Fig. 27.1. Geographic distribution and structural framework of the Bol’shoy

Patom Fm. Key: 1, Jurassic System and post Jurassic deposits; 2, Ordovician
System; 3, Cambrian System; 4 –8 Neoproterozoic: 4, Tinnaya and Zherba
Fms.; 5, Zhuya Gr.; 6, Upper and middle parts of Dal’naya Tayga Gr.; 7,
Bol’shoy Patom Fm. (lower formation of Dal’naya Tayga Gr.); 8, Ballaganakh
Gr.; 9, main faults; a, Ura Anticline; b, Zheday Anticline.

Valyukhta Fm. is replaced here by siltstones and limestones (Ura

Fm.) and its upper part is replaced by stromatolitic and oncolitic
dolomites and limestones of the carbonate platform (Kalancha
Fm.). Similar lateral facies changes in the lower and upper parts
of the Dal’nyaya Tayga Gr. can also be observed on uplifted
parts of the eastern and western flanks of the Patom arch (Dzhe-
linda and Bol’shaya Chuya Rivers, accordingly) and at the
western edge of the Aldan Shield beside the Patom arch (Chara
River). Carbonate platforms of the upper part of the Zhuya Gr.
drape all older deposits from the Patom arch up to the edge of
the Aldan shield, thinning in this direction from 600 m to 200 m.
Glaciogenic deposits and associated strata
The thickness of the Bol’shoy Patom Fm. is up to 1100 m. Differ-
ent kinds of diamictites make up c. 60– 80% of the sections as well
as subordinate sandstones, mudstones, siltstones, conglomerates
and conglo-breccias. It is impossible to describe here the detailed
succession of the formation owing to its great thickness and the
frequent intercalation of rocks of different types (Fig. 27.3). In
general, the formation can be divided into three informal members.
Following geological sections of the Bol’shoy Patom Fm and
enclosing deposits in the Ura Uplift (from top to base):
Lower part of Barakun Fm.
Member II: black shales with intercalations of black limestones
(.200 m)
Member I: dark grey, thinly bedded pelitomorphic dolomite that
conformably overlies unit 4 (c. 6 m)
Bol’shoy Patom Fm.
Member III: at the upper boundary of the member, thin polymictic
silty, fine sandstone occurs with rare coarse grains (1 mm);
below are dark-grey massive and sometimes faintly bedded dia-
mictites; rare slump structures, small sandstone and conglomer-
ate lenses and interbeds (120 –180 m)
Fig. 27.2. Lithostratigraphic section, palaeontological data and C- and
Sr-isotope variations in carbonate rocks of the upper part of the Patom SGr. and
lowermost Cambrian system. Key: 1, diamictites (glacial and reworked glacial
deposits); 2, carbonate conglo-breccias; 3, conglomerates; 4, sandstones; 5,
siltstones; 6, limestones; 7, sandy limestones; 8, oolitic limestones; 9,
stromatolitic limestones; 10, dolomites; 11, marls; 12, Pertatatataka microfossil
assemblage; 13, Nemakit-Daldyn small shelly fossils; 14, Early Cambrian
fossils (sunnaginicus biozone); 15, erosional boundaries; 16–18, d13C
variations in sections of the Patom SGr. (Pokrovsky et al. 2006): 16, Ura Uplift;
17, Zhuya R.; 18, Boreholes at the left bank of Lena R.; d13C variations in late
Cryogenian –Cambrian (Vendian-Cambrian) sections: 19, Ura Uplift (Sochava
et al. 1996; Pelechaty 1998); 87Sr/86Sr: – 0.70725 – 0.70855 (Gorokhov et al.
1995; Vinogradov et al. 1996; Pokrovsky et al. 2006).
Member II: frequent intercalation of graded bedded sandstones,
siltstones, carbonate conglo-breccias and dark grey faintly
bedded diamictites with slump structures, rare massive diamic-
tites, graded diamictites; there are intercalations of cross-bedded
fine sandstones; lenses and scours filled by sandstones and con-
glomerates; rare dropstones, slump structures (150 –320 m)
Member I: intercalation of dark grey massive and faintly bedded
diamictites with graded bedded sandstone and mudstones beds
and small sandstone lenses. Rare slump and small cross
bedding structures (450 –700 m)
Upper part of Mariinskiy Fm.
Grey sandy limestone (.200 m)
 BOL’SHOY PATOM FORMATION
Table 27.1. Succession of the Patom SGr. in the Ura Uplift
Zhuya Gr.
Chencha Fm. White, pale and pink micritic, stromatolitic and oolitic Limestones, silty and sandy limestone in upper part 450– 600 m
Nikol’skoe Fm. Red and green thin-bedded marls with intercalations of limestone; sandstone member at base 500– 600 m
Dal’nyaya Tayga Gr.
Kalancha Fm. Dark grey limestone and pale stromatolitic, oncolitic and oolitic dolomites; siltstone intercalations 550– 60 m
Ura Fm. Grey bedded mudstones, siltstones with intercalations of dark limestone in upper part and fine sandstones in lower part 350 m
Barakun Fm. Black limestone, partly oolitic and brecciated in upper part; black shales with intercalation of black limestone in lower part; thin-bedded dark
dolomite at the base 600 m
Bol’shoy Patom Fm. Diamictites, shales, sandstones, minor conglomerates 900–1100 m
Ballaganakh Gr.
Mariinsky Fm. Grey limestones, sandy limestones and black shales 500 m
Bugarikhta Fm. Sandstones, grey shales, minor gravelly sandstones .1000 m
Diamictites
Diamictites have a grey or dark grey (almost black) fine matrix
with randomly dispersed, usually paler clasts. The matrix ranges
between silty sandstone to sandy mudstone. Diamictites with a
mudstone matrix are darker. Matrix is arkosic (quartz grains,
30– 70%; plagioclase, 20–58%; microcline, up to 18%). Minor
components include quartzites (1 –12%), gneisses and granitoids
(2 –10%), carbonates (1 –6%) and mica (1– 5%). Roundness of
sand grains varies, but in general is low (less than 1.5 on the
five-number scale of Khabakov 1933, and 0.45 on the scale of
Krumbein & Sloss 1951). Grains are cemented by sericite, rarely
carbonate, chlorite and detrital biotite.
Clast content varies from 1 to 15%, commonly 3 –10%. Small
and medium-sized pebbles are most typical, cobbles and boulders
are also common with a few blocks reaching 1.1–1.7 m across.
Clasts are randomly scattered in the matrix, often forming nests
of 3–4 stones and occasionally thin stone lags. Clast composition
is quite uniform: granitoids (45 –65%), granite-gneisses (20– 35%)
and gneisses (5 –30%). Other clast types (5 –15%) include lime-
stones, dolostones, sandstones, amphibolites, quartzites, quartz
and schists. The source area for the granitoids and gneisses was
311
the northwestern Aldan Shield, which at that time would have
been situated no less than 100 km to the east of the Ura uplift
and crystalline basement of the southern part of Siberian craton.
Clast roundness varied between 0 and 4, but mean roundness is
generally between 0.7 to 1.2 (using the Khabokov 1933 scale).
In general, crystalline stones are more rounded (1 –3), although
some angular and subangular slabs of granites and gneisses are
found too. Two main types of diamictites can be identified:
massive and stratified (Chumakov & Krasil’nikov 1991).
Massive diamictites. Massive diamictites are the most typical rocks
of the Bol’shoy Patom Fm., particularly in the lower and upper
members. They form 50 –70% of formation sections. The thick-
nesses of individual units of these diamictites vary from 3 to
70 m. Only indistinct clast concentrations can be recognized some-
times as diffuse horizons or lenticular bodies. Large boulders occur
individually or as nests at certain stratigraphic levels. Small lenses
(0.1 –0.3  1– 2 m) of sandstones or sandy siltstones of slightly
concavo-convex shape occur rarely in some massive diamictite
units. Commonly, the lenses have a gradational lower boundary
and a sharp upper boundary sometimes with very thin silt
Fig. 27.3. Composite sections of the
Bol’shoy Patom Formation (parts of
sections not limited by line from right side
are based on eluvial debris and small
dispersed outcrops, white indicates absence
of outcrops). Key: 1, massive diamictites; 2,
faint bedded diamictites and diamictites
with relicts of bedding; 3, carbonate
conglo-breccies; 4, conglomerates; 5,
gritstones; 6, gravelly sandstones; 7,
sandstones; 8, sandstones with traces of
thermal metamorphism; 9, siltstones; 10,
sandy limestones, sandstones and siltstones
containing many carbonate grains; 11,
limestones; 12, dolomites; 13, crystalline
and sedimentary clasts; 14, carbonate clasts;
15, cross-bedding; 16, wavy bedding; 17,
normal graded bedding; 18, reverse graded
bedding; 19, sandstone lenses; 20, erosional
channels and scours; 21, ripple marks; 22,
striated boulders; 23, slump rolls; 24, slump
boudins; 25, slumpfolds; 26, dropstones; 27,
lonestons; 28, stone nests; 29, broken
stones; 30, sole marks; 31, stone fabrics
(hatch lines from Lungersgauzen 1963); 32,
dominant direction of cross-bedding (hatch
lines from Lungersgauzen 1963).
312 N. M. CHUMAKOV ET AL.
laminae. Some lenses have an erosional lower boundary with basal
grits, pebble conglomerates or single boulders (Fig. 27.4a).
Some of the massive diamictites contain rare small fragments of
sandstone or siltstone beds partly reworked by slumps as folds,
rolls, boudins, ragged fragments and other relicts of sedimentary
beds. Some slump fragments have faint and gradational boundaries
with the matrix, forming clods, clots and ‘noses’, implying slumps
of semi- and unlithified sediments. Massive diamictites containing
slump structures constitute 4– 30% of sections.
Stone nests and broken stones are common. Flat surfaces of
single quartzite boulders show one or two systems of longitudinal
sub-parallel grooves and striae of different sizes. Elongated stones
in some massive diamictites had primary orientations in one predo-
minant direction and one or more crossing it in subdominant direc-
tion(s) (Fig. 27.3). At Lena River, the predominant direction has a
NW –SE or north –south, or more rarely a NE –SW or WNW – ESE
orientation. At Ura River, the predominant direction has a north –
south orientation.
Stratified diamictites. Stratified diamictites are of two types: bedded
and graded-bedded. Bedded diamictites have matrices that are
muddy, silty or sandy to varying degrees, and exhibit a range of
clast sizes. Some bedded diamictites have a laminated muddy
matrix. The thicknesses of individual laminae vary from 0.01 to
1 cm. Rare pebbles in these rocks are angular or subangular.
Another form of bedded diamictites has a silty to sandy matrix
with bed thicknesses from 0.1 to 0.4 m or more. Most such beds
have a lenticular or irregular form (Fig. 27.4b).
Graded-bedded diamictites are characterized by clear vertical
changes in the granulometry of matrix and clasts. The thicknesses
of graded-bedded diamictites vary from 1 to 60 cm. These diamictites
are often related to the lower or middle parts of cyclic alternations of
diamictites, sandstones and siltstones (Fig. 27.4c, d). The thickness
of individual cycles varies from 0.5 to 3 m, while the total thickness
of the cyclic units can reach tens of metres. Some cycles have normal
grading (Fig. 27.4c), whereas others have reverse grading in the lower
parts and normal grading in the upper parts of the cycles (Fig. 27.4d).
In this latter case there are often rip-up structures with abundant
mudstone chips at the base of cycles. Large stones or fragments of
sedimentary rocks, similar to rafts of debris flow, occur sometimes
in the upper or middle parts of the cycles.
Associated sedimentary rocks
Beside diamictites, the Bol’shoy Patom Fm. contains subordinate
rocks: sandstones and grits, conglomerates, siltstones and
conglo-breccias.
Sandstones and grits. These make up between 10 and 25% of sec-
tions. The mineral composition of sandstones is similar to that of
the diamictite matrix. Limestone grains occur only in the middle
member of the formation on the western limb of the Zheday
fold, where beds of fine sandstones occur with between 10 and
30% limestone clasts. The thicknesses of sandstone beds generally
range from 2 to 10 m. In the Lena River section, fine- and medium-
grained sandstone forms units up to 40 m thick. There are three
types of sandstones. The first main type of sandstones exhibits pro-
nounced normally graded bedding: from grits and coarse sand-
stones at the base to medium and fine-grained sandstones at the
top. Thin siltstone laminae occasionally overlie these sandstones.
Graded-bedded sandstones are often associated with graded-
bedded diamictites. Thin beds of reverse graded-bedded sand-
stones occur rarely.
The second type of sandstones has restricted development in
the middle member of the Bol’shoy Patom Fm. They are
thinly bedded silty, fine-grained sandstones, often with small
climbing ripple cross-bedding. These sandstones form individual
units up to 40 m thick. The cross-beds dip generally to the south
at Ura River and to the NW at Lena River. The third type of
sandstones includes fine, thinly bedded sandstones with abun-
dant carbonate grains.
Conglomerates. These occur rarely as thin lenses. It seems that they
are residual, a result of the winnowing of diamictites by bottom
currents. Other rare types of coarse sandstones, grits and conglom-
erates fill erosion channels 0.5–10 m deep and several tens of
metres wide. In places, the channels have rather steep slopes.
Mudstones and siltstones. These form beds, and more rarely units,
of between a few metres and a few tens of metres thick. The
rocks exhibit rhythmic lamination and contain dispersed lone-
stones and small till pellets. Some lonestones can be identified as
dropstones, displaying splash features, with rupture of laminae
below and draping of laminae above the clasts (Fig. 27.4e).
Carbonate conglo-breccies. These form a number of beds from 0.1
to 0.5 m thick in the middle member of the formation. They have a
siliciclastic, carbonate-bearing sandstone matrix and contain abun-
dant (from 30 to 70%) angular and subangular fragments (from
small pebbles to blocks 1.2 m across) of oolitic and stromatolitic
dolomites and limestones. The matrix of the conglo-breccies has
normal and reverse graded bedding. Some large stones occur in
the middle or upper parts of the beds and are similar to rafts of
debris flows.
Fig. 27.4. Some typical structures in
diamictites of the Bol’shoy Patom
Formation. (a) Sandstone lens with boulders
in silty diamictites. (b) Two thin irregular
beds of silty-sandy diamictites occur in silty
diamictite. (c) Normal grading of cycle. (d)
Reverse grading in lower and normal
grading in upper part of cycle. (e) Dropstone
and till pellets in laminated mudstone.1,
stones of granite and sandstone; 2, massive
dimictites; 3, bedded diamictites; 4, shale
chips; 5, sandstone; 6, shale.
 BOL’SHOY PATOM FORMATION
Boundary relations with overlying and underlying
non-glacial units
The upper boundary of the Bol’shoy Patom Fm. is conformable. It
was observed in two places. The best locality is situated 7.5 km
from the mouth of the Ura River on its northern bank, and the
second one is located at the beach on the southern side of the
Lena River 6 km upstream of the Bol’shoy Patom River mouth.
At Ura River, diamictite with a sandy mudstone matrix and rare
clasts gradually change upsection, at first to silty coarse sandstone
with dispersed pebbles and then to silty fine polymictic sandstone
with rare coarse grains. Transitional beds are c. 1 m thick. Fine
sandstone is conformably overlain by the first dolomite member
of the Barakun Fm. This dark grey or black laminated dolomite
is peloidal, with small asymmetric anticline-like structures
(0.5  0.5 m) with sharp angular aggrading crests resembling
‘tepees’.
The lower contact of the Bol’shoy Patom Fm. has not been
observed in outcrop, but the lower boundary of the Dal’nyaya
Tayga Gr. is believed to be erosional, at least along the basin
margins. This opinion is based on the existence in the Bol’shoy
Patom Fm. and in the lower part of the coeval ‘Dzhemkukan’
and Nichatka Fms. of big carbonate clasts of rocks of the Mariins-
kiy Fm. (Chumakov 1993; Ivanov et al. 1995 and others).
Chemostratigraphy
The C-isotope profile of carbonates of the Patom SGr. exhibits an
alternation of high and low d13C intervals (Fig. 27.2) with two
extraordinary features: large-amplitude d13C variation, exceeding
21‰ (– 13.5 to 8.3), and an enormous thickness (up to 1000 m) of
13
C-depleted strata (d13C, , –7.5) Zhuya Gr. (Pokrovsky et al.
2006).
The Mariinskiy Fm., underlying the Bol’shoy Patom (Dzhemku-
kan) Fm., exhibits a 600 m positive ‘plateau’ around d13C  7–
8‰. Negative d13C values have been found in the carbonate-
bearing sandstone of the upper part of the Bol’shoy Patom Fm.
at the Ura River (28.8‰) and in the lower part of Dzhemkukan
Fm. at the Zhuya River (29.1‰) where diamictites are absent.
Cap dolostones at the Ura River have moderately low d13C
values from –3.3 to –4.2‰. Upsection there is a second positive
‘plateau’ within the Barakun (d13C  6–8‰) and Valyukhta
(d13C  4–6‰) Fms. with an overall thickness near 1.5 km,
which is overlain by the 300-m-thick Nikol’skoe Fm. consisting
of variegated micritic and peloidal limestones, marls and oncolitic
limestones, which are all depleted in 13C, with d13C values as low
as –13.5‰. The strongly 13C-depleted (in the range d13C  –7.5
to 10‰) carbonates continue upwards, where they also character-
ize the Chencha Fm., a c. 600–700-m-thick unit of micritic lime-
stones, stromatolitic limestones and pink oncolitic limestones.
Ultrahigh-d13C carbonates of the Dal’niyaya Tayga Gr., and
ultralow-d13C of the Zhuya Gr. have been traced around the
Patom Basin to the east of this basin (at Chara River and the Chara-
Tokko watershed) without significant variation in C-isotope com-
position. The thickness of high-d13C dolomites of the Sen’ Fm.
(correlative of Ura and Kalancha Fms.) at the Chara River does
not exceed 200–250 m, and low-d13C limestones of the Torgo
Fm. (correlative of Chencha Fm.) extend over 300– 350 m.
Further to the east, the Torgo Fm. (150 m) partly consists of dolo-
mites with the same d13C ( –7.5 to –10), but higher d18O (up to
30‰; Pokrovskii et al. 2006).
High Sr contents are typical for limestones of the Patom SGr.
They reach 11 700 ppm in the Chencha Fm., 2500 ppm in the
Barakun and Valukhta Fms., 2200 ppm in the Mariinskiy Fm.;
only in the Nikol’skaya Fm. do they not exceed 400 ppm. The
range of measured 87Sr/86Sr ratios (data from Gorokhov et al.
1995; Vinogradov et al. 1996; Pokrovskii et al. 2006 and their
unpublished data) in the limestones of the Chencha Fm. is
313
relatively small (0.7079 – 0.7087), but is greater lower down in
the succession: Nikol’skaya Fm. (0.7079 –0.7094); Valukhta Fm.
(0.70765 –0.7092); Barakun Fm. (0.7025 –0.7102); Mariinskiy
Fm. (0.7075 –0.7092). Taking into account the low measured
Rb –Sr (,0.01), Mn –Sr (,0.1) and Fe –Sr (,2) ratios, we con-
sider that 87Sr/86Sr (min.) ¼ 87Sr/86Sr (primary) for all formations
excluding, probably, the Mariinskiy Formation.
In general, isotope stratigraphic data from the upper part of
the Patom SGr. show an increase in lowermost 87Sr/86Sr ratios
from high-Sr limestones of from 0.70725 in the Barakun Fm. to
0.7079 –0.7080 in the Nikol’sk and Chencha Fms.
Palaeolatitudes and palaeogeography
Systematic palaeomagnetic investigations of the Patom SGr. were
made twice at the Paleomagnetic Laboratory of the Geological
Institute of the RAS, Moscow (by S. V. Shipunov) and then at
the Paleomagnetic Laboratory of the US Geological Survey in
Flagstaff (by D. Elston). All rocks of the supergroup appear to
have been remagnetized during Palaeozoic times.
Geochronological constraints
From the beginning of the 1960s up to 2005, the Patom SGr. was
dated on the basis of stromatolites and microphytolites to be
Middle and Late Riphean (Mesoproterozoic-Cryogenian) (Zhurav-
leva et al. 1961; Keller 1963; Chumakov 1993; and others) or Late
Riphean (Tonian-Cryogenian) (Khomentovskiy et al. 1998). Pre-
sumed correlation between the Russian stratigraphic chart on
Late Precambrian and the stratigraphic chart of the International
Commission on Stratigaphy is shown in Table 27.2.
First suggestions of a Vendian (late Cryogenian-Ediacaran) age
for the upper part of the Patom SGr. were made by J.K. Sovetov,
who correlated tillites of the Prisayan’ya (South Siberia) with the
Bol’shoy Patom Fm. (Sovetov 2002). The Zherba and Tinnaya for-
mations overlying the Patom SGr. were suggested to be Vendian
(Ediacaran –lowermost Cambrian) by all investigators of the
region. This opinion was recently confirmed by the discovery of
small shelly fossils of the Nemakit-Daldyn stage in the uppermost
part of the Tinnaya Fm. and fossils of the Cambrian Sunnaginicus
Biozone of the Tommotian Stage in the lowermost part of the
Nokhtuysk Fm. (Khomentovsky et al. 2004).
There are no precise radiometric age data pertaining to the
Patom SGr. Numerous data were obtained using the K –Ar and
Rb –Sr methods on whole sedimentary rocks of the upper part of
Dal’nyaya Tayga Gr., revealing strong epigenetic rejuvenation
(452 –531 Ma; Vinogradov et al. 1996). A wide range of ages
was obtained on gabbro-dolerite sheets intruded between the
Bol’shoy Patom and Barakun Fms. The hypabyssal nature of the
sheets implies that they were formed at depths of 2 –4 km.
Dating by the K –Ar whole-rock method shows ages between
520 and 633 Ma (Oleynikov et al. 1983). These data can represent
only very approximate age constraints. Taking into consideration
the thickness of overlying deposits, it is possible to suggest that
Table 27.2. Correlation between Vendian and Ediacaran Systems
Cambrian Lower Tommotian Stage
Cambrian Lower
Nemakit-Daldynian Horizon
Upper Kotlin Horizon
Vendian Redkino Horizon
Ediacaran
Laplandian
Glusk Gaskiers
Lower Horizon
Upper Blon'
Lower Blon'
Cryogenian
Upper Riphean
glacial horizons
314 N. M. CHUMAKOV ET AL.
the age of the sheets is close to the age of the upper part of the
Patom SGr. or deposits immediately overlying. A wide range of
ages was also obtained using the K –Ar glauconite method for
the Zherba Fm. overlying the Patom SGr., giving 570–604 Ma
(Ivanov et al. 1995, corrected with constants from Steiger &
Jäger 1977).
New biostratigraphic and chemostratigraphic data allow us to
reinterpret radically the ages of the Dal’nayay Tayga and Zhuya
Groups. Ediacaran fossils Beltanelloides sorichevae were found
in the lower part of the Barakun Fm. at Ura River 8.5 km from
its mouth (Leonov & Rud’ko in press). New maceration and rein-
vestigation of very rich microfossil assemblages from shales of the
Ura Fm. collected by us on the northern bank of the Ura River,
3.5 km from its mouth, point to similar age (Vorob’eva et al.
2007). This assemblage contains abundant large acanthomorphic
acritarchs of genera Ericiasphaera, Tanarium, Appendisphaera,
Meghystrichosphaeridium, Sinosphaera and taxa very similar to
the genera Alicesphaeridium, Variomargosphaeridium, Dicrospi-
nasphaera and others. Ericiaspaera adspersa dominates the
biota, while Tanarium conoideum, Appendisphaera tenuis, ?Sino-
sphaera rupina and Meghistrichosphaeridium hadianensis also
occur. There are also numerous morphologically complicated,
new acanthomorphic forms. Most of the listed taxa are known
from a number of post-glacial formations: the post-Olympic
(?post-Marinoan) Pertatataka Fm. of Central Australia (Gray
2005), the post-Nantuo Doushantuo Fm of South China (Zhou &
Xiao 2007), the post-Blaini Infrakrol Fm. of the Lesser Himalayas
(Tiwary & Knoll 1994), as well as the pre-Redkino Vychegda Fm.
of the Russian Plate (Veis et al. 2006; Vorob’eva et al. 2006) and
the upper part of the Nepa Horizon of the Siberian Plate. Associ-
ations of such microfossils are usually referred to as ‘Pertatataka-
type biota’. If confirmed, the Ura biota would appear to be similar
to the second microfossil zone of ECAP of K. Grey of Central Aus-
tralia (Vorob’eva et al. 2007). The presence of acanthomorphic
acritarchs in the Ura microbiota has been reported previously by
some other researchers. They considered them to be Palaeozoic
forms penetrating into the Ura Formation, or interpreted the micro-
biota as Late Riphean (Cryogenian) but with many Vendian
elements (for a review see Nagovitsin et al. 2004; Vorob’eva
et al. 2007). The conclusion on the Late Riphean (Cryogenian)
age of the Ura microbiota was made by Nagovitsin et al. (2004),
particularly from the finding of the form described incorrectly as
Trachyhystrichosphaera aff. aimika. In the opinion of Vorob’eva
et al. (2007), the forms described in this paper (Nagovitsin et al.
2004) are really fragments of Early Vendian taxa, such as Tanar-
ium, Alicesphaeridium and others.
The assignment of the Dal’naya Tayga Group to the Early
Vendian or Late Cryogenian –Early Ediacaran has been confirmed
by a minimal age of 600 + 10 Ma from detrital zircons from the
upper part of the Khomolkho Fm. (U –Th – Pb, LA-ICPMS,
Meffre et al. 2008). The Khomolkho Fm. is a stratigraphic ana-
logue of the Valyukhta Fm. (Zhadnova 1961; Keller et al. 1967;
Ivanov et al. 1995, and many others). Khomentovsky & Postnikov
(2001) even suggest that the Khomolkho Formation is a strati-
graphic analogue of the Ura Fm. This date therefore indicates a
maximal age of Valyukhta and Ura Fm.
These data allow us to suggest correlation of the Bol’shoy Patom
Fm. with the Nantuo, Blaini and Olympic Fms. In this case, the posi-
tive d13C anomaly in the upper part of the Dal’nyaya Tayga Gr. may
be correlated with a similar anomaly in the lower part of Doushantuo
Fm., while the negative anomaly in the Zhuya Gr. may be correla-
tive with a similar anomaly ‘Dounce’ in the uppermost member of
the Doushantuo Fm. (Zhou & Xiao 2007; Zhu et al. 2007).
87
Sr/86Sr ratios of the uppermost part of the Dal’nyaya Tayga Gr.
and of the Doushantuo Fm. (Yang et al. 1999) are consistent with
this interpretation (minimum values of 0.70765 and 0.7077, respect-
ively). Correlations with the Nantuo and the Doushantuo formations
allows us to constrain the age of the Dal’nyaya Tayga and Zhuya
Groups to between 636/660 and 550 Ma.
Ultra-low-d13C carbonates of the Zhuya Gr. could be correlated,
besides anomaly ‘Dounce’, with the negative d13C anomalies of
the Shuram and Buah formations (Oman) and probably the
similar anomaly in the Wonoka Fm. (S. Australia). Sedimentologi-
cal and geochemical parameters suggest that the Zhuya Gr. had
preserved its primary C-isotope composition and was formed
over at least 5 million years (a similar duration is suggested for
the Dounce anomaly). The negative C-isotope anomaly of the
Zhuya Gr. exceeds all known C-isotope anomalies both in ampli-
tude and thickness of strata, providing evidence of unprecedented
‘contamination’ of the water column by low-13C organic carbon.
If the Nantuo Formation is really coeval with the Yerelina Sub-
group, which remains to be proven, then the upper part of the
Dal’yaya Tayga Gr. and the Zhuya Gr are Ediacaran in age.
According to the Russian stratigraphic chart, both the Dal’yaya
Tayga and Zhuya groups belong to the Lower and Middle Vendian.
Discussion
Three rock associations can be recognized in the Lower Vendian
(Late Criogenian) Bol’shoy Patom Fm. The first and main rock
associations comprise massive diamictites with subordinate
bedded diamictites and laminated mudstones and siltstones. This
association is typical of the lower and upper members of the
Bol’shoy Patom Fm. Intercalation with laminated mudstones and
siltstones, bedded diamictites and the existence of sandstone
lenses related to winnowing of diamictites by local bottom currents
implies that massive diamictites were deposited in a rather quiet
basin. Laminated mudstones and siltstones undisturbed by waves
and traces of slumping show that the basin floor was rather deep
and sloped. It was a large basin, probably open to the sea, as the
Bol’shoy Patom Fm. is gradually replaced towards the south and
SW by the thick Dzhemkukan Fm. comprising distal turbidite
slates with subordinate sandstones (c. 1000 m).
Unsorted matrix, a very wide range of size and roundness of dis-
persed erratic clasts, the presence of glacial grooves on boulders,
dropstones and till pellets in laminated mudstones and siltstones
are all evidence for intense ice-rafting in the basin. A wide size
range and low mean roundness of stones are not typical for
debris rafted by seasonal or perennial sea ice, which commonly
transports well-rounded and well-sorted stones picked up from
coastal beaches. Therefore, it is possible to conclude that the
first rock association was deposited mainly by glacier and
iceberg rafting of material derived from shelf glaciers and outlet
glaciers. Glacier-like clast fabrics in some massive diamictites
point to the presence of grounded shelf glaciers moving in
north –south and NW – SE directions. The basement of the Siberian
Craton was the main source of clastic deposits in the Bol’shoy
Patom Fm. This is clear from the predominance of granitoids,
granite-gneisses and gneisses, which make up almost all of the
cobbles and boulders in the massive diamictites. The source of car-
bonate conglo-breccias was from carbonate platforms bordering
the Patom basin in Mariinsky time. Their erosion was related to
a fall in sea level during the Bol’shoy Patom glaciation. Cross-
bedding in sandstones allows us to suggest that in the Lena –
Bol’shoy rivers area, glacial meltwaters flowed towards the
north and NW and in the Ura River area to the SW and south.
The presence of wet-based, tidal and outlet glaciers was probably
the main cause of the rather restricted areal distribution of iceberg
facies in the Patom Basin. In such environments, the main dis-
charge of glacial debris from shelf glaciers and icebergs occurs
in a proximal position. In general, the first rock association
implies glacial and proximal glacial marine palaeoenvironments.
The second and third rock associations are found predominantly
in the middle member of the Bol’shoy Patom Fm. The second rock
association comprises intercalations of turbidite, debris and grain
flow sandstones and massive or bedded diamictites. Rare cross-
bedded coarse sandstones and conglomerates filled channels cut
 BOL’SHOY PATOM FORMATION
into graded-bedded strata. Laminated mudstones with dropstones
also occur. The second rock association was probably deposited
as submarine fans. Their formation may be related to high-activity
debris, grain and turbid flows related to sub-glacial or glacial rivers
during glacial retreat.
The third rock association consists of thinly bedded, fine-
grained sandstones with small-scale cross-bedding of climbing
ripple-type and finely laminated sandstones with abundant carbon-
ate grains. These deposits are probably related to interchannel
interfans and distal parts of submarine fans with bottom currents.
The sandstones with abundant carbonate grains could be distal
turbidites transported by carbonate conglo-breccia debris flows.
The third rock association may be related to a rise in sea level.
The above-mentioned data allow us to conclude that the
Bol’shoy Patom Fm. was deposited in the marine Patom basin
during glaciation by and with the strong influence of ice sheets,
iceberg rafting and glacial melt waters. The glaciogenic origin of
the Bol’shoy Patom Fm. is supported by the presence of a
typical cap dolomite conformably overlying it (first member of
Barakun Fm., see section ‘Boundary relations with overlying and
underlying non-glacial units’) that is very similar to such dolomites
elsewhere in Asia, Svalbard, West Africa, Australia and other
regions of the world and by well-founded correlation of this for-
mation with the Nichatka Fm., which crops out on the southeastern
flank of the folded arch and western margin of the Aldan shield. In
Nichatka Fm., continental glacial facies have been discovered
(Chumakov 1993). During deposition of the Bol’shoy Patom
Fm., the southern part of the Siberian Craton was covered by ice
sheets flowing into the Patom Sea.
We thank M. I. Buyakayte for help in field and isotopic researches, and N. G. Vor-
ob’eva and V.N. Sergeev for the investigation of microfossils from our collection.
This research was supported by grants of the Russian Fund of Basic Investigations
(nos 08-05-00433 and 10-05-00294) and Program No 24 of the Presidium of
Russian Academy of Sciences. This represents a contribution of the IUGS- and
UNESCO-funded IGCP (International Geoscience Programme) project #512.
References
Bobrov, A. K. 1979. Stratigraphy and paleogeography of Upper Pre-
cambrian deposits of South Yakutiya. Yakutian Publishers, Yakutsk
(in Russian).
Chumakov, N. M. 1959. Stratigraphy and tectonics of the south-western
part of the Viluy Depression. In: Shatsky, N. S. (ed.) Tektonika
SSSR. IV. AN SSSR, Moscow, 345– 460 (in Russian).
Chumakov, N. M. 1965. Certain Precambrian tillite-like rocks of the
USSR. Izvestiya Academii Nauk SSSR, Seriya geologigeskaya, 2,
83 – 101 (in Russian).
Chumakov, N. M. 1993. Riphean Middle Siberian Glaciohorizon. Strati-
graphy and Geological Correlation, 1, 17 – 28.
Chumakov, N. M. & Krasil’nikov, S. S. 1991. Lithology of Riphean
tilloids; Ura Uplift. Lithology and Mineral Resourses, 3, 58 –78.
Chumakov, N. M., Pokrovsky, B. G. & Melezhik, V. A. 2007. Geologic
history of Patom Supergroup, Late Precambrian, Middle Siberia.
Doclady Earth Sciences, 413, 379–383.
Gorokhov, I. M., Semikhatov, M. A. et al. 1995. Strontium isotopic
composition of Riphean, Vendian and Lower Cambrian carbonate
rocks of Siberia. Stratigraphy and Geological Correlation, 3, 3– 33.
Gray, K. 2005. Ediacaran palynology of Australia. Association of Austra-
lian Palaeontologists. Memoir 31. Association of Australian Palaeon-
tologists, Canberra, 439.
Ivanov, A. I., Lifshits, V. I. et al. 1995. Precambrian of the Patom
Upland. Nedra, Moscow (in Russian).
Keller, B. M. (ed.) 1963. Stratigraphy of SSSR. Upper Precambrian.
Nedra, Moscow, 716 (in Russian).
Keller, B. M., Semikhatov, M. A. & Chumakov, N. M. 1967. Upper
Proterozoic of Siberian Craton and its frame. In: Stratigraphy of
Precambrian and Cambrian of Middle Siberia. Krasnoyarsk
knizhnoe izdatel’stvo, Krasnoyarsk, 71– 83 (in Russian).
315
Khabakov, A. V. 1933. A Brief Guide to Field Studies of Conglomerates.
Nauchno-Tekhnicheskoe Izdatel’stvo, Moscow, 11 (in Russian).
Khomentovskiy, V. V., Postnikov, A. A. & Fayzulin, M. Sh. 1998. Bai-
kalian of stratotype locality. Geology and Geophysics, 39, 1505–1517.
Khomentovskiy, V. V. & Postnikov, A. A. 2001. Neoproterozoic evol-
ution Baikal-Vilyu branch of Palaeoasian Ocean. Geotectonics, 3,
3– 21.
Khomentovskiy, V. V., Postnikov, A. A., Karlova, G. A., Kochnev,
B. B., Yakshin, M. S. & Ponomarchuk, V. A. 2004. Vendian of
Baikal-Patom Upland (Siberia). Geology and Geophysics, 45,
465– 484.
Krumbein, W. C. & Sloss, L. L. 1951. Stratigraphy and Sedimentation.
Freeman, San Francisco.
Leonov, M. V. & Rud’ko, S. V. 2010. The find of Ediacarian fossils in
Dal’nyaya Tayga Group (Ediacaran-Vendian of Patom Highland),
in press.
Lungersgauzen, G. F. 1963. Tillites and tillit-like deposits (in Russian).
In: Keller, B. M. (ed.) Stratigraphy of SSSR. Upper Precambrian.
Nedra, Moscow, 566–577 (in Russian).
Meffre, S., Large, R. R. et al. 2008. Age and pyrite Pb-isotopic
composition of the giant Sukhoi Log sediment-hosted gold
deposit, Russia. Geochimica et Cosmochimica Acta, doi: 10.1016/
j.sca.2008.03.005.
Nagovitsin, K. E., Faizullin, M. Sh. & Yakshin, M. S. 2004. New
forms of Baikalian acanthomorhytes from the Ura Formation of
Patom Uplift, East Siberia. Novosti paleontologii i stratigrafii, 6– 7,
7– 19.
Oleynikov, B. V., Tomshin, M. D. & Kopylova, A. G. 1983. Basic rocks
of Ura anticlinorium (in Russian). In: Oleynikov, B. V. (ed.) Petrol-
ogy and Geochemistry Late Precambrian Intrusive Basic Rocks of
Siberian Platform. Nauka, Novosibirsk, 146– 167 (in Russian).
Pelechaty, S. M. 1998. Integrated chronostratigraphy of the Vendian
System of Siberia: implications for a global stratigraphy. Journal of
Geological Society of London, 155, 957– 973.
Pokrovsky, B. G., Melezhik, B. A. & Buyakayte, M. I. 2006.
Geochemistry of isotopes C, O, Sr and S, chemostratigraphy and
environments of sedimentation of Late Precambrian deposits of
Patom trough. Lithology and Mineral Resourses, 5, 505–530.
Sochava, A. V., Podkovyrov, V. N. & Vinogradov, D. P. 1996. Vari-
ation of carbon and oxygen isotopes in Vendian– Lower Cambrian
carbonate rocks of the Urinsky anticlinorium (South Siberian Plat-
form). Lithology and Mineral Resources, 31, 248–257.
Sovetov, J. K. 2002. Vendian foreland basin of the Siberian cratonic
margin: Paleopangean accretionary phases. Russian Journal of
Earth Sciences, 4, 363–387.
Steiger, R. H. & Jäger, E. 1977. Subcommission on geochronology:
convention on the use ofdecay constants in geo- and cosmochronol-
ogy. Earth and Planetary Sciences Letters, 36, 359– 362.
Tiwary, M. & Knoll, A. H. 1994. Large acanthomorphic acritarchs
from the Infrakrol Formation of the Lesser Himalaya and their
stratigraphic significance. Journal of Himalayan Geology, 5,
193– 201.
Veis, A. F., Vorob’eva, N. G. & Golubkova, E. Y. 2006. First find of
Lower Vendian microfossils at Russian plate: taxonomy composition
and biostratigraphic significance. Stratigraphy and Geological Cor-
relation, 14, C28 – 46.
Vinogradov, V. I., Pichugin, L. P., Bikhover, V. N., Golovin, D. I.,
Muraviev, V. I. & Bujakayte, M. I. 1996. Isotope features and
dating of epigenetic alterations of Upper-Precambrian deposits of
the Urinsky Uplift. Lithology and Mineral Resources, 31, 60 –69.
Vorob’eva, N. G., Sergeev, V. N. & Semikhatov, M. A. 2006. Unicum
Lower Vendian Kel’tma microbiota of Timan: new data on palaeon-
tology of Vendian and its global features Doclady Academii nayk,
410, 366–371.
Vorob’eva, N. G., Sergeev, V. N. & Chumakov, N. M. 2007. New
occurrences of Lower Vendian microfossils in Ura Formation:
problem Patom Supergroup of Middle Siberia. Doclady Earth
Science, 419, 782– 787.
Yang, J., Sun, W., Wang, Z., Xue, Y. & Tao, X. 1999. Variatios in Sr
and C isotopes and Ce anomalies in successions from China: evidence
for the oxygenation of Neoproterozoic seawater? Precambrian
Research, 93, 215– 233.
316 N. M. CHUMAKOV ET AL.

Zhadnova, T. P. 1961. Stratigraphy of north-east part of Patom Highland.
Trudy Tsentral’nogo nauchno-issledovatel’skogo geologorasvedoch-
nogo institute, 38, 103– 123 (in Russian).
Zhou, G. & Xiao, S. 2007. Ediacaran d13C chemostratigraphy of South
China. Chemical Geology, 237, 89 – 108.
Zhu, M., Strauss, H. & Shields, G. A. 2007. From snowball earth to the
Cambrian bioradiation: calibration of Ediacaran-Cambrian earth
history in South China. Palaeogeography, Palaeoclimatology,
Palaeoecology, 254, 1 –6.
Zhuravleva, Z. A., Komar, V. A. & Chumakov, N. M. 1961. Succession
and correlation of Upper Precambrian deposits of West Yakutiya. In:
Bobrov, A. K. (ed.) Data on Geology and Mineral Resources of
Yakutiya ASSR. Issue 13. Yakutiya Geological Survey, Yakutsk,
12 – 28 (in Russian).
 Chapter 28

Late Cryogenian (Vendian) glaciogenic deposits in the Marnya Formation, Oselok Group,
in the foothills of the East Sayan Range, southwestern Siberian Craton

J. K. SOVETOV
Trofimuk Institute of Petroleum Geology and Geophysics, Russian Academy of Sciences, Siberian Branch,
Koptyug Av. 3, Novosibirsk, 630090, Russian Federation (e-mail: SovetovYK@ipgg.nsc.ru)

Abstract: The glaciogenic deposits of the Ulyakha Member (Mb.) referred to as Ulyakha diamictite (tillite), and the associated Late
Cryogenian (Early Vendian) glaciofluvial deposits, are the lower strata of the Marnya Formation (Fm.) of the Oselok Group (Gr.) in
the southwestern Siberian craton, which rest erosionally over the Cryogenian (Late Riphean) Karagassy Group. The Oselok Group depos-
its fill the Peri-Sayan foredeep and their stratigraphic equivalents immediately overlie the basement of the Siberian Platform. The gla-
ciogenic deposits, including diamictite, breccia, boulder conglomerate and sandstone, had multistage sedimentation, were widespread,
and underlie a cap dolomite of the lower Ozerki Mb. The Late Cryogenian (Early Vendian) age of the Ulyakha Mb. is defined by traces of
shallow-marine soft-bodied animals in the middle part of the Ozerki Mb., and by findings of Ediacaran-like Metazoa in the overlying
Bolshaya Aisa Mb. of the Marnya Fm. Deposition of the Ulyakha tillites and associated rocks has been correlated with the Late Cryo-
genian (Marinoan) glaciation based on the d13C patterns in carbonate deposits in the Marnya and Uda formations and their stratigraphic
position much below the Pre-Manykai and Zhuya d13C negative signatures, which have been equated to the Wonoka anomaly.

Glaciogenic and associated deposits in the Sayan region occur at
the base of the Oselok Group in the Biryusa province (Fig. 28.1).
Their stratigraphic equivalents are the basal strata of the Olkha
Fm. in the Irkutsk province (Baikal Group of the Baikal region),
and of the Taseeva and Chapa Groups in the Yenisei Range
(Sovetov et al. 2007a). All these sediments fill Late Cryogenian-
Ediacaran (Vendian) foredeeps echeloned along the southwestern
margin of the Siberian craton. The glacial facies correlate with
poorly sorted sediments in the lowermost sections of the Nepa
and Vanavara Groups in the craton interior and of the Dalnyaya
Taiga Group in the Patom passive margin (Sovetov 2002a, b;
Sovetov & Komlev 2005; Sovetov at al. 2007a). Thus, Late
Cryogenian (Early Vendian) glacial events left their signature
over about 1.8 million square kilometres.
The Oselok Group was originally defined as a formation in the
Sayan region (Khomentovsky 1950) and was recognized as a
group after detailed stratigraphic studies in the 1960s when it
was divided into the Marnya, Uda and Aisa formations (Dubin
et al. 1969; Khomentovsky et al. 1972). These three units are
distinguished from one another and from the underlying rocks by
diamictite and breccia at the base of the Marnya Fm., arkose and
orthoquartzite conglomerate and sandstone at the base of the
Uda Fm. and lithoclastic massive sandstone near the bottom of
the Aisa Fm. The succession of the three formations appears to
persist continuously from the Tagul River in the NW to the Iya
River in the SE of the Biryusa area (Fig. 28.1). The best outcrops
of the Marnya Fm. (400 – 660 m) are found along the sides of the
Uda valley between the Gladkiy Mys site and the Marnya inlet
(Fig. 28.1). Berzin (1967) discovered unsorted mixitite at the
base of the Marnya Fm. along the Uda River but found no traces
of glacial origin in the diamictite. Dubin et al. (1969) and Khomen-
tovsky et al. (1972) first distinguished a member of the Oselok
basal breccia and conglomerate lying over deeply eroded Late
Precambrian (Cryogenian or Upper Riphean) strata of the
Karagassy Group.
The original subdivision of the Marnya Fm. into lithological
members 1 –4 (Dubin et al. 1969) and I –V (Bragin 1985) was
later updated (Sovetov 2002a), and the members received their
proper names and genetic identification. Diamictite was jointly
assigned to the Ulyakha Mb. and faceted-boulder breccia to the
lower Plity Mb. Diamictite and faceted-boulder breccia in the
two members bear numerous traces of glacial origin, including
striated boulders and abrupt facies changes, which has prompted
distinguishing local lithological members (Sovetov & Komlev
2005). Besides sedimentology, the Marnya Fm. (and the Oselok
Group as a whole) was sampled for palaeontology in search
of Ediacaran biota and traces of soft-bodied organisms, and for
C- and O-isotope compositions in cap carbonates. On this basis,
along with global Late Neoproterozoic correlations, the glacio-
genic deposits of the area have been identified as Late Cryogenian
tillite (Sovetov & Komlev 2005; Sovetov 2007).
The discussion below concerns the sedimentology of the
Marnya Fm. (Oselok Group) in its representative sections, compo-
sition of glaciogenic deposits, erosional features, pathways
of clastics transported by ice and meltwater, biostratigraphic
and chemostratigraphic constraints, and palaeomagnetic data
with implications for Late Cryogenian –Early Ediacaran (Early
Vendian) palaeogeography of the Siberian craton.
Structural framework and Neoproterozoic sedimentary
basins of the Sayan region
The study area of the Sayan region occupies the Biryusa catchment
along the southwestern margin of the Siberian craton where the
basement is exposed or locally buried under Neoproterozoic and
Cambrian sediments (Fig. 28.1). The craton in this region
borders a metamorphic belt of Palaeozoic rocks lying at 15 –208
to the basement along the Main Sayan Fault. The province of
Palaeozoic rocks corresponds to the East Sayan uplift and includes
blocks with Precambrian continental crust. The Main Sayan Fault
is a zone of left-lateral strike– slip faults and mylonites, which
apparently initiated in the Neoproterozoic (Cryogenian) when
several terranes rifted off the Siberian craton (Berzin 1967). The
zone experienced at least two episodes of compression during
the Late Precambrian Baikalian (Cadomian) and Early Palaeozoic
orogenies that produced a Late Cryogenian (Early Vendian;
Sovetov 1977, 2002b; Sovetov & Blagovidov 2004) and Middle
Cambrian –Early Ordovician (Anatol’eva 1972; Sovetov unpub-
lished) molasse foreland basins.
The sediments –basement contact is marked by a structural and
metamorphic unconformity. The stratigraphic gap between the
basement and the Cryogenian Karagassy Group reaches 1–
0.8 Ga in the Biryusa and Urik-Iya fault blocks and 2.9– 2.0 Ga
within the Sharyzhalgai terrane where Archaean rocks underlie

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 317– 329. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.28
318 J. K. SOVETOV

Fig. 28.1. Precambrian geology of the Sayan region on the south-western Siberian Craton. (a) Geological map of the region showing structural elements and
lithostratigraphy. (b) Stratigraphic scheme of the Neoproterozoic deposits in the Foothills of the East Sayan Range and their correlation with deposits in the adjacent
regions. Supersequences: Sq 1, Marnya; Sq 2, Greben; Sq 3, Ust-Tagul.
 MARNYA FORMATION, OSELOK GROUP
the basal diamictite of the Late Cryogenian –Early Ediacaran
Olkha Fm.
The oldest basement rocks (granulite, gneiss and schist) belong
to the Archaean Sharyzhalgai Group in the Sharyzhalgai terrane
(Sez’ko 1988). The Early Proterozoic Biryusa, Kan and Dzhu-
glyma terranes are composed of migmatized biotite, amphibole,
garnet granulite and gneiss, granite gneiss and amphibolite.
The Palaeo-Mesoproterozoic Urik-Iya trough basin is filled
with 10 000 m of biotite-quartz phyllite, shale, metasandstone,
conglomerate, tuff, and volcanic rocks (Berzin 1967; Sez’ko
1988). The metamorphic complexes of the Biryusa terrane are
intruded by Palaeoproterozoic granites of the two-stage Sayan
complex dated between 1870 –1734 Ma (Donskaya et al. 2002;
Turkina et al. 2003). Unmetamorphosed Neoproterozoic sedi-
ments fall into two geodynamically different sequences: the
rift-related deposits of the Cryogenian Karagassy Group and
the Late Cryogenian –Ediacaran Oselok Group, which filled a
foreland basin.
The Karagassy Group lies, with a large structural unconfor-
mity, over metamorphic and igneous complexes of the Biryusa
fault block and over low-grade metasediments and volcanic com-
plexes of the Urik-Iya basin. The Oselok Group oversteps deeply
eroded surfaces of the Karagassy formations in the Biryusa catch-
ment; its stratigraphic equivalents (Olkha and Moty Groups) rest
upon the Archaean Sharyzhalgai metamorphics only outside the
rift basin in the Irkut catchment. According to recent data
(Sovetov & Blagovidov 2004, 2006; Sovetov et al. 2007a), the
Karagassy sediments fill the Cryogenian Iya-Tumanshet failed
rift (aulacogen), whereas the Oselok Group is the fill of the
Late Cryogenian –Ediacaran Peri-Sayan foredeep. The fault-
bounded rifts formed at the same time and in the same geody-
namic environment as the Teya-Chapa and Vorogovka failed
rifts 500 km to the NW in the Yenisei Ridge (Sovetov 1997;
Sovetov et al. 2007a). The Karagassy rocks are intruded through-
out the area by small bodies of gabbro dolerite sills, dykes and
stocks of the Nersa complex with an Ar/Ar plagioclase age of
741 + 2 Ma (Gladkochub et al. 2006).
The Late Cryogenian –Ediacaran Peri-Sayan foredeep initiated
upon the Iya-Tumanshet rift and the adjacent craton edge after
regression and deep glacial erosion during the Late Cryogenian
glaciation (Sovetov 2002b; Sovetov & Blagovidov 2004;
Sovetov & Komlev 2005). The Late Cryogenian –Ediacaran sedi-
mentary fill of the Peri-Sayan, Peri-Baikal and Yenisei foredeeps,
together with the coeval terrigenous and terrigenous-carbonate
platform cover, belong to an enormous foreland basin on the
western and southwestern margins of the Siberian craton. The
Peri-Sayan foredeep basin evolved in two stages recorded in
the Late Cryogenian– Early Ediacaran Marnya (1) and Late Edia-
caran Greben (2) regional supersequences (Sovetov et al. 2007a).
The Marnya supersequence consists of glaciogenic deposits,
shallow-marine terrigenous and carbonate rocks, and less abun-
dant fluvial facies. The Early Ediacaran was associated with
transgressions from the SW onto the southern and southwestern
parts of the Siberian craton as a result of the closing Palaeoasian
ocean and clastic transport from the craton interior. During the
Late Ediacaran (Greben) syncollisional stage, the basin experi-
enced transgressions from the NE, and the clastic transport was
from flanking orogens in the western, southwestern and southern
craton margins. The Cambrian history of the Siberian craton
began with the break-up of the Ediacaran (Late Vendian)
orogens, formation of a terrigenous hanging-wall basin, followed
by the development of evaporite and carbonate basins (Sovetov
et al. 2007a).
The Karagassy and Oselok sediments were deformed by post-
Cambrian to pre-Devonian west –east compression (Berzin
1967) that split the sedimentary basins into blocks displaced
along strike –slip, thrust and reverse faults. This deformation
event produced shear zones and related folds, but was not
accompanied by metamorphism.
319
Stratigraphy
The Karagassy and Oselok Groups (Fig. 28.1) in the Biryusa
watershed were interpreted jointly as a single complex (Khomen-
tovsky 1950) before a stratigraphic gap between them and a
complex stacking pattern was discovered, providing grounds for
their subdivision (Dubin et al. 1969).
Karagassy Group
The Karagassy Group, varying in thickness from c. 800 to
c. 2000 m, consists of the Shangulezh, Izan (currently referred to
as Tagul; Egorova et al. 1971; Bragin 1986; Shenfil 1991;
Sovetov & Blagovidov 2006) and Ipsit formations (Khomentovsky
et al. 1972). The Shangulezh Fm. (250 –600 m) lies with an
angular unconformity over the Urik-Iya folded metamorphosed
sediments and volcanic rocks and over high-grade metamorphic
rocks and Palaeoproterozoic granitoids of the Sayan complex.
The three formations are second-order sequences and are separated
from one another by gaps and low-stand erosional surfaces. The
sequences record facies changes from alluvial fans to a terrigenous
carbonate tidal plain and a carbonate tidal shelf with abundant
stromatolite buildups, microphytolite dolarenite and dolomicrite.
Of broad occurrence are the structures of supratidal, subtidal and
peritidal deposits with bimodal flow directions of terrigenous-
carbonate and silt material, minor scouring forms, flat stromatolite
mats, mud cracks and halite glyptomorphs.
Oselok Group
The Oselok Group overlies the Karagassy Group without structural
unconformity but with an erosional contact (locally about 1000 m
of the Ipsit and partly Tagul deposits have been removed). The
group includes the Marnya, Uda and Aisa formations (Dubin
et al. 1969; Khomentovsky et al. 1972; Sovetov & Komlev
2005). The eroded top of the Aisa Fm. underlies, without structural
unconformity, the basal conglomerates of the Ust’-Tagul Fm.
Marnya Formation. The general succession of members, their
replacement and pinching out, have been inferred from correlation
and are reported below following Sovetov & Komlev (2005), with
later additions (Fig. 28.2). The new subdivision updates the earlier
stratigraphic interpretations (members I to V) by Dubin et al.
(1969) and Bragin (1984, 1985).
The Karapchatui Member (I), the oldest in the Marnya Fm., was
discovered by Bragin (1984), and its stratigraphy was updated
through studies by Sovetov in 2006 along the right side of the
Uda River (Karapchatui Brook mouth). The section comprises
0.1– 0.2 m of ferruginous silicified cataclastic rocks at the base
overlain by (1) 3.2 m of massive quartzite faceted-boulder
breccia and white cross-bedded orthoquartzite; (2) 22 m of white
and yellow cross-bedded orthoquartzite with debris in small
fluvial – lacustrine cycles with thin-bedded orthoquartzite and stro-
matolite dolomite; (3) 30–35 m of variegated thinly laminated
quartz sandstone and stromatolitic dolomite; (4) 20– 30 m of
quartz and massive cross-bedded feldspar-quartzose sandstone.
Orthoquartzites give way to breccia in the SW, and also in the
NE, towards the sides of the underlying channelized unconformity.
The Nersa Mb. (18.7 m thick) (Mb I, Bragin 1984), found on the
west side of the Uda River in the Nersa and Kremenshet interfluve,
consists of massive fluvial boulder conglomerate and coarse
pebbly sandstone in the upper section, with clasts of the Karagassy
arkose quartzitic sandstone, stromatolite dolomite, chert and doler-
ite. A c. 22-m-thick sill of the Nersa complex was found 2.1 m
below the Nersa-Ipsit contact.
The Ulyakha Mb. stratotype occurs along the right side of the
Uda River 0.3– 1 km upstream of the Ulyakha mouth, where the
 (after Bragin 1985)*

320
Gladky Mys Ulyakha Kirei Ozerki Karapchatui Plity Bogatyr Kedrovyi

Ikei Fm.
Glacio-fluvial quarzite sandstone

Muksut Mb.
with detritus

500
Dolostone of ephemeral
postglacial lake

Mb. Kagat Mb.

Glacio-fluvial litharenite with
single pebbles
Pescherny

Laminated black and grey-green

mudstone with sandstone
lenses
(after Bragin 1985)

400 Eolian (?) quartzite sanstone dunes

Unyl Mb.
Uda Fm.

Laminated cap-dolomite

Current-rippled dolostone
Nizhnyaya Uda Mb.

Plity
Mb. Grainstone sand waves
300 N=31
Quarzite sand waves and bars

Kedrovy
Marnya Fm. (after Dubin et al. 1969)

Quarzite foreshore bars

Bol’shaya Aisa Mb.

Mb.

J. K. SOVETOV
Plity Ozerki Mb. Sandstone-siltstone tempestite
N=53
N=19 Plity Ipsit Ulyakha Mb.
200 Mb. Fm.
Tide-related channelized sandstone
Mb.

170 m -siltstone and micrite limestone

30 m Nersa Mb. tempestite
20 m
N=52 Gravelly sandstone tempestite
Ipsit
Tygnei Mb.

20 m Fm.
Karapchatui Mb.
Ognit Mb.

Hurricane cross-bedded
Ulyakha Mb.

N=27
N=10 sandstone
20 m
Fluvial cross-bedded gravel
100 Ipsit Palaeocurrent directions and sandstone quartzite and
Tagul
Fm.
Tagul
arkose
Fm. Fm.
Delta plain thin bedded siltstone
Red stromatolitic dolostone with gravel and sandstone
Ozerki Mb.

lenses
Kedrovyi 99 00 Sandy tempestite and Ooids and stromatolitic
70
Nizhneudinsk sand-waves with slumps limestone
Sandstone coarse cross-bedded
Dolerite sills bars and channels of deep-water
m0 Clay-rich diamictite with rivers
Uda River

Siberian
r
ive

predominant quarzitic cobbles Trace-fossils

Craton
aR

60 and boulders
Bogatyr
Dolomitic diamictite with Moulds and imprints of soft-body
us

a b c
54 30 predominant dolomite cobbles animals: a.Cyclozoa, b. Bilaterian,
ry

Plity
Bi

and boulders c. Problematics

Karapchatui
Faceted-boulder breccia
al

Ozerki Glacial erosion surface

ik
Ba

Kirei
25 km
ke

Gladky Mys Ulyakha

La

100 110
Boulder conglomerate 20 m Non-cropped parts of sections

Fig. 28.2. Stratigraphy and depositional systems of the Marnya and Uda formations at different locations along the Uda and Biryusa rivers, showing the lateral variability and stratigraphic relationships of the members from
south to north. Note that the Ikei Formation includes the Aisa Fm. and Kagat and Muksut members, which are differentiated on the basis of colour. The Ikei Fm. is generally no longer referred to in the literature. Inset maps
show localities of stratigraphic logs as well as average palaeo-ice flow direction based on sedimentary structures in glacial and glaciofluvial sediments.
 MARNYA FORMATION, OSELOK GROUP
Member erosively overlies the Tagul Fm. The contact is perfectly
exposed in the Uda side 5 km upstream of the Marnya mouth. The
Ulyakha Mb. (up to 53 m thick) (Mb I, Bragin 1985), comprises
lenses and upright veins of greenish-grey, dark grey (to black) or
red coarse diamictite and dolomite breccia, and less abundant
channelized coarse sandstone (Figs 28.2 & 28.3). The member is
recognizable by its boulders and basement erratic boulders, abun-
dant glacial striation on clasts, grooves and other traces of erosion
on the surface where diamictite overlies the Tagul Fm.
The Tygnei Mb. (80 –90 m thick) (Mb IIa, Bragin 1985), com-
posed of black bituminous pyrite siltstone with thin stromatolite
dolomite in the upper part and sandstone interbeds and greenish-
grey laminated siltstone in the lower part, crops out along the
sides of the Uda River.
The Plity Mb. (20 –35 m thick) (Mb IIb, Bragin 1985), consists
of faceted-boulder quartzite breccia and yellowish-white feldspar-
quartzose sandstone under massive fluvial and eolian (?)
cross-bedded sandstone.
The Kedrovy Mb. (40 m thick), oversteps the Ulyakha diamictite
on the west side of the Uda River (near Kedrovy Village) and con-
sists of grey coarse extralitharenite sandstone with cobbles and
pebbles at the base of fluvial cycles. It underlies white coarse
arkose sandstone of the Ognit Mb. and is a stratigraphic equivalent
of the Plity Mb.
The Ozerki Mb. (50– 60 m thick) (Mb II, Bragin 1985), lies
conformably over various members of the Marnya Fm. at differ-
ent localities (Fig. 28.2). It pinches out in the Sayan region where
the pre-Marnya erosion was moderate. The sections of the
member are either (i) yellow laminated, wave rippled and tidal-
current cap dolomite with numerous cavities after pyrite
nodules or (ii) laminated stromatolite and cross-bedded sandy
dolomite, and feldspar-quartzose sandstone in the middle part
with abundant traces of soft-bodied animals (Sovetov 2007;
Sovetov & Solovetskaya 2008). At one locality (Ulyakha,
Fig. 28.2), the base of the carbonate Ozerki Mb. overlies the
diamictite-bearing Ulyakha Mb. and is thought to correlate
with the base of the Ediacaran.
The Ozerki Mb. is overlain by the Ognit and Bolshaya Aisa
members. The Ognit Mb. (up to 84 m thick) (Mb IV, Bragin
1985), consists of dark-grey and black coarse orthoquartzite and
arkoses, whereas the Bolshaya Aisa Member (105 –269 m thick)
(Mb V, Bragin 1984) consists of shoreface greenish-grey cross-
bedded sandstone and siltstone and shelf HCS tempestites and tida-
lites, with less abundant black microphytolite and silty limestone.
The Bolshaya Aisa Mb. contains Ediacaran-like biota (Sovetov &
Komlev 2005; Sovetov 2007; Sovetov & Solovetskaya 2008).
Uda Fm.
The stratigraphic extent of the Uda Fm. long remained unclear
after the lowermost grey rocks of the Oselok Group, originally
attributed to it, had been included into the Marnya Fm. According
to Bragin (1985), its lower section comprised the Ognit and
Bolshaya Aisa members but Khomentovsky et al. (1972) and
Sovetov & Komlev (2005) identified its lower boundary by
arkose and quartz gravelstone of the Nizhnyaya Uda Mb. Together
with the latter, the Uda Fm. is currently divided into three members
(Fig. 28.2).
The Nizhnyaya Uda Mb. (VI, Bragin 1985), most often
17– 30 m, is composed of white and yellow fine-pebble quartz
conglomerate, gravelstone and coarse gravelly cross-bedded
fluvial sandstone. The member rests upon the deeply (to 4 m)
eroded surface of the Bolshaya Aisa Mb. and to the SW its lower
40 m (out of a 70 m) section comprises coastal bar and
sandwave deposits.
The Unyl Mb. (VI, Bragin 1985), 25 –100 m, is genetically
linked to the Nizhnyaya Uda Mb. and consists of grey and
321
dark-grey siltstone with lenses of fine to coarse cross-bedded sand-
stone deposited in subsidiary channels and river mouth bars.
The Peschernyi Mb. (VIII, Bragin 1985), 7 –30 m, is composed
of dark-grey and black stromatolitic and microphytolitic limestone
and limestone tempestite, is locally replaced by epigenetic dolo-
mite, and varies strongly in thickness, pinching out cratonward.
Aisa Fm.
The Aisa (Ikei, Bragin 1985) Fm. is composed of continental
molasse of two lithofacies groups: (i) fluvial channel litharenite
sandstone and (ii) bar and overbank sandstone, siltstone and,
rarely, mudstone. Estimates of its total thickness vary from
1800 –2000 m (Egorova et al. 1971) to 1300 –1600 m (Khomen-
tovsky et al. 1972), or 1300 –1400 m according to measurements
by Sovetov in the most complete section on the sides of the
Biryusa and Tagul rivers. The formation has a prominent base
along the roof of the Peschernyi limestone and a very sharp top at
the base of the overlying Ust’-Tagul Fm. conglomerate attributed
to the Late Vendian Nemakit-Daldyn regional stage (Ediacaran-
Early Cambrian) (Sovetov et al. 2007a). The Aisa Fm. includes
four members of different thicknesses: Kagat, Muksut, Serebrovo
and Katalchikov. Overbank deposits contain abundant Arumberia
plant remnants (Sovetov 2006).
Glaciogenic deposits of the Marnya Fm. and associated strata
Diamictite of the Ulyakha Mb. was studied in outcrops along the
Uda River between the Marnya and Ulyakha mouth areas and
near Kedrovyi Village on the Biryusa River (Nersa section). The
maximum thickness of diamictite measured from its lower
erosion surface to the base of the Ozerki Mb. cap dolomite is
53 m. The matrix of boulder diamictite (mixtites) of the Ulyakha
Mb. (c. 85% of the rock volume) has a sand-silt-clay composition.
The rocks bear evident lithological signatures of tillite (Sovetov
2002a; Sovetov & Komlev 2005; Sovetov & Donskaya 2006),
namely: (1) iron- and bullet-shaped stones, which are faceted,
striated, polished and range in size from gravel to clasts up to
70 cm (occasionally 1.5 m) across; (2) orientation of elongate
stones; (3) lodgment stones pressed into dolomite of the Tagul
Fm.; (4) dropstones in sandy dolomite of outwash lakes; (5) steep-
walled subglacial channels, boulder beds and clustered stones; (6)
up to 1.5-m-deep nearly vertical crevasses with breccia; (7) inner
stratification of diamictite in the form of four nested lenses, differ-
ing in coloration, composition of stones, and containing meltout
intercalations; (8) associated fluvial sandstone with debris laterally
replaced by breccia; (9) stones of rocks typical of the underlying
Tagul and the Ipsit formations (Karagassy Group) and less fre-
quent erratics of gneiss, granite, biotite schist and metarhyolite
derived from the Early Proterozoic basement; (10) erosional
forms, such as narrow (a few decimetres wide) and shallow
furrows, scours tens of metres wide and 3– 4 m deep, local (tens
to hundreds of metres across) and regional (tens of kilometres
wide) erosional valleys; (11) blunt-end contacts of diamictite
with sidewalls of channelized troughs; (12) folds and rupture
features at the top of the underlying rocks of the Ipsit and Tagul
formations; (13) association with laminated cap dolomite with
sharp lower contact; (14) association with laminated siltstone
and shale alternated with gravelly-sandy deposits of gravity
flow; (15) association with boulder conglomerate containing
striated pebbles; (16) absence of ash fall, impact events, lahars,
volcaniclastic debris flows or turbidity flows on the continental
slope and rise; (17) absence of indications typical of tectonic
thrusting or melange; (18) occurrence of diamictite above a
regional unconformity.
The architectural elements in Marnya Fm. diamictite were dis-
tinguished by mapping outcrops. Figure 28.3 shows cross-sections
 322
I II III IV V VI VII VIII IX X XI XII XIII XIV XV

Ozerki Mb.
(a)

IV
I
2m II
III
III
10 m
in Ulakha river location
Glaciogenic deposits

S dolomite

in diamictite
Ulyakha Member
n=111
Marnya Formation

n=141 n=226 n=112

chert

stones
n=649 N arkose and orthoquartzite
20 m basement rocks
(b)

2m
Tagul

10 m
Fm. Ozerki Mb.

N Profile direction 325

n=209
I n=575 II III IV V VI VII
S

J. K. SOVETOV
Plity Member

(c)
Glaciogenic deposits in Ozerki location
Marnya Formation

III

IX
2,5 m VIII
Tygnei Member

5m VIII IX X XII
XI
(d)

20 m

VIII 200

Profile direction 50 Vergence of glacial dislocations

Ulyakha Mb.

Black cross-bedded sandstone in Deglacial fine detrital Outwash-plain cross-bedded glaciofluvial

Key to symbols:

Black argillaceous diamictite and sandstone with cobbles (zandur) III Diamictite lenses separated
stones 20 cm intraglacial and supraglacial channels diamictite by a shear zone
Green-grey argillaceous diamictite Elongation direction of cobbles and Pebbly sandstone from currents Reference sections of outcrop logs
stones
Laminated cap dolomite
(underwater fans and fan-deltas) XI
Tagul
Fm.

Red-brown sand-clay mixitite, Elongation direction of lodgment Massive and layered stromatolite Glacial shears
no stones cobbles and boulders and grainstone dolomite
Shear zones and boundaries Dollutite Percentages of different lithologies of stones
Dolomite diamictite and breccia
of diamictite lenses 4-40 mm
Intraglacial and supraglacial Laminated siltstone and
breccias and gravelstone in Basal shear zone Closed parts of section
mudstone in ephemeral lake
channels and crevasses

Fig. 28.3. Outcrop logs and field sketch showing different facies and lithology of diamictite clasts in the upper and lower Ulyakha Member along the Uda River. (a) Vertical section of an outcrop on the left side of the Uda River
1 km upstream of Ulyakha Brook. (b) Enlarged left side of section (a) showing the geometry of sandstone and conglomerate lenses within black argillaceous diamictite. (c, d) Single vertical section of the lower Ulyakha Mb.
at the Ozerki locality on the left side of the Uda River, 5 km upstream of the Marnya River.
 MARNYA FORMATION, OSELOK GROUP
and logs of the upper and lower Ulyakha Mb. along the Uda River.
For complete stratigraphy of the member, see Figure 28.2. The
upper Ulyakha Mb. (Fig. 28.3a) consists of four nested lenses of
diamictite with shear boundaries (I to IV). They are named black
and green argillaceous diamictite after their matrix colour and
siliciclastic composition. Diamictite consist of outsized clasts
from sand- and fine-gravel-sized rocks to elongate iron-shaped
boulders with striation oriented at 100 –2158. Lenses III and IV
are cut with 3–17-m-wide channels filled with either dolomite dia-
mictite, conglomerate or breccia. The lenses are massive or have
inclined bedding with inclusions of black coarse cross-bedded
sandstone; the channels have steep or overhanging walls, some
resemble crevasses (Fig. 28.3b). Palaeocurrent direction measured
on cross-bedding and channel-bed slopes suggests currents moved
towards 230 –3308.
The Ulyakha Mb. at the Ozerki outcrop consists of three lenses
separated by shear zones or shear planes (Fig. 28.3c, d). These
deposits host abundant boulders of the Ipsit Fm. quartzite with
striation, less abundant boulders of the Tagul Fm., silicified stro-
matolitic dolomite, and a few erratic boulders of basement
granite and gneiss. The boulders and their striation are oriented
mainly at 150– 2408. The direction of glacier advance has been
inferred from the 150– 2708 orientation of lodgment boulders
pressed into the Tagul dolomite. A similar interpretative orien-
tation (2008) has been found in folds of the Tagul dolomite and
siltstone beds beneath the diamictite. Small folds only occur at
the Ozerky location above a horizontal shear zone 8 –10 m
below the Ulyakha Mb., and the underlying part of the Tagul
Fm. is rather undeformed. The upper Ulyakha Mb. includes
black siltstone and lenses of dolomite diamictite and breccia in
U-shaped channels. According to correlation (no visible contact),
Fig. 28.4. Cryogenian (Early Vendian) glacial erosional forms in the southwestern Siberian Craton. See Figure 28.2, inset, for map of localities.
323
diamictites are thought to be buried under black bituminous silt-
stone deposited in ephemeral lakes (Tygnei Mb.), which in turn
are overlain by laminated, wave-rippled and stromatolite dolomite
of the Ozerki Mb.
The Karapchatui Mb. fills a 60–70-m-deep valley, incised into
the Ipsit and Tagul Formations (Figs 28.2 & 28.4). The valley is
filled with cross-bedded sandstone, sandy stromatolitic dolomite
and faceted-boulder breccia. The breccia occur in the valley
floor and its east side, and the centre is occupied by an outwash-
plain fan oriented at 340– 08 in the NW – north direction, according
to numerous measurements of oblique foresets in streams, large
ripples and dunes.
Boundary relations with overlying and underlying
non-glacial units
All members of glaciogenic deposits at the base of the Marnya Fm.
(diamictite, faceted-boulder breccia, conglomerate and sandstone)
have sharp contacts with the underlying deposits of the Ipsit and
Tagul formations in the Sayan region, with local scouring features
(Figs 28.2–28.4). The deepest erosion exceeds 500 m in the area
between the Biryusa and Uda rivers (Fig. 28.4) and is c. 1000 m
throughout the Sayan region (Dubin et al. 1969; Bragin 1984;
Sovetov & Komlev 2005).
The erosive contact of the Ulyakha Mb. black argillaceous dia-
mictite with the underlying dolomite of the Tagul Fm. (Karagassy
Group) was mapped on the left side of the Uda River at the Ozerki
locality (Fig. 28.3c). The profile runs across a 90–110-m-long and
4.5-m-deep flat-bottomed channel, which is incised into the Tagul
324 J. K. SOVETOV
Fm. and is filled with dolomite diamictite from the lower Ulyakha
Mb. The total erosion depth at the locality approaches 9 m.
Grooves and scours with sharp contacts on the surface of the
Tagul dolomite, as well as terraces, are oriented at 190– 1958
(Fig. 28.4).
A large erosional valley is exposed on the right side of the Uda
River at the mouth of the Karapchatui Brook. The Karapchatui Mb.
overlies the Tagul Fm., with 10–158 unconformity, in the centre of
an erosional valley (Figs 28.2 & 28.4). The eastern valley side dips
at 30 –508 and strikes at 3008, and its upper part is composed of the
Ipsit Fm. rocks (Fig. 28.4). Another exposed erosional valley
occurs along the Uda River at the Plity locality (Fig. 28.2). The
15-m-deep north – south valley is incised into the Ipsit Fm. and
filled with faceted-boulder breccia and minor cross-bedded
sandstone of the lower Plity Mb. Thick-bedded cross-bedded sand-
stone of the upper Plity Mb. covers the eastern side of the valley
and lies over the Ipsit quartzose arkose siltstone with a sharp
contact. The dip of oblique cross-bedding foresets indicates
north –south transport of material along the valley. The same
relationship between the Plity Mb. and the Ipsit Fm. is found in
outcrops along the right side of the Tagul River 6 km upstream
of Georgievka Village. The distinct base of the glaciofluvial
boulder conglomerate of the Nersa Mb. (Marnya Fm.), with silt-
stone of the Ipsit Fm., crops out along the Biryusa River (near
Kedrovyi Village) (Fig. 28.2).
The upper boundary of the Marnya glaciogenic deposits is well
exposed along the sides of the Uda River at the Ulyakha and Ozerki
localities. The Ulyakha Mb. rocks underlie laminited cap dolomite
along a sharp contact (Fig. 28.3). The same prominent boundary
between the Plity sandstone and cap dolomite is exposed at the
Ozerki locality (Fig. 28.2). Dolomite of the Ozerki Mb. pinches
out along the sides of the Uda River (Plity and Gladkiy Mys
sites and at Bogatyr’ Hill) and leaves sandstone tempestite of the
overlying Bolshaya Aisa Mb. Overlying the Plity Mb. (Fig. 28.2).
Chemostratigraphy
Tentative estimates of 87Sr/86Sr ratios were obtained in the
Peschernyi Mb. limestone. These Sr ratios are lower than those
reported from the Wonoka Fm. (Calver 2000), and the correlated
Shuram Fm. of Oman (Burns et al. 1994) (0.70796 – 0.70848
v. 0.7087 –0.7088 and 0.7085 –0.7088, respectively; Fig. 28.5).
C- and O-isotope compositions were studied in carbonate rocks
of the Marnya and Uda formations (Fig. 28.5; Sovetov & Komlev
2005; Sovetov et al. 2007b). The resulting C-isotope curve
includes five characteristic segments A –E (Fig. 28.5; Sovetov &
Komlev 2005). The d13C values in segment A, which correspond
to samples from the middle and upper parts of the Tagul Fm.
(Karagassy Group), vary from – 1.5 to þ3‰. All d13C values in
dolomitic diamictite of the Marnya Fm. (segment B) are positive
and within þ1.6‰. A negative anomaly occurs in segment C
corresponding to the overlying Ozerki dolomite, where d13C
gradually decreases to –5.7‰ and then increases smoothly to
zero towards the top of the Ozerki Mb. A positive d13C excursion
is found in the upper third of the Bolshaya Aisa Mb. (Marnya Fm.;
from þ6.3‰ decreasing upward to þ4‰) as well as in the
Peschernyi Mb. (Uda Fm.; þ3.1 to þ7.5‰). Based on these
data, the Ulyakha diamictite was correlated with the Late
Cryogenian Varanger glacial deposits in Norway (Sovetov &
Komlev 2005).
The chemostratigraphic chart as a whole has the following fea-
tures (Fig. 28.5): (i) a d13C low in cap carbonates over the Ulyakha
Mb., (ii) transition to positive d13C signatures spanning the
Bolshaya Aisa and Peschernyi members in the Sayan area
(Sovetov & Komlev 2005, figs 6–8), comparable to the anomalies
found in the Uluntui Fm. of the Baikal Group, southwestern Baikal
area (Khabarov & Ponomarchuk 2005), (iii) a large d13C low in the
Kachergat Fm. of the Baikal Group, to – 6‰ (Sovetov et al.
2007b), similar to a low of –10‰ in the Oskoba Fm., Baikit
region (Vinogradov et al. 1994; Khomentovsky et al. 1998;
Sovetov et al. 2007b), which both appear to correlate with the
Zhuya and partly the N2 negative anomalies (Pelechaty 1998;
Pokrovsky et al. 2006) in the Patom region and with the
Wonoka anomaly identified in various parts of the world
(Halverson et al. 2005). The grounds for this correlation are
discussed below.
Palaeolatitude and palaeogeography
Palaeomagnetic directions were measured in diamictite samples
from the Ulyakha Mb. (Marnya Fm.) (54.338N, 98.878E) and the
overlying Tygnei black siltstone (54.338N, 98.898E) collected by
Sovetov at the Ozerki locality (Uda River). Diamictite and dolo-
mite exhibit nearly horizontal bedding planes at the locality.
Measurements were performed using the standard procedure
(Butler 1992) of stepwise thermal demagnetization on a magnet-
ometer at the Paleomagnetic Centre of International Petroleum
Geology and Geophysics (Novosibirsk, Russia).
According to published and preliminary data, the primary
palaeomagnetic directions in the rocks of the Marnya Fm. have
been overprinted by Early Cambrian geodynamic events (Metelkin
et al. 2005; D. V. Metelkin, pers. comm.). The available palaeo-
magnetic evidence for Neoproterozoic and Early Palaeozoic
rocks from the Siberian craton (Smethurst et al. 1998; Pisarevsky
& Natapov 2003; Shatsillo et al. 2005; Metelkin et al. 2007)
suggest a sub-equatorial location.
Geochronological constraints
The minimum age of glaciation comes from Ediacaran (Late
Vendian) –Early Cambrian small-shell fauna found within the
Ostrovnoi Fm., (Khomentovsky et al. 1998), which is correlated
with the upper part of the Ust’Tagul Fm. (Fig. 28.6). The latter ero-
sively overlies the Oselok Group, while its equivalents overlie the
Taseeva and Chapa groups in the Yenisei Ridge (Sovetov 2002b,
fig. 6; Sovetov & Blagovidov 2004). Furthermore, Metazoan
trace fossils were found in quartzite in the middle of the Ozerki
Mb. and imprints and moulds of Ediacaran-like organisms have
been reported in the Bol’shaya Aisa Mb. (Sovetov & Komlev
2005; Sovetov 2007; Sovetov & Solovetskaya 2008).
The maximum age for the Marnya Fm. glaciogenic deposits
in the Sayan region has been constrained by two dates: (i) the
741 + 2 Ma 39Ar/40Ar age of plagioclase from the Nersa gabbro
dolerite sills (Gladkochub et al. 2005) that intrude the Shangulezh
Fm. of the underlying Karagassy Group (gabbro dolerite clasts are
also found in the Ulyakha Member diamictite), and (ii) the
39
Ar/40Ar 696 + 8.5 Ma age of biotite from dolerite and tuff
(Postnikov et al. 2005) among diamictite of the Chivida Fm.
(middle section of Chingasan Group; Yenisei Ridge) that lies
stratigraphically lower than the Oselok Gr.
Discussion
Main structural and morphological features of the Marnya basal
strata were discussed in Sovetov & Komlev (2005) and Sovetov
(2008). The formation includes several major glaciogenic sedi-
mentary structures and architectural elements (Miall 1985),
namely (i) diamictite with striated boulders and poorly rounded
blocks, (ii) faceted-boulder breccia and (iii) intradiamictite- and
supra-diamictite glacial channels and crevasses. The subsidiary
elements are (iv) outwash –plain fluvial debris sandstone with
thin layers and lenses of stromatolite dolomite, (v) sandstone and
gravelstone deposits of alluvial plains, fans and fan deltas, (vi)
laminated fine silt and mudstone deposited in ephemeral stagnant
lakes. Glaciogenic deposition was accompanied by deep erosion of
 MARNYA FORMATION, OSELOK GROUP 325

δ13C δ13C Depositional systems

10 8 6 4 2 0 2 4 6 8 10 10 8 6 4 2 0 2 4 6 8 10
Braided alluvial and delta plain
Baikal region, Kurtun river
(isotope data by Sovetov, et al.,

Phytolithic bioherms and

Continental slope and shelf

Сhencha Fm.
Kachergat Fm.=Aisa Fm.

storm-dominated deposits
2007 b)

Nikolsky Fm.
Contour currents deposits

Passive continental margin

(isotope data by Pokrovsky et al.,2006)
Patom Upland Ura river
Deep-water delta
Valukhta Fm.
87 86
Sr/ Sr Siliciclastic
0.70796-0.70848 deep-water fan
Uda Fm.

Pescherny Mb. E
Epicontinental sea, strand plain

Borakun Fm. Siliciclastic and carbonate

(isotope data by Sovetov, Komlev,

Continental rise
Unyl Mb.
Sayan region Uda river

Nizhnyaya Uda Mb. ? deep-water fans

Bol. Aisa Mb. D
Ognit Mb.

nian glaciation
2005)

Late Cryoge-
Marnya Fm.

C Siliciclastic
Ozerki Mb.
Dzhemkukan Fm. deep-water fan
and diamictite
Onset glacigenic deposits

Ulyakha Mb. B Mariinka Fm.

Onset glacigenic deposits
Tagul Fm.
A
Fig. 28.5. Chemostratography of the Oselok Gp. and correlation with deep-water deposits of the Patom Upland. Carbon-isotope curve for the Marnya and Uda formations
is complemented with some data from the Kachergat Fm., which is a stratigraphic equivalent of the Aisa Fm. The uppermost negative anomaly in the Nikolsky and Chencha
formations is thought to correlate with the Wonoka anomaly (Halverson et al. 2005) and the Pre-Manykai anomaly identified by Pokrovsky & Missarzhevsky (1993).

the substrate and multiform scouring, and dislocation of rocks in the Baikal Group of the Baikal region, and the Nepa and Tira for-
the underlying bed (Fig. 28.4). Judging by their morphology, the mations of the Baikit and Nepa-Botuobiya forebulges (Sovetov
channel-like stone clusters formed in ice during deglaciation and et al. 2007a). The regional stratigraphy and correlations are con-
can be identified as kames. Diamictite have all the features of trolled by the following marker units and surfaces (Fig. 28.6):
tillite, and are thought to have been produced by rock crushing (1) transgression and flooding surface at the base of the Katanga
and transport of clasts by ice and through sub-ice channels by melt- and Ostrovnoi formations and its equivalents at the base of the
water. Deglaciation is recorded by deposition of black siltstone (to Kurtun Fm. and in the middle of the Shamanka, Ust’-Tagul, and
0.5 m) in ephemeral lakes and, later, by deposition of dark-grey Uglovoi formations, (2) base of the Ust’-Tagul supersequence
and black laminated cap dolomite of the Ozerki Mb., which with river incision, (3) base of the Greben supersequence with
grade up the section into stromatolitic dolomite, dolomite tidalites river incision, (4) base of the Marnya supersequence defined by
and sand –grainstone sandwaves. a deeply eroded surface under glacial and glaciofluvial deposits.
The Ulyakha tillite, with the associated faceted boulder breccia, Subsidiary correlation levels are associated with third-order
boulder conglomerate and outwash sand, make up a complex suc- sequences (Sovetov 2002b; Sovetov & Komlev 2005). The
cession (Sovetov & Komlev 2005). The palaeocurrent directions Marnya glaciogenic deposits have been correlated with breccia,
suggest the existence of two major glaciers in the Siberian diamictite at the base of the Olkha Fm. (Mordvin 1972; Chumakov
craton: a smaller one marked by the Karapchatui outwash fan in 1993), Baikal Gr. (Sovetov & Komlev 2005) and the base of the
the south and a larger one evident from the Ulyakha Mb. channels Chapa Gr. (Sovetov et al. 2007a), conglomerate and breccia at
and striated lodgment boulders in the NE of the area in the central the base of the Taseeva Gr. (Sovetov et al. 2007a), as well as the
Siberian Craton. The extent of the latter glacier is recorded in Vanavara, and Nepa Formations (Tyschenko 1980) (Fig. 28.6).
clastic transport, erratic stones from the basement of the Siberian Carbonate members in the Marnya, Uda, Goloustnaya and
Platform and in sedimentary structures observed in the Sayan Uluntui formations pinch out cratonward from foredeeps, while
and Baikal regions (Sovetov & Donskaya 2006). carbonate members of the Oskoba and Tira formations are replaced
Lithostratigraphic evidence available for deposits of the in the foredeeps by terrigenous molasse (Fig. 28.6). Sedi-
Precambrian –Cambrian transition from different areas of the mentological studies of the upper Oselok Group showed that depo-
Siberian craton bears some signatures that may serve as ties in sition of the Peschernyi marine limestones was followed by a new
regional correlations. Figure 28.6 shows lithostratigraphic corre- evolutionary stage of the foreland basin, with orogens and the
lation of the Oselok Group with the Taseeva Group of the bordering alluvial plains originating around the southwestern
Yenisei Ridge, the Olkha and Moty Groups of the Irkutsk province, edge of the Siberian craton (Sovetov & Blagovidov 2004;
 326
J. K. SOVETOV
Fig. 28.6. (a) Correlation of Late Neoproterozoic deposits in the southwestern Siberian Craton and stratigraphic position of the tillite units, complemented and modified after Sovetov et al. (2007a). Formation name
abbreviations: Ayan, Ayankan Fm.; Bol, Bol’shoi Lug Fm.; Bor, Borakun Fm.; Ch, Chencha Fm.; Kal, Kalancha Fm.; Khuzh, Khuzhir Fm.; Kul, Kulekin Fm.; M, Mariinka Fm.; Nik, Nikol’sky Fm.; Shan, Shanhar Fm.; Sham,
Shamanka Fm.; Tb, Tirbes Mb.; Ul, Uluntui Fm.; Ud, Uda Fm.; Vor, Vorogovka Grp.; Mr, Marnya Fm.; Gol, Goloustnaya Fm.; Tb Mbr, Tirbess Mb. Logs: 1, by (Sovetov et al. 2007a), with additions from Karpinskaya,
Karpinsky & Ustalov (pers. comm.); 2, modified after Vinogradov et al. (1994); 3, after Tyschenko (1980); 4, by Sovetov et al. (2007a) with addition from Shenfil (1991). (b) Map showing location of stratigraphic logs shown in
(a). (c) Map showing distribution of late Cryogenian diamictite and inferred glacial extent for the Siberian Platform. Localities as follows: a, Kutukas river, Enisei Ridge, Chapa Group, Stolbovaya Formation (this paper); b,
Chapa and Teya rivers, Enisei Ridge, Chapa Group, Pod’em Formation (this paper); c, Tagul River, Cis’Sayan region, Oselok Group, Marnya Fm. (Sovetov unpublished data); d, Birusa and Uda rivers, Cis’Sayan region, Oselok
Group, Marnya Fm. (this paper); e, Irkut River, Irkut Cis’Sayan region, Olkha Fm. (Mordvin 1972); f, Buguldeika River, southwestern Cis’Baikal region, Baikal Group, Goloustnaya Fm. (this paper); g, Right Ulkan River,
northwestern Cis’Baikal region, Baikal Group Goloustnaya Fm. (Korobeinikov & Semeikina 1978); h, Patom River, Patom Upland, Dal’nyaya Taiga Group, Dzhemkukan Fm. (this paper); i, Patom River, Patom Upland,
Dal’nyaya Taiga Group, Dzhemkukan Fm.; j, Nichatka Lake, Aldan Shield, Nichatka Fm.; k, Siberian Platform, Markha River, Borehole 225–0, Conglomerate suite (Kraevsky & Shishkin 2000).
 MARNYA FORMATION, OSELOK GROUP
Sovetov et al. 2007a). Continental molasse deposited during
the Ediacaran (Late Vendian) collision belongs to the Greben
regional stratigraphic stage.
Glaciogenic deposits of the Marnya Fm., interpreted as tillites,
and stratigraphic equivalents of the formation belong to the upper
diamictite unit in the succession of two regional stratigraphic
stages (Sovetov et al. 2007a). The lower unit was deposited
during a period that can be correlated with the Early Cryogenian
glacial in a setting of rifting, volcanism and subsidence of deep
basins (Sovetov 1997). Deposition of the Marnya glacial deposits
is thought to be coeval with the early stage of the Late Cryogenian
(Early Vendian) foreland basin (Sovetov 2002b; Sovetov & Komlev
2005) and with formation of post-glacial shallow fringing seas in the
southwestern part of the craton. The Ulyakha glaciogenic deposits
can be equated to a Late Cryogenian glaciation (Halverson et al.
2005), for the following reasons: (i) higher stratigraphic position
of the Ulyakha tillite relative to the Chivida diamictite (tillite) of
the Yenisei Range (Sovetov 1997; Sovetov et al. 2007a) constrained
by the coeval c. 700 Ma volcanism (Postnikov et al. 2005); (ii)
occurrences of Vendian-Ediacaran Metazoan in the Ozerki Mb.
(Sovetov & Komlev 2005; Sovetov 2007; Sovetov & Solovetskaya
2008); (iii) post-glacial regional transgression (Sovetov 2007) and
correlation of related terrigenous deposits (Bolshaya Aisa Mb.)
with the Nepa Fm. containing microfossil assemblages as in the
Ediacaran Pertatatak Fm. in central Australia (Moczydłowska
et al. 1993; Golubkova & Raevskaya 2007).
Chemostratigraphy suggests that the glaciation on the Siberian
craton recorded in the Marnya Fm. can be equated to the Marinoan
epoch, which, according to the modern d13C time scale, is slightly
older than the base of Doushantuo cap dolomite 635.2 + 0.6 Ma
(Condon et al. 2005). The evidence that the Ulyakha glacial
deposits are stratigraphically lower that the Gaskiers tillite is pro-
vided by: (i) correlation of the Aisa syncollisional molasse in the
Sayan region with the Oskoba Fm. in the Baikit region
(Fig. 28.6), with the latter marked by an intense d13C low (referred
to as the Pre-Manykai negative anomaly by Pokrovsky 1996
and Sovetov et al. 2007b) that correlates with the Zhuya d13C nega-
tive anomaly (Pokrovsky et al. 2006) in the Patom region and with
the Wonoka anomaly (Halverson et al. 2005) in South Australia; (ii)
position of a d13C positive anomaly in the Peschernyi Member rocks
(Uda Fm.), which is thought to be correlated with a similar anomaly
below the Gaskiers glaciation; (iii) tectonothermal events in terranes
on the craton SW related to their post-glacial accretion at 600–
550 Ma (Nozhkin et al. 2007), which is confirmed by the available
87
Sr/86Sr ratios in the Peschernyi Mb. limestone.
Based on the widespread nature of diamictite and breccia at the
base of the Ediacaran cover of the Siberian Platform, a thick ice
sheet occupied most of the southwestern craton area (Fig. 28.6).
Erosion of Neoproterozoic sediments was locally deeper than
1000 m; erosional valleys reached widths up to 50 km and radiated
centrifugally to the west, SW and south, following the advance of
the glaciers. At the same time, another glaciation centre, inferred
from the palaeocurrent directions in an outwash fan of the Karap-
chatui Mb., was located in the extreme southwestern basement
inliers in the Sayan region, and may record the complex structure
of a single ice sheet. The continental ice sheet extended as far as
remnant deep seas on the Patom passive margin, discharging
gravity flows to produce the deep-water fan deposits of the
Dzhemkukan Fm. (Chumakov et al. 2006).
Funding for the research was provided by the Russian Foundation for Basic
Research (grants nos 04-05-65299 and 08-05-00959), which was indispensable
for regional sedimentological and stratigraphical studies, and by the Earth
Science Department (ESD) of the Russian Academy of Sciences (Project 18 of
the ESD Program ‘Biosphere origin and evolution’ and Project 10.1 of the
RAS-ESD Program ‘Central Asian orogen: geodynamics and stages of continen-
tal crust formation’). I wish to thank D. Metelkin who performed the palaeomag-
netic analysis of samples from the Marnya Fm. and my field companions,
postgraduate students D. Komlev, M. Medvedev and A. Kulikova. This represents
327
a contribution of the IUGS- and UNESCO-funded IGCP (International
Geoscience Programme) project #512.
References
Anatol’eva, A. I. 1972. Pre-Mesozoic Redbed Deposits, Trudy Inst.
Geol. Geophiz. Sibir. Divis. Akad. Nauk SSSR, Nauka, Novosibirsk,
190, 323 (in Russian).
Berzin, N. A. 1967. Main Fault Zone in the East Sayan. Nauka, Moscow
(in Russian).
Bragin, S. S. 1984. On relationship of the Precambrian Karagassy and
Oselok Groups in the Sayan region. In: Khomentovsky, V. V.
(ed.) The Late Precambrian and the Early Palaeozoic Stratigraphy.
Central Siberia. IGG, Novosibirsk, 133–147 (in Russian).
Bragin, S. S. 1985. Late Precambrian Oselok Group in the Sayan
region (stratigraphic division and correlation). In: Khomentovsky,
V. V. (ed.) The Late Precambrian and the Early Palaeozoic of
Siberia. The Vendian and the Riphean. IGG, Novosibirsk, 44 –57
(in Russian).
Bragin, S. S. 1986. Stratigraphy of the Late Riphean Karagassy Group in
the Sayan region: Some problems. In: Khomentovsky, V. V. &
Shenfil, W. Yu. (eds) The Late Precambrian and the Early Palaeo-
zoic of Siberia. Stratigraphy and Palaeontology. IGG, Novosibirsk,
32– 39 (in Russian).
Burns, S. J., Haudenschild, U. & Matter, A. 1994. The strontium
isotopic composition of carbonates from the late Precambrian
(c. 560– 540 Ma) Hukf Group of Oman. Chemical Geology, 111,
269– 282.
Butler, R. F. 1992. Palaeomagnetism: Magnetic Domains to Geologic
Terrains. Blackwell Science, Oxford.
Calver, C. R. 2000. Isotope stratigraphy of the Ediacaran (Neoproterozoic
III) of the Adelaida Rift Complex, Australia, and the overprint of water
column stratification. Precambrian Research, 100, 121–150.
Chumakov, N. M. 1993. Middle Siberian Riphean glaciohorizon. Strati-
graphy. Geological Correlation. Vol. 1. No 1, 21– 34 (in Russian).
Chumakov, N. M., Pokrovsky, B. G. & Melezhik, V. A. 2006. Patom
Complex geological history, Late Precambrian, Middle Siberia.
Doklady Earth Sciences, 413, 379–383 (in Russian).
Condon, D., Zhu, M., Bowring, S., Wang, W., Yang, A. & Jin, Y. 2005.
U–Pb ages from the Neoproterozoic Doushantuo Fm., China.
Science, 308, 95– 98.
Donskaya, T. V., Salnikova, E. B. et al. 2002. Early Proterozoic post-
collisional magmatism at the southern flank of the Siberian craton:
new geochronological data and geodynamic implications. Doklady
Earth Sciences, 382, 125–129 (in Russian).
Dubin, P. V., Khomentovsky, V. V. & Yakshin, M. S. 1969. New
data on Late Precambrian geology of the Sayan region. In:
Sokolov, B. S. (ed.) Early Cambrian and Late Precambrian Strati-
graphy of the Southern Siberian Craton. Nauka, Moscow, 86 –101 (in
Russian).
Egorova, O. P., Urumov, Yu. D., Volynets, Yu. N. & Vernoslova,
Z. S. 1971. New data on geology and phosphate potential of Upper
Riphean strata in the Iya-Tumanshet province of the Sayan region.
In: Fainshtain, G. H. (ed.) Geology and Mineral Deposits of the
Siberian Craton. Nedra, Moscow, 69– 84 (in Russian).
Gladkochub, D. P., Donskaya, T. V., Mazukabzov, A. M., Pono-
marchuk, V. A. & Stanevich, A. M. 2005. Neoproterozoic
gabbro dolerites of the Biryusa terrane (southern Siberian craton) as
possible indicators of the Rodinia dispersal, In: Sklyarov, E. V.
(ed.) Geodynamic Evolution of the Lithosphere in the Central Asian
Orogen (From Ocean to Continent). Workshop Proceedings. Book
1. IZK, Irkutsk, 59 –62 (in Russian).
Gladkochub, D. P., Wingate, M. T. D., Pisarevsky, S. A., Donskaya,
T. V., Mazukabzov, A. M., Ponomarchuk, V. A. & Stanevich, A.
M. 2006. Mafic intrusions in southwestern Siberia and implications
for a Neoproterozoic connection with Laurentia. Precambrian
Research, 147, 260–278.
Golubkova, E. Yu. & Raevskaya, E. G. 2007. Lower Vendian complex
of microfossils from the interior part of the Siberian platform. In:
Semikhatov, M. A. (ed.) The Rise and Fall of the Vendian (Edia-
caran) Biota. Origin of the Modern Biosphere. Transactions of the
328 J. K. SOVETOV
International Conference of the IGSP 493. GEOS, Moscow, 39– 42
(in Russian).
Halverson, G. P., Hoffman, P. F., Schrag, D. P., Maloof, A. C. &
Rice, A. H. N. 2005. Toward a Neoproterozoic composite carbon-
isotope record. GSA Bulletin, 117, 1181–1207.
Khabarov, E. M. & Ponomarchuk, V. A. 2005. Carbon isotopes in the
Upper Riphean deposits of the Baikal Group in western Cisbaikalia:
stratigraphic implications. Russian Geology and Geophysics, 46,
1019– 1027 (in Russian).
Khomentovsky, A. S. 1950. Some data on Geology of Tumanshet salt-
bearing region (Eastern Siberia). Bulletin of Moscow Society of
Nature Researchers, 15, 65 – 79 (in Russian).
Khomentovsky, V. V., Shenfil’, V. Yu., Yakshin, M. S. & Butakov,
E. P. 1972. Late Precambrian and Early Cambrian Reference
Sections in the Siberian Platform. Nauka, Moscow (in Russian).
Khomentovsky, V. V., Faizulin, M. Sh. & Karlova, G. A. 1998. The
Vendian Nemakit-Daldyn stage in the southwestern Siberian craton.
Doklady Earth Sciences, 362, 813– 815 (in Russian).
Konstantinova, L. N. 2005. Katanga area. Section II. Vendian. In:
Melnikov, N. V. (ed.) Stratigraphy of Oil and Gas Basins of
Sibera. Riphean and Vendian of Siberian Platform and its Plaited
Border. Academic Publishing House ‘Geo’, Novosibirsk, 187– 193
(in Russian).
Korobeinikov, N. K. & Semeikina, L. K. 1978. The nature of inter-
relation of Baikal Group and Middle Proterozoic volcanic – plutonic
complex in the north-western Cis’Baikal region. In: Khomentovsky,
V. V. (ed.) News in Stratigraphy and Palaeontology of Late Precam-
brian in the Eastern and Northern Regions of Siberia. Institute
Geology and Geophysics of SB Academy of Sciences of USSR,
Novosibirsk, 134– 146 (in Russian).
Kraevsky, B. G. & Shishkin, B. B. 2000. Study state and direction of
stratigraphy investigation of Riphean deposits in the inner regions
of Siberian Platform. In: Krasnov, V. I. (ed.) Stratigraphy and
Palaeontology of Siberia. Siberian Scientific Research Institute of
Geology, Geophysics and Mineral Deposits, Novosibirsk, 23 –31
(in Russian).
Metelkin, D. V., Belonosov, I. V., Gladkochub, D. P., Donskaya, T.
V., Mazukabzov, A. M. & Stanevich, A. M. 2005. Palaeomagnetic
directions from Nersa intrusions of the Biryusa terrane, Siberian
craton, as a reflection of tectonic events in the Neoproterozoic.
Russian Geology and Geophysics, 46, 395–411 (in Russian).
Metelkin, D. V., Vernikovsky, V. A. & Kazansky, A. Yu. 2007. The
Neoproterozoic stage of the Rodinia evolution in the light of new
palaeomagnetic data from the western margin of the Siberian
craton. Russian Geology and Geophysics 48, 42– 59 (in Russian).
Miall, A. D. 1985. Architectural-element analysis: A new method of
facies analysis applied to fluvial deposits. Earth-Science Review,
22, 261– 308.
Moczydłowska, M., Vidal, G. & Rudavskaya, V. A. 1993. Neoproter-
ozoic (Vendian) phytoplankton from the Siberian platform, Yakutia.
Palaeontology, 36, 495–521.
Mordvin, A. P. 1972. The Khuzhir ore occurrence of gold-bearing con-
glomerates, In: Fainshtain, G. H. (ed.) Geology and Gold Potential
of the Riphean and Vendian Strata in the Southern Periphery of the
Irkutsk Amphitheatre. Vostochno-Sibirskoe Izdatelstvo, Irkutsk,
154– 189 (in Russian).
Nozhkin, A. D., Turkina, O. M., Sovetov, Yu. K. & Travin, A. V.
2007. The Vendian accretionary event in the southwestern margin
of the Siberian craton. Doklady Earth Sciences, 415A, 866–871 (in
Russian).
Pelechaty, S. M. 1998. Integrated chronostratigraphy of the Vendian
System of Siberia: implications for a global stratigraphy. Journal of
the Geological Society, London, 155, 957– 973.
Pisarevsky, S. A. & Natapov, L. M. 2003. Siberia in Rodinia. Tectono-
physics, 375, 221–245.
Pokrovsky, B. G. 1996. Boundary between Proterozoic and Palaeozoic:
isotope anomalies in the sedimentary sections of Siberian Platform
and global environments change. Lithologia i Poleznye Iskopaemye,
4, 376–392 (in Russian).
Pokrovsky, B. G. & Missarzhevsky, V. V. 1993. Isotope correlation of
boundary Precambrian/Cambrian beds of the Siberian Platform.
Doklady Earth Sciences, 329, 768– 771 (in Russian).
Pokrovsky, B. G., Melezhik, V. A. & Buyakaite, M. I. 2006. Carbon,
oxygen, and strontium isotope compositions of Late Precambrian
rocks in the Patom complex, Central Siberia. Paper 1. Results, isoto-
pic stratigraphy, and dating problems. Lithologia i Poleznye Iskopae-
mye, 5, 1 –26 (in Russian).
Postnikov, A. A., Nozhkin, A. D. et al. 2005. New data on the age of
Neoproterozoic deposits of the Chingasan and Vorogovka Groups in
the Yenisei Ridge. In: Sklyarov, E. V. (ed.) Geodynamic Evolution
of the Lithosphere in the Central Asian Orogen (From Ocean to Con-
tinent). Workshop Proceedings, Book 2. IZK, Irkutsk, 71– 74 (in
Russian).
Sez’ko, A. I. 1988. Principal stages of continental crust forming in the
Cis-Saya. In: Letnikov, F. A. (ed.) Evolution of Crust in Precam-
brian and Palaeozoic. Sayan-Baikal Mountain Region. Nauka Sib.
Division, Novosibirsk, 7 –41 (in Russian).
Shenfil, V. Yu. 1991. Late Precambrian Stratigraphy of the Siberian
Platform. Nauka, Novosibirsk (in Russian).
Shatsillo, A. V., Didenko, A. N. & Pavlov, V. E. 2005. Two competing
palaeomagnetic directions in the Late Vendian: new data for the SW
region of the Siberian Platform. Russian Journal of Earth Sciences, 7,
3 –24.
Smethurst, M. A., Khramov, A. N. & Torsvik, T. H. 1998. The
Neoproterozoic and Palaeozoic palaeomagnetic data for the Siberian
Platform: From Rodinia to Pangea. Earth Science Research, 43,
1 –24.
Sovetov, J. K. 1977. Upper Precambrian Sandstones in the Southwestern
Siberian Platform. Nauka, Novosibirsk (in Russian).
Sovetov, J. K. 1997. Late Riphean rifting and the Baikal geodynamic
cycle of the Siberian Platform. In: Koroteev, V. A. (ed.) The
Riphean of Northern Eurasia: Geology and General Problems of
Stratigraphy. Institute of Geology and Geochemistry, Ekaterinburg,
223– 230 (in Russian).
Sovetov, J. K. 2002a. Vendian glaciation on the Siberian craton. In: Let-
nikov, F. A. (ed.) Geology, Geochemistry, and Geophysics at the
Turn of the 20th-to-21st Century. Proceedings, Russian Conference.
IZK, Irkutsk, 122–124 (in Russian).
Sovetov, J. K. 2002b. Vendian foreland basin of the Siberian cratonic
margin: Palaeopangean accretionary phases. Russian Journal of
Earth Sciences, 4, 363– 387.
Sovetov, J. K. 2006. Vegetation along Vendian rivers, climate zonation,
and palaeogeography of the Siberian craton in the Late Vendian. In:
Sklyarov, E. V. (ed.) Geodynamic Evolution of the Lithosphere in
the Central Asian Orogen (From Ocean to Continent). Proceedings
of Workshop on ESD RAS Integration Research Programs, 2. IZK,
Irkutsk, 143– 146 (in Russian).
Sovetov, J. K. 2007. New occurrence of soft body Vendian–Ediacaran
like Metazoa in south-west of the Siberian Platform, Oselok group,
Biruysa Cis Sayan. In: Semikhatov, M. A. (ed.) The Rise and Fall
of the Vendian (Ediacaran) Biota. Origin of the Modern Biosphere.
Transactions of the International Conference of the IGSP 493.
GEOS, Moscow, 33 –37 (in Russian).
Sovetov, J. 2008. Marinoan glaciation in the Siberian craton: locality,
erosional forms, deposits and constraints to age. CGC-04 Neoproter-
ozoic ice ages: Quo vadis? – Part 2. International Geological
Congress. Oslo. 6– 14 August. Abstracts. File://E:/33IGC/
1343311.html.
Sovetov, J. K. & Blagovidov, V. V. 2004. Reconstruction of sedimen-
tary basin: an example of Vendian foredeep – foreland basin in south-
west of Siberian Platform. In: Leonov, Yu. G. & Volozh, Yu. A.
(eds) Sedimentary Basins: Methods of Research, Structure and Evol-
ution. Transactions, 543. Scientific World, Moscow, 159– 210 (in
Russian)
Sovetov, J. K. & Komlev, D. A. 2005. Tillite at base of the Oselok Group
in the Sayan region and the position of the lower boundary of the
Vendian in the southwestern Siberian Platform. Stratigraphy and
Geological Correlation, 13, 3– 34.
Sovetov, J. K. & Blagovidov, V. V. 2006. Late Riphean sedimentation
in Iya-Tumanshet aulacogene (south-west of the Siberian Platform):
supersequences and correlation with Riphean stratotype in the
Bashkiriyan anticlinorium. In: Maslov, A. V. (ed.) Lithological
Aspects of Geology Stratified Environments. Materials of 7th Ural
Lithological Meeting. Institute Geology and Geochemistry of Ural
 MARNYA FORMATION, OSELOK GROUP
Division of Russian Academy of Sciences, Ekaterinburg, 248–250
(in Russian).
Sovetov, J. K. & Donskaya, T. V. 2006. Basement in south-west of
Siberian Platform by analysis of erratic stones in the Early Vendian
tillite. In: Sklyarov, E. V. (ed.) The Geodynamic Evolution of the
Lithosphere in Central Asian Orogen (From Ocean to Continent).
Proceedings of Workshop on ESD RAS Integration Research Pro-
grams, 2. IZK, Irkutsk, 147– 150 (in Russian).
Sovetov, J. & Solovetskay, L. 2008. Oldest Vendian-Ediacaran
fossils in the Oselok Group: Contribution to Late Neoproterozoic
(Ediacaran) age of sea transgression and origin of the Siberian plat-
form cover. HPF-07 Rise and Fall of the Ediacaran (Vendian)
Biota. International Geological Congress, Oslo, 6– 14 August.
Abstracts. File://E:/33IGC/1343889.html.
Sovetov, J. K., Kulikova, A. E. & Medvedev, M. N. 2007a. Sedi-
mentary basins in the southwestern Siberian craton: Late Neoproter-
ozoic –Early Cambrian rifting and collisional events. In:
Linnemann, U., Nance, R. D., Kraft, P. & Zulauf, G. (eds) The
Evolution of the Rheic Ocean; From Avalonian-Cadomian Active
329
Margin to Alleghenian– Variscan Collision. Geological Society of
America, Special Paper 423, 549– 578.
Sovetov, J. K., Blagovidov, V. V. & Talibova, A. G. 2007b. Carbon
isotopes in Vendian carbonates in the southwestern Siberian craton:
correlation with geodynamic and palaeoclimate events and palaeo-
geography. In: Halimov, E. M. (ed.) Proceedings of the Vinogradov
XVIII Symposium on Isotope Geochemistry. Moscow, 245– 246 (in
Russian).
Turkina, O. M., Bibikova, E. V. & Nozhkin, A. D. 2003. Stages and
geodynamic settings of Early Proterozoic granite formation on the
southwestern margin of the Siberian craton. Doklady Earth Sciences,
389, 159–163 (in Russian).
Tyschenko, L. F. 1980. Regional correlation of Moty Formation deposits
in the Irkutsk amphitheatre. In: Karagodin, Yu. N. (ed.) The Pro-
blems of Lithostratigraphy. Nauka, Novosibirsk, 149–158 (in
Russian).
Vinogradov, V. I., Pokrovsky, B. G. et al. 1994. Isotopic geochemistry
and age of Upper Precambrian deposits in the West Siberian Platform.
Lithologia i Poleznye Iskopaemye, 4, 49 –76 (in Russian).
 Chapter 29

The Tsagaan Oloom Formation, southwestern Mongolia

FRANCIS A. MACDONALD
Department of Earth and Planetary Sciences, Harvard University, Cambridge, MA 02138, USA (e-mail: fmacdon@fas.harvard.edu)

Abstract: The Tsagaan Oloom Formation (Fm.) in southwestern Mongolia contains two Neoproterozoic glacial deposits, with diamic-
tite in the Maikhan Ul Member (Mb.) and in the Khongoryn Mb., which are separated by over 500 m of limestone. The Maikhan Ul Mb.
ranges in thickness between 5 m and greater than 300 m, expanding in deeper-water sections towards the SW, where it is composed of
two massive diamictites separated by over 100 m of sandstone, siltstone and shale. The basal 10 m of the overlying Tayshir Mb. of the
Tsagaan Oloom Fm. consists of a fine-laminated, dark grey limestone. The Khongoryn Mb. is composed primarily of limestone clasts in a
shale matrix, and is between 0 and 23 m thick. The overlying Ol Mb. contains sedimentary structures characteristic of basal Ediacaran cap
carbonates including micropeloids, tubestone stromatolites, giant wave ripples and former aragonite crystal fans. U– Pb evaporation ages
from zircons in the underlying Dzabkhan Volcanics constrain the Tsagaan Oloom Fm. to ,773 Ma, and tuffs within the Maikhan Ul and
Tayshir members testify to the potential for additional geochronology. The Cryogenian organic-rich limestone of the Tayshir Mb., which
lies between the two glacial deposits, is ideally suited for geochemical studies and has been the subject of several carbon, strontium and
rare earth element investigations. Limited palaeomagnetic studies suggest a mid- to low-latitude position of the Dzabkhan platform
during deposition of the glaciogenic strata, and additional studies are in progress.

Neoproterozoic diamictites are present on the Dzabkhan platform
(also referred to as the Zavkhan basin) of southwestern Mongolia,
in a .100 km NW – SE trending belt, with additional discontinu-
ous exposures further north. The most complete exposures of the
Tsagaan Oloom Fm. are between the Khasagty-Nuru ridge and
the Dzabkhan River (Fig. 29.1). The geology of the Dzabkhan plat-
form was first described by Bezzibetsev (1986), who divided the
stratigraphy into three formations (the Dzabkhan, Tsagaan Oloom
and Bayan Gol). Subsequent work focused on the early Cambrian
palaeontology of the Bayan Gol Fm., with an eye for correlation
with Siberia; the results of these studies were published entirely
in Russian (for a list of these references see Brasier et al. 1996b).
The first descriptions in English came in 1996 with the publication
of a Geological Magazine issue dedicated to the Neoproterozoic-
Cambrian stratigraphy of southwestern Mongolia (Brasier et al.
1996a). The studies reported therein were the product of two inter-
national field excursions, one in 1991 as part of the 21st Joint
Soviet – Mongolian Palaeontological Expedition (Zhegallo &
Zhuravelev 1991), and a second in 1993 sponsored by IGCP
Project 303 and the Mongolian Academy of Sciences (Dorjnamjaa
et al. 1993). The results of these excursions included the translation
of geological maps and measured sections into English (Khomen-
tovsky & Gibsher 1996), a reconnaissance chemostratigraphic
characterization of the Tsagaan Oloom and Bayan Gol formations
(Brasier et al. 1996b), and a detailed stratigraphic study of the
Maikhan Ul diamictite at Tsagaan Gol (Lindsay et al. 1996).
Recently, Macdonald et al. (2009) conducted detailed chemo-
and litho-stratigraphic studies on previously unstudied sections
and discovered an additional diamictite higher in the succession.
This work supported the earlier conclusion of Brasier et al.
(1996b) that the Maikhan Ul member is early Cryogenian in age
and established that the Khongoryn diamictite is an end Cryogen-
ian glacial deposit. With the new C-isotope chemostratigraphic
correlations and the documentation of a low-angle unconformity,
a .40 million year hiatus was identified within the Tsagaan
Oloom Formation, above the Khongoryn diamictite but below
the phosphorite horizon (Macdonald et al. 2009).
At Tsagaan Gol, the Maikhan Ul Mb. contains two diamictites
separated by over 100 m of sandstone, siltstone and shale
(Lindsay et al. 1996). This creates a bit of confusion in the litera-
ture, particularly with the discovery of an additional diamictite
higher in the Tsagaan Oloom Fm., because the two diamictites
within the Maikhan Ul Mb. have also been referred to as the
upper and lower diamictites of the Tsagaan Oloom Fm.
(Khomentovsky & Gibsher 1996; Lindsay et al. 1996). Macdonald
et al. (2009) grouped the lower two diamictites and the intervening
clastic units together in the Maikhan Ul Mb., while referring to the
diamictite c. 500 m higher in the sequence as the Khongoryn
member. Levashova et al. (2010) informally referred to the
Maikhan Ul diamictite as the Tayshir Fm. Here, we do not
follow this nomenclature because it creates unnecessary confusion,
particularly as the overlying carbonates have been previously
called the Tayshir Mb. (Macdonald et al. 2009).
Structural framework
The Dzabkhan terrane (also referred to as the Baydaric microcon-
tinent when grouped with the Baidrag terrane, Fig. 29.1a) is a com-
posite Precambrian terrane, hosting a heterogeneous Archaean and
Proterozoic crystalline basement intruded by c. 805 –770 Ma con-
tinental arc volcanism (Badarch et al. 2002; Zhao et al. 2006).
Based on similarities in the Neoproterozoic stratigraphy, radio-
metric ages in the underlying basement (Badarch et al. 1998),
and the continuity of aeromagnetic anomalies associated with
the fringing Neoproterozoic ophiolites (Buchan et al. 2002), the
southwestern margin of the Dzabkhan basin can be traced to the
western margin of the Khubsugul basin along the Tuva-Mongolia
border (Fig. 29.1a). The tectonic events that transformed the south-
western and western margins of the Dzabkhan and Khubsugul ter-
ranes from continental arcs to thermally subsiding passive margins
remain unclear. On the southern margin of the Dzabkhan terrane,
on the south side of the Khasagty-Nuru ridge, the Dzabkhan and
the Tsagaan Oloom Formations are separated by as much as
2 km of canabalizing, rift-related sediments with inter-fingering
basalt that are referred to as the Shargyngol complex (Ruzhentsev
& Burashnikov 1996). Facies patterns and the orientation of cross-
beds in the Tsagaan Oloom Fm. indicate deepening to the SW
(Macdonald et al. 2009). In the latest Ediacaran to early Cambrian,
the rifted passive margin began to subside again after a deposi-
tional hiatus of .40 Ma. It has been proposed that this accommo-
dation space was created by flexure with the arrival of the
Khantayshir-Dariv arc (Macdonald et al. 2009). With the early
Cambrian arc – continent collision, the Neoproterozoic stratigra-
phy was shortened and repeated in thrust blocks with a basal
detachment beneath the Dzabkhan Fm. The deformation is
largely brittle and thin-skinned, without involvement of the base-
ment. Early Palaeozoic granites and narrow NW-trending grabens

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 331– 337. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.29
332
Fig. 29.1. (a) Tectonic map of western Mongolia modified from Badarch et al. (2002) and Windley et al. (2007). Teeth on faults indicate the inferred dip of subduction
zones. Key: K-D Arc, Khantayshir-Dariv Arc; AZ, accretionary zone including the arc, metamorphic rocks and ophiolitic assemblages. (b) Geological map of the Tayshir
region, highlighting the members of the Tsagaan Oloom Fm. and the position of measured sections.
cut the Cambrian NNE-vergent structures. On the outcrop scale,
the diamictite and carbonate rocks of the Tsagaan Oloom Fm.
are little deformed with no apparent strain. Sedimentary structures
are typically preserved in limestone, but are often obfuscated by
recrystallization in dolomite.
Stratigraphy
The stratigraphy of the Dzabkhan basin (Fig. 29.2) begins with
.2 km of silicic to intermediate volcanic rocks of the Dzabkhan
Fm. On the south side of the Khasagty-Nuru ridge, the Dzabkhan
Fm. is succeeded by as much as 2 km of sandstone turbidites and
conglomerate that are referred to as the Shargyngol complex
(Ruzhentsev & Burashnikov 1996). North of the Khasagty-Nuru
ridge, the Shargyngol complex is ,100 m thick and often absent,
with the Maikhan Ul Mb. of the Tsagaan Oloom Fm. non-
conformably overlying the Dzabkhan Fm. The Maikhan Ul Mb. is
composed of diamictite, sandstone and shale, and varies in thickness
from 5 m to .275 m. The Maikhan Ul Mb. is overlain with a knife-
sharp contact by the Tayshir Mb., which consists of c. 570 m of lime-
stone that included three super-sequences (Macdonald et al. 2009).
The Tayshir Mb. is succeeded by the Khongoryn diamictite,
which is composed of limestone cobble to boulder lonestones in a
shale matrix and varies from 0 to 23 m in thickness. The Khongoryn
diamictite is in turn overlain by micropeloidal dolostones of the Ol
Mb. (Macdonald et al. 2009). In the subsequent transgression, for-
merly aragonite crystal fans are developed at the dolostone–
limestone transition. Above the post-glacial transgression, the Ol
Mb. shallows upwards from grey limestone rhythmite into c. 10 m
of limestone grainstone. The overlying Ulaan Bulagyn Mb. is up
to 500 m thick and is composed primarily of massive weathering
dolomite; however, in more distal sections, the Ulaan Bulagyn
Mb. thins to less than 100 m and is composed largely of limestone.
The Zunne Arts Mb. begins with distinct pink-coloured colum-
nar stromatolites (Boxonia grumulosa) that overly a karstic surface
with metre-scale relief (Macdonald et al. 2009). The Boxonia bio-
herms are overlain by 10 –20 m of violet and green shale that are
variably phosphatized and interbedded with lenses of dolomite
and microcrystalline to nodular phosphorite. This phosphatic
shale is overlain by more than 100 m of blue limestone rhythmite
F. A. MACDONALD
and ribbonite that include nodular black chert and bed parallel,
meandering ichnogenera (Goldring & Jensen 1996). Above the
Zunne Arts Mb. of the Tsagaan Oloom Fm., the early Cambrian
Bayan Gol Fm. is composed of c. 1000 m of mixed carbonate
and siltstone with a rich diversity of ichnogenera, small shelly
fossils and calcimicrobial patch reefs (Kruse et al. 1996).
Glaciogenic deposits and associated strata
The Maikhan Ul Member
The Maikhan Ul Mb. progressively thickens to the SW (Fig. 29.3),
but also displays considerable variability on individual thrust
blocks. For example, at the easternmost exposures on the
Tayshir Block (F718), the Maikhan Ul Mb. is only 6.7 m thick
and is composed predominantly of a massive cobble –boulder
clast diamictite, whereas just 1 km to the west (F713) it thickens
to 81.6 m with multiple diamictite units separated by 57 m of
massive, fine to coarse-grained sandstone. These sandstone
bodies are composed of graded centimetre- to metre-thick beds
and contain no evidence of tidal influence (Fig. 29.3).
Further south, in more distal sections, the Maikhan Ul Mb. con-
tinues to thicken. On the Khongoryn Block (F701), the Maikhan Ul
Mb. fills palaeo-canyons and varies in thickness between 160 and
283 m. One palaeo-canyon, directly west of F701, is c. 125 m deep
and 0.6 km wide. This palaeo-canyon has an erosive base that is
mantled with a volcanic-clast, cobble conglomerate and is filled
with stratified diamictite units with dropstones and striated clasts
(Macdonald et al. 2009), thin-bedded sandstone beds and a
c. 0.5-m-thick carbonate bed. Two massive to bedded diamictite
units lie above the canyon fill, separated by 62 m of siltstone and
sandstone with rare cobble lonestones and two additional
c. 0.5-m-thick carbonate beds.
At Tsagaan Gol, where the member measures 304 m, again,
two diamictite units are separated by a thick sequence of flat-
bedded shale, siltstone and sandstone (Lindsay et al. 1996).
Cobble lonestones are present in both the basal and upper metre
of this clastic succession, between the two massive diamictites.
Khomentovsky & Gibsher (1996) also reported a measured
section from Urtor Tsakhir Mountain, c. 120 km west of Tayshir,
 TSAGAAN OLOOM FORMATION 333

Bayan Gol Fm. c. 1 km of where the Maikhan Ul Mb. is even thicker but still preserves this
general stratigraphic pattern of two diamictite units separated by
BG
clastic & carbonate rocks
-8 -4 0 4 8 sandstone and siltstone. In this area, mudcracks are also well devel-
oped near the top of these intervening clastic units.
In both the upper and lower diamictite units of the Maikhan Ul
Zunne Arts Member
Mb., the most common lithology comprises a matrix-dominated
diamictite with shale and sandstone encasing sub-rounded cobble
derived from the underlying Dzabkhan Formation; granite, meta-
1400 morphic and carbonate clasts of unknown origin are also present.
Also, near the base of the Maikhan Ul Mb. at Tsagaan Gol,
C clasts of deformed soft sediment have been reported (Lindsay
P et al. 1996). Clast size varies from grit to blocks .2 m across.

1200 The Khongoryn Member

OO
Ulaan Bulagyn Member

OO The Khongoryn Mb. is thickest one gully east of Tsagaan Gol

(F723); however, like the Maikhan Ul Mb., there are significant
facies changes both from north to south and from east to west
OO
(Fig. 29.4). East of Tsagaan Gol, the diamictite is 23 m thick and
composed of pebble- to boulder-sized clasts of blue-grey limestone
1000 from the underlying Tayshir member in a dark grey shale matrix
that becomes more marly and lighter coloured up-section. Striated
clasts and limestone clasts with soft sedimentary deformation are
also present. Just 6 km west, near Tsagaan Gol, the diamictite
Tsagaan Oloom Formation

is nearly absent and only 2 m of recessive shale are preserved.

The Khongoryn Mb. is also well developed on the Khongoryn
C block, south of Bayan Gol (F708), where it consists of 14.7 m of
800 Maieberg
K Ol

sub-rounded limestone pebbles, cobbles and boulders in a grey

anomaly shale matrix. Both laterally and up-section, clasts are irregularly
OO distributed, varying from clast-poor facies to boulder nests. To
the NE of the Khongoryn block, the Khongoryn Mb. is either
thin or absent.

600
Tayshir anomaly Associated carbonate rocks
Tayshir Member

The basal 10 m of the Tayshir Mb., which overlies the Maikhan

C Ul Mb., is composed of a dark grey, millimetre-laminated lime-
OO
stone. Overall, the Tayshir Mb. consists of ,650 m of limestone
OO that record three regionally extensive sequences. The base of the
400 OO first sequence is defined by a c. 10-m-thick, dark grey (weather-
ing to tan), millimetre laminated limestone that is succeeded by
c. 100 m of limestone marl and rhythmite, shoaling up-section to
c. 20 m of grainstone. The second sequence begins with c. 10 m
of limestone marl and rhythmite followed by c. 200 m of mas-
C sively bedded, blue grainstone and microbialaminite. The third
200 Rasthof sequence begins with c. 50 m of limestone rhythmite and debris
anomaly flows with numerous black chert beds and nodules, and then
shallows up-section to c. 210 m of dark, fetid limestone micro-
-8 -4 0 4 8 bialaminite and minor grainstone with giant ooids (.0.5 cm
diameter).
MU

Maikhan Ul diamictite The Ol Mb., which overlies the Khongoryn diamictite, begins
with 7–40 m of buff to pink coloured, largely recrystallized,
0 micropeloidal dolostone. Low-angle cross-stratification (Aitken
(m) vv Dzabkhan Formation
vv 1991), tubestone stromatolites (Corsetti & Grotzinger 2005),
DV

vv c. 2 km of voclanic & and giant wave ripples (Allen & Hoffman 2005) are also present
vv
volcanoclastic rocks in the Ol Mb. dolomite (Fig. 29.4). The Ol Mb. transgresses
upwards into limestone ribbonite and then rhythmite with
limestone C chert
dolomite P phosphorite
grainstone & OO giant ooids
microbialaminite Fig. 29.2. Composite carbon chemo- and lithostratigraphy of the Dzabkhan
ribbonite stromatolite basin, modified from Macdonald et al. (2009). K, Khongoryn diamictite. The
trace fossils Rasthof anomaly was first well-documented in Namibia (Yoshioka et al. 2003)
rhythmite/shale and, like the Maieberg anomaly (Halverson et al. 2005), has now been reported
debris flow globally. The Tayshir anomaly was first documented in Mongolia (Macdonald
diamictite flooding et al. 2009) and a potentially correlative mid-Cryogenian anomaly is also
siltstone/ present in the Bonahaven Fm. of the Dalradian Supergroup in Scotland and
sandstone exposure
Ireland (Prave et al. 2009).
334 F. A. MACDONALD

Tayshir Block Khongoryn Tsagaan below the Maikhan Ul Mb., indicating only a limited hiatus.
Block, F701 Block,F724 However, it is not clear if this sandstone is part of the Shargyngol
E 1 km W suite or should be included within the Maikhan Ul Mb. South of
Tsagaan Gol, the clastic units between the Dzabkhan Fm. and
the lower Maikhan Ul diamictite unit thicken to over 100 m and
lack any evidence of glacial influence on sedimentation. Conver-
sely, to the east and north of Tayshir, both the Maikhan Ul Mb.
and the Dzabkhan Fm. thin, with the diamictites of the Maikhan
250 Ul resting on an erosional contact with the Dzabkhan Fm. or the
basement rock.
Contact between the Maikhan Ul Mb. and the overlying Tayshir
Mb. is very sharp. The Tayshir Mb. rests conformably on a later-
ally persistant, c. 10-cm-thick layer of red clay that marks the
top of the Maikhan Ul Mb.
The Khongoryn diamictite typically lies above blue-grey, giant
200 ooid grainstones of the lower limestone of the Tsaagan Oloom
Fm.; however, on the Khongoryn Block (F708), there is an
additional 7.7 m of black shale and rhythmite preserved above
the ooids. The erosion of this shale likely provides the detrital
matrix for the Khongoryn diamictite. The Khongoryn diamictite
is overlain with a sharp yet conformable contact by dolostone
and limestone of the Ol Mb.

150 F718 F713

Chemostratigraphy
volcanic Strontium isotope values rise from 0.7067 to 0.7073 in the lime-
massive stones of the Tayshir Mb. In the Ulaan Bulagyn Mb. 87Sr/86Sr
diamictite values rise from 0.7073 to 0.7077, and then in the Zunne Arts
Mb. from 0.7078 to over 0.7080 (Brasier et al. 1996b; Shields
matrix composition et al. 2002).
100 Carbonate d13C values in the dark-grey laminated limestone
sandstone
above the Maikhan Ul Fm. are moderately negative with values
siltstone increasing upwards through the overlying pink marls to þ8‰
(Fig. 29.2). Values plummet abruptly at the flooding surface in
shale the middle of the Tayshir Mb., reaching a low of – 7.5‰. Mac-
limestone donald et al. (2009) refer to this sudden drop in d13C values as
clast size the Tayshir anomaly. From this nadir, d13C values increase
smoothly to þ9‰ for the upper Tayshir Mb., with values reported
50 boulder as high as þ11‰ (Brasier et al. 1996b). Shields et al. (2002) also
cobble measured d13C of organic matter in the Tayshir Mb. of the Tsagaan
gravel Oloom Fm. and found that trends roughly followed those exhibited
by the d13C in carbonate.
clast composition Overlying the upper diamictite, d13C values in the Ol Mb. begin
granite around –1‰ and follow a sigmoidal profile (Fig. 29.2). Values
clastic return to c. –1‰ at the top of the dolostone, and then decrease
again at the limestone – dolomite transition, reaching a nadir of
carbonate –6‰. Above the Ol Mb., d13C values oscillate around þ3‰
0
volcanic for most of the Ulaan Bulagyn Mb., returning to 0‰ below the
metres
sub-Zunne Arts Mb. karstic surface.
Fig. 29.3. Stratigraphy of the Maikhan Ul Mb. Locations of sections are shown Sheilds et al. (1997, 2002) reported a Ce anomaly in the Tayshir
in Figure 29.1b. Section F724 is modified from Lindsay et al. (1996). Symbols Mb. from samples collected at Tsagaan Gol. In this section, the
used that are not in the legend are in Figure 29.2. recessive strata bearing the C-isotope anomaly are not exposed,
and thus, they did not document the transgressive sequence or
the negative C-isotope values.
c. 5-cm-tall former aragonite crystal fans present at the limestone –
dolostone transition. Crystal fans are present both as individual
blades growing upwards into the sediment, and as crystal fan Palaeolatitude and palaeogeography
shrubs that are over 10 cm across.
Recent palaeomagnetic studies on the 805–770 Ma Dzabkhan Fm.
indicate that the Dzabkhan terrane was located at a latitude of
Boundary relations with overlying and underlying 478 þ 168/–128 (Levashova et al. 2010). From palaeomagnetic
non-glacial units studies on peri-Siberian terranes, including the early Cambrian
Salaany Gol Fm. on the Dzabkhan terrane, Kravchinsky et al.
The Maikhan Ul Mb. rests with an erosive base on the Shargyngol (2001) concluded that the Tuva-Mongolia belt was at low latitude,
suite and the Dzabkhan Formation (Khomentovsky & Gibsher adjacent to Siberia throughout the Ediacaran and Cambrian.
1996), and fills palaeo-topography with conglomerates lining However, this study lacked a robust confidence test (i.e. only a
palaeo-valleys (Fig. 29.3). According to Lindsay et al. (1996), at reversal test with few samples and low resolution). Moreover, an
Tsagaan Gol, soft-sedimentary deformation is present in sandstone earlier study on the Salaany Gol Fm. gave entirely different
 TSAGAAN OLOOM FORMATION
Tayshir Block Khongoryn Block
E 18 km 3 km 10 km W E 18 km W
50
C C
C C
C
25
C C
(metres)
-25
F710 F716 F715 F706 F725 F708
micropeloidal dolomite giant wave ripples
tubestone stromatolites
results (Evans et al. 1996), but it was also compromised by uncer-
tainty in the relative ages of the folds used in the fold test and poss-
ible magnetic overprints. Further palaeomagnetic studies on the
Dzabkhan terrane are necessary, and are in progress (Gregory
et al. 2007). Nonetheless, as non-skeletal carbonate is preferen-
tially produced in the warmest parts of the surface ocean (Broecker
& Peng 1982), and the Tsaagan Oloom Fm. is dominated by
shallow-water carbonates, it is likely that the Dzabkhan terrane
was situated at low latitudes (less than 308) throughout the Cryo-
genian and Ediacaran.
Along with other peri-Siberian terranes, it has been suggested
that the Dzabkhan terrane occupied a Precambrian position
between Siberia and Laurentia (Gladkochub et al. 2006), and
rifted away from Siberia in the late Neoproterozoic (Sengor &
Natal’in 1996; Kuzmichev et al. 2001; Kuzmichev et al. 2005).
Sengor & Natal’in (1996) further posit that throughout the late
Neoproterozoic and early Palaeozoic, the Dzabkhan terrane was
attached to the Central Mongolian Block, which along with other
terranes, stretched to the present day Sea of Okhotsk. Both the
Tuva-Mongolia (including the Dzabkhan terrane) and Central
Mongolian Blocks host Cambrian trilobites endemic to Siberia
(Astashkin et al. 1995) and Silurian brachiopods characteristic of
the peri-Siberian realm (Hou & Boucot 1990). Alternatively, citing
similarities in SHRIMP ages on zircons, Zhao et al. (2006) and
Demoux et al. (2009) have suggested that the Baydrag and Dzab-
khan terranes originated from the northern margin of Gondwana.
This reconstruction is supported by the palaeomagnetic results of
Levashova et al. (2010), which point to Neoproterozoic connec-
tions with India, South China, Tarim or Australia.
Geochronological constraints
Although the diamictites of the Dzabkhan basin have not been
directly dated radiometrically, maximum age constraints on the
glacial deposits are provided by zircons from rhyolites within
the Dzabkhan Formation of 777 + 6 Ma (Zhao et al. 2006),
803.4 + 8.0 and 773.5 + 3.6 Ma (U– Pb laser evaporation, Leva-
shova et al. 2010).
Discussion
A glacial origin of the Maikhan Ul diamictite units is indicated by
the presence of faceted and striated clasts, and bullet-shaped
335
Tsagaan Block
E 15 km 6 km W
C
C C
Fig. 29.4. Stratigraphy of the Khongoryn
diamictite and overlying Ol Mb. Positions
F726 F723 F724 of measured sections are in Figure 29.1b.
Symbols used that are not in the legend are
crystal fans
in Figures 29.2 and 29.3.
dropstones that penetrate laminated beds. At Tsagaan Gol,
cobble dropstones are also present in both the basal and upper
metre of the clastic succession, between the two massive diamictite
units. Moreover, in more proximal settings, such as on the Khon-
goryn and Tayshir blocks, rare lonestones are present within the
sandstone beds. These observations indicate that the deposition
of the clastic units was influenced, at least in part, by glaciation.
Macdonald (2009) inferred a pro-glacial environment, including
emergent conditions and proglacial lakes, for both the diamictite
and the clastic units of the Maikhan Ul Mb. from the presence of
mud cracks and 0.5-m-thick carbonate beds. A pro-glacial
environment is further supported by high lateral facies variability.
Within this context, the intervening clastic units can be interpreted
as a step-back of the ice-line, and the upper diamictite as an
ice-advance.
The rise in 87Sr/86Sr from 0.7067 to 0.7073 in the limestone of
the Tayshir Mb. is mirrored in the Rasthof Fm. in Namibia and the
Keele Fm. in NW Canada, suggesting that the underlying Maikhan
Ul diamictites are early Cryogenian glacial deposits (Halverson
et al. 2007). The black laminated cap carbonate above the
Maikhan Ul diamictites also contains a modest negative
C-isotope anomaly similar to the Rasthof Fm. (Yoshioka et al.
2003); the extremely enriched values of the Tayshir Mb. are also
consistent with a Cryogenian age (Hoffman & Schrag 2002;
Halverson et al. 2005). The Tayshir anomaly (Macdonald et al.
2009) can be correlated to the moderately negative 13dC values
obtained from the exposure-surface riddled Gruis Fm. of northern
Namibia (Halverson et al. 2005) and the Cryogenian Bonahaven
Dolomite of the British-Irish Caledonides (McCay et al. 2006),
or to the Trezona anomaly in Australia (McKirdy et al. 2001)
and Namibia (Halverson et al. 2005).
A glacial origin of the Khongoryn diamictite is indicated by
the presence of striated clasts and dropstones that penetrate lami-
nated beds. The Khongoryn diamictite is thin or absent on the
most proximal sections to the NE of the map area (Fig. 29.1). In
more distal sections to the SW, the diamictite is composed of
cobble to boulder clasts of the underlying limestone within a
weakly bedded shale to marl matrix. This shale matrix was
likely derived via erosion of the shale unit in the upper portion
of the Tayshir Mb., which is only present on the Khongoryn and
Tsagaan blocks. The lack of stratigraphic architecture within the
deposit, the irregular distribution of ice-rafted debris, such as
boulder nests, and the conformable overlying contact with the Ol
Mb. indicate that this deposit formed as a single rainout during
the terminal deglaciation.
336 F. A. MACDONALD
The overlying basal dolostone of the Ol Mb. is composed of
fine-laminated micropeloids and contains tubestone stromatolites,
giant wave ripples and pseudomorphosed crystal fans. These
peculiar sedimentary structures, their specific order, and the
distinct, sigmoidal C-isotope profile are characteristic of basal
Ediacaran cap carbonates globally (Hoffman et al. 2007). This
suggests that the underlying Khongoryn diamictite is an
end-Cryogenian glacial deposit (Macdonald et al. 2009), with
the termination bracketed elsewhere by U –Pb ages of
635.51 + 0.54 Ma and 635.23 + 0.57 Ma (Condon et al. 2005).
The phosphorites in the Zunne Arts Mb. rest above a low-angle
unconformity. Ediacaran chemostratigraphic correlations indicate
that this surface represents a .40 Ma depositional hiatus, and as
such is unrelated to the glacial deposits in the Khongoryn
member (Macdonald et al. 2009).
Further sedimentological studies on the Maikhan Ul Mb. are
needed to understand the significance of the interbedded sandstone
and siltstone, and to determine if the Maikhan Ul diamictites
represent a single episode of deglaciation or multiple ice advances
and retreats. Stratigraphic studies are also needed to better under-
stand the basin dynamics of the Neoproterozoic– Cambrian
margins of the Dzabkhan terrane, and the relationships with
other Mongolian terranes. The palaeogeography of the Peri-
Siberian terranes also remains speculative. It is clear, however,
that island arcs surrounded the Dzabkhan terrane for much of the
Neoproterozoic and Cambrian, and therefore there is excellent
potential for U – Pb zircon geochronology studies in the Dzabkhan
basin. Furthermore, the low-grade and high organic content of
the limestone in the Tayshir Mb. is ideally suited for multi-proxy
studies to better constrain the geochemical evolution of Cryogen-
ian oceans.
I thank field assistants B. Boldoo, E. Oyun, T. Adiya, J. Otgonhuu and U. Bold. I
also thank Bayassa and A. Amaglaan for help with logistics. I thank D. Jones and
P. Hoffman for comments and help in the field. I am grateful to P. Hoffman and
NSF grant EAR-0417422 (to PFH) for support of this work. This represents a con-
tribution of the IUGS- and UNESCO-funded IGCP (International Geoscience
Programme) project #512.
References
Aitken, J. D. 1991. The Ice Brook Formation and Post-Rapitan, Late
Proterozoic glaciation, Mackenzie Mountains, Northwest Territories.
Geological Survey of Canada Bulletin, 404, 1 –43.
Allen, P. A. & Hoffman, P. F. 2005. Extreme winds and waves in
the aftermath of a Neoproterozoic glaciation. Nature, 433, 123– 127.
Astashkin, V. A., Pegel, T. V. et al. 1995. The Cambrian System of the
Foldbelts of Russia and Mongolia. International Union of Geological
Sciences, London.
Badarch, G., Byamba, J. et al. 1998. Geological Map of Mongolia.
Mineral Resources Authority of Mongolia, Ulaan Baator.
Badarch, G., Cunningham, W. D. & Windley, B. 2002. A new terrane
subdivision for Mongolia: implications for the Phanerozoic
crustal growth of Central Asia. Journal of Asian Earth Sciences,
21, 87 –110.
Bezzubetsev, V. V. 1986. On the Precambrian –Cambrian stratigraphy
of the Dzabkhan River Basin. Materials on the Geology of MPR,
Gostopotekhizdat, 1963, 29– 42.
Brasier, M. D., Dorjnamjaa, D. & Lindsay, J. F. 1996a. The Neoproter-
ozoic to early Cambrian in southwest Mongolia: an introduction.
Geological Magazine, 133, 365–369.
Brasier, M. D., Shields, G., Kuleshov, V. N. & Zhegallo, E. A.
1996b. Integrated chemo- and biostratigraphic calibration of early
animal evolution: Neoproterozoic – early Cambrian of southwest
Mongolia. Geological Magazine, 133, 445– 485.
Broecker, W. S. & Peng, T. H. 1982. Tracers in the Sea. LDEO Press,
Lamont-Doherty Earth Observatory, Palisades, NY.
Buchan, C., Pfander, J. et al. 2002. Timing of accretion and collisional
deformation in the Central Asian Orogenic Belt: implications
of granite geochronology in the Bayankhongor Ophiolite Zone.
Chemical Geology, 192, 23 –45.
Condon, D. J., Zhu, M., Bowring, S. A., Wang, W., Yang, A. & Jin, Y.
2005. U– Pb ages from the Neoproterozoic Doushanto Formation,
China. Science, 308, 95– 98.
Corsetti, F. A. & Grotzinger, J. P. 2005. Origin and significance of tube
structures in Neoproterozoic post-glacial cap carbonates: example
from Noonday Dolomite, Death Valley, United States. Palaios, 20,
348– 363.
Demoux, A., Kroener, A., Badarch, G., Jian, P., Tomurhuu, D. &
Wingate, M. T. D. 2009. Zircon ages from the Baydrag block and
the Bayankhongor ophiolite zone: time constraints on late Neoproter-
ozoic to Cambrian subduction- and accretion-related magmatism in
central Mongolia. Journal of Geology, 117, 377–397.
Dorjnamjaa, D., Bat-Ireedui, Y. A., Dashdavaa, Z. & Soelmaa, D.
1993. Precambrian–Cambrian Geology of the Dzavkhan Zone.
Earth Sciences Department, Oxford.
Evans, D. A. D., Zhuravlev, A. Y., Budney, C. J. & Kirschvink, J. L.
1996. Palaeomagnetism of the Bayan Gol Formation, western
Mongolia. Geological Magazine, 133, 487–496.
Gladkochub, D. P., Wingate, M. T. D., Pisarevsky, S. A., Donskaya,
T. V., Mazukabzov, A. M., Ponomarchuk, V. A. & Stanevich, A.
M. 2006. Mafic intrusions in southwestern Siberia and implications
for a Neoproterozoic connection with Laurentia. Precambrian
Research, 147, 260– 278.
Goldring, R. & Jensen, S. 1996. Trace fossils and biofabrics at the
Precambrian-Cambrian boundary interval in western Mongolia.
Geological Magazine, 133, 403– 415.
Gregory, L. C., Meert, J. G., Levashova, N. & Gibsher, A. S. 2007.
Paleomagnetic and geochronologic data from Central Asia: infer-
ences for Early Paleozoic tectonic evolution and timing of worldwide
glacial events. American Geophysical Union, Fall Meeting,
GP43C –1487.
Halverson, G. P., Hoffman, P. F., Schrag, D. P., Maloof, A. C. & Rice,
A. H. N. 2005. Toward a Neoproterozoic composite carbon-isotope
record. Geological Society of America Bulletin, 117, 1181–1207.
Halverson, G. P., Dudás, F. O., Maloof, A. C. & Bowring, S. A. 2007.
Evolution of the 87Sr/86Sr composition of Neoproterozoic Sea-
water. Palaeogeography, Palaeoclimatology, Palaeoecology, 256,
103–129.
Hoffman, P. F. & Schrag, D. P. 2002. The snowball Earth hypothesis;
testing the limits of global change. Terra Nova, 14, 129–155.
Hoffman, P. F., Halverson, G. P., Domack, E. W., Husson, J. M.,
Higgins, J. A. & Schrag, D. P. 2007. Are basal Ediacaran (635
Ma) post-glacial ‘cap dolostones’ diachronous? Earth and Planetary
Science Letters, 258, 114– 131.
Hou, H. F. & Boucot, A. J. 1990. The Balkhash-Mongolia-Okhotsk
Region of the Old World Realm. Geological Society, London,
Memoir, 12, 297– 303.
Khomentovsky, V. V. & Gibsher, A. S. 1996. The Neoproterozoic –
Lower Cambrian in northern Govi-Altai, western Mongolia: regional
setting, lithostratigraphy and biostratigraphy. Geological Magazine,
133, 371– 390.
Kravchinsky, V. A., Konstantinov, K. M. & Cogne, J.-P. 2001.
Palaeomagnetic study of Vendian and Early Cambrian of South
Siberia and Central Mongolia: was the Siberian platform assembled
at this time? Precambrian Research, 110, 61 –92.
Kruse, P. D., Gandin, A., Debrenne, F. & Wood, R. 1996. Early Cam-
brian bioconstructions in the Zavkhan Basin of western Mongolia.
Geological Magazine, 133, 429– 444.
Kuzmichev, A., Bibikova, E. V. & Zhuravlev, D. Z. 2001. Neoproter-
ozoic (800 Ma) orogeny in the Tuva-Mongolia Massif (Siberia):
island arc-continent collision at the northeast Rodinia margin.
Precambrian Research, 110, 109– 126.
Kuzmichev, A., Kroener, A., Hegner, E., Dunyi, L. & Yusheng, W.
2005. The Shishkhid ophiolite, nothern Mongolia: a key to the
reconstruction of a Neoproterozoic island-arc system in central
Asia. Precambrian Research, 138, 125–150.
Levashova, N. M., Kalugin, V. M., Gibsher, A. S., Yff, J., Ryabinin,
A. B., Meert, J. & Malone, S. J. 2010. The origin of the Baydaric
microcontinent, Mongolia: constraints from paleomagnetism and
geochronology. Tectonophysics, 485, 306–320.
 TSAGAAN OLOOM FORMATION
Lindsay, J. F., Brasier, M., Shields, G., Khomentovsky, V. V. &
Bat-Ireedui, Y. A. 1996. Glacial facies associations in a Neoproter-
ozoic back-arc setting, Zavkhan Basin, western Mongolia. Geologi-
cal Magazine, 133, 391–402.
Macdonald, F. A. 2009. Neoproterozoic stratigraphy of Alaska and
Mongolia. PhD, Harvard University.
Macdonald, F. A., Jones, D. S. & Schrag, D. P. 2009. Stratigraphic and
tectonic implications of a new glacial diamictite–cap carbonate
couplet in southwestern Mongolia. Geology, 37, 123–126.
McCay, G. A., Prave, A. R., Alsop, G. I. & Fallick, A. E. 2006. Glacial
trinity: Neoproterozoic Earth history within the British-Irish
Caledonides. Geology, 34, 909– 912.
McKirdy, D. M., Burgess, J. M. et al. 2001. A chemostratigraphic over-
view of the late Cryogenian interglacial sequence in the Adelaide
fold –thrust belt, South Australia. Precambrian Research, 106,
149– 186.
Prave, A. R., Fallick, A. E., Thomas, C. W. & Graham, C. M. 2009.
A composite C-isotope profile for the Neoproterozoic Dalradian
Supergroup of Scotland and Ireland. Journal of the Geological
Society of London, 166, 1 –13.
Ruzhentsev, S. V. & Burashnikov, V. V. 1996. Tectonics of the
western Mongolian Salairides. Geotectonics, 29, 379– 394.
Sengor, A. C. & Natal’in, B. A. 1996. Paleotectonics of Asia: fragments
of synthesis. In: Yin, A. & Harrison, M. (eds) The Tectonic Evol-
ution of Asia. Cambridge University Press, Cambridge, 486– 640.
337
Shields, G., Stille, P., Brasier, M. & Atudorei, N.-V. 1997. Stratified
oceans and oxygenation of the late Precambrian environment: a post
glacial geochemical record from the Neoproterozoic of W. Mongolia.
Terra Nova, 9, 218–222.
Shields, G. A., Braiser, M. D., Stille, P. & Dorjnamjaa, D. 2002.
Factors contributing to high d13C values in Cryogenian limestones
of western Mongolia. Earth and Planetary Science Letters, 196,
99– 111.
Windley, B. F., Alexeiev, D., Xiao, W., Kroener, A. & Badarch, G.
2007. Tectonic models for accretion of the Central Asian
Orogenic Belt. Journal of the Geological Society of London, 164,
31– 47.
Yoshioka, H., Asahara, Y., Tojo, B. & Kawakami, S. 2003.
Systematic variations in C, O, and Sr isotopes and elemental concen-
trations in Neoproterozoic carbonates in Namibia: implications
for a glacial to interglacial transition. Precambrian Research, 124,
69– 85.
Zhao, Y., Song, B. & Zhang, S. H. 2006. The Central Mongolian micro-
continent: its Yangtze affinity and tectonic implications. In: Jahn,
B. M. & Chung, L. (eds) Symposium on Continental Growth and
Orogeny in Asia. Taipei, Taiwan, 135– 136.
Zhegallo, L. & Zhuravelev, A. Y. 1991. Guidebook for the
International Excursion to the Vendian– Cambrian Deposits
of the Dzabkhan Zone of Mongolia. Palaeontological Institute,
Moscow.
 Chapter 30

The Khubsugul Group, Northern Mongolia

FRANCIS A. MACDONALD* & DAVID S. JONES

Department of Earth and Planetary Sciences, Harvard University, Cambridge, MA 02138, USA
*Corresponding author (e-mail: fmacdon@fas.harvard.edu)

Abstract: The Khubsugul Group of northern Mongolia contains diamictites in the Ongoluk and Khesen formations that are succeeded
by a stratiform phosphorite deposit and .2 km of early Cambrian dolomite. The stratigraphy of the Khubsugul Group, including the two
diamictites, can be correlated with that of the Dzabkhan platform in southern Mongolia. By correlation, the Ongoluk diamictite is an early
Cryogenian glacial deposit. A glaciogenic origin is inferred from the presence of striated clasts and bed-penetrating dropstones. The
younger Khesen diamictite consists predominantly of a massive carbonate-clast diamictite, but also contains bed-penetrating dropstones
in rare stratified facies, and is inferred to be end Cryogenian in age. The two diamictites are separated by as much as 250 m of allodapic
carbonate. The phosphorite in the upper Khesen Formation (Fm.) is likely latest Ediacaran to early Cambrian in age and is separated from
the glacial deposits by a major hiatus. Consequently, no links can be made between the phosphogenesis and the glacial deposits. Only
limited geochemical, geochronological and palaeomagnetic results from the Khubsugul basin have been reported to date, but work is
ongoing and there is strong potential for future studies.

Diamictites in the Ongoluk and Khesen formations of the
Khubsugul Group of northern Mongolia are exposed discontinu-
ously in a c. 250 km north –south belt (Fig. 30.1). The most com-
plete exposures of the two diamictites and the overlying carbonate
and phosphorite occur along the Khesen and Ongoluk Gols (tr.
Rivers), on the west side of Lake Khubsugul (Fig. 30.2;
50842.50 N, 1008110 E and 50844.30 N, 100812.20 E).
Most of the studies in the Khubsugul basin have focused on the
phosphorite deposits and the regional tectonics, with the diamic-
tites mentioned only in passing (e.g. Ilyin 1990, 1998). Although
much of this work is in the Russian literature, several key
manuscripts from Litologiya i Poleznye Iskopaemye have been
translated to English in Lithology and Mineral Resources (e.g.
Osokin & Tyzhinov 1998; Ilyin 2004). Additionally, a Russian
field guide of the Khubsugul Basin was produced for an IGCP
excursion to the phosphorite localities (Ilyin & Byamba 1980).
This field trip spawned the hypothesis that ice rings orbited
Precambrian Earth, that the shadow of these rings initiated
the Neoproterozoic glaciations, and that their collapse led to
phosphogenesis and precipitated the Cambrian radiation
(Sheldon 1984).
Geological work commenced in the Khubsugul basin in the
mid-1960s with the discovery of ore-grade phosphorites (Donov
et al. 1967). The Khubsugul Group was originally described in
detail, including the identification of diamictites in the Ongoluk
Fm., by Ilyin (1973). Osokin & Tyzhinov (1998) later differen-
tiated a second diamictite in the basal Khesen Fm. and documented
the presence of both diamictites throughout the Khubsugul basin.
The diamictites in the Khubsugul basin have not been formally
named, and the formations within which they occur have been
named differently in Mongolia and Siberia (see Chumakov
2011). The diamictite at the base of the Ongoluk Fm. in the Khub-
sugul basin is likely equivalent to the diamictite in the Khushatai
Fm. of the Sarkhoi Group (Osokin & Tyzhinov 1998). Otherwise,
the Khubsugul Group in Mongolia is largely correlative with the
Boxon Group in Siberia, with the Khesen Fm. roughly equivalent
to the Zabit Fm. Correlations of specific diamictites in the Khesen
and Zabit formations are complicated by multiple conglomeratic
horizons within both formations (Kheraskova & Samygin 1992;
Osokin & Tyzhinov 1998).
Ilyin (1973, 2004) referred to the upper c. 50 m of the lower
diamictite as the ‘perforated shales’ after the holes left behind
from eroded carbonate clasts, and he defined the diamictite as
the basal member of the Khubsugul Group; however, the diamic-
tite is underlain by an additional several hundred metres of
clastic rocks that he included with the underlying Arasan Fm.
Following Osokin & Tyzhinov (1998), we include these clastic
rocks with the Ongoluk Fm. of the Khubsugul Group, and refer
to the diamictite in the Ongoluk Fm. as the Ongoluk diamictite.
The upper diamictite is in the basal Khesen Fm. of the Khubsugul
Group and is herein referred to as the Khesen diamictite. The two
diamictites are separated by 100– 250 m of carbonate of the upper
Ongoluk Fm.
Structural framework
The Khubsugul terrane is a composite Precambrian terrane, host-
ing a heterogeneous Archaean and Proterozoic crystalline base-
ment intruded by c. 800 Ma continental arc volcanism (Badarch
et al. 2002). Based on similarities in Neoproterozoic stratigraphy,
radiometric ages in the underlying basement (Badarch et al. 2002)
and the continuity of aeromagnetic anomalies associated with
fringing Neoproterozoic ophiolites (Buchan et al. 2002), the south-
western margin of the Dzabkhan platform can be traced to the
western margin of the Khubsugul basin along the Tuva-Mongolia
border (Fig. 30.1, Macdonald 2011). The eastern boundary of the
Dzabkhan terrane is obscured by Palaeozoic intrusions (Badarch
et al. 1998) and, consequently, pre-Ordovician connections with
the Baidrag terrane remain ambiguous. Overlap assemblages indi-
cate that the Dzabkhan, Khubsugul, Baidrag and Tarvagatay
terranes had amalgamated into a single continental mass by the
Devonian (Badarch et al. 2002).
On both the Khubsugul and Dzabkhan terranes, Palaeoprotero-
zoic basement is overlain by thick volcanic –volcaniclastic succes-
sions (Badarch et al. 2002). On the Khubsugul terrane, the Sarkhoi
volcanic rocks have been interpreted as having a continental arc
affinity (Kuzmichev et al. 2001). These are unconformably over-
lain by the rift-related volcanic rocks and clastic sediments of
the Dharkhat Group (Ilyin 1990). The Khubsugul terrane trans-
formed from a continental arc to a thermally subsiding passive
margin after the c. 800 Ma Shishkhid arc accreted to its western
margin and prior to rifting along its eastern margin (Kuzmichev
et al. 2005).
Ilyin (2004) documented a deepening to the west in the
phosphorite-bearing strata of the Khesen Fm. and the overlying

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 339– 345. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.30
340 F. A. MACDONALD & D. S. JONES
Fig. 30.1. Tectonic map of western Mongolia modified from Badarch et al.
(2002) and Windley et al. (2007). Teeth on faults indicate the inferred dip of
subduction zones. K-D Arc, Khantayshir-Dariv Arc; AZ, accretionary zone
including arcs, metamorphic rocks and ophiolites; NP, Neoproterozoic; Ord,
Ordovician; Carb, Carboniferous.
early Cambrian shelf carbonates, and suggested these were depos-
ited in a rift graben. Macdonald et al. (2009) alternatively posited
that most of the Neoproterozoic sediments on the Dzabkhan and
Khubsugul terranes were deposited on a thermally subsiding
rifted margin, but that the late Ediacaran to early Cambrian phos-
phorites and overlying early Cambrian deposits formed in a fore-
deep basin in response to the Salarian orogeny (Ruzhentsev &
Burashnikov 1996).
Exposures of Cambrian carbonates on the west side of Lake
Khubsugul are folded in tight, south-plunging synclines and are
cut by Palaeozoic granites (Fig. 30.3), perhaps also related to an
extension of the early Cambrian Salarian orogeny (Ruzhentsev &
Burashnikov 1996) accompanying the collision between the
Agradag arc and the Khubsugul terrane. Palaeozoic intrusions are
particularly common in the south of the Khubsugul basin where
they effectively obliterate the host stratigraphy (Badarch et al. 1998).
Lake Khubsugul and the Darkhat depression formed as a
southern arm of the Neogene Baikal rifting episode. Neoprotero-
zoic –Cambrian stratigraphy on the west side of Lake Khubsugul
was uplifted and exposed along a rift shoulder. Rifts commonly
followed Precambrian structures and were accompanied by volu-
minous basaltic volcanism (Logatchev 1984).
Stratigraphy
The Khubsugul basin sequence begins with rift-related volcanic
and clastic rocks of the Darkhat Group that rest unconformably
on Precambrian basement and meta-sediments (Ilyin 1973, 1990,
2004). These volcanic rocks are overlain with hundreds of
metres of clastic and carbonate rocks that are variably preserved
under the sub-Khubsugul Group unconformity (Osokin & Tyzhi-
nov 1998). The Khubsugul Group begins with as much as 100 m
of argillite, unsorted sandstone and minor limestone of the basal
Ongoluk Fm. and grades upwards into the Ongoluk diamictite,
with clasts becoming larger and more abundant upwards. The
Ongoluk diamictite is overlain by 100 –250 m of allodapic dolo-
mite of the upper Ongoluk Fm. The Khesen Fm. begins with a
second diamictite that is c. 50 m thick and carbonate clast-
dominated (Osokin & Tyzhinov 1998). The Khesen diamictite is
capped with a 3– 5 m dolostone, the top of which hosts
centimetre-scale barite fans. Stratiform and granular phosphorite
rests above a major flooding surface, less than 50 m above the
basal Khesen diamictite (Ilyin 2004). Poorly sorted, carbonate
clast conglomerates and syn-sedimentary folding are also
common in the uppermost Khesen formation. The upper c. 2 km
of the Khubsugul Group consists of late Ediacaran and Cambrian
platformal carbonates of the Erkhelnur Fm.
Glaciogenic deposits and associated strata
The Ongoluk diamictite
Near Lake Khubsugul, the Ongoluk diamictite is composed of a
matrix-supported, stratified diamictite that ranges in thickness
from c. 200 m to 415 m, thinning to the south and west. The thick-
est and most complete section of the Ongoluk Fm. is on the ridge
north of Khesen Gol (Fig. 30.4). There, the basal contact of the dia-
mictite is gradational with gravel-sized lonestones becoming more
common upwards, both in an unsorted sandstone matrix and in a
laminated argillite matrix, and, as such, the base is difficult to
define. Sub-angular to sub-rounded gravel clasts of dolomite and
quartzite are most common, with occasional cobbles of granite.
Clasts become larger and more common in the upper c. 100 m of
the diamictite (the ‘perforated shale’ member of Ilyin 1973,
2004), consisting predominantly of dolomite cobbles in an argillite
or siltstone matrix.
At Ongoluk Gol, the base of the Ongoluk diamictite is not
exposed, but c. 130 m of clast poor siltstone and very coarse sand-
stone are present that are very similar to the lower c. 300 m at
Khesen Gol. The ‘perforated shale’ member is c. 50 m thick at
Ongoluk Gol, beginning with a 5-m-thick, massive, clast-
supported dolomite diamictite, and continuing upwards with a
dark, laminated argillite matrix with sub-rounded to angular,
boulder-sized clasts of granite, quartzite, volcanic and meta-
morphic rocks, and gravel to boulder-sized clasts of dolomite.
Quartzite and volcanic clasts are commonly faceted and striated
(Osokin & Tyzhinov 1998).
The Khesen diamictite
The Khesen diamictite tends to be thinner than the Ongoluk dia-
mictite, and is composed of a massive, carbonate clast-dominated
diamictite. The thickness of the Khesen diamictite ranges from 10
to 65 m, with no systematic geographical trend. The thickest
measured section of the Khesen diamictite in the Lake Khubsugul
area is exposed on the ridge north of Ongoluk Gol (Figs 30.3 &
30.5) and consists predominately of a massive, carbonate-clast dia-
mictite. Clasts are composed of angular to sub-rounded dolomite
and limestone (including giant ooid and stromatolite clasts) that
are commonly imbricated, with sizes ranging from pebble to
boulders. The Khesen diamictite has a yellow-weathering dolomite
matrix, with the massive, unbedded deposits broken only by thin,
lenticular, rhythmically bedded marls. At Khesen Gol, quartzite
and volcanic pebbles are also present, and the massive diamictite
is interrupted with multiple fine-laminated beds that are penetrated
by outsized clasts.
Boundary relations with overlying and underlying
non-glacial units
Near Lake Khubsugul the Ongoluk diamictite has a gradational
basal contact with clast-size and abundance decreasing gradually
down-section until completely disappearing from the siltstone
and sandstone (Osokin & Tyzhinov 1998). At Khesen Gol, the
underlying clastic units consist of nearly 100 m of millimetre-
laminated argillite and siltstone, with 0.5 m interbeds of unsorted
very coarse sandstone and gravel conglomerate, and at least two
beds of texture-less limestone. These units rest unconformably
 THE KHUBSUGUL GROUP
on dolostone of the Arasan Group. The Ongoluk diamictite is sep-
arated from overlying dolomite by a sharp contact, although
regionally, deposition appears to be uninterrupted.
Along the Bokson River in Siberia, rocks that are equivalent to
the Ongoluk diamictite rest unconformably on the Darkhat volca-
nics with a 15-m-thick weathering crust at the base (Kuzmichev
2001). A similar, though thinner breccia has been documented in
341
Fig. 30.2. Geology of the western shores of
Lake Khubsugul, showing the positions of
measured sections.
the Darkhat region of Mongolia and along the Sarkhoi River
(Osokin & Tyzhinov 1998).
The yellow-weathering Khesen diamictite rests disconformably
on a sharp contact with the underlying light blue dolomites. The
upper contact, however, is commonly broken with a couple of
metres of dolomite breccia between the massive diamictite and
the thin dolomite overlying the Khesen diamictite. The dolomite
342 F. A. MACDONALD & D. S. JONES

Khesen Gol, above a disconformable flooding surface (Ilyin 2004) that cuts
down into the underlying units (Fig. 30.5).
M602, M613
c. 2km Chemostratigraphy
not shown
Erkhelnur Formation C Chemostratigraphic studies have not been reported for the carbon-
ate units bounding the two diamictites, although work is in pro-
40 0 Ongoluk Gol, M607, gress. Carbon-isotope values through the phosphatic interval of
M608, M609 the Khesen Fm. at Ongoluk Gol range from – 7% to þ5% (Ilyin
& Kiperman 2000; Ilyin 2004); however, as these data have not
P yet been reproduced, their utility for correlation is questionable.
Strontium-isotope values of c. 0.7080 have also been reported
400
P from carbonate interbedded with the Khubsuglul phosphorite
C deposit (Shields et al. 2000).
300
P
P Palaeolatitude and palaeogeography
ce
r fa

Cocks & Torsvik (2007) provide a review of the palaeomagnetic and

su

Palaeozoic fauna affinity studies of Siberia and the peri-Siberian

Khubsugul Group

al

300 terranes. However, there are very few reliable palaeomagnetic

iat
dh

data on the Khubsugul terrane, particularly in the Neoproterozoic,

200 P
re

so the palaeolatitudes presented are highly speculative.

Khesen Fm.

er

From palaeomagnetic studies on peri-Siberian terranes, Krav-

inf

chinsky et al. (2001) concluded that the Tuva-Mongolia belt was

P at low latitude, adjacent to Siberia throughout the Ediacaran and
Cambrian. However, this study lacked a robust confidence test
(i.e. only a reversal test with few samples and low resolution).
200 Along with other peri-Siberian terranes, it has been suggested
100 that the Khubsugul terrane occupied a Precambrian position
between Siberia and Laurentia (Gladkochub et al. 2006), and
Ongoluk Formation

rifted away from Siberia in the late Neoproterozoic (Sengor &

Natal’in 1996; Kuzmichev et al. 2001, 2005). Sengor & Natal’in
(1996) further posit that throughout the late Neoproterozoic and
early Palaeozoic, these terranes were attached to the Central Mon-
100 golian Block, which along with other terranes, stretched to the
present day Sea of Okhotsk. However, this reconstruction is incon-
0 (m ) sistent with the presence of Ordovician accretionary zones on the
NW margins of the Baidrag and Dzabkhan terranes and a Late
Cambrian Dhizda arc on the west margin of the Khubsugul
c. 450 m terrane (Badarch et al. 2002). The Khubsugul and Tavargatay ter-
not shown ranes host Cambrian trilobites endemic to Siberia (Astashkin et al.
C 1995) and Silurian brachiopods characteristic of the peri-Siberian
0 (m) realm (Hou & Boucot 1990). Thus, although there is a paucity of
Darkhat

reliable palaeomagnetic constraints on the Khubsugul terranes,

Arasan

c. 200 m
vv several lines of evidence indicate that they were adjacent to
c. 2 km not shown
vv Siberia in the Neoproterozoic and early Palaeozoic. Pisarevsky
vv not shown et al. (2000) present a strong c. 615 Ma pole on red beds along
vv the Lena River, pinning Siberia at equatorial latitudes in the Neo-
proterozoic. Further palaeomagnetic studies on the Khubsugul
grainstone rhythmite & terrane are necessary, and are in progress (J. Meert pers. comm.).
siltstone C chert
ribbonite diamictite P phosphorite
allodapic flooding barite fans Geochronological constraints
carbonate surface
Although the diamictites of the Khubsugul Group have not been
Fig. 30.3. Stratigraphy of the Khubsugul Group along the Khesen and Ongoluk directly dated, there are at least two radiometric constraints on
Gols. The locations of measured sections are shown in Figure 30.2.
the maximum age of the deposits. In Siberia, the Bokson Group
overlies the Shishkhid arc, which contains magmatic zircons
from rhyolites with a concordant U – Pb SHRIMP age of
is only c. 1 m thick at Ongoluk Gol with centimetre-scale bladed 800 + 2 Ma (Kuzmichev et al. 2005). The Bokson Group also
barite fans at the upper dolomite –limestone transition. Approxi- overlies the Sorkhoi Group, which contains volcanic rocks that
mately 20 km north, the overlying white dolomite reaches a have been dated with whole-rock Rb –Sr at 718 + 30 Ma (Buya-
maximum thickness of 6 m, and is interspersed with bed-parallel kaite et al. 1989). In Mongolia, the volcanic rocks of the Sorkhoi
cements (Fig. 30.5). A transgression continues above the dolomite Group are stratigraphically equivalent to the Darkhat and Dzab-
into as much as 100 m of dark grey limestone rhythmite. The khan volcanics (Macdonald 2011), which have been dated at
phosphorite-bearing strata is composed largely of dolomite with 850 + 2 and 750 + 3 Ma (Pb/Pb zircon, Burashnikov 1990),
common olistostromes and mass flow deposits, and comes in and more recently, at 777 + 6 Ma (U – Pb SHRIMP zircon, Zhao
 THE KHUBSUGUL GROUP 343

Khesen Gol, Ongoluk Gol, Khirbisteg Gol N. Ongoluk Gol Khesen Gol N. Khesen Ridge Bakha Gol
M611 M607, M609 M602 M615, M616 M618
M613, M614 M605, M610 P P
450 P
P C
P P P
200 100 P
P
v P P
m
P P
400 P P
150 P
P
v

50
350 v
v

100

300
50
0 (m)

250 2.5 km 3.3 km 1.0 km 14.0 km

S N
0 (m) Fig. 30.5. Detailed stratigraphy of the Khesen diamictite and overlying
basal Khesen Fm. on the western shores of Lake Khubsugul. Positions of
dolomite facies measured sections are in Figure 30.2. For legend see Figures 30.3 and 30.4.
grainstone
200 allodapic et al. 2006), and 803.4 + 8.0 and 773.5 + 3.6 Ma (laser evapor-
rhythmite ation zircon, Levashova et al. 2010). The uppermost Khesen For-
mation is thought to be latest Ediacaran in age by correlation with
diamictite facies the Zabit Fm. in Siberia, which contains Cloudina and Renalsis
(Kheraskova & Samygin 1992). The Khesen Fm. is overlain by
massive the early Cambrian, Archaeocyathid-bearing Erkhelnur Fm.
stratified matrix (Ilyin & Zhuraveleva 1968), providing a robust minimum age on
150 the diamictites.
sandstone
siltstone
Discussion
argillite
clast size Although the Dzabkhan platform of southern Mongolia may have
100 been geographically separated from the Khubsugul basin, many
boulder units in the Tsagaan Oloom Fm. (Macdonald 2011) can be corre-
cobble lated with the Khubsugul Group. Both successions are underlain
pebble by riftogenic volcanic rocks and begin with interbedded clastic
rocks and diamictites. The Maikhan Ul diamictite is an early Cryo-
clast composition genian glacial deposit (Brasier et al. 1996), and like the basal
granite Ongoluk diamictite in the Khubsugul Group, it is commonly
50 clastic over 100 m thick and dominated by siltstone and coarse sand
(Lindsay et al. 1996). The upper Ongoluk Fm. can be correlated
carbonate with the Tayshir member of the Tsagaan Oloom Fm., and the
Ongoluk Formation carbonate-rich basal Khesen diamictite can be correlated with
the Khongoryn diamictite (Macdonald et al. 2009). Chemostrati-
Arasan Formation graphy indicates that the phosphorites on both the Khubsugul
0 (m) and Dzabkhan terranes were deposited in the latest Ediacaran to
early Cambrian above a major Ediacaran hiatus (Shields et al.
Fig. 30.4. Detailed stratigraphy of the Ongoluk diamictite along the Khesen 2000; Macdonald et al. 2009).
and Ongoluk Gols. The locations of measured sections are shown in Figure 30.2. Unconformities in the Ongoluk Fm. developed near basement
m, metamorphic clasts; v, volcanic clasts. highs on rift shoulders that were active at least until the onset of
344 F. A. MACDONALD & D. S. JONES
deposition of the Ongoluk diamictite (Osokin & Tyzhinov 1998).
While the Khubsugul basin was undergoing extension a large
thickness of diamictite accumulated near the present Lake
Khubsugul. Evidence for a glaciogenic origin of the Ongoluk
diamictite includes exotic clasts with a mixed lithology, bed-
truncating lonestones, and faceted and striated clasts (Osokin &
Tyzhinov 1998). The contact with the overlying light-blue alloda-
pic dolostone is sharp, and as they appear regionally conformable,
a rift –drift transition is inferred within the Ongoluk diamictite.
Although striated clasts have not been observed in the younger
carbonate-rich Khesen diamictite, and exotic clasts are rare, a
glacial origin of the massive deposit is inferred from the presence
of bed-penetrating lonestones exposed along Khesen Gol, the geo-
chemistry signature in associated carbonates (unpublished data)
and the presence of overlying barite fans, which are present
above basal Ediacaran cap dolostones in Australia (Kennedy
1996), Mauritania (Deynoux & Trompette 1976), NW Canada
(Hoffman & Schrag 2002) and south China (Jiang et al. 2003).
A basal Ediacaran age of the carbonate immediately overlying
the Khesen diamictite is inferred from stratigraphic correlation
with the Dzabkhan platform (Macdonald et al. 2009), and from
the occurrence of the barite fans.
Kheraskova & Samygin (1992) rejected a glaciomarine origin
of diamictites in the Zabit Formation exposed on the Siberian
side of the border, arguing that these deposits represent rift-related
submarine slumps and debris flows. They further suggested that the
diamictites in the Zabit Fm. (and correlative Khesen Fm.) are latest
Ediacaran to early Cambrian in age, citing the presence of
Cloudina and Renalsis in the Zabit Fm. On the Mongolian side
of the border, there is no evidence of a glacial origin for the diamic-
tite at the base of the Khesen Fm., or the carbonate and carbonate
clasts conglomerate associated with the phosphorite in the upper-
most Khesen Fm., which were interpreted as olistostromes and
slumps. It is possible that, like the Dzabkhan platform to the south
(Macdonald et al. 2009), there is a major hiatus in the Ediacaran and
that Kheraskova & Samygin (1992) are miscorrelating mass flows
in the latest Ediacaran to early Cambrian upper Khesen Fm. with
the end Cryogenian Khesen diamictite described here from the
base of the Khesen Fm. Dobretsov (1985) considered the Siberian
diamictites to be ‘nappe thrust olistostromes’ related to the Salarian
orogeny, which has been stratigraphically constrained to the early
Cambrian on the Dzabkhan terrane (Ruzhentsev & Burashnikov
1996). Again, it seems likely that there is a conflation between
the Khesen diamictite in the lower Khesen Fm., which is interpreted
here as glaciogenic, and the conglomerates interpreted as olistos-
tromes and allodapic deposits in the upper Khesen Fm.
Sheldon (1984), Osokin & Tyzhinov (1998) and Ilyin (2004)
have suggested a genetic relationship between the diamictite and
phosphorite. However, Ilyin (2004) documented a disconformity
at the base of the phosphorite (Fig. 30.5). Chemo- and lithostrati-
graphic studies in the correlative Dzabkhan platform (Macdonald
et al. 2009; Macdonald 2011) suggest a major hiatus and flooding
surface between the Khesen diamictite and the phosphorite series,
casting doubt on any genetic relationship.
Further chemo- and lithostratigraphic studies are needed to
better constrain the basin dynamics and depositional setting of
Neoproterozoic strata in Mongolia. The palaeogeography of the
peri-Siberian terranes also remains speculative. It is clear,
however, that island arcs surrounded the Khubsugul terrane for
much of the Neoproterozoic and early Cambrian, and thus, there
is excellent potential for geochronology in the Khubsugul basin.
We thank our field assistants U. Bold, J. Otgonhuu and Eerie. We also thank
A. Bayasgalan and the Mongolian University of Science and Technology for
making our fieldwork possible, D. Schrag for use of the Harvard University
Laboratory for Geochemical Oceanography, and G. Eischeid for help in the lab-
oratory. We thank the National Science Foundation for funding, and P. Hoffman
for inspiration and support. This represents a contribution of the IUGS- and
UNESCO-funded IGCP (International Geoscience Programme) project #512.
References
Astashkin, V. A., Pegel, T. V. et al. 1995. The Cambrian System of the
Foldbelts of Russia and Mongolia. International Union of Geological
Sciences, London.
Badarch, G., Byamba, J. et al. 1998. Geological Map of Mongolia.
Mineral Resources Authority of Mongolia, Ulaan Baator.
Badarch, G., Cunningham, W. D. & Windley, B. 2002. A new terrane
subdivision for Mongolia: implications for the Phanerozoic crustal
growth of Central Asia. Journal of Asian Earth Sciences, 21, 87 – 110.
Brasier, M. D., Shields, G., Kuleshov, V. N. & Zhegallo, E. A. 1996.
Integrated chemo- and biostratigraphic calibration of early animal
evolution: Neoproterozoic –early Cambrian of southwest Mongolia.
Geological Magazine, 133, 445– 485.
Buchan, C., Pfander, J. et al. 2002. Timing of accretion and collisional
deformation in the Central Asian Orogenic Belt: implications of
granite geochronology in the Bayankhongor Ophiolite Zone. Chemi-
cal Geology, 192, 23 –45.
Burashnikov, V. V. 1990. Tectonics of the Urgamal Zone, Early Calido-
nides of Western Mongolia. Russian Academy of Sciences, Moscow.
Buyakaite, M. I., Kuzmichev, A. B. & Sokolov, D. D. 1989. 718 Ma
Rb –Sr errorchron of the Sorkhoi Group in the East Sayan. Doklady
Akademii Nauk SSSR, 309, 150–154.
Chumakov, N. M. 2011. Glacial deposits of Bokson Group, East Sayan
Mountains, Buryatian Republic, Russian Federation. In: Arnaud,
E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological
Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 285–288.
Cocks, L. & Torsvik, T. H. 2007. Siberia, the wandering northern terrane,
and its changing geography through the Palaeozoic. Earth Science
Reviews, 82, 29 –74.
Deynoux, M. & Trompette, R. 1976. Late Precambrian mixtite: glacial
and/or non-glacial? Dealing especially with the mixtite of West
Africa. American Journal of Science, 276, 117–125.
Dobretsov, N. L. 1985. Overthrust tectonics of the East Sayans. Geotec-
tonics, 19, 26– 34.
Donov, N. A., Edemsky, H. B. & Ilyin, A. V. 1967. Cambrian phos-
phorites of Mongolia Popular Republic. Sovetskaya Geologia, 3,
55 – 60.
Gladkochub, D. P., Wingate, M. T. D., Pisarevsky, S. A., Donskaya,
T. V., Mazukabzov, A. M., Ponomarchuk, V. A. & Stanevich, A.
M. 2006. Mafic intrusions in southwestern Siberia and implications
for a Neoproterozoic connection with Laurentia. Precambrian
Research, 147, 260– 278.
Hoffman, P. F. & Schrag, D. P. 2002. The snowball Earth hypothesis;
testing the limits of global change. Terra Nova, 14, 129–155.
Hou, H. F. & Boucot, A. J. 1990. The Balkhash-Mongolia-Okhotsk
Region of the Old World Realm. Geological Society, London,
Memoir, 12, 297– 303.
Ilyin, A. V. 1973. Khubsugul Phosphorite-Bearing Basin. Geologiches-
kiy Institut, Akademiya Nauk SSSR, Moscow.
Ilyin, A. V. 1990. Proterozoic supercontinent, its latest Precambrian
rifting, breakup, dispersal into smaller continents, and subsidence
of their margins: Evidence from Asia. Geology, 18, 1231– 1234.
Ilyin, A. V. 1998. Rare-earth geochemistry of ‘old’ phosphorites and
probability of syngenetic precipitation and accumulation of phos-
phate. Chemical Geology, 144, 243–256.
Ilyin, A. V. 2004. The Khubsugul phosphate-bearing basin: new data and
concepts. Lithology and Mineral Resources, 39, 454–467.
Ilyin, A. V. & Byamba, J. 1980. Handbook for the Excursion ‘Phosphor-
ites of the Khubsugul Basin in the Mongolian Peoples’s Republic’.
Geologicheskiy Institut, Akademiya Nauk SSSR, Moscow.
Ilyin, A. V. & Kiperman, Y. A. 2000. Mass accumulation of biogenic
rocks at the Vendian/Cambrian boundary and carbon isotopic
anomalies. Soveremenny voprosy geologii (Modern Problems of
Geology). Nauchnyi Mir, Moscow.
Ilyin, A. V. & Zhuraveleva, I. T. 1968. On the boundary between the
Cambrian and the Precambrian at Prikhusugulie (Mongolian PR).
Doklady Akademii Nauk SSSR, 182, 1164–1166.
Jiang, G., Kennedy, M. J. & Christie-Blick, N. 2003. Stable isotopic
evidence for methane seeps in Neoproterozoic postglacial cap car-
bonates. Nature, 426, 822– 826.
 THE KHUBSUGUL GROUP
Kennedy, M. J. 1996. Stratigraphy, sedimentology, and isotope geochem-
istry of Australian Neoproterozoic postglacial cap dolostones:
deglaciation, d13C excursions, and carbonate precipitation. Journal
of Sedimentary Research, 66, 1050– 1064.
Kherzaskova, T. N. & Samygin, S. G. 1992. Tectonic conditions in the
East Sayan Vendian –Middle Cambrian terrigenous carbonate associ-
ation. Geotectonics, 26, 445– 458.
Kravchinsky, V. A., Konstantinov, K. M. & Cogne, J.-P. 2001.
Palaeomagnetic study of Vendian and Early Cambrian of South
Siberia and Central Mongolia: was the Siberian platform assembled
at this time? Precambrian Research, 110, 61 – 92.
Kuzmichev, A. 2001. Early Baikalian tectonic events in the Tuva-
Mongolia Massif: arc-microcontinent collision. Geotectonics, 35,
185– 198.
Kuzmichev, A., Bibikova, E. V. & Zhuravlev, D. Z. 2001. Neoproter-
ozoic (800 Ma) orogeny in the Tuva-Mongolia Massif (Siberia):
island arc-continent collision at the northeast Rodinia margin.
Precambrian Research, 110, 109– 126.
Kuzmichev, A., Kroener, A., Hegner, E., Dunyi, L. & &Yusheng, W.
2005. The Shishkhid ophiolite, nothern Mongolia: a key to the
reconstruction of a Neoproterozoic island-arc system in central
Asia. Precambrian Research, 138, 125–150.
Levashova, N. M., Kalugin, V. M., Gibsher, A. S., Yff, J., Ryabinin,
A. B., Meert, J. & Malone, S. J. 2010. The origin of the Baydaric
microcontinent, Mongolia: constraints from palaeomagnetism and
geochronology. Tectonophysics, 485, 306–320.
Lindsay, J. F., Brasier, M., Shields, G., Khomentovsky, V. V. &
Bat-Ireedui, Y. A. 1996. Glacial facies associations in a Neoproter-
ozoic back-arc setting, Zavkhan Basin, western Mongolia. Geologi-
cal Magazine, 133, 391–402.
Logatchev, N. A. 1984. The Baikal rift system. Episodes, 7, 38 –42.
Macdonald, F. A. 2011. The Tsagaan Oloom Formation, southwestern
Mongolia. In: Arnaud, E., Halverson, G. P. & Shields-Zhou, G.
345
(eds) The Geological Record of Neoproterozoic Glaciations.
Geological Society, London, Memoirs, 36, 331–337.
Macdonald, F. A., Jones, D. S. & Schrag, D. P. 2009. Stratigraphic and
tectonic implications of a new glacial diamictite-cap carbonate
couplet in southwestern Mongolia. Geology, 37, 123–126.
Osokin, P. V. & Tyzhinov, A. V. 1998. Precambrian Tilloids of the Oka-
Khubsugul phosphorite-bearing basin (Eastern Sayan, Northwestern
Mongolia). Lithology and Mineral Resources, 33, 142– 154.
Pisarevsky, S. A., Komissarova, R. A. & Khramov, A. N. 2000. New
palaeomagnetic result from Vendian red sediments in Cisbaikalia
and the problem of the relationsip of Siberia and Laurentia in the
Vendian. Geophysics Journal International, 140, 598–610.
Ruzhentsev, S. V. & Burashnikov, V. V. 1996. Tectonics of the
western Mongolian Salairides. Geotectonics, 29, 379–394.
Sengor, A. C. & Natal’in, B. A. 1996. Palaeotectonics of Asia: frag-
ments of synthesis. In: Yin, A. & Harrison, M. (eds) The Tectonic
Evolution of Asia. Cambridge University Press, Cambridge,
486– 640.
Sheldon, R. P. 1984. Ice-ring origin of the Earth’s atmosphere and hydro-
sphere and late Proterozoic – Cambrian phosphogenesis, Phosphorite,
Geological Survey of India Special Publication, 17, Udaipur,
Rajasthan, India.
Shields, G., Stille, P. & Brasier, M. 2000. Isotopic records across two
phosphorite giant episodes compared: the Precambrian –Cambrian
and the Late Cretaceous – recent. SEPM Special Publications, 66,
102–115.
Windley, B. F., Alexeiev, D., Xiao, W., Kroener, A. & Badarch, G.
2007. Tectonic models for accretion of the Central Asian Orogenic
Belt. Journal of the Geological Society of London, 164, 31 –47.
Zhao, Y., Song, B. & Zhang, S. H. 2006. The Central Mongolian micro-
continent: its Yangtze affinity and tectonic implications. In: Jahn,
B. M. & Chung, L. (eds) Symposium on Continental Growth and
Orogeny in Asia. Taipei, Taiwan, 135– 136.
 Chapter 31

The Blaini Formation of the Lesser Himalaya, NW India

JAMES L. ETIENNE1,2*, PHILIP A. ALLEN3, ERWAN LE GUERROUÉ4, LARRY HEAMAN5,

SUMIT K. GHOSH6 & RAFIQUE ISLAM6
1
Present address: Neftex Petroleum Consultants Ltd, 97 Milton Park, Abingdon, Oxfordshire OX14 4RY, UK
2
Geologisches Institut, Departement Erdwissenschaften, ETH-Zentrum, Haldenbachstrasse 44, CH-8092,
Zürich, Switzerland
3
Department of Earth Sciences and Engineering, Imperial College London, South Kensington Campus, London SW7 2AZ, UK
4
Géosciences Rennes (UMR 6118 – CNRS), 263 Avenue du General Leclerc. CS 74205, Université de Rennes 1,
35042 Rennes Cedex, France
5
Department of Earth & Atmospheric Sciences, 1 – 26 Earth Sciences Building, University of Alberta, Edmonton, Alberta,
T6G 2E3, Canada
6
Wadia Institute of Himalayan Geology, 33 General Mahadeo Singh Road, Dehra Dun, Uttarakhand, India
*Corresponding author (e-mail: james.etienne@neftex.com)

Abstract: Neoproterozoic glaciogenic deposits crop out widely across the Lesser Himalaya fold and thrust belt in NW India. Underlain
by the siliciclastic Simla and Jaunsar Groups, the Blaini Formation (Fm.) includes at least two thick and regionally extensive diamictite
units, separated by siliciclastics and argillites and capped by a pink microcrystalline dolomite. A glaciogenic origin is supported by the
presence of relatively abundant striated clasts and the local preservation of polished and striated pavement on underlying Simla Group
clastics. The cap dolostone is isotopically light with respect to both 13C and 18O, which show strong covariance. The Blaini is ubiqui-
tously deformed, incorporated in regional-scale folds and thrusts, and exhibits locally intensive intra-formational deformation. Until
recently, geochronological constraints have remained poor, but new detrital zircon ages from diamictite samples provide a maximum
age limit of 692 + 18 Ma (207Pb/206Pb). Reliable palaeomagnetic data are required to constrain the position of this important passive
continental margin in palaeogeographical reconstructions.

The Lesser Himalaya fold and thrust belt in NW India comprises
up to 10 km of variably deformed low-grade Neoproterozoic to
Lower Palaeozoic metasediments (Schelling 1992). Originally
described as the ‘Blaini Conglomerate’ from the type section
at Baliana Nala in Himachal Pradesh (Medlicott 1864), the
Blaini Fm. is an important stratigraphic marker in the Lesser
Himalaya and crops out across .300 km of the fold belt
(Fig. 31.1). Some of the best exposures occur in the Mussoorie
syncline, between Māldeota and Dhanaulti, in the Giri River
valley near Dadahu and around Simla. A panoply of designa-
tions occur in the literature, including Blaini Conglomerate,
Blaini Group, Blaini Boulder Bed, Blaini Series, Blaini diamic-
tites, Blaini North, Blaini South and Blaini Fm., but the latter
is preferred here given the unit’s extensive mappable nature
(Pilgrim & West 1928; Fuchs & Sinha 1978) and use in
recent publications (e.g. Jiang et al. 2002, 2003a; Kaufman
et al. 2006).
A glaciogenic origin for diamictites of the Blaini Fm. has long
been recognized (e.g. Oldham 1887; Holland 1908; Pilgrim &
West 1928; Auden 1946; Saxena & Pande 1969; Gaur & Dave
1971; Bhargava & Bhattacharyya 1975; Bhatia & Prasad 1975;
Jain & Varadaraj 1978; Bhatia & Prasad 1981; Jain 1981);
however, in the absence of good biostratigraphic data, an
assumed correlation with the Permo-Carboniferous Talchir tillites
of peninsular India has hampered recognition of the true age and
significance of these deposits, despite a Proterozoic age frequently
being posited (e.g. see Bhatia & Kanwar 1975 and references
therein; Brookfield 1994). Recently published biostratigraphic
data and detailed work on the overlying Krol and Tal Groups
have firmly established a Proterozoic age for the Blaini Formation
(Fig. 31.2). General overviews, including textural descriptions of
diamictite lithofacies, are included in Bhargava & Bhattacharyya
(1975), Bhatia & Prasad (1975), Jain & Varadaraj (1978), Bhatia
& Prasad (1981) and Jain (1981).
Structural framework
The Blaini Fm. crops out around the margins of a series of tectoni-
cally complex synclines and in thrust block slices between the
Main Boundary Fault and Main Central Thrust. Resting upon a
succession of continental deltaic to shallow marine clastics
(Nagthat Fm.) and more distal facies of the Simla Group (Ghosh
1991), the Blaini Fm. acts as a key stratigraphic marker horizon
in this part of the Himalayas.
The Blaini Fm. is overlain by the Infra Krol Fm. and the Krol
and Tal groups, a succession that is thought to represent the
inner part of a north-facing passive continental margin (Brookfield
1993). Based upon palaeogeographical reconstructions (e.g.
Torsvik 2003) and isotopic and sequence stratigraphic similarities,
this succession may have opposed the Chinese south-facing
passive margin represented by the sedimentary cover of the
South China block (Jiang et al. 2003a; Kaufman et al. 2006;
McFadden et al. 2008). Palaeocurrent indicators and facies
relationships in the underlying Jaunsar Group (Ghosh 1991)
suggest a sediment source to the south/SE, and a basin margin
oriented broadly NE –SW. Sequence stratigraphic analysis of the
Krol Group illustrates a similar trend for the Ediacaran carbonate
platform (Jiang et al. 2002; Kaufman et al. 2006).
The Blaini Fm. is ubiquitously deformed, and faulted and folded
on a regional scale, exhibiting penetrative cleavage and evidence
for pressure solution. For these reasons, gravel clast macrofabrics
may locally reflect the tectonic grain rather than primary deposi-
tional fabrics. In contrast, clast surface features such as glacial

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 347– 355. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.31
348 J. L. ETIENNE ET AL.

SOLAN 4 5 7 8
Siwalik Zone
3
9
1 Subathu-Dagshai
2 CHOR
6 Tal Group
10

11 Carbonate formations (Krol, Shali, Deoban etc)

12 13 15

14 Black slate formations (Infra Krol, Blaini, Shali, Sor)

MUSSOORIE
17 18 Fossiliferous Upper Palaeozoics of the
16
19 Landsdowne syncline

DEHRA DUN Blaini/Mandhali

20
21
Jaunsar Group
22 (Nagthat/Chandpur undivided)
23
24
Chail
RISHIKESH
KAUDIYALA
Basic metavolcanic rocks

Chail/Nagthat undivided
500 km
SRINAGAR
M.
B.T Simla slates
. STUDY
SHILLONG
AREA
DELHI
30 N Crystalline
basement

INDIA
CALCUTTA 20 N
MUMBAI

MADRAS NAINI TAL

0 10 20 30 40
10 N km 25
70 E 80 E 90 E

Fig. 31.1. Geological map of the Lesser Himalaya fold-and-thrust belt in NW India. Redrawn from Fuchs & Sinha (1978). Numbers indicate location of measured
sections (Table 31.1; Fig. 31.3).

striae and polish have been preserved and are distinct from tectonic
striae such as slickenlines. Brookfield (1987) noted the occurrence
of shear structures within diamictite lithofacies, which is evident
from pressure shadows and augen adjacent to rotated clasts.
Although a tectonic origin for these features needs to be con-
sidered, it is possible that the shearing reflects soft-sediment defor-
mation generated during sediment transport either in the subglacial
environment, where large shear stresses can be generated beneath
overriding ice (Boulton 1996), or during downslope resedimenta-
tion of debris, which results in similar microfabrics.
Stratigraphy
The Blaini Formation is generally stratigraphically simple, broadly
characterized by a tripartite diamictite –shale –diamictite succes-
sion that is variable in thickness across 300 km of the fold belt
between Simla and Nainital. Additional diamictite intervals
locally occur in the NW of the region and have also been reported
from the area around Nainital (Jiang et al. 2003a). Diamictite units
are generally less than 50 m thick (Fig. 31.3) and contain relatively
abundant glacioclastic debris bearing cross-cutting glacial striae,
facets and polish. The Blaini is capped by a pink microcrystalline
dolomite and is overlain by the Infra-Krol Fm. In the upper part,
the cap contains thin grey and red shale partings.
Glaciogenic deposits and associated strata
Massive and laminated diamictite lithofacies
Diamictites occur as massive or weakly stratified clast-poor to
clast-rich units with poorly sorted silty to sandy matrices. Thick-
nesses vary across the fold belt, but are typically on the order of
10–40 m (Fig. 31.3). Although exposure is poor, sheet-like geo-
metries best explain the regional distribution of diamictite
beneath the cap dolostone. Lateral continuity between stratigraphi-
cally lower occurrences of diamictite is more difficult to demon-
strate, given the few available sections where the base of the
formation may be observed. In the eastern part of the Mussoorie
syncline, the coarse sediment fraction of the basal diamictite
fines northwards, and on the northwestern limb, the upper
surface of the uppermost diamictite is characterized by a win-
nowed transgressive lag immediately beneath the cap dolostone.
 THE BLAINI FORMATION 349

Fig. 31.2. Lithostratigraphic and biostratigraphic subdivision of the Neoproterozoic– Cambrian succession of the Lesser Himalaya, summarized from Jiang et al. (2002),
Hughes et al. (2005) and this study. Carbon–isotope data from Kaufman et al. (2006) and this study. For details on biostratigraphy, refer to Prasad et al. (1990), Tiwari
(1999), Mazumdar & Banerjee (1998), Brasier & Singh (1987), Bhatt & Mathur (1990), Kumar et al. (1983), Mathur & Srivastava (1994), Kumar et al. (1987), Joshi et al.
(1989), Jell & Hughes (1997), Mathur & Joshi (1989a, b), Tripathi et al. (1984, 1986), Singh & Rai (1983), Banerjee & Narain (1976), Bhargava et al. (1998), Mathur
et al. (1988), Rai (1987), De et al. (1994), Bhargava (1984), Joshi & Mathur (1987), Tiwari & Knoll (1994), Shanker et al. (1997), Mathur & Shanker (1989, 1990), Bhatt
(1991) and Hughes et al. (2005).

In the Solan region, at Rahed (section 3, Table 31.1, Fig. 31.3), the
lower diamictite is erosionally based, and overlies a striated and
polished surface of Simla Group argillites.
Within this lithofacies, clast types include subrounded to suban-
gular quartzite, microscale laminated siltstone, chert, vein quartz,
gneiss, limestone, sandstone, dolomite, basalt, shale and slate.
Striated and polished clasts are relatively abundant at most localities.
Statistical clast shape and roundness data have previously been pre-
sented in Bhargava & Bhattacharyya (1975) and Bhatia & Prasad
(1975). Some petrographic and heavy mineral data may also be
found in Jain & Varadaraj (1978) and Bhatia & Prasad (1981).
Interpretation
Numerous interpretations have been posited for the genesis of
the Blaini diamictites, emphasized by discussions in Bhatia &
Kanwar (1975, and references therein) which revolve primarily
350 J. L. ETIENNE ET AL.

Table 31.1. Locations of measured sections (Figs 31.1 and 31.3) these data support an extra-basinal provenance for some of the
material in the diamictites.
Section no. Locality Northing Easting Where diamictites directly overlie striated pavement, a primary
(Figs 1, 3) tillite interpretation is considered most likely; however, across
much of the fold belt, this association is not observed. If the intra-
1 Baliana Nala 30859.6280 77802.4660 formational shear structures observed within the diamictites are
2 Deothal from 30850.4900 7789.8540 syn-sedimentary in origin, they may reflect either subglacial pro-
to 30850.5450 7789.9200 cesses of sediment transport or sediment redistribution by debris
3 Rahed 30854.8410 77812.3260 flows, and are therefore not diagnostic palaeoenvironmental indi-
4 Giri River Valley from 308540 01.5000 778130 52.7500 cators. There is little evidence for significant glaciotectonic struc-
to 308550 00.5900 778140 49.8600 turation, stratigraphic complexity, lithofacies heterogeneity or
5 Sargaon (i) 30856.1090 77816.2460
ice-proximal processes typical of terrestrial proglacial environ-
6 Deoria from 308500 24.9300 778110 03.9700
ments or glaciomarine grounding-line fan assemblages (cf. Benn
to 308500 23.9200 778110 11.6200
& Evans 1998; Powell & Cooper 2002) and a more distal setting
7 Sargaon (ii) 30856.5550 77816.7940
8 Rajgarh-Giripool Road 30853.2230 77814.2290
from the ice margin may be more likely. Coarse-tail grading
Section locally exhibited in the basal diamictite in the Mussoorie syncline
9 Pervi River Valley 30853.6530 77814.3970 may indicate downslope reworking of glaciogenic debris distal
10 Mareog 308520 13.0800 778120 53.0000 from the ice margin, but the locally reworked tops of younger dia-
11 Dadahu-Sangar Road 30839.3110 77826.1350 mictites beneath the cap carbonate probably indicates deposition
Section within neritic water depths. The diamictite lithofacies thus encom-
12 Sangar 30841.9150 77825.2630 passes a range of different glacially influenced facies types, includ-
13 Dadahu 30837.4760 77827.8000 ing primary tillites and glaciogenic debrites, which form part of a
14 Dubrah-Mahipur 30820.6720 78808.2960 continuum of deposits interpreted by Brookfield (1987) as basal
15 Narendranagar 30809.2070 78817.9580 tills, flow tills and subaqueous meltout tills.
16 Shivpuri Bridge 308080 11.2700 788230 16.3000
17 Bhanswari 30828.6790 78810.8030
18 Dhanaulti 30825.8160 78814.2280 Sandstones, siltstones and shales
19 Dhanaulti-Raipur Road 30823.9590 78817.9550
Section Between the lower and upper diamictite units, a series of associated
20 Renuka-Sataun Road 30835.3750 77829.2690 lithofacies occur that include massive amalgamated sandstone
Section beds, flaggy siltstones and laminated silty shales. Sandstones
21 Maldeota 30820.4420 78808.1840 overlie the diamictites at the base of the Blaini Fm. and either
22 Shilla (Sataun-Shilai 30837.3040 77842.0350 pass directly upwards into shales or through stacked flaggy silt-
Road) stones into shales. Dark grey flaggy siltstones with sharp bases
23 Bhatoli (Kempte Falls– 30830.5980 78800.0360 and tops are best developed in the Dhanaulti-Raipur (section 19,
Yamuna Bridge) Table 31.1, Fig. 31.3) and Sataun-Renuka road sections. These
24 Gular Ghatti 30807.4010 78825.4380 thinly bedded (centimetre-scale) units bear no depositional
25 Nainital– Bhowali 29823.0890 79829.6610 structures, but in the Dhanaulti sections show evidence for
Section near Kaliakhan
thickening-upward, thinning-upward cycles throughout the section.
The shales achieve thicknesses in the order of at least several
tens of metres, but given an absence of key marker beds (sand-
stones rarely re-appear in section until just beneath the upper dia-
around glacial v. non-glacial processes such as turbidity currents mictite unit), and the local extent of deformation, stratigraphic
or mass failure events (see also discussions in Rupke 1968; thicknesses may locally exceed this. In the Solan region shales
Valdiya 1973). However, an enhanced understanding of modern are of limited stratigraphic thickness and sandstone beds have
glaciomarine depositional systems now recognizes the impor- thin (10 mm thick) pebbly lag bases.
tant role played by debris flows in resedimenting glaciogenic
debris in these environments and the fact that the two are not Interpretation
mutually exclusive (cf. Dowdeswell et al. 1996, 1998; Taylor
et al. 2002). The lower portion of the Blaini Fm. is characterized by a fining-
Diamictites of the Blaini Fm. are interpreted as glaciogenic in upward motif recorded by a transition from diamictite through
origin on the basis that (i) the lithofacies contains abundant glacio- sandstones and siltstones into shales. This retrogradational stacking
clastic debris (striated, faceted and polished clasts) and (ii) the pattern indicates a transition to more distal conditions of sedimen-
basal diamictite overlies striated and polished pavement (near tation, with massive sandstones and siltstones probably deposited
Rahed, section 3, Table 31.1, Fig. 31.3). below the storm-weather wave base. More proximal conditions of
Several publications have suggested that the bulk of macroclasts sedimentation are indicated up-section as the lithofacies stacking
within the Blaini diamictites are attributable to local sources pattern turns around and shales pass up into sandstones and diamic-
derived from the underlying passive margin sequence (e.g. tites. Coarser clastics and only thinly developed shales indicate
Auden 1934; Rupke 1968; Niyogi & Bhattacharya 1971; Valdiya more proximal conditions in the NW around Solan compared
1973; Jain & Varadaraj 1978), including Brookfield (1987), who with thicker developed shale successions in the SE (e.g. in the
noted a direct compositional influence on the basal diamictite Mussoorie syncline). Additional support for more proximal con-
depending on which units of the Simla Group it overlies. Petro- ditions in the NW is provided by the preservation of striated pave-
graphic and heavy mineral analyses presented in Jain and Vara- ment at Rahed indicative of grounded ice conditions.
daraj (1978) are consistent with these interpretations; however,
the observation of orthoclase, microcline, perthite, myrmekite,
zircon, tourmaline and spinel by those authors are indicative of Dolomite
provenance from acid igneous rocks, although the degree of
reworking needs consideration. Because crystalline basement Dolomite locally cements siltstones and parts of the uppermost dia-
does not crop out in this part of the fold belt, it is possible that mictite unit in the Giri Valley to the east of Solan, and has also been
 THE BLAINI FORMATION 351

Fig. 31.3. Lithostratigraphic sections and d13C chemostratigraphy of the Blaini Formation in the Lesser Himalaya of NW India. Numbered sections are located in
Figure 31.1 and Table 31.1.
352 J. L. ETIENNE ET AL.
observed as discrete interbeds within equivalent units in the
Mussoorie syncline (Kaufman et al. 2006), but is most extensively
developed as a regional thinly bedded, laminated pink microcrys-
talline dolomite (Fig. 31.3). Across the fold belt, the dolostone
rarely exceeds 9 m in thickness. Parallel or ‘crinkly’ lamination
dominate, with the exception of a small patch of brecciated carbon-
ate observed at Dhanaulti in the Mussoorie syncline. Like other cap
carbonates, the Blaini dolostones contain some enigmatic ‘tepee-
like’ structures. In the Nigalidhar and Mussoorie synclines, these
are similar in form to ripple bedforms, with a wavelength of
c. 1 m and amplitudes of the order of 0.2 m. Where observed,
these structures occur 4–4.5 m above the base of the cap and are
underlain and overlain by planar laminated dolomite. Laminae
are locally pinched out at ripple crests, but have not been observed
to onlap against the ripple trunks. Cracks run through and penetrate
the ripple crests in the axial plane. Some millimetre-scale displace-
ment occurs along these fractures, but no significant brecciation of
the carbonates occurs in the axial zones.
Interpretation
Most exposures of the Blaini cap carbonate are dominated by
planar or crinkly laminated dolomite. The latter may indicate an
algal origin, but no distinct structured mats, stromatolitic or
other biohermal forms have been observed that could provide
more definitive evidence of shallow-water sedimentation. With
the exception of very locally developed breccia observed at
Dhanaulti, a general lack of intraclasts or sedimentary breccias
suggests that deposition occurred largely below the storm-weather
wave base. The enigmatic tepee structures, which resemble ripple
bedforms, are identical to examples observed elsewhere from cap
carbonates (e.g. Allen & Hoffman 2005; Alvarenga et al. 2007);
however, their interpretation is far from straightforward. No
clear examples of laminae onlapping the ripple forms have been
observed that require a sedimentary origin, and a tectonic origin
has to be considered. Stromatactis-like structures associated with
similar features in the Doushantuo Fm. in South China do not
appear to occur (which might imply linkage to methane gas or
fluid escape; cf. Jiang et al. 2006). Tepees described elsewhere
in the literature have been variably interpreted as resulting from
different syndepositional to early diagenetic processes of sedimen-
tation, cementation and deformation, a discussion of which may
be found in Jiang et al. (2006).
Boundary relations with overlying and underlying
non-glacial units
Across the fold belt, the Blaini Fm. is significantly deformed and
heavily vegetated at altitudes up to c. 2.5 km above sea level.
Rheological contrasts between the different lithofacies of the for-
mation and its bounding strata have often led to deformation along
contact surfaces, which presents some challenges in characterizing
boundaries. Nevertheless, there are a number of key exposures
across the fold belt where stratigraphic relationships may be
properly established.
For example, sections at Māldeota show the diamictites to dis-
conformably overlie Nagthat Fm. quartzites by a slickenlined
faulted contact. Further north, along the road between Māldeota
and Dhanaulti, the contact between the basal diamictite and under-
lying quartzites remains sharp, but with no evidence for tectonic
dislocation. Striated pavement has previously been reported at
this contact, but present exposures do not appear to retain glacial
striae. However, in the northwestern part of the fold belt, where
the diamictites overlie softer mudrocks of the Simla Group,
diamictites directly overlie polished striated pavement (e.g. near
Rahed). Multiple sets of cross-cutting glacial striae provide
evidence for subglacial abrasion and the base of the formation in
this instance is clearly erosional.
Recent reports of biostratigraphically significant protoconodant
assemblages from the Gangolihat Dolomite have led Azmi & Paul
(2004) to adopt a modified version of Valdiya’s (1995) lithostrati-
graphic scheme for the region, suggesting a correlation between
the Deoban and Krol Group carbonates, and invoking a major
regional unconformity at the base of the Blaini. This is controver-
sial given the widely held view that the Deoban carbonates lie
stratigraphically beneath the Jaunsar Group, so further data are
required to test these competing hypotheses.
Chemostratigraphy
Abundant d13C and d18O stable isotope data are available for the
Blaini cap dolostone (Kaufman et al. 2006; this study) and some
strontium and magnesium values have been published (Kaufman
et al. 2006). Kaufman et al (2006) identified a negative C-isotopic
signature with up-section trends towards more negative values,
which recover to near 0‰ in the Garhwal syncline.
We collected c. 350 samples from 10 localities across the fold belt
to better characterize the nature of the isotopic character and facili-
tate correlation between sections (Fig. 31.3). As with most cap
carbonates, the Blaini dolostones are depleted in both carbon and
oxygen, which show strong co-variation. In this sample set,
d13Ccarb values range from –0.44 to –6.12, and d18O values vary
between –5.49 and –13.27. In Figure 31.3 we present a first
attempt at a composite d13Ccarb curve throughout the cap carbonate.
The curve was generated by matching absolute values and trends in
values between 10 sections in the Solan region, Nigalidhar and
Mussoorie synclines. No correction has been attempted to account
for stratigraphic condensation or expansion, which may have
resulted from variation in sedimentation rates or differential com-
paction. Our data confirm the trend observed by Kaufman et al.
(2006), with a pronounced negative excursion to –5‰ d13Ccarb fol-
lowed by a recovery towards less negative values (Fig. 31.3).
Kaufman et al (2006) also report negative values for carbonate
interbedded with the upper diamictites in the Mussoorie syncline.
Significance of isotopic analyses
Neoproterozoic cap carbonates are renowned for being isotopi-
cally light with respect to 13C, and the use of their negative
isotope excursions as chemostratigraphic markers has become a
standard tool in both the correlation of glacially influenced strata
and the construction of a composite Precambrian C-isotope
curve (Halverson 2005).
Whether the values observed here reflect secular changes in the
primary isotopic composition of seawater is difficult to establish. A
lack of pelagic carbonate-secreting organisms makes establishing
a proxy for primary seawater composition challenging, particularly
if the original grains have been destroyed by recrystallization. The
strong covariation of d13C and d18O raises the probability of diage-
netic alteration associated with fluids flushing through the cap
carbonate. Some support for fluid migration is evident given the
ubiquitous occurrence of sparry calcite-filled veins in the cap.
If the up-section trends in values reflect variations in the primary
composition of seawater, then absolute values are likely to differ
between localities given the transgressive nature of the carbonates
deposited during post-glacial eustatic recovery. Given the poor
continuity between exposures, an absence of distinct markers or
fauna to facilitate correlation, sections may only be compared by
trends in values (Fig. 31.3). The composite curve that results is
interesting in that it broadly honours our understanding of the
basin geometry, with progressive onlap (and therefore younging)
from NW to SE, normal to the basin margin (see earlier discussion
and Kaufman et al. 2006, fig. 2), and little variability between more
 THE BLAINI FORMATION
closely spaced sections that occur along strike from one another.
Some perturbations to this trend may indicate palaeobathymetric
variation resulting from deglaciation of the shelf.
Although numerous measurements were made to characterize
the trends across the fold belt, extremely negative values have
not been observed that may implicate local methane hydrate desta-
bilization (cf. Jiang et al. 2003b; McFadden et al. 2008), although
this does not negate dissociation events as a driving mechanism for
negative excursions. The trend towards negative values, and recov-
ery towards more positive values is similar to that recorded
from the Keilberg cap on the South Congo Craton and generally
exhibited by ‘Marinoan’ (c. 635 Ma) cap carbonates elsewhere
(Kennedy et al. 1998). The observation of negative values in inter-
diamictite carbonates by Kaufman et al. (2006) differ from the
general thesis of Kennedy et al. (2001), although some matrix
samples from the Ghaub Fm. on Fransfontein Ridge exhibit nega-
tive values below the Keilberg cap (Kennedy et al. 2001, fig. 1).
Palaeolatitude and palaeogeography
Palaeomagnetic data for the Malani Igneous Suite place Rajasthan
along the western margin of Rodinia in mid to high latitudes at
c. 750 Ma (Torsvik et al. 2001). Palaeopoles for the Blaini Fm.
also indicate a mid- to high-latitude position for the lower diamic-
tite (46.58N, 1098E), but tropical palaeolatitudes for the overlying
cap dolostone (3.08N, 98.58E; Klootwijk 1979). Although these
palaeolatitudes sit comfortably with Phanerozoic analogues of
northern-hemisphere glaciation, positive fold tests are required
to demonstrate primary remnant magnetization. Better geochrono-
logical control is also required to constrain the position of this
important glaciated passive continental margin in palaeogeo-
graphic reconstructions.
Geochronological constraints
A Neoproterozoic (as opposed to the formerly accepted view of
Permo-Carboniferous) age for the Blaini Fm. is supported by the
occurrence of Lower Cambrian trace fossils, trilobites, small
shelly fauna and acanthomorphic acritarchs in basal Tal Group
chert and phosphorites (e.g. Banerjee & Narain 1976; Brasier &
Singh 1987; Bhatt 1989, 1991; Tiwari & Knoll 1994; Tiwari
1999; Hughes et al. 2005; Fig. 31.2). A SHRIMP U –Pb detrital
zircon age of 525 + 8 Ma from Tal Group sediments at Gopichand
ka Mahal in the Mussoorie syncline supports palaeontological data
indicative of a Cambrian age, and thus reflects the youngest
isotopic age limit for the Blaini Fm. (Myrow et al. 2003). In the
353
Fig. 31.4. Probability–age plot for detrital
zircons extracted from massive diamictite
sample JE-1-002 (Māldeota, Mussoorie
syncline).
Krol Belt, the P]/] boundary is located either in the uppermost
part of the Krol Group (Krol E) or at the base of the Tal Group
(Banerjee et al. 1997).
Crystalline basement is not exposed in the region, so maximum
age constraints are poorly defined. However, new detrital zircon
ages for samples of the diamictite at the base of the Blaini Fm.
(at Māldeota in the Mussoorie syncline) provide, to our knowl-
edge, the first direct radiometric data for the unit, with a youngest
subpopulation of three grains giving a mean weighted-average age
of 692 + 18 Ma (207Pb/206Pb; Fig. 31.4). Four grains in the same
sample define a slightly older subpopulation with a mean
weighted-average age of 770 + 24 Ma (207Pb/206Pb). A number
of older detrital populations are evident, which are also recorded
in other samples. The full results of the detrital zircon analysis
will be published elsewhere.
Conclusions
The Blaini Fm. records widespread glaciomarine conditions of
sedimentation in a passive continental margin setting. At least
two major periods of ice advance are recorded by the widespread
distribution of glaciogenic diamictites, which include primary sub-
glacial tillites deposited where the ice was grounded on the conti-
nental shelf, and glaciogenic debris flows that redistributed
subglacial debris onto submarine fans in outer neritic to slope
environments. The huge volumes of glacioclastic debris and
their widespread distribution point towards the existence of an
extensive continental ice sheet, probably with polythermal charac-
teristics. The limited geochronological data do not presently allow
the duration of these glacial events to be constrained, but both
periods of ice sheet growth were probably younger than
692 + 18 Ma. The cap dolomite bears stable isotopic character-
istics typical of Marinoan (c. 635 Ma) post-glacial cap carbonates
elsewhere; however, further data are required to better constrain
the age of these deposits.
This research was funded by Schweizerischer National Fonds Grant 103502.
M. Papp, M. Faccenda and R. Singh are kindly thanked for their assistance in the
field. Stable isotope analyses were undertaken by T. Venneman at the University
of Lausanne and D. Mrofka at the University of California Riverside. Laser
Ablation MC ICP-MS detrital zircon analyses were undertaken at the University
of Alberta, Edmonton. This study has benefited from ongoing discussions with
M. Kennedy (UCR), R. Rieu (Repsol YPF), G. Jiang (University of Nevada) and
A. Cozzi (ENI), and represents a contribution towards the IGCP 512 project.
S. K. Ghosh and R. Islam are grateful to the Director, Wadia Institute of Himalayan
Geology, for providing the necessary facilities. Special thanks go to D. Banerjee
and A. J. Kaufman for constructive reviews, which improved this manuscript.
354 J. L. ETIENNE ET AL.
References
Allen, P. A. & Hoffman, P. F. 2005. Extreme winds and waves in the
aftermath of a Neoproterozoic glaciation. Nature, 433, 123– 127.
Alvarenga, C. J. S. De, Dardenne, M. A. et al. 2007. Isotope strati-
graphy of Neoproterozoic cap carbonates in the Araras Group,
Brazil. Gondwana Research, 13, 469–479.
Auden, J. B. 1934. The Geology of the Krol Belt. Records of the Geologi-
cal Survey of India, 69, 123–167.
Auden, J. B. 1946. Blaini-Talchir. Current Science, 12, 346–348.
Azmi, R. J. & Paul, S. K. 2004. Discovery of Precambrian – Cambrian
boundary protoconodants from the Gangolihat Dolomite of Inner
Kumaun Lesser Himalaya: implication on age and correlation.
Current Science, 86, 1653–1660.
Banerjee, D. M. & Narain, M. J. 1976. Trace fossils in the Lower Tal
Formation of Mussoorie and their environmental significance.
Journal of Sedimentary Petrology, 46, 235–239.
Banerjee, D. M., Schidlowski, M., Siebert, F. & Brasier, M. D. 1997.
Geochemical changes across the Proterozoic –Cambrian transition in
the Durmala phosphorite mine section, Mussoorie Hills, Garhwal
Himalaya, India. Palaeogeography, Palaeoclimatology, Palaeoecol-
ogy, 132, 183– 194.
Benn, D. I. & Evans, D. J. A. 1998. Glaciers and Glaciation. Arnold,
London.
Bhargava, O. N. 1984. Trace fossils from the ?Cambrian Tal Group,
Sirmur District H.P. and proposed redefinition of the Tal. Journal
of the Palaeontological Society of India, 29, 84 –87.
Bhargava, O. N. & Bhattacharyya, B. K. 1975. The Blaini Formation
of Himachal Pradesh and Uttar Pradesh. Bulletin of the Indian
Geologists’ Association, 8, 71 – 99.
Bhargava, O. N., Singh, I., Hans, S. K. & Bassi, U. K. 1998. Early
Cambrian trace and trilobite fossils from the Nigali Dhar Syncline
(Sirmaur District, Himachal Pradesh), lithostratigraphic correla-
tion and fossil content of the Tal Group. Himalayan Geology, 19,
89 –108.
Bhatia, M. R. 1981. Appendix: Late Palaeozoic diamictites of Simla
Hills, Lesser Himalaya, India; chronostratigraphic age and geo-
chemistry. In: Hambrey, M. J. & Harland, W. B. (eds) Earth’s
Pre-Pleistocene Glacial Record. Cambridge University Press,
Cambridge, 293– 294.
Bhatia, S. B. & Kanwar, R. C. 1975. Blaini and related formations.
Special Issue. Bulletin of the Indian Geologists’ Association, 8, 279.
Bhatia, M. R. & Prasad, A. K. 1975. Some sedimentological, lithostrati-
graphic and genetic aspects of the Blaini Formation of parts of Simla
Hills, Himachal Pradesh, India. Bulletin of the Indian Geologists’
Association, 8, 162– 185.
Bhatia, M. R. & Prasad, A. K. 1981. Evolution of Late Paleozoic glacial
marine sedimentation in the Simla Hills, Lesser Himalaya, India.
Neues Jahrbuch fur Palaontologie Mh, 267– 288.
Bhatt, D. K. 1989. Small shelly fossils, Tommotian and Meishucunian
stages and the Precambrian –Cambrian boundary – implications of
the recent studies in the Himalayan sequences. Journal of the
Palaeontological Society of India, 34, 55 – 68.
Bhatt, D. K. 1991. The Precambrian –Cambrian transition interval in
Himalaya with special reference to small shelly fossils – a review
of current status of work. Journal of the Palaeontological Society
of India, 36, 109–120.
Bhatt, D. K. & Mathur, A. K. 1990. Small shelly fossils of the
Precambrian Cambrian boundary beds from the Krol-Tal succession
in the Nainital Syncline, Lesser Himalaya. Current Science, 59,
218– 222.
Boulton, G. S. 1996. Theory of glacial erosion, transport and deposition
as a consequence of subglacial sediment deformation. Journal of
Glaciology, 140, 43 –62.
Brasier, M. D. & Singh, P. 1987. Microfossils and Precambrian –
Cambrian boundary stratigraphy at Maldeota, Lesser Himalaya.
Geological Magazine, 124, 323–345.
Brookfield, M. E. 1987. Lithostratigraphic correlation of Blaini For-
mation (late Proterozoic, Lesser Himalaya, India) with other late Pro-
terozoic tillite sequences. Geologische Rundschau, 76, 477–484.
Brookfield, M. E. 1993. The Himalayan passive margin from Precam-
brian to Cretaceous times. Sedimentary Geology, 84, 1 – 35.
Brookfield, M. E. 1994. Problems in applying preservation, facies and
sequence models to Sinian (Neoproterozoic) glacial sequences in
Australia and Asia. Precambrian Research, 70, 113–143.
De, C., Das, D. P. & Andraha, P. K. 1994. Ichnostratigraphic and
palaeoenvironmental significance of trace fossils from Tal Formation
of Nigali Dhar Syncline, Sirmur District, Himachal Pradesh, India.
Indian Journal of Geology, 66, 77 –90.
Dowdeswell, J. A., Kenyon, N. H., Elverhøi, A., Laberg, J. S.,
Hollender, F.-J., Mienert, J. & Siegert, M. J. 1996. Large-scale
sedimentation on the glacier-influenced Polar North Atlantic
margins: long-range side-scan sonar evidence. Geophysical Research
Letters, 23, 3535– 3538.
Dowdeswell, J. A., Elverhøi, A. & Spielhagen, R. 1998. Glacimarine
sedimentary processes and facies on the polar North Atlantic
Margins. Quaternary Science Reviews, 17, 243–272.
Fuchs, G. & Sinha, A. K. 1978. The tectonics of the Garhwal-Kumaun
Lesser Himalaya. Jahrbuch der Geologischen Bundesanstalt, 121,
219– 241.
Gaur, G. C. S. & Dave, V. K. S. 1971. Blaini tillites near Rishikesh and
their origin. Journal of the Geological Society of India, 12, 164– 172.
Ghosh, S. K. 1991. Palaeoenvironmental analysis of the Late Proterozoic
Nagthat Formation, NW Kumaun Lesser Himalaya, India. Sedimen-
tary Geology, 71, 33 –45.
Halverson, G. P. 2005. A Neoproterozoic chronology. In: Xiao, S. (ed.)
Neoproterozoic Geobiology. Kluwer Academic Publishers, Delft,
Netherlands.
Holland, T. H. 1908. On the occurrence of striated boulders in the Blaini
Formation of Simla with discussion on the geological age of the beds.
Records of the Geological Survey of India, 37, 129–135.
Hughes, N. C., Peng, S. et al. 2005. Cambrian biostratigraphy of the Tal
Group, Lesser Himalaya, India, and early Tsanglangpuan (late early
Cambrian) trilobites from the Nigali Dhar syncline. Geological
Magazine, 142, 57 –80.
Jain, A. K. 1981. Stratigraphy, petrography and palaeogeography of the
Late Paleozoic diamictites of the Lesser Himalaya. Sedimentary
Geology, 30, 43 –78.
Jain, A. K. & Varadaraj, N. 1978. Stratigraphy and provenance of Late
Palaeozoic diamictites in parts of Garhwal Lesser Himalaya, India.
Geologische Rundschau, 67, 49 – 72.
Jell, P. A. & Hughes, N. C. 1997. Himalayan Cambrian trilobites.
Special Papers in Palaeontology, 58, 1– 113.
Jiang, G., Christie-Blick, N., Kaufman, A. J., Banerjee, D. M. &
Rai, V. 2002. Sequence stratigraphy of the Neoproterozoic Infra
Krol Formation and Krol Group, Lesser Himalaya, India. Journal
of Sedimentary Research, 72, 524– 542.
Jiang, G., Sohl, L. E. & Christie-Blick, N. 2003a. Neoproterozoic
stratigraphic comparison of the Lesser Himalaya (India) and Yangtze
block (south China): Paleogeographic implications. Geology, 31,
917–920.
Jiang, G., Kennedy, M. J. & Christie-Blick, N. 2003b. Stable isotopic
evidence for methane seeps in Neoproterozoic postglacial cap
carbonates. Nature, 426, 822–826.
Jiang, G., Kennedy, M. J., Christie-Blick, N., Wu, H. & Zhang, S.
2006. Stratigraphy, sedimentary structures, and textures of the late
Neoproterozoic Doushantuo cap carbonate in South China. Journal
of Sedimentary Research, 76, 978– 995.
Joshi, A. & Mathur, V. K. 1987. Report of Cruziana type trace fossils
from the Arenaceous Member of Tal Formation, Mussoorie
Synform. Indian Minerals, 41, 61 –65.
Joshi, A., Mathur, V. K. & Bhatt, D. K. 1989. Discovery of redlichid
trilobites from the Arenaceous Member of the Tal Formation,
Garhwal Syncline, Lesser Himalaya, India. Journal of the Geological
Society of India, 33, 538– 546.
Kaufman, A. J., Jiang, G., Christie-Blick, N., Banerjee, D. M. & Rai,
V. 2006. Stable isotope record of the terminal Neoproterozoic Krol
platform in the Lesser Himalayas of northern India. Precambrian
Research, 147, 156– 185.
Kennedy, M. J., Runegar, B., Prave, A. R., Hoffman, K.-H. &
Arthur, M. A. 1998. Two of four Neoproterozoic glaciations?
Geology, 26, 1059–1063.
Kennedy, M. J., Christie-Blick, N. & Prave, A. R. 2001. Carbon isoto-
pic composition of Neoproterozoic glacial carbonates as a test of
 THE BLAINI FORMATION
paleoceanographic models for snowball Earth phenomena. Geology,
29, 1135– 1138.
Klootwijk, C. T. 1979. Review of palaeomagnetic data from the Indo-
Pakistani fragment of Gondwanaland. In: Farah, A. & De Long,
K. A. (eds) Geodynamics of Pakistan. Geological Survey of Pakistan,
Queta, 41– 80.
Kumar, G., Raina, B. K., Bhatt, D. K. & Jangpangi, B. S. 1983. Lower
Cambrian body- and trace-fossils from the Tal Formation, Garhwal
Synform, Uttar Pradesh, India. Journal of the Palaeontological
Society of India, 28, 106– 11.
Kumar, G., Bhatt, D. K. & Raina, B. K. 1987. Skeletal microfauna of
Meishucunian and Qiongzhusian (Precambrian– Cambrian bound-
ary) age from the Ganga Valley, Lesser Himalaya, India. Geological
Magazine, 124, 167– 171.
Mathur, V. K. & Joshi, A. 1989a. Record of inarticulate brachiopods
from the Arenaceous Member of the Tal Formation, Garhwal
Syncline, Lesser Himalaya, India. Current Science, 58, 446–448.
Mathur, V. K. & Joshi, A. 1989b. Record of redlichiid trilobite from
the Lower Cambrian Tal Formation, Mussoorie Syncline, Lesser
Himalaya, India. Journal of the Geological Society of India, 33,
268– 270.
Mathur, V. K. & Shanker, R. 1989. First record of Ediacaran fossils
from the Krol Formation, Nainital syncline. Journal of the Geological
Society of India, 34, 245– 254.
Mathur, V. K. & Shanker, R. 1990. Ediacaran medusoids from
Cambrian Tal Formation, Himachal Lesser Himalaya and the Krol
Formation, Naini Tal syncline. Journal of the Geological Society of
India, 36, 74 – 78.
Mathur, V. K. & Srivastava, M. C. 1994. Record of microgastropod
from the Arenaceous Member of the Tal Formation, Garhwal
Syncline, Lesser Himalaya, India. Current Science, 66, 228–229.
Mathur, V. K., Joshi, A. & Kumar, G. 1988. Trace fossils from
Cambrian Tal Formation Himachal Lesser Himalaya, India, and
their stratigraphic significance. Journal of the Geological Society of
India, 31, 467– 475.
Mazumdar, A. & Banerjee, D. M. 1998. Siliceous sponge spicules in
the Early Cambrian Chert-phosphate Member of the Lower Tal
Formation, Krol belt, Lesser Himalaya. Geology, 26, 899– 902.
McFadden, K. A., Huang, J. et al. 2008. Pulsed oxidation and biological
evolution in the Ediacaran Doushantuo Formation. Proceedings of the
National Academy of Sciences of the United States of America, 15,
3197– 3202.
Medlicott, H. B. 1864. On the geological structure and relation of the
southern portion of the Himalayan ranges between the rivers
Ganges and Ravee. Memoirs of the Geological Survey of India, 3,
1 – 212.
Myrow, P. M., Hughes, N. C. et al. 2003. Integrated tectonostrati-
graphic analysis of the Himalaya and implications for its tectonic
reconstruction. Earth and Planetary Science Letters, 212, 433–441.
Niyogi, D. & Bhattacharya, S. C. 1971. A note on the Blaini boulder
beds of the Lower Himalaya. Himalayan Geology, 1, 111– 122.
Oldham, R. D. 1887. Notes on some points in Himalayan Geology.
Records of the Geological Survey of India, 20, 155–161.
Pilgrim, G. E. & West, W. D. 1928. The structure and correlation of
Simla rocks. Memoirs of the Geological Survey of India, 53, 1– 140.
Powell, R. D. & Cooper, J. M. 2002. A glacial sequence strati-
graphic model for temperate, glaciated continental shelves.
355
In: Dowdeswell, J. A. & O. Cofaigh, C. (eds) Glacier-Influenced
Sedimentation on High-Latitude Continental Margins. Geological
Society, London, Special Publications, 203, 215– 244.
Prasad, B., Maithy, P. K., Kumar, G. & Raina, B. K. 1990. Precam-
brian– Cambrian acritarchs from the Blaini-Krol-Tal sequence of
Mussoorie Syncline, Garhwal Lesser Himalaya, India. Memoirs of
the Geological Society of India, 16, 19 – 32.
Rai, V. 1987. Additional trace fossils from the Tal Formation (Early
Cambrian) Mussoorie hills, Uttar Pradesh, India. Journal of the
Palaeontological Society of India, 32, 53 –59.
Rupke, J. 1968. Note on the Blaini Boulder Bed of Tehri Garhwal,
Kumaon Himalayas. Journal of the Geological Society of India, 9,
131– 133.
Saxena, M. N. & Pande, C. 1969. The Blaini Tillite from the type area
of Simla Himalayas. Bulletin of the Indian Geological Association,
2, 57 – 64.
Schelling, D. 1992. The tectonostratigraphy and structure of the eastern
Nepal Himalaya. Tectonics, 11, 925– 943.
Shanker, R., Mathur, V. K., Kumar, G. & Srivastava, M. C. 1997.
Additional Ediacaran biota from the Krol Group, Lesser Himalaya,
India and their significance. Geoscience Journal, 18, 79– 94.
Singh, I. B. & Rai, V. 1983. Fauna and biogenic structures in Krol-Tal
succession (Vendian– Early Cambrian), Lesser Himalaya: their
biostratigraphic and palaeoecological significance. Journal of the
Palaeontological Society of India, 28, 67 –90.
Taylor, J., Dowdeswell, J. A., Kenyon, N. H. & O ´ ’Cofaigh, C. 2002.
Late Quaternary architecture of trough-mouth fans: debris flows and
suspended sediments on the Norwegian margin. In: Dowdeswell,
J. A. & O’Cofaigh, C. (eds) Glacier-Influenced Sedimentation on
High-Latitude Continental Margins. Geological Society, London,
Special Publications, 203, 55 –71.
Tiwari, M. 1999. Organic-walled microfossils from the Chert–
phosphorite Member, Tal Formation, Precambrian –Cambrian
Boundary, India. Precambrian Geology, 99, 99– 113.
Tiwari, M. & Knoll, A. H. 1994. Large acanthomorphic Acritarchs from
the Infrakrol Formation of the Lesser Himalaya and their stratigraphic
significance. Journal of Himalayan Geology, 5, 193– 201.
Torsvik, T. H. 2003. The Rodinia jigsaw puzzle. Science, 300,
1379–1381.
Torsvik, T. H., Carter, L. M., Ashwal, L. D., Bhushan, S. K., Pandit,
M. K. & Jamtveit, B. 2001. Rodinia refined or obscured: palaeomag-
netism of the Malani igneous suite (NW India). Precambrian
Research, 108, 319–333.
Tripathi, C., Jangpangi, B. S., Bhatt, D. K., Kumar, G. & Raina, B. K.
1984. Early Cambrian brachiopods from ‘Upper Tal’, Mussoorie
syncline, DehraDun District, Uttar Pradesh, India. Geophytology,
14, 221– 227.
Tripathi, C., Kumar, G., Mehra, S., Bhatt, D. K., Mathur, V. K. &
Joshi, A. 1986. Additional Early Cambrian (Botomian) brachiopod
fossil localities in Tal Formation, Lesser Himalaya, India, and their
significance. Current Science, 55, 585–588.
Valdiya, K. S. 1973. Blaini conglomerates of Himachal Pradesh and
Garhwal. Recent Researchs in Geology, 1, Hindustan Publishing
Corporation, Delhi.
Valdiya, K. S. 1995. Proterozoic sedimentation and Pan-African
geodynamic development in the Himalaya. Precambrian Research,
74, 35– 55.
 Chapter 32

Neoproterozoic glacial records in the Yangtze Region, China

QI-RUI ZHANG*, XUE-LEI CHU & LIAN-JUN FENG

State Key Laboratory of Lithospheric Evolution, Institute of Geology and Geophysics, Chinese Academy of Sciences,
PO Box 9825, Beijing 100029, China
*Corresponding author (e-mail: qrzhang@mail.iggcas.ac.cn)

Abstract: The Nanhua rift occurred at c. 820 Ma. The last rifting stage was likely associated with the Neoproterozoic glaciations. The
glacially influenced sequence, in ascending order, comprises the Jiangkou Group (c. 720– 663 Ma), the interglacial Datangpo Formation
(Fm.) (c. 663 Ma) and the Nantuo Fm. (,656– 635 Ma). In deep-water environments the glacially influenced units exhibit conformable
contacts with a transitional lower boundary and a sharp upper boundary. In shallow marine environments, the lower boundary is always
unconformable and erosive and the upper boundary is sharp but conformable.
The Jiangkou glaciogenic deposits likely comprise the waxing, the maximum and the waning stages of glaciation. The lower part of the
Chang’an Fm. corresponds to the waxing stage. It initially inherited the pre-glacial deep-water environment, and later, the thickening and
coarsening upward trend likely indicates a lowering of sea level and ice-sheet advance. The predominant, massive diamictite in the upper
Chang’an Fm. probably highlights the maximum glacial stage, and likely resulted from an amalgamation of debris flow deposits, suspen-
sion and ice-rafted debris. The Fulu Fm. records the waning stage; its lower member is characterized by arkosic sandstone and greywacke
indicative of an interstadial between glacial events, while the upper member represents the final glacial event of the Jiangkou glaciation.
The carbonaceous black shale and Mn carbonate at the base of the interglacial Datangpo Fm. were likely indicative of sedimentary star-
vation associated with Jiangkou deglaciation. The Nantuo glaciation is characterized by an abrupt onset and termination with negligible
waxing and waning stages. The sediments likely resulted from subglacial deposition, sedimentary gravity flows and ice-rafted debris with
reworking by currents.
The onset of Jiangkou glaciation took place around c. 720 Ma and terminated at c. 663 Ma. The Nantuo glaciation started at
,c. 656 Ma and terminated at 635 Ma. The palaeolatitude during the Jiangkou glaciation is unconstrained and was 37 + 78 during
the Nantuo glaciation. The palaeogeographic reconstruction for the Yangtze Block is controversial: it was either located between
Australia and Laurentia or to the NW of Australia and to the north of India.
Values of d34Spyrite from the top of the Fulu and the base of Datangpo formations exhibit extraordinarily high values up to þ60‰. Values
of d13Ccarb from the base of the Datangpo Fm. range from –7.4‰ to –13.0‰ and from the cap carbonate of the Doushantuo Fm. are gen-
erally lower than –5‰; however, methanogenic values as low as –48‰ have been reported. From the Datangpo Fm., dinosterane has been
detected, and phytane, triterpanes and hopane-type pentacyclic triterpanes are also abundant. Biomarker data from glacially influenced rocks
indicate that the photosynthetic process never ceased and that the marine environment was probably normal during glaciations.
The banded iron formation (BIF) at the base of the Fulu Fm. is either basaltic, tuffaceous, siliciclastic or carbonate-hosted. Rhodochrosite
at the base of the Datangpo Fm. was probably associated with submarine volcanic and hydrothermal activity, but a detrital oxide origin
cannot be excluded.
The difference between the Jiangkou and the Nantuo glaciations is significant. Tectonics and associated topography are speculated to be
the main controlling factors on the Neoproterozoic glaciations in the Yangtze region.

The Neoproterozoic glacially influenced (GI) rocks in China
were first recognized by Willis et al. (1907). Wang et al. (1981)
summarized the Neoproterozoic GI record in China and Lu et al.
(1985) edited the first memoir of the Neoproterozoic GI rocks in
China. By the 1990s, a series of geological memoirs on the
regional geology in each province reported the GI successions in
detail. This paper focuses on the Neoproterozoic GI records in
the Yangtze region of South China.
The stratigraphic classification scheme of Neoproterozoic GI
deposits used in this paper is shown in Figure 32.1b. The strati-
graphic nomenclature changes considerably, so brief explanations
are necessary.
The ‘Sinian’ System and Period
The term ‘Sinian’ first appeared in the nineteenth century
(Richthofen 1882), and since 1983 it has been restricted to the
Neoproterozoic in South China (Liu et al. 1991). More recently,
however, it was redefined as the stratigraphic unit between the
Nantuo Fm. and the Cambrian (NCS 2001), and coincides with
the Ediacaran Period (Knoll et al. 2004).
The Nanhuan System and Period
The Nanhuan ‘period’ ranges between 800 Ma and 680 Ma (NCS
2001), and so almost coincides with the chronometrically defined
Cryogenian Period (850 –650 Ma) (Remane 2000). However, this
paper redefines its range as from c. 720 Ma to 635 Ma on the basis
of the most recent geochronological results.
The Jiangkou Group and the Jiangkou glaciation
The Jiangkou Group, formerly the Jiangkou Fm. (Yang 1985), is a
term for the lower GI deposits and encompasses the Chang’an and
Fulu formations (BGMR-Hunan 1997; Zhang et al. 2003).
Before the 1980s, the term ‘Chang’an glaciation’ (Wang et al.
1981) was used to represent the early glaciation, and Lu et al.
(1985) proposed the term ‘Gucheng glaciation’ to replace it.
However, neither is appropriate, so the term Jiangkou glaciation
is preferred (Zhang et al. 2003).
The Fulu Fm.
The term ‘Fulu Fm.’ has been redefined at least twice. It was first
defined as an interglacial unit between the Nantuo and the
Chang’an formations (Fig. 32.2b, A) (BGMR-Guangxi 1985).
The upper boundary of the Fulu Fm. was then lowered to the
base of the later established Datangpo Fm. (Lu et al. 1985).
Recently, the upper boundary of the Fulu Fm. was again lowered
to the base of the ‘Gucheng Fm.’ (Fig. 32.2b, A), which was
emphasized to be the record of a single glaciation (BGMR-Hunan

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 357– 366. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.32
358
Fig. 32.1. (a) Outcrops of Neoproterozoic glacial deposits in the Yangtze region (adapted from Liu et al. 1991): (i) outline of the mainland of China and (ii) the inset
rectangle in (i). The star in (ii) is the location of Zhaoxing section and the pentagon the location of Shixian section. (b) A comprehensive Neoproterozoic stratigraphic
succession in southeastern Guizhou (adapted from BGMR-Guizhou 1987). 1, Condon et al. 2005; 2, Zhou et al. 2004; 3, Zhang et al. 2008a; 4, Zhou et al. 2007a; 5, Wang
et al. 2003a; *, Hao & Zhai 2004.
1997; Zhu et al. 2007). This paper adopts the definition given by
Lu et al. (1985).
By 1985, based on the occurrence of BIF and rhodochrosite, the
Fulu Fm. was considered to be an interglacial unit sandwiched
between the ‘Chang’an glaciation’ and the Nantuo glaciation.
After the establishment of the interglacial Datangpo Fm., the
former ‘interglacial’ nature of the Fulu Fm. became invalid and
in fact it records the waning stage of the Jiangkou glaciation.
Structural framework
At c. 1140 –900 Ma, the Yangtze and the Cathaysian plates col-
lided along the southeastern margin of the Yangtze region (Li
et al. 2008). The Nanhua rift started at c. 820 Ma (Fig. 32.1b) at
the marginal area of the plate (Wang & Li 2003). At c. 830–
795 Ma and c. 780 –745 Ma, two pulses of bimodal anorogenic
magmatic activity occurred (Li et al. 2003). The early pulse
began prior to, and peaked at the onset of the rift; the later pulse
of volcanism accompanied the main rifting stage. The first three
rifting phases were likely pre-glacial, and the fourth phase associ-
ated with glaciation (Wang & Li 2003).
The surface below the Neoproterozoic GI deposits is important
and mostly represents a characteristic stratigraphic break. It prob-
ably resulted from the uplift and erosion of the rift shoulder. The
pre-glacial units immediately below the erosive surface become
older towards the north (Fig. 32.2b), while units above the
erosive surface become younger (Fig. 32.2b), probably indicating
that rift-induced subsidence persisted throughout the Nanhuan
period. The Doushantuo cap dolostone likely marks the end of
rifting and the onset of thermal subsidence.
Magmatism during the Nanhuan was relatively weak and
probably represents the final episode of magmatism and
rifting. The rifting may reconcile the great contrast between the
Q. R. ZHANG ET AL.
uplift and erosion in the interior of the region and the accommo-
dation created in the rift basin. Syn-sedimentary faulting is
evident from abrupt changes in thickness of the GI units. The
manganese deposits of the Datangpo Fm. were likely related to
hydrothermal and submarine volcanic activity (Liu 1994; Yang
et al. 1997).
The great thickness (.2000 m, Fig. 32.3) of the Nantuo Fm. in
the SW of Hunan (BGMR-Hunan 1988) and SE of Guizhou
(BGMR-Guizhou 1987) probably signifies the final filling of the
rift basin. The ‘Nd isotopic shift’ (c. 765 Ma to 635 Ma) revealed
by Li & McCulloch (1996) also reflects the final filling stage of
the Nanhua Rift.
Stratigraphy
A comprehensive stratigraphic column of the Neoproterozoic is
given in Figure 32.1b. The GI units (c. 720– 635 Ma) bracketed
by the Ediacaran and the ‘Yangzian’ (Hao & Zhai 2004) are
defined as Nanhuan (Cryogenian) in this paper. There are two GI
units: the Jiangkou Group and the Nantuo Fm. (Fig. 32.1b).
The Jiangkou Group comprises two formations: the Chang’an
and the Fulu. The Chang’an Fm. occurs only in the western parts
of the SE margin (Figs 32.2 & 32.3). The Fulu Fm. is composed
of two members: the Liangjiehe and the Gucheng (sections A to
D in Fig. 32.2); the contact between them is generally conformable
(section A & B in Fig. 32.2). The Nantuo Fm. (Fig. 32.1b) is the
most widespread Neoproterozoic GI unit (Figs 32.2 and 32.3).
Sandwiched between the two GI units is the interglacial Datangpo
Fm., which is thinner and finer grained. In the east of the Yangtze
region, this tripartite structure is also typical, but the two lower
units are much thinner. Generally, the Nanhuan in the Yangtze
region is better preserved because of the tectonic simplicity and
low degree of metamorphism.
 THE YANGTZE REGION 359

Fig. 32.2. Neoproterozoic lithostratigraphy in the Yangtze region. (a) Variation along a north–south direction (locations of sections are given as stars in Fig. 32.3). The
distance between the two end sections is c. 607 km. Distances between adjacent sections are also given. Vertical scale is the same except for section A. (b) Schematic
stratigraphic break structure. Section A is adapted from BGMR-Guangxi (1985). Sections E, F and G are adapted from Sha et al. (1963). Section G shows no
Neoproterozoic glacial deposits, and section A is complete.

Glaciogenic deposits and associated strata
The sedimentological description and facies analysis of the GI
deposits are updated regularly, with the most recent and detailed
version being provided by Dobrzinski & Bahlburg (2007).
Figure 32.3 shows the distribution of the Nantuo Fm., in which
the dashed line outlines the distribution of the Jiangkou Group.
with cross-stratification and loading structures, and lenses of con-
glomerate and breccia (Fig. 32.4a). Usually it is grey to greenish in
colour with 60 to 80% clay minerals. Clast content is less than 5%,
with sizes ranging from 0.2 cm to 5 cm and maximum 80 cm, sub-
angular to sub-rounded. Rock types are quartz, sandy mudstone,
sandstone, quartzite-like sandstone, minor volcanic rocks and,
rarely, striated clasts.

The Chang’an Fm. The Fulu Fm.

In deep-water environments, the contact between the Chang’an
and the pre-glacial Gongdong formations is generally conformable
(BGMR-Guangxi 1985; BGMR-Guizhou 1987). The transitional
base of the Chang’an Fm. is distinguished from the pre-glacial
grey to greenish mudstone by sandy mudstone with clasts.
The Chang’an Fm. can be divided into two parts. The majority
of the lower part (619 m thick) is mudstone (Fig. 32.4a), with
cycles showing 0.1– 0.8 mm stratification. There are also coarse
to fine-grained, massive to graded sandstones, with few current
ripples and sole flutes, stratified and massive diamictites and
minor granular conglomerate at the base. Clast size varies from
0.2 to 5 cm, clast shape from sub-rounded to angular, and the
clast types are mainly quartz, sandstone and slate, with minor
fragments of basic to acidic igneous rocks. Locally, there are
lonestones (likely dropstones), striated clasts and mudstone
breccia. Reverse grading is occasionally seen in coarse-
grained intervals.
The upper part (1279 m) comprises massive diamictite (92%
of whole), with subordinate stratified mudstone, sandstone
The Fulu Fm. comprises two members (Fig. 32.4b): the Liangjiehe
and the Gucheng (Zhang et al. 2003).
The Liangjiehe Member (583 m) comprises mainly massive,
fine-grained arkosic arenite or greywacke with lonestones, and
occasionally with beds and lenses of granular conglomerate and
diamictite. Cross-stratification and ripple marks, as well as
graded bedding, are present. Locally at the base there are layers
of purple to reddish banded ironstone with graded bedding and
cross-stratification. Beds or lenses of micritic dolostone are
also reported.
From the sedimentary characteristics of the Liangjiehe
Member in the Zhaoxing section (Fig. 32.4b), it is hard to deter-
mine its climatic character. However, there are sections where
the Liangjiehe Member is typified by lonestone and beds of
massive diamictite. Therefore, it is reasonable to suggest that
it is interstadial.
The Gucheng Member in the type section in Changyang County
of Hubei Province is only 5.6 m thick, and comprises beds of
diamictite with about 30– 40 vol% clasts, and lenses of sandy
360
Fig. 32.3. Distributions of Jiangkou Group (outlined by the black dashed line; adapted from BGMR-Guangxi 1985 and Wang & Li 2003) and Nantuo Fm. (grey areas;
adapted from Wang & Li 2003). The contours show thickness of the Nantuo Fm. only.
conglomerate, sandy mudstone with lonestones (interpreted as
dropstones), and striated clasts (Ma & Wang 1983). In the deep-
water Zhaoxing section the Gucheng Member is about 217 m
thick (Fig. 32.4b).
The Fulu Fm. is also characterized by its calcareous composition
(Dobrzinski & Bahlburg 2007) and dolomite mounds or lenses
(Zhou et al. 2007b).
The Datangpo Fm.
This is named after a section in northeastern Guizhou Province,
with more than 500 m (Fig. 32.5) of fine-grained muddy
sandstone, muddy siltstone, mudstone and about 5 m of black
carbonaceous manganese shale at the base (Ma & Wang
1983). A range of sedimentary structures (gas holes, diapirs,
seeping tubes, soft sediment deformation and mud volcanoes)
are reported from layers and lenses of rhodochrosite associated
with dolomite lenses (Zhou et al. 2007b). The upward coarsen-
ing and thickening trend is obvious. To the SE of the Yangtze
region, it thins to a thickness of c. 12 m of dark carbonaceous
shale (Fig. 32.4c) (BGMR-Guangxi 1985). However, there are
sections without black carbonaceous shale but only grey-green
massive mudstone.
The Nantuo Fm.
The maximum thickness of the Nantuo Fm. is more than 2000 m
(Fig. 32.3); it is composed predominantly of diamictite with
scarce mudstone, sandstone and conglomerate. Dropstones, and
Q. R. ZHANG ET AL.
striated and bullet-shape clasts (Fig. 32.4c) are frequently observed
(BGMR-Guangxi 1985). Because more than 90% of its thickness is
massive diamictite, it looks similar to the upper part of the
Chang’an Fm.; however, its clasts are larger, more common and
lithologically more variable (e.g. granite, diabase, volcanic rocks,
carbonate etc.), especially in its upper part. Allochthonous sedi-
mentary blocks, in cases up to a maximum thickness of 10 m, fre-
quently occur on shelf and upper slope environments. Diamictites
with sharp base and deformed stratification, and occasionally
gravel lag deposits, can be recognized in fresh outcrops.
The cap carbonate of the Doushantuo Fm.
The cap carbonate represents the first member of the Doushantuo
Fm., typically 3– 6 m thick and extending basin wide with the
Nantuo Fm. (Fig. 32.3). It can be classified into three intervals:
C1, the strongly disrupted and cemented basal layer; C2, the
middle laminated layer with local tepee-like structures; and
C3, the thinly laminated silty and shaly limestone and dolomite
(Jiang et al. 2006). More than 90% of the rock minerals are
micritic and microcrystalline dolomite with little quartz, feldspar
and muscovite; calcite occurs mainly as filling materials in
fissures (Wang et al. 2005). On shelf environments the cap
carbonate is thicker, less deformed, with cemented breccias,
tepee-like structures, stromatactis-like cavities, sheet cracks
and barite fans. On the outer shelf or upper slopes it is thinly
laminated and deformed, and in the basal part (C1 of Jiang
et al. 2006). The centimetre-scale dolostone layers frequently
intercalate with lenses or layers of silicate. The cap carbonate
is missing in the lower slope and basin environments.
 THE YANGTZE REGION 361

Fig. 32.4. Stratigraphic columns of the (a) Chang’an and (b) Fulu formations along the Zhaoxing section (the star in Fig. 32.1a,ii; adapted from BGMR-Guizhou 1987)
and (c) the Datangpo and Nantuo formations along the Shixian section (the pentagon in Fig. 32.1a,ii; adapted from BGMR-Guangxi 1985).

Boundary relations with overlying and underlying
non-glacial units
The boundaries of the GI units are conformable in slope to
basin environments; in shelf environments the upper boundary
remains conformable but the lower boundaries are always
unconformable.
The upper boundary of the Jiangkou Group
The boundary is conformable but lithologically sharp. Below the
surface is diamictite (gravel bearing sandy mudstone) of the
Gucheng Member and above it is usually the carbonaceous black
shale and locally grey-greenish mudstone of the Datangpo Fm.

The lower boundary of the Nantuo Fm.

The lower boundary of the Jiangkou Group
This is mostly unconformable and in contact with different units;
however, it is conformable in the SW deep-water environment
(Fig. 32.2b), for example, in the Zhaoxing section (the star in
Fig. 32.1a(ii)) it is transitional. The top part of the Gongdong
Fm. (Fig. 32.1b) comprises thinly bedded, grey sandy mudstone
in which a single layer (c. 4 cm thick) is usually composed of a
parallel laminated lower part and a structureless upper part; this
is likely to correspond to the Td and Te of the Bouma divisions,
respectively. The base of the Chang’an Fm. is, however, char-
acterized by a thickly bedded greywacke, c. 20 m thick, with
sporadic pebbles (,5 mm), on top of which is laminated sandy
mudstone, where dropstones and clast clusters are occasionally
found.
On shelf and upper slope environments, the contact between
Nantuo Fm. and the Datangpo Fm. is lithologically sharp and
erosive. In deep-water environments, however, it is locally transi-
tional from mudstone to inversely graded diamictite beds as seen in
the Pancao section of the Guzhang County in Hunan Province and
in the Dacao section of the Ningguo County in Southern Anhui
Province.
The upper boundary of the Nantuo Fm.
The contact between the Nantuo and Doushantuo formations is
sharp, conformable and occasionally pyrite-bearing (Jiang et al.
2006). The top of the Nantuo Fm. usually comprises diamictite
362 Q. R. ZHANG ET AL.
in shelf environments; however, in slope and basin environments
there could exist a thin interval of sandy mudstone or
muddy siltstone.
Isotope stratigraphy
34
d Spyrite
Fulu Fm. This is about 30 m thick in the Minle section of Hunan
Province. d34Spyrite values of þ24.1‰ and þ23.5‰ at c. 1 m
and 10 m from the base and þ41.8‰ to þ58.4‰ at c. 23 m and
29 m are reported by Li et al. (1999). Zhou et al. (2007b) speculate
that these high positive d34Spyrite values were likely associated with
cold methane seepage.
Datangpo Fm. Super heavy values of d34Spyrite (40 –60‰) are
reported from the vicinity of rhodochrosite mineralization (Li
et al. 1999, 2006; Tang & Liu 1999) and are attributed alternatively
to the anoxic conditions, low level of dissolved sulphate S (Liu
et al. 2006) or restricted basin (Li et al. 1999). However, Wang
et al. (2005) suggest that these extraordinarily high d34S values
were probably associated with cold methane seepage, as they are
similar to the values of c. 47.8– 67.1‰ from ODP Site 1146 in
the South China Sea.
Cap carbonate of the Doushantuo Fm. Values of d34Spyrite ranging
from 36.35‰ to 50.62‰ are reported by Wang et al. (2005),
with the high values interpreted as recording sulphate reduction
at cold methane seeps.
d13Ccarb
Fulu Fm. In the upper part of the Fulu Fm., values of d13Ccarb from
dolomite and pebbly dolomitic sandstone are higher than –2.86‰
and were proposed to be associated with cold spring carbonate
(Zhou et al. 2007b).
Fig. 32.5. The type section of Datangpo
Fm. in Songtao County in the NE of
Guizhou Province (adapted from Yang
et al. 2002).
Datangpo Fm. At the base of the Datangpo Fm., values of d13Ccarb
of rhodochrosite range from –7.4‰ to –13.0‰ (Tang & Liu 1999;
Yang et al. 2002), so values of d13Ccarb from the base of Datangpo
Fm. are probably lower than –5‰ (Chu et al. 2003).
Doushantuo Fm. Carbon-isotope data from the cap carbonate of
Doushantuo Fm. are abundant. Jiang et al. (2003a) and Wang
et al. (2008a) reported very negative C-isotope values as low
as –48‰ in the Yangtze Gorges area and in Changyang
County, interpreted as evidence for methanogenesis. Generally
there is a negative excursion in the cap carbonate of the
Doushantuo Fm. (Jiang et al. 2007; Zhou & Xiao 2007; Zhu
et al. 2007).
Nd depleted mantle model ages (tDM)
The tDM ages of sediments from the Neoproterozoic to lower
Danzhou Group (c. 820– 765 Ma, Fig. 32.1b) are stable at
c. 1.8 Ga (Li & McCulloch 1996). In the upper Danzhou Group
(c. 765–720 Ma, Fig. 32.1b) the tDM ages drop to 1.3–1.4 Ga,
and during the Nanhuan (c. 720–635 Ma, Fig. 32.1b) return to
c. 1.8 Ga. From the Ediacaran onwards (,635 Ma), the tDM ages
stabilize at c. 1.8 Ga again. This ‘Nd isotopic shift’ (Li &
McCulloch 1996, fig. 3) probably reveals the final filling of the
Nanhua rift by predominant juvenile crustal materials.
Other characteristics
Banded iron formation (BIF)
BIF appears at the base of the Fulu Fm. (Fig. 32.4), and can be
traced from the northern Guangxi Region in the west to the
middle of the Jiangxi Province in the east of the Yangtze region.
Tang et al. (1987) classify the deposits into four types: basaltic,
tuffaceous, siliciclastic and carbonate interbedded.
 THE YANGTZE REGION
Manganese deposits
The black shale-hosted manganese deposits of the Datangpo Fm.
are important in China, although the deposits are relatively small
(Roy 2006). They are mostly associated with faults and formed
in local rift or graben basins. The content of volcanic materials
may reach 25–50% (Yang & Pang 2006). Submarine volcanic
and hydrothermal activities are considered to be the major factor
in the formation of the deposits (Yang et al. 1997; Qin et al.
2005; Yang & Pang 2006). They are also closely associated with
deglacial transgression after the Jiangkou glaciation, so Liu et al.
(2006) speculate that they resulted from diagenetic remobilization
of lateritic soil residues deposited in restricted basins by low-
latitude glaciation. These two factors are not mutually exclusive,
and a combined model is proposed (Tang & Liu 1999).
Biomarkers
Wang et al. (2003b, 2008b) studied biomarkers from samples of
the Nanhuan System. The total organic content (TOC) of diamic-
tites and cap carbonate range from 0.01 to 0.13%, 1–2 orders of
magnitude lower than those of the non-GI sedimentary rocks.
The positive correlation trend between TOC and the concen-
trations of phytane and/or pristane implies that photosynthetic
autotrophs were probably the primary organisms contributing to
the sedimentary organic matter. The unusually low concentrations
of phytane plus pristane (0.005 –0.076 ng g21) in diamictites
reveal a very weak photosynthetic process, and likely indicate
that the photosynthetic process never ceased, allowing the survival
of photosynthetic eukaryotes and other organisms. The GI sedi-
mentary environment was probably normal marine (Wang et al.
2003b, 2008b).
Dinosterane has been detected from black shale of the Datangpo
Fm. (Meng et al. 2003), indicating that dinoflagellates and other
eukaryotic algae had evolved by the mid-Neoproterozoic. Xie
et al. (1999) report n-alkane distribution curves for manganese
deposits of the Datangpo Fm. Phytane is predominant, while the
Pr– Ph ratios (0.71 –0.83) indicate a reducing environment. Triter-
panes and hopane-type pentacyclic triterpanes are abundant, and
the organic matter originated from algae and microbes was
likely associated with methane seeps (Zhou et al. 2007b).
Palaeolatitude and palaeogeography
Palaeopoles
Li et al. (1996) report a palaeolatitude of 38 + 88 for the Liantuo
Fm. (766 + 18 Ma) in Hubei Province. Zhang & Piper (1997)
achieve a similar palaeolatitude of 37 + 78 for the Nantuo Fm.
(635 Ma) in Yunnan Province.
The proximity of the palaeopoles of Liantuo and Nantuo
formations indicates that either the palaeopoles remained almost
unchanged from the Liantuo stage to the Nantuo stage, or
that there was a loop in the apparent polar wander path from
c. 766 –635 Ma.
Palaeogeographical position
The relative position of the Yangtze region is controversial. Li
et al. (2008) places the Yangtze region (South China Block)
between Australia and Laurentia during the assembly of the
supercontinent Rodinia at c. 900–780 Ma and proposes that it
moved to the NE of Australia during the break-up of Rodinia at
c. 750–630 Ma. Kirschvink (1992), Zhang & Piper (1997), Jiang
et al. (2003b) place the South China Block to the NW (present
coordinate) of Australia and to the north of India.
363
Geochronological constraints
Onset of the Jiangkou glaciation
A sample from the top unit of the Banxi Group (Fig. 32.2b, B) pro-
vides a SHRIMP U –Pb zircon age of 725 + 10 Ma (Zhang et al.
2008a). Because the Danzhou Group (Fig. 32.2b, A) is transitional
upward into the Jiangkou Group, this age becomes a new
maximum age constraint on the onset of the Jiangkou glaciation,
and the age of the onset is reasonably estimated to be c. 720 Ma.
Termination of the Jiangkou glaciation
The U –Pb zircon age of 663 + 4 Ma from a sample near the base
of the Datangpo Fm. (Zhou et al. 2004) approximates to the ter-
mination of the Jiangkou glaciation (Fig. 32.1b).
Onset of the Nantuo glaciation
A recent U– Pb SHRIMP zircon age of 656 + 3 Ma from the top of
the Dantangpo Fm. in west Hunan indicates that the onset of the
Nantuo glaciation should be younger than c. 656 Ma (Zhang
et al. 2008b). Considering the shelf setting in the sampled
section and significant erosion at the Nantuo/Datangpo boundary,
the Nantuo glaciation could be significantly younger than 656 Ma.
Termination of the Nantuo glaciation
After the dating of 635.2 + 0.6 Ma from samples near the base of
the cap carbonate on top of the Nantuo Fm. (Condon et al. 2005),
the age of the termination of Nantuo glaciation is constrained to
c. 635 Ma (Ogg & Gradstein 2008).
Discussion
There has been a prolonged debate on the correlation of the
Liantuo Fm. with various GI units (Yin et al. 2003 among
others), which has brought great difficulty to the investigation of
the evolution of Neoproterozoic glaciations in South China.
However, a newly achieved age of 725 + 10 Ma (Zhang et al.
2008a) from the top of the Banxi Group, the recalculated age of
766 + 18 Ma of the former 748 + 12 Ma of the Liantuo Fm.
(Zheng 2003) and 765 + 14 Ma from the Danzhou Group (Zhou
et al. 2007a) seem to have solved the problem and confirm that
the Liantuo Fm. is not a GI unit and can only be correlated with
the middle of the Banxi (Danzhou) Group.
Based on the above evidence, it is almost certain that in the
Yangtze region there were two Neoproterozoic glaciations: the
Jiangkou (c. 720 to 663 + 4 Ma) and the Nantuo (,656+3 to
635.2 + 0.6 Ma). The salient decrease in tDM ages from c.
765 Ma to c. 720 Ma probably indicates that the uplifted rift
shoulder was unroofed, weathered and eroded, and supports the
proposition that the atmospheric content of CO2 was significantly
drawn down, triggering glaciations (Goddéris et al. 2007). The ‘Nd
isotopic shift’ reveals that a great amount of juvenile mantle-
derived material was incorporated into the sedimentary prove-
nance (Li & McCulloch 1996). The peculiar hiatus architecture
of the GI stratigraphy in the Yangtze region (Fig. 32.2b) testifies
independently to the uplifting, erosion and filling of the rift
basin. It is therefore the tectonics and greatly evolved topography
that finally controlled the geological records of the glaciations.
The depositional environment
It was the Nanhua rift that provided the accommodation necessary
for GI deposition. The northwardly expanding distribution of
364 Q. R. ZHANG ET AL.
younger GI units (Fig. 32.2b) likely signifies the protracted
subsidence of the basin. Except for a few outcrops in the west of
the region, the Neoproterozoic GI deposits are believed to
be glaciomarine.
The Jiangkou glaciations. The maximum thickness is 2698 m
(BGMR-Guizhou 1987) and it can be divided into three stages.
The first or waxing stage corresponds to the lower part of the
Chang’an Fm. (Fig. 32.4a). The laminated sandy mudstones with
dropstones and clast-clusters at the base indicate an inherited
deep-water environment with the addition of glacially derived
sediments. The thickening and coarsening upward trend probably
signifies the lowering of the sea level and advance of ice sheets.
The predominant massive diamictite with graded sandstones and
mudstone in the upper part of the Chang’an Fm. (Fig. 32.4a) prob-
ably records the amalgamation of debris-flow deposits and rainout
in an ice-distal environment during the second or main glacial
stage. The shallow ice-proximal records of these two stages were
likely deposited originally, but were not preserved because of
erosion by the later glacial events. In the waning or third stage,
BIF and massive and cross-stratified greywacke are probably
indicative of an interstadial, where the frequent occurrence of
lonestones indicates the existence of icebergs and ice sheets. The
diamictite in the Gucheng Member was pro-glacial, as described
by Dobrzinski & Bahlburg (2007), and likely subglacial or ice-
proximal too in certain sections.
The Datangpo interglacial. The thin starved sediments of the
Datangpo Fm. in the Zhaoxing-Shixian area (the star and pentagon
in Fig. 32.1a(ii)) are probably indicative of a deep-water environ-
ment. With few exceptions, nothing coarser than muddy fine-
grained sandstone is reported from the unit in this region, so it
probably represents a distal marine environment. The basal black
shale and Mn dolomite were probably deposited in a restricted
shelf environment at the early transgression induced by the degla-
ciation of the Jiangkou glaciation.
The Nantuo glaciation. The Nantuo Fm. in the Yangtze Gorges area
was originally believed to be terrestrial (Wang et al. 1981, among
others) and was known as the Nantuo Tillite for more than half a
century. In fact, the Nantuo Fm. in the Yangtze Gorges area
(Zhang 1995), as well as in the Nanhua rift, was glaciomarine.
The predominant diamictite was likely subglacial and ice-proximal
in an inner shelf environment, including sediment gravity flow and
rainout. In ice-distal environments, amalgamation of debris flow
deposits, rainout diamictite and disturbed deposition are dominant.
The thin layer of lag deposits (Liu et al. 1999) likely indicates the
existence of a winnowing current. The inversely graded pebbly
mudstone at the transitional lower boundary was likely hemipela-
gic, and the sandstone or sandy mudstone at the top was turbiditic
in origin; however, the waxing and waning stages are not
significant.
Jiang et al. (2006) consider that the cap carbonate of the Doush-
antuo Fm. shares morphological and petrographic attributes with
modern and ancient methane seeps, as a result of post-glacial
warming of the ocean, and is indicative of relatively deep-water
deposition, most likely below storm wave base.
Comparison between the Jiangkou and the Nantuo glaciations
The differences between the two glaciations are obvious. The most
conspicuous difference is in stratigraphic complexity. The Jiangkou
glaciation can be divided into three stages, the waxing, the
maximum and the waning. The preservation of the Jiangkou
glacial units is highly limited in the southeastern marginal region
of the Yangtze block, because the complete succession is restricted
to the deep-water environment. The Nantuo glaciation was relatively
simple, usually with negligible or no waxing and waning records.
The reasons for these differences are controversial. Dobrzinski
& Bahlburg (2007) emphasize the different reaches of the ice
sheets and ascribe the differences to the palaeolatitude, intensity
and extension of the two glaciations. In fact, the role of the tectonic
control implied in Figure 32.2b was likely crucial, and the
difference of the evolving topography in different rift stages was
probably a secondary controlling factor. The Jiangkou glaciation
happened within a background of contrasting relief, while the
Nantuo glaciation occurred within a glacially levelled topography.
Sedimentary structures (Liu et al. 1999), redox proxies
(Dobrzinski et al. 2004), and biomarkers (Wang et al. 2003b,
2008b) probably suggest that hydrological cycle was not shut
down during glaciations in the Yangtze region.
This paper was supported by project nos 40532012 and 40373011 of the Natural
Science Foundation of China (NSFC) and project KZCX3-SW-141 of the Chinese
Academy of Sciences (to X.-L. Chu). We thank G. A. Shields and E. Arnaud for
their encouragement and help in writing this paper. Detailed and insightful
reviews by G.-Q. Jiang, C.-M. Zhou, H. Bahlburg and G. A. Shields-Zhou are
gratefully acknowledged. This is a contribution to International Geological
Correlation Program Project 512.
References
BGMR-GUANGXI (BUREAU OF GEOLOGY AND MINERAL
RESOURCES OF GUANGXI ZHUANG AUTONOMOUS
REGION) 1985. Regional Geology of Guangxi Zhuang Autonomous
Region. Geological Publishing House, Beijing, 853 (in Chinese with
English abstract).
BGMR-GUIZHOU (BUREAU OF GEOLOGY AND MINERAL
RESOURCES OF GUIZHOU PROVINCE) 1987. Regional
Geology of Guizhou Province. Geological Publishing House,
Beijing, 698 (in Chinese with English abstract).
BGMR-HUNAN (BUREAU OF GEOLOGY AND MINERAL
RESOURCES OF HUNAN PROVINCE) 1988. Regional Geology
of Hunan Province. Geological Publishing House, Beijing, 719 (in
Chinese with English abstract).
BGMR-HUNAN (BUREAU OF GEOLOGY AND MINERAL
RESOURCES OF HUNAN PROVINCE) 1997. Multiple Classifi-
cation of the Stratigraphy of China (43): Stratigraphy (Lithostratic)
of Hunan Province. China University of Geosciences Press,
Wuhan, 292 (in Chinese).
Chu, X.-L., Zhang, Q.-R., Zhang, T.-G. & Feng, L.-J. 2003. Sulfur and
carbon isotopic variations in Neoproterozoic sedimentary rocks from
southern China. Progress in Natural Science, 13, 875–880.
Condon, D., Zhu, M.-Y., Bowring, S., Wang, W., Yang, A. & Jin, Y.
2005. U– Pb ages from the Neoproterozoic Doushantuo Formation,
China. Science, 308, 95– 98.
Dobrzinski, N. & Bahlburg, H. 2007. Sedimentology and environ-
mental significance of the Cryogenian successions of the Yangtze
Platform, South China block. Palaeogeography, Palaeoclimatology,
Palaeoecology, 254, 100–122.
Dobrzinski, N., Bahlburg, H., Strauss, H. & Zhang, Q.-R. 2004. Geo-
chemical climate proxies applied to the Neoproterozoic glacial suc-
cession on the Yangtze platform, south China. In: Jinkin, G. S.,
McMenamin, M. A. S., McKay, C. P. & Sohl, L. (eds) The
Extreme Proterozoic: Geology, Geochemistry and Climate. Ameri-
can Geophysical Union, Washington DC, Geophysical Monograph,
146, 13– 32.
Goddéris, Y., Donnadieu, Y. et al. 2007. Coupled modeling of global
carbon cycle and climate in the Neoproterozoic: links between
Rodinia breakup and major glaciations. Comptes Rendus Geoscience,
339, 212– 222.
Hao, J. & Zhai, M. 2004. Jinning movement and Sinian System in China:
their relationship with Rodinia Supercontinent. Chinese Journal of
Geology, 39, 139–152 (in Chinese with English abstract).
Jiang, G.-Q., Kennedy, M. J. & Christie-Blick, N. 2003a. Stable
isotopic evidence for methane seeps in Neoproterozoic postglacial
cap carbonates. Nature, 426, 18– 25.
Jiang, G. Q., Sohl, L. E. & Christie-Blick, N. 2003b. Neoproterozoic
stratigraphic comparison of the Lesser Himalaya (India) and
 THE YANGTZE REGION
Yangtze block (south China): Paleogeographic implications.
Geology, 31, 917–920.
Jiang, G. Q., Kennedy, M. J., Christie-Blick, N., Wu, H.-C. & Zhang,
S.-H. 2006. Stratigraphy, sedimentary structures, and textures of the
late Neoproterozoic Doushantuo cap carbonate in South China.
Journal of Sedimentary Research, 76, 978–995.
Jiang, G. Q., Kaufman, A. J., Christie-Blick, N., Zhang, S.-H. &
Wu, H.-C. 2007. Carbon isotope variability across the Ediacaran
Yangtze platform in South China: implications for a large
surface-to-deep ocean d13C gradient. Earth and Planetary Science
Letters, 261, 303–320.
Kirschvink, J. L. 1992. A paleogeographic model for Vendian and
Cambrian time. In: Schopf, J. W. & Klein, C. (eds) The Proterozoic
Biosphere: A Multidisciplinary Study. Cambridge University Press,
Cambridge, 569–581.
Knoll, A. H., Walter, M. R., Narbonne, G. M. & Christie-Blick, N.
2004. A new period for the geologic time scale. Science, 305, 621–622.
Li, X.-H. & McCulloch, M. T. 1996. Secular variation in the Nd isotopic
composition of Neoproterozoic sediments from the southern margin
of the Yangtze Block: evidence for a Proterozoic continental collision
in southeast China. Precambrian Research, 76, 67– 76.
Li, Z.-X., Zhang, L. & Powell, C. Mc. A. 1996. Positions of the east
Asian cratonic blocks in the Neoproterozoic supercontinent
Rodinia. Australian Journal of Earth Science, 43, 593–604.
Li, R.-W., Chen, J., Zhang, S., Lei, J., Shen, Y. & Chen, X. 1999.
Spatial and temporal variations in carbon and sulfur isotopic compo-
sitions of Sinian sedimentary rocks in the Yangtze platform, South
China. Precambrian Research, 97, 59– 75.
Li, Z.-X., Li, X.-H., Kinny, P. D., Wang, J., Zhang, S. & Zhou, H. 2003.
Geochronology of Neoproterozoic syn-rift magmatism in the Yangtze
Craton, South China and correlations with other continents: evidence
for a mantle superplume that broke up Rodinia. Precambrian
Research, 122, 85– 109.
Li, Z.-X., Bogdanova, S. V. et al. 2008. Assembly, configuration, and
break-up history of Rodinia: a synthesis. Precambrian Research,
160, 179– 210.
Liu, J.-S. 1994. Early Sinian sedimentary formations, turbidite and
manganese ore genesis in Hunan Province. Geotectonica et Metallo-
genia, 18, 173–182 (in Chinese).
Liu, H.-Y., Dong, R.-S. et al. 1991. The Sinian System of China. Scien-
tific Press, Beijing (in Chinese).
Liu, H.-Y., Hao, J. & Li, Y.-J. 1999. The Late Precambrian Stratigraphy
and Geological Evolution of Middle-Eastern China. Science Press,
Beijing.
Liu, T.-B., Maynard, J. B. & Alten, J. 2006. Superheavy S isotopes
from glacier-associated sediments of the Neoproterozoic of south
China: oceanic anoxia or sulfate limitation? In: Kesler, S. E. &
Ohmoto, H. (eds) Evolution of Early Earth’s Atmosphere, Hydro-
sphere – Constraints from Ore Deposits. Geological Society of
America Memoir, 198, 205–222.
Lu, S.-N., Ma, G.-G., Gao, Z. J. & Lin, W.-X. 1985. Primary research on
glacigenous rocks of Late Precambrian in China. In: PRECAM-
BRIAN GEOLOGY EDITORIAL COMMITTEE (ed.) Precambrian
Geology No. 1, The Collected Works of Late Precambrian Glacigen-
ous Rocks in China. Geological Publishing House, Beijing, 1 –86 (in
Chinese with English abstract).
Ma, G.-G. & Wang, Y.-G. 1983. Discussion on the glacial geology of the
early Sinian in the area adjacent to Hubei, Hunan, Sichuan and
Guizhou Provinces of China. Bulletin of Yichang Institute of
Geology and Mineral Resources, Chinese Academy of Geological
Sciences, 7, 43– 51 (in Chinese with English abstract).
Meng, F.-W., Yuan, X. L., Zhou, C.-M. & Chen, Z.-L. 2003. Dinoster-
ane from the Neoproterozoic Datangpo black shales and its biological
implications. Acta Micropalaeontologica Sinica, 20, 97 –102.
NCS (NATIONAL COMMITTEE ON STRATIGRAPHY) 2001. Strati-
graphic Guide in China and Explanation (revised). Geological Pub-
lishing House, Beijing, 59 (in Chinese).
Ogg, J. G. & Gradstein, F. M. 2008. The Concise Geologic Time Scale.
Cambridge University Press, Cambridge.
Qin, Y., Zhou, Q. & Zhang, S. 2005. Elementary properties of manga-
nese of Nanhuan in the Northeastern Guizhou. Guizhou Geology,
122, 246– 251.
365
Richthofen, F. von. 1882. China, 2. Ergebnisse Eigener Reisen und
Darauf Gegrundeter Studien, Berlin.
Remane, J. 2000. International Stratigraphic Chart, with Explanatory
Note. Sponsored by ICS, IUGS and UNESCO (distributed at the
31st International Geological Congress, Rio de Janeiro 2000), 16.
Roy, S. 2006. Sedimentary manganese metallogenesis in response to the
evolution of the Earth system. Earth-Science Reviews, 77, 273– 305.
Sha, Q.-A., Liu, H.-Y., Zhang, S.-S. & Chen, M.-E. 1963. Nantuo tillite
in the eastern Yangtze Gorges area. Scientia Geologica Sinica, 3,
139– 148 (in Chinese).
Tang, J.-F., Fu, H.-Q. & Yu, Z.-Q. 1987. The stratigraphic horizon, type,
and formation condition of Precambrian siliceous iron formation in
south China. Ore Geology, 6, 1 – 10 (in Chinese).
Tang, S.-Y. & Liu, T.-B. 1999. Origin of the early Sinian Minle manga-
nese deposit, Hunan Province, China. Ore Geology Reviews, 15,
71– 78 (in Chinese).
Wang, J. & Li, Z.-X. 2003. History of Neoproterozoic rift basins in South
China: implications for Rodinia break-up. Precambrian Research,
122, 141–158.
Wang, Y.-L., Lu, S.-N., Gao, Z.-J., Lin, W.-X. & Ma, G.-G. 1981. Sinian
tillites of China. In: Hambrey, M. J. & Harland, W. B. (eds)
Earth’s Pre-Pleistocene Glacial Record. Cambridge University
Press, 386–401.
Wang, J., Li, X.-H., Duan, T., Liu, D., Song, B., Li, Z. & Gao, Y. 2003a.
Zircon SHRIMP U– Pb dating for the Cangshuipu volcanic rocks and
its implications for the lower boundary age of the Nanhua strata in
South China. Chinese Science Bulletin, 48, 1663–1669.
Wang, T.-G., Wang, C.-J., Zhang, W.-B., Shi, Q., Zhu, L. & Chen,
J.-Y. 2003b. Initial organic geochemical investigation on Late
Neoproterozoic– Early Cambrian sediments in the Yangtze region,
China. Progress in Natural Science, 13, 936– 941.
Wang, J.-S., Gan, H.-Y., Wei, Q., Hu, G.-W. & Ge, Q. 2005. Stable
isotopes of carbon and sulfur of cap dolomite in the Three Gorges
and its mechanism discussion. Geoscience, 19, 14 –20 (in Chinese
with English abstract).
Wang, J.-S., Jiang, G.-Q., Xiao, S.-H., Li, Q. & Wei, Q. 2008a. Carbon
isotope evidence for widespread methane seeps in the c. 635 Ma
Doushantuo cap carbonate in south China. Geology, 36, 347– 350.
Wang, T.-G., Li, M.-J., Wang, C.-J., Wang, G.-L., Zhang, W.-B., Shi,
Q. & Zhu, L. 2008b. Organic molecular evidence in the Late
Neoproterozoic Tillites for a palaeo-oceanic environment during
the snowball Earth era in the Yangtze region, southern China.
Precambrian Research, 162, 317– 326.
Willis, B., Blackwelder, E. & Sargent, R. H. 1907. Research in
China. Vol. 1 part 1. The Carnegie Institution of Washington,
Washington, DC.
Xie, Q.-L., Chen, D.-F. & Chen, X.-P. 1999. Characteristics of sedimen-
tary organic matter in Songtao manganese deposits, Guizhou. Acta
Sedimentologica Sinica, 17, 280– 284.
Yang, Y.-J. 1985. The glacigenous strata of lower Sinian in Hunan
Province. In: PRECAMBRIAN GEOLOGY EDITORIAL COMMIT-
TEE (ed.) Precambrian Geology No. 1: The Collected Works of Late
Precambrian Glacigenous Rocks in China. Geological Publishing
House, Beijing, 224–241 (in Chinese with English abstract).
Yang, S.-X. & Pang, K.-T. 2006. Mineralization model for the manganese
deposits in northwestern Hunan – an example from Minle manganese
deposit in Huayuan, Hunan. Sedimentary Geology and Tethyan
Geology, 26, 72 –80.
Yang, J.-D., Xue, Y.-S., Sun, W.-G., Tao, X.-C., Wang, Z.-Z. & Zhou,
C.-M. 1997. The origin and age of the manganese ore of Nantuo stage.
Chinese Science Bulletin, 42, 1538– 1541.
Yang, R.-D., Ouyang, Z.-Y., Zhu, L.-J., Wang, S.-J., Jiang, L.-J.,
Zhang, W.-H. & Gao, H. 2002. A new understanding of manganese
carbonate deposits in early Sinian Datangpo stage. Acta Mineralogica
Sinica, 22, 329– 336.
Yin, C.-Y., Liu, D.-Y., Gao, L.-Z., Wang, Z.-Q., Xing, Y.-S., Jian, P. &
Shi, Y.-R. 2003. Lower boundary age of the Nanhua System and the
Gucheng glacial stage: evidence from SHRIMP dating. Chinese
Science Bulletin, 48, 1657– 1662.
Zhang, Q.-R. 1995. The origin of the Sinian Nantuo Formation in
Yichang County, Hubei Province. Scientia Geologica Sinica, 30,
147– 152 (in Chinese with English abstract).
366 Q. R. ZHANG ET AL.
Zhang, Q. R. & Piper, J. D. A. 1997. Palaeomagnetic study of
Neoproterozoic glacial rocks of the Yangzi Block: palaeolatitude
and configuration of South China in the late Proterozoic Superconti-
nent. Precambrian Research, 85, 173– 199.
Zhang, Q.-R., Chu, X.-L., Bahlburg, H., Feng, L.-J., Dobrzinski, N.
& Zhang, T.-G. 2003. Stratigraphic architecture of the Neoproter-
ozoic glacial rocks in the ‘Xiang-Qian-Gui’ region of the central
Yangtze Block, South China. Progress in Natural Science, 13,
783– 787.
Zhang, Q.-R., Li, X.-H., Feng, L.-J., Huang, J. & Song, B. 2008a. A
new age constraint on the onset of the Neoproterozoic glaciations
in the Yangtze Platform, South China. Journal of Geology, 116,
423– 429.
Zhang, S.-H., Jiang, G.-Q. & Han, Y.-G. 2008b. The age of the Nantuo
Formation and Nantuo glaciation. Terra Nova, 20, 289– 294.
Zheng, Y.-F. 2003. Neoproterozoic magmatic activities and global
evolution. Kexue Tongbao, 48, 1705–1720 (in Chinese).
Zhou, C.-M. & Xiao, S.-H. 2007. Ediacaran d13C chemostratigraphy of
South China. Chemical Geology, 237, 89– 108.
Zhou, C.-M., Tucker, R., Xiao, S.-H., Peng, Z.-X., Yuan, X.-L. &
Chen, Z. 2004. New constraints on the ages of Neoproterozoic
glaciations in south China. Geology, 32, 437– 440.
Zhou, J.-B., Li, X.-H., Ge, W.-C. & Li, Z.-X. 2007a. Age and origin of
middle Neoproterozoic mafic magmatism in southern Yangtze
Block and relevance to the break-up of Rodinia. Gondwana Research,
12, 184– 197.
Zhou, Q., Du, Y.-S., Yan, J.-X., Zhang, M.-Q. & Yin, S.-L. 2007b.
Geological and geochemical characteristics of the cold seep
carbonates in the Early Nanhua System in Datangpo, Songtao,
Guizhou Province. Earth Science – Journal of China University of
Geosciences, 32, 845–852 (in Chinese with English abstract).
Zhu, M.-Y., Zhang, J.-M. & Yang, A.-H. 2007. Integrated Ediacaran
(Sinian) chronostratigraphy of South China. Palaeogeography,
Palaeoclimatology, Palaeoecology, 254, 7 – 61.
 Chapter 33

Neoproterozoic glaciogenic diamictites of the Tarim Block, NW China

MAOYAN ZHU* & HAIFENG WANG

State Key Laboratory of Palaeobiology and Stratigraphy, Nanjing Institute of Geology and Palaeontology,
Chinese Academy of Sciences, Nanjing, 210008, China
*Corresponding author (e-mail: myzhu@nigpas.ac.cn)

Abstract: Neoproterozoic glaciogenic outcrops are mostly limited to the southwestern and northwestern margins of the Tarim basin and
the Tianshan Mountains. Only two glaciogenic diamictites are recorded in the Neoproterozoic successions from the Tielikeli and Aksu-
Wusi areas of the Tarim basin. The lower diamictite from the Polong Formation in the Tielikeli area is c. 800 m thick. The upper dia-
mictite in the area is very thin (Yutang diamictite, 10 m; Yulmeinak diamictite, 51 m). In contrast to the Tarim basin, the Neoproterozoic
successions along the northern margin of the Tarim Block in the Tianshan Mountains record three or four glaciogenic diamictites, which
are well developed and crop out in the Qurugtagh area. Although the glaciogenic nature of the oldest Beiyixi diamictite remains ques-
tionable, all other diamictites (Altungol, Tereeken and Hankalchough) exhibit distinct glaciogenic sedimentary features. SHRIMP zircon
U –Pb dating demonstrate that the Bayisi diamictite was deposited between 740+7 Ma and 725+10 Ma, the Altungol and Tereeken
diamictites between 725+10 Ma and 615+6 Ma, respectively, and the Hankalchough diamictite between 615+6 Ma and c. 542 Ma.
Carbonates atop the Altungol, Tereeken and Hankalchough diamictites show distinct C-isotopic values that are typical for those recorded
in ‘cap carbonates’ on other continents. The striking similarity between the cap carbonate of the Tereeken and Marinoan-age Nantuo
diamictites further support a 635 Ma ‘Marinoan’ age for the Tereeken glaciation. Unlike the 582 Ma Gaskiers glaciations of Newfound-
land, the Hankalchough diamictite is possibly ,551 Ma, as suggested by C-isotope chemostratigraphy and biostratigraphy, indicating
that post-Marinoan glaciations on different continents may be diachronous.

Supplementary material: Data are available at: http://www.geolsoc.org.uk/SUP18469.

Neoproterozoic glaciogenic diamictites on the Tarim Block were
first recognized by a Swedish geologist, E. Norin, following a
joint Chinese –Swedish expedition 1927–1935 in the Qurugtagh
area (Norin 1937). Although several detailed studies have been
carried out since and three Neoproterozoic glaciogenic diamictites
had already been reported in the early 1980s (see Gao et al. 1985a
and references therein), these diamictites only recently received
wide attention (Xiao et al. 2004). Various research groups have
now investigated the Neoproterozoic successions of the Tarim
basin margin, particularly in the Qurugtagh region, focusing on
the study of Neoproterozoic tectonics, magmatism, lithologies,
geochronology, chemostratigraphy and palaeontology.
The present chapter aims to review both old and recent studies,
focusing on the stratigraphy of the Neoproterozoic diamictites in
four palaeogeographical areas located on or close to the Tarim
Block (Fig. 33.1). Using detailed descriptions and recent geochro-
nological and chemostratigraphic data, this chapter discusses cor-
relations and ages of the Neoproterozoic diamictites of the Tarim
Block. Published palaeomagnetic data and their palaeogeographic
implications are also reviewed.
Glaciogenic diamictites in the Tielikeli area,
SW Tarim Basin
The Tielikeli area is located at the southwestern margin of the
Tarim basin, where Neoproterozoic outcrops are well exposed
along the northern fringes of the western Kunlun Mountains
(Fig. 33.1(1) & Fig. 33.2) (Ma et al. 1991; Wang et al. 2004).
The Neoproterozoic strata and tillites of the area, which had orig-
inally been mapped as part of the Devonian Qizlafu Group, were
first recognized by the local geologist Ma and his colleagues in
1978 and described in detail a decade later (Ma et al. 1989).
Stratigraphy
The Cryogenian and Ediacaran successions in the area (Fig. 33.3)
start with c. 230 m of conglomerate (Yalaguz Fm.), which
unconformably overly the Sukulok Group. Interbedded within
the massive conglomerate unit are lenticular or bedded feldspathic
sandstones and thinly bedded silty mudstone. The conglomerate is
dominated by sub-rounded or rounded, well-sorted clasts of chert
and porphyry.
The Yalaguz Fm. is overlain by the Qakmaklik Group which
consists in ascending order of the Polong, Kilix and Yutang for-
mations. The Polong Fm. is dominated by purplish-brown
massive diamictite of c. 880 m thick. The diamictite is underlain
by c. 200 m of grey laminated cherts and silicified shale. The
Kilix Fm. consists of interbeds of purplish-green laminated mud-
stone and siltstone with thin beds of silty micritic limestone (c.
312 m) in the lower part, purple quartz sandstone (42.9 m) in the
middle, and purple interbeds of feldspathic sandstone and con-
glomerate (86.8 m) at the top. The feldspathic sandstone exhibits
large-scale cross-stratification and horizontal stratification. The
Yutang Fm. is subdivided into three parts. The lower part (c.
50 m) consists of grey-greenish laminated muddy cherts and mud-
stone with a conglomerate 0.1–0.5 m thick at the base. The middle
part (c. 10 m) consists of purplish brown diamictite. The upper part
(121.9 m) of the Yutang Fm. consists of reddish and brownish
feldspathic sandstone and conglomerate showing large-scale
cross-stratification and wave ripples.
Overlying the Qakmaklik Group is the Kurkak Fm., which is
composed of four parts. A 10-m-thick laminated dolomite marks
the contact between the Qakmaklik Group and the Kurkak Fm.
The lower part of the Kurkak Fm. is c. 110 m in thickness and
consists of black shale. This is succeeded by 100 m of grey
quartz sandstone and conglomerate in the middle, and dark grey
shale with interbeds of siltstone and rare phosphorite nodules/
concretions or lamina in the upper part of this unit.
The Kzisuhum Fm. is composed of four parts in ascending
order. The lower interval is 55.6 m thick and consists of dark
green quartz sandstone with thin interbeds of feldspathic sandstone
and pyrite-bearing siltstone in the lower part, and dark grey silt-
stone, greyish-green sandstone and grey coarse-grain quartz sand-
stone with three phosphorite beds in the upper part. The overlying
interval comprises 86.1 m of yellowish brown sandy dolostone
with interbeds of purple siltstone and minor sandy siderite

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 367– 378. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.33
368 M. ZHU & H. WANG

Fig. 33.1. Geological map of the Tarim Block and adjacent areas (modified from Lu et al. 2008). Key: PH, Phanerozoic rocks; NP, Neoproterozoic rocks; MP,
Mesoproterozoic rocks; PP, Palaeoproterozoic rocks; AR, Archaean rocks; RNP, early Neoproterozoic granitoids; F, Faults; IF, Inferred faults; Diamictites, Cryogenian
and Ediacaran diamictites; Q, Quaternary desert and sedimentary deposits; Square: 1, Tielikeli area, SW Tarim Basin; 2, Aksu-Wusi area, NW Tarim Basin; 3, Quruqtagh
area, southern margin of Eastern Tianshan Mts.; 4, Guozigou-Keguqing Mt. area, western range of Northern Tianshan Mountains.

layers. Approximately 52.2 m quartz sandstone with large-scale
planar cross-stratifications overlies the dolostone interval. The
upper part ofthe Kzisuhum Fm. is dominated by dolostone
(177.4 m). The contact to the overlying Devonian strata is sharp
and unconformable.
Description of the diamictites
There are two diamictite intervals in the Tielikeli area. Both are
interpreted as glaciogenic diamictites or tillites. The lower diamic-
tite interval of the Polong Fm. is dominated by massive diamictite
with 5 –30% randomly organized, angular, sub-angular or
sub-rounded clasts of porphyry, quartz porphyry, chert, quartzite,
dolostone, granite, gneisses, diabase and so on. Clasts are domi-
nated by pebbles of 2 –10 cm in diameter, with a maximum diam-
eter of 100 cm. Glaciogenic features include striated marks, snicks
and notches on the surface of cobble and pebble size clasts, and
parallel cracks of clasts. The diamictite at the base and top of the
interval is characterized by 30–50% poorly sorted clasts with
traces of bedding. Laminated mudstone and siltstone with drop-
stones are observed in the middle and top of the interval.
The upper diamictite interval of the Yutang Fm. is only 10 m
thick, showing similar sedimentary features and a similar clast
composition as the lower diamictite. Similar dropstone-bearing
laminated mudstone also occurs at the top of the upper diamictite.
It is interesting to note that both diamictite intervals are
underlain by an interval consisting of grey laminated muddy

Fig. 33.2. Geological map showing

outcrop of the Tielikeli area, SW Tarim
Basin (modified from Ma et al. 1989).
 THE TARIM BLOCK, NW CHINA
Fig. 33.3. Composite stratigraphic log of the Cryogenian and Ediacaran in the
Tielikeli area, SW Tarim Basin (modified from Ma et al. 1989). Key as in
Figure 33.10.
cherts and silicified shale, demonstrating a similar depositional
sequence.
Glaciogenic diamictites in the Aksu-Wusi area,
NW Tarim Basin
Neoproterozoic outcrops exposed in the Aksu-Wusi area, NW
Tarim basin, are easily accessible (Figs 33.1, 33.2 & 33.4). Glacio-
genic diamictite in the area was first documented in local geologi-
cal reports during the 1950s and has been well studied by Gao &
Qian (1985) and Gao & Chen (1993).
369
Stratigraphy
The Neoproterozoic succession in the area (Fig. 33.5) consists of
the Qiaoenblaq and Wushinanshan groups; the former overlies
the metamorphic rocks of the Aksu Group unconformably. The
Aksu Group is older than 962 + 12 Ma (whole-rock Rb –Sr
dating) and is dominated by blueschist (Liou et al. 1989;
Gao et al. 1993). The Qiaoenblaq Group, overlying the Aksu
blueschists, consists of (in ascending order) the Xifangshan, Dong-
qiaoenblaq, Muyangtan and Dongwu formations. The Xifangshan
Fm. is c. 1715 m thick and is characterized by rhythmic beds of
grey-green feldspathic sandstone, feldspathic quartz sandstone
and siltstone. The rhythmites exhibit sandstone with grading fea-
tures in the lower part, and siltstone with horizontal and cross-
stratification in the upper part. The rhythmic beds range from
1 cm to 4 m in thickness, but are predominantly 50– 100 cm.
They have been interpreted as gravity flow deposits or proximal
turbidites (Gao et al. 1993).
The Dongqiaoenblaq Fm. consists of 15 –20-m-thick grey-green
conglomerate with interbeds of sandstone in the lower and massive
diamictite with a maximum thickness of 70 m in the upper part.
Overlying the diamictite is the Muyangtan Fm., which is
c. 107 m thick and is composed of grey-green calcareous siltstone
and quartz sandstone with parallel and cross-stratification. The
overlying Dongwu Fm., c. 176 m thick, is composed of grey-green
and grey-brownish coarse-grained feldspathic sandstone. There is
a 1– 2-m-thick conglomerate bed at the base of the Dongwu Fm.
The conglomerate is dominated by sub-rounded or rounded
cobbles and pebbles of granite.
The Wushinanshan Group consists of the Yulmeinak, Sugetblaq
and Qigeblaq formations. The Yulmeinak Fm. overlies sandstone
of the Dongwu Fm. above an angular unconformable contact and
is composed of purplish massive diamictite with a maximum thick-
ness of 70 m, but the diamictite gradually decreases and eventually
pinches out towards the east. The overlying Sugetblaq Fm. consists
of two members. The lower member is c. 728 m thick is and com-
posed of brownish coarse-grained feldspathic sandstone with inter-
beds of yellow-green siltstone and sandstone in the lower part, and
red, grey and brownish ferrous quartz sandstone with thin interbeds
of hematite and a 15-m-thick unit of peridotite basalt in the upper
part. The uppermost part of the lower member of the Sugetblaq
Fm. is characterized by 79 m of dark green and brownish amygda-
loidal olivine basalt. The upper member of the Sugetblaq Fm. is
composed of brownish and greyish thin-bedded sandy limestone
with edgewise clasts and interbedded thin siltstone in the lower
part and yellow to green, thin-bedded, glauconitic feldspathic
sandstone in the upper part. The overlying Qigeblaq Fm. is a
characteristic interval of thick-bedded dolostone with a thickness
of 186 m. This unit is overlain by early Cambrian cherts and phos-
phorite, which contain small shelly fossils (Gao et al. 1981; Qian &
Xiao 1984; Yao et al. 2005).
Description of the diamictites
As in the Tielikeli area, there are two diamictite intervals inter-
preted as glaciogenic in the Aksu-Wushi area. The lower diamic-
tite interval of the Dongqiaoenblaq Fm. is composed of 20–80%
randomly organized sub-rounded clasts of felsite, dacite, granite,
jasper, quartzite, sandstone, andesite, diabase and so on. Clasts
are dominated by c. 10-cm-sized pebbles with a maximum diam-
eter of 60 cm. The matrix is composed of feldspathic sandstone
and siltstone with a mud component, showing soft deformation
features and contorted bedding in some beds. Although clasts are
randomly organized, and striated marks have been observed on
the surface of some clasts from the diamictite of the Dongqiaoen-
blaq Fm., the high degree of sorting, contorted bedding and sub-
rounded shapes of the clasts, which are similar in lithology to
the clasts from the underlying conglomerate, led Gao et al.
370 M. ZHU & H. WANG
(1993) to reinterpret the diamictite as having been rapidly depos-
ited from submarine debris flows.
The upper diamictite interval of the Yulmeinak Fm. is a discon-
tinuous interval with an unconformable basal surface. The diamic-
tite shows distinctly glaciogenic characteristics, including a large
abraded surface at its base with notching, scouring and striations,
which is a clear indication of ice movement or glacial pavement.
The diamictite is massive and matrix-supported with 20– 30%
clasts of various lithologies including granites and volcanic rocks,
blueschists, sandstone and siltstone originally derived from the
underlying strata. The clasts vary in size, with the largest ones at
the base (maximum diameter of 6 m), and show a poor degree of
sorting. Dominant clasts are subangular with a muddy or calcar-
eous envelope, but characteristic selliform (saddle-shaped) and
other irregularly shaped clasts are also observed. Striation marks,
nicks and notches are common on the surface of clasts. Dropstones
of various lithologies occur in laminated siltstone in the middle
of the diamictite interval.
Glaciogenic diamictites in the Quruqtagh area, southern
margin of the eastern Tianshan Mountains
The Quruqtagh area is located at the northeastern margin of the
Tarim basin and belongs to the southern margin of the Eastern
Tianshan, covering an area of about 30 000 km2 (Figs 33.1–
33.3). The Quruqtagh area is customarily subdivided into three dis-
tricts. The area west of Xindi is called the East Quruqtagh, the area
between Xindi and Xinger is the Middle Quruqtagh, while the area
east of Xinger is the West Quruqtagh. The Neoproterozoic out-
crops and glaciogenic diamictites are well exposed in all three dis-
tricts (Fig. 33.6) and were first reported by a Swedish geologist,
E. Norin, after a joint 1927 –1935 Chinese – Swedish expedition
in the area (Norin 1937). The Neoproterozoic stratigraphy of the
area has been well established by local geological surveys during
the 1950s and has been documented in detail by Gao & Zhu
(1984). Since then, the latest Neoproterozoic successions have
been investigated using various methods (e.g. Zhong & Hao
1990; Xiao et al. 2004; Duan et al. 2005; Huang et al. 2005; He
et al. 2007a; Liu et al. 2007; Zhang et al. 2007; Shen et al.
2008; Kou et al. 2008; Zhu et al. 2008; Xu et al. 2009).
Stratigraphy
The latest Neoproterozoic successions in the Quruqtagh area
(Fig. 33.7) are known as the Quruqtagh Group, and overlie a
Fig. 33.4. Geological map showing the
Cryogenian and Ediacaran outcrop of the
Aksu-Wusi area, NW Tarim Basin
(modified from Gao et al. 1985b).
c. 800-m-thick carbonate unit, the Paergangtagh Fm., with an
unconformable contact (Zhao et al. 1985; Gao et al. 1993). The
Quruqtagh Group is subdivided (in ascending order) into nine
formations: the Bayixi, Zhaobishan, Altungol, Huangyanggou,
Tereeken, Zhamoketi, Yukkengol, Shuiquan and Hankalchough
formations (Gao & Zhu 1984; Xiao et al. 2004; Kou et al. 2008).
The best continuous outcrops are located in the Middle
Quruqtagh area.
The Bayixi Fm. This consists predominantly of shallow-water
marine siliciclastic and volcanic rocks with several intervals of
diamictite. The overall thickness of the formation ranges from
640 to 1670 m. It starts with a basal conglomeratic interval with
predominant clasts of dolostone and limestone from the underlying
Paergangtagh Fm. The clasts are angular to subangular in shape,
with the largest clasts being 30 cm in diameter. The lower part
of the Bayixi Fm. consists of volcanic rocks, slates and sandstones
associated with diamictite. The middle part of the formation con-
sists of sandstone, and siltstone interbedded with diamictite and
slates. The upper part of the formation consists of two diamictite
intervals of c. 200 m thickness each, separated by interbedded
slates and siltstones.
The Zhaobishan Fm. This is a characteristic interval with thick-
nesses ranging from 358 to 570 m, consisting predominantly of
sandstones and calcareous siltstones in the lower part and slates
with interbedded thin calcareous siltstones and sandstones in the
upper part. The sandstones in the lower part of the formation are
lithic, coarser-grained (0.2 –2 mm), better rounded and sorted
and show large cross-bedding, whereas sandstones in the upper
part are more fine-grained (0.1 –1 mm) and show a lower degree
of textural maturity. The overall sequence of the Zhaobishan
Fm. represents a deepening-upward sequence.
The Altungol Fm. This is characterized by three massive diamictite
intervals associated with sandstones, siltstones and shales. The
thickness of the formation increases from East Quruqtagh
(764 m) to West Quruqtagh (2395 m). In the South Quruqtagh
area, the formation is only 8 m thick. The basal and upper
massive diamictites mark the basal and top boundaries of the
Altungol Fm. The thickness of the three diamictite intervals
ranges from 50 to 100 m. Associated sandstones, siltstones and
shales exhibit rhythmic beddings with average bed thicknesses
of 40–60 cm.
 THE TARIM BLOCK, NW CHINA
Fig. 33.5. Composite stratigraphic log of the Cryogenian and Ediacaran in the
Aksu-Wusi area, NW Tarim Basin (after Gao et al. 2005b). Key as in
Figure 33.10.
371
The Huangyanggou Fm. This is a newly erected formation repre-
senting an interval between the uppermost diamictite of the Altun-
gol Fm. and the diamictite of the Tereeken Fm. (Cao 1991; Kou
et al. 2008). The interval without diamictite was originally the
upper part of the Altungol Fm., and was interpreted to represent
an interglacial interval between the Altungol and Tereeken glacia-
tions (Kou et al. 2008). The base of the Huangyanggou Fm. is
marked by c. 2 m of grey-purplish massive micritic dolostone,
which represents a cap carbonate above the Altungol diamictite.
The lower part of the formation consists of meta-andesite, clinker-
ing breccia and silty tuff, while the upper part consists of silty mud-
stone and muddy siltstones with thin interbeds or lenses of
limestone. A 1–3-m-thick limestone unit marks the top boundary
of the Huangyanggou Fm., underlying the base of the Tereeken
Fm. A c. 50-m-thick dolostone interval reported from the Altungol
Fm. by Xiao et al. (2004) at the Xishankou section may correspond
to the Huangyanggou Fm.
The Tereeken Fm. This is characterized by grey massive diamictites
with thicknesses ranging from 689 to 1845 m. The diamictite is
separated into several units by thin interbeds of finely laminated
silty mudstones and carbonate.
The Zhamoketi Fm. This is a siliciclastic interval consisting of
metre-scale rhythmites of sandstone, siltstone and mudstone,
with thicknesses ranging from 297 m to 793 m. The base of the for-
mation is marked by a 2 –10-m-thick unit of dolostone with char-
acteristics of basal Ediacaran cap carbonates, overlying the
Tereeken diamictites. The top of this formation is marked by a c.
20– 80-m-thick diabase sill. The rhythmites have been interpreted
as turbidites, showing distinct structures of the Bouma sequences
deposited on a continental slope (Li & Dong 1991; Kou et al.
2008). The rhythmites exhibit grading at their base with overlying
laminated and wavy or cross-bedded intervals; convoluted lami-
nations or soft-deformation have been observed within some
beds. Sole marks such as flute casts, groove casts and load casts
are common at the base of the rhythmites.
The Yukkengol Fm. This consists of green-greyish finely laminated
siltstones and shales with thicknesses ranging from 80 to 583 m,
representing condensed sequences of deeper- water facies.
The Shuiquan Fm. This is characterized by laminated carbonates
and silty shales. Its variable thicknesses between 22 and 307 m
was interpreted to result from erosion at the top of the formation
(Gao & Zhu 1984). Centimetre-scale ribbonites, stromatolites
and microbially laminated dolostone were reported in the Shui-
quan Fm., suggesting sedimentation in the euphotic zone (Xiao
et al. 2004). The wormiform carbonaceous fossils – filaments c.
2 mm wide and up to 40 mm long – from the Shuiquan Fm.
reported by Gao et al. (1985) were reinterpreted as Vendotaenid
fossils (Xiao et al. 2004).
The Hankalchough Fm. This consists of light grey diamictite. Its
thickness decreases eastward from 467 m to less than 10 m. A
dolostone unit (1 –6 m thick) at the base of the Xishanblaq Fm.
marks the top of the Hankalchough Fm.
The Xishanblaq Fm. consists of black cherts and cherty phos-
phorites associated with volcanic rocks. Cambrian microfossils
have been reported from the phosphorites (Yao et al. 2005).
Description of the diamictites
The four major intervals of diamictite in the Quruqtagh Group
have been assigned to three glaciations (Gao & Zhu 1984).
Below is a summary of these four diamictites based on detailed
descriptions by Gao & Zhu (1984), Xiao et al. (2004) and Kou
et al. (2008).
372 M. ZHU & H. WANG
The Bayixi diamictite. The Bayixi diamictites are distributed across
only a limited area largely near the Middle Quruqtagh. The thick-
ness of the diamictite units in the Bayixi Fm. varies from a few
metres to more than 100 m. Poorly sorted and angular clasts (5 –
20%) occur in the silty and muddy matrix of the diamictites (see
Supplementary Material, Fig. 1S-a). Clasts are mostly derived
from older metamorphic and igneous rocks, including gneiss,
granite, diabase, dolostones or marbles, and quartz; siliciclastic
clasts are rare. Clast sizes are predominantly ,6 cm in diameter.
Some boulders are perpendicularly or obliquely oriented relative
to bedding plane. The Bayixi diamictites were interpreted as gla-
ciomarine in origin by Gao & Zhu (1984), supported by the vari-
ety of lithic compositions, the angular and irregular shapes of the
clasts, scratch marks and non-unidirectional striations on the
clast surfaces. However, glacial striations were not confirmed by
Xiao et al. (2004). In addition, no cap carbonate has been reported.
The Altungol diamictites. Based on Kou et al. (2008), there are three
diamictite units within the newly defined Altungol Fm. However,
as documented by Gao & Zhu (1984), the Altungol diamictite units
appear unevenly distributed because only one unit has been
observed in some sections. The Altungol diamictites are massive,
have a matrix of low maturity and include clasts of quartzite,
granite, cherts and quartz. The clasts are generally .1 cm with
the largest being c. 15 cm in diameter (see Supplementary
Material, Fig. 1S-b). Clasts with stepped fractures and glaciogenic
striations have been observed (Kou et al. 2008). Gao & Zhu (1984)
claimed that the Altungol diamictites have a closer affinity with the
overlying Tereeken diamictites and argued that the Altungol and
Tereeken diamictites jointly represent a mid-Quruqtagh ice age.
However, Xiao et al. (2004) suggest the alternative interpretation
that the Altungol diamictites resemble more closely the Bayixi dia-
mictites in clast composition, degree of metamorphism and thick-
ness of individual diamictite beds. Xiao et al. (2004) argued that if
all the diamictites in the Bayixi and Altungol formations were
found to be glaciogenic, it would be more likely that the Altungol
and Bayixi diamictites represent pulses of a single ice age. Xiao
et al.’s (2004) hypothesis is further supported by the observations
of Kou et al. (2008) who confirmed the existence of an interglacial
interval, the Huangyanggou Fm., separating the Altungol diamic-
tites from the Tereeken ice age. A 2 –4-m-thick, grey-purplish car-
bonate unit with negative values of d13C at the base of the
Huangyanggou Fm. was considered to be a cap carbonate above
the Altungol diamictites (Kou et al. 2008). However, in opposition
to Xiao et al.’s (2004) hypothesis, Kou et al. (2008) assigned the
Altungol diamictites to a distinctly separate ice age.
Fig. 33.6. Geological map showing the
Cryogenian and Ediacaran outcrop of the
Quruqtagh area, southern margin of the
Eastern Tianshan Mountains (modified
from Gao & Zhu 1984).
The Tereeken diamictite. The Tereeken diamictite is the most con-
tinuous and distinct interval in the Quruqtagh area, showing unam-
biguous glaciogenic features, striated clasts and dropstones
throughout. Clasts in the diamictites are dominated by carbonate,
granite and other igneous rocks; marble and siliciclastic clasts
are more abundant in comparison to the Bayixi and Altungol dia-
mictites (see Supplementary Material, Fig. 1S-c). Striated clasts
are common (Fig. 1S-d). Three to eight individual diamictite
units can be recognized within the Tereeken Fm. (Gao & Zhu
1984). These units are intercalated with non-glaciogenic, lami-
nated silty marine mudstones and carbonates. Several thin carbon-
ate laminae (,10 mm thick and laterally continuous for more than
100 m) occur within the silty mudstone in the lower Tereeken Fm.
at the Heishan –Zhaobishan section (Xiao et al. 2004). These
laminae consist of vertically oriented, upward-growing calcite
crystals, and are typically inundated by overlying siltstone. In
addition, a bedded carbonate unit occurs in the lower Tereeken
Fm. between two massive diamictite units, at the same section.
As suggested by Xiao et al. (2004), the existence of laminated silt-
stones and bedded dolostones between Tereeken diamictite units
indicates that there was an active hydrological system during the
Tereeken ice age.
A 2–10-m-thick carbonate unit above the Tereeken diamictites
represents a distinct lithostratigraphic marker, and was recognized
as the ‘Yukkengol limestone’, defining the boundary between the
Tereeken diamictite and overlying strata (Norin 1937). The same
boundary was subsequently adapted to define the boundary
between the Lower and Upper Sinian Systems (Gao & Zhu
1984). The carbonate unit in the Yukkengol area has been
described in detail as a cap carbonate atop the Tereeken diamictite
by Xiao et al. (2004). According to observations from a section
between Xidashan and Moheshan, similar to the Yukkengol area,
the carbonate unit consists of three subunits (see Supplementary
Material, Fig. 1S-e, f ). The lower subunit (c. 1.5 m) is a massive
dolostone with angular allochthonous clasts of various compo-
sitions. The middle subunit (c. 2.5 m) consists of thinly bedded,
finely laminated dolostone with thin interbeds of shales. The
upper subunit (c. 1 m thick) consists of medium-bedded dolostone.
It is interesting to note that the uppermost part of the Tereeken
diamictite underlying the cap carbonate shows well-defined
bedding and represents deposits of a deglaciation interval.
The Hankalchough diamictite. The Hankalchough diamictite is
composed of light grey and greenish-grey massive diamictite
(see Supplementary Material, Fig. 2S-a, b, c). The clast compo-
sition differs from that of the underlying three diamictite intervals,
 THE TARIM BLOCK, NW CHINA 373

consisting of angular clasts dominated by carbonate and granite.

Siliciclastic clasts from underlying strata are abundant. Some car-
bonate boulders contain overturned stromatolite, probably derived
from older stromatolitic dolostones in the Paergangtagh Group,
which underlies the Quruqtagh Group (Zhao et al. 1985). The Han-
kalchough diamictite was originally considered to be terrestrial in
origin (Gao & Zhu 1984), but has been reinterpreted as glaciomar-
ine by Xiao et al. (2004) due to the occurrence of dropstones.
A distinct 1– 5-m-thick grey-greenish or light grey, finely lami-
nated muddy-silty dolostone or calcareous mudstone unit above
the Hankalchough diamictite (see Supplementary Material,
Fig. 2S-d) contains common lithic fragments and small clasts
(3 –5%) with disseminated pyrite granules (5 –10 mm). It was orig-
inally interpreted as glaciolacustrine in origin (Gao & Zhu 1984)
and reinterpreted as an atypical type of cap dolostone (Xiao
et al. 2004).

Glaciogenic diamictites in the Guozigou-Keguqingshan

area, western range of North Tianshan Mountains

The Guozigou-Keguqingshan area belongs to the western range of

the North Tianshan Mts. (Figs 33.1– 33.4). The well-exposed late
Precambrian successions in the area (Fig. 33.8) is composed of
three intervals of diamictite, which were first recognized by local
geologists in 1976 and subsequently documented in detail by
Wang et al. (1983) and Gao et al. (1985b) based on fieldwork in
1978 –1979. Since then, no more new information has been
published.

Stratigraphy

The Cryogenian and Ediacaran Kailaketik Group in the area

(Fig. 33.9) is c. 700–1000 m thick and is subdivided into six
formations (in ascending order): the Kulutieliekti, Tulasu,
Biexibastao, Keyindi, Tarqat and Talisayi formations. The
outcrop quality and thicknesses of the succession in the area
vary from west to east. The lower part of the succession is well
exposed in the Kuguqingshan area and the upper part of the succes-
sion is well developed in the Guozigou area. The succession is
overlain disconformably by the Lingkuanggou Fm., which con-
tains early Cambrian phosphatized small shelly fossils, and is
underlain by the carbonate of the Kusongmqiek Group with
unconformable contact.

The Kulutieliekti Fm. This is c. 115 m thick. A basal conglomerate

interval of 7.2 m is composed predominantly of limestone and
dolostone clasts derived from the underlying Kusongmqiek
Group. The well-sorted clasts have rounded or sub-rounded shapes
with diameters ranging from 1 to 25 cm. The basal conglomerate is
overlain by a 9-m-thick greenish tuff bed. The tuffaceous bed con-
tains c. 30% volcanic clasts (0.03 –0.3 cm in diameter) and carbon-
ate clasts (0.5 –1.5 cm in diameter). The subsequent strata are
composed of dark-greyish, massive amygdaloidal basalt with a
thickness of c. 90 m; the basalt occasionally shows subangular
carbonate and volcanic clasts (2 –5 cm in diameter). Overlying
the basalt is a 9.4-m-thick greyish, massive diamictite.

The Tulasu Fm. This is composed of laminated, black, carbon-

aceous, silicified mudstone with intercalations of black siltstone
with a thickness of 105 m. The siltstone layers increase up-section
and form rhythmic thin mudstone –siltstone couplets. Organic
microfossils are reported from this interval.

Fig. 33.7. Composite stratigraphic log of the Cryogenian and Ediacaran in The Biexibastao Fm. This consists of a lower member (25 m) of silt-
the Quruqtagh area, southern margin of the Eastern Tianshan Mountains stone and sandstone and an upper member of diamictite (30 m).
(after Gao & Zhu 1984; He et al. 2007; Kou et al. 2008). Legends as in Limestone beds occur within the well-bedded siltstone and
Figure 33.10. sandstone interval.
374 M. ZHU & H. WANG
The Keyindi Fm. This is a thin interval in the Kuguqing Mountains
area and becomes thicker in the Guozigou area, where it is more
than 115 m thick, but the basal part is unexposed. The formation
is composed of laminated, dark-greyish muddy siltstone and
black silty mudstone.
The Tarqat Fm. This is also composed predominantly of laminated,
muddy siltstone but with grey-greenish or purplish colour and
cross-stratification in some layers. Thin lenticular limestone beds
(0.7– 3 cm) occur in some sections.
The Talisayi Fm. This is a distinct interval consisting of greyish or
purplish, massive diamictite with a thickness ranging from 79 to
424 m. Several dark-greyish, laminated siltstone layers are recog-
nized within the diamictite. The diamictite is overlain by Cambrian
phosphorite of the Linkuanggou Fm., which contains early small
shelly fossils and trilobites. A conglomerate bed at the base of
the Linkuanggou Fm. and an irregular contact marks the unconfor-
mity between the Talisayi and the Linkuanggou formations.
Description of the diamictites
The three diamictite intervals in the area were first interpreted by
Wang et al. (1983) as representing three glaciations. These inter-
vals were used for correlation with the Quruqtagh area.
The Kulutieliekti diamictite. This diamictite consists of 40– 50%
clasts of dolostone, limestone, marbles, cherts and basalt. The
diameter of these clasts varies from 1 to 25 cm, and the largest
diameter reaches 35 cm. However, the diamictite lacks convincing
glaciogenic features, and its glaciogenic provenance remains
uncertain.
The Biexibastao diamictite. This is not a continuous interval but
often changes laterally to silty dolostone. The diamictite shows
distinct glaciogenic features, such as striated clasts. The diamic-
tite contains 5 –20% clasts with average diameters ranging from
0.2 to 15 cm. The diameters of the largest clasts reach up to
100 cm. The clasts are generally sub-angular and sub-rounded,
dominated by carbonate, chert and volcanic rocks. The laterally
equivalent silty dolostone also contains 5% angular or subangular
clasts with average diameters of 0.3–3 cm. The diameters of the
largest clasts within the dolostone range up to 20 cm. Similar to
the laterally equivalent diamictite, the clast composition is also
dominated by carbonate. The associated mudstone, which occurs
either below the diamictite or is laterally equivalent, contains
5–25% sand grains and 5% angular or sub-angular clasts with
Fig. 33.8. Geological map showing the
Cryogenian and Ediacaran outcrop of the
Guozigou-Keguqingshan area, western
range of Northern Tianshan Mountains
(modified from Wang et al. 1983).
an average diameter of 0.2–5 cm. The large clasts (up to 13 cm in
diameter) exhibit characteristics of dropstone within the finely
laminated mudstone, providing solid evidence of glaciogenic pro-
venance of these diamictite, dolostone and mudstone units.
The Talisayi diamictite. This contains 10% clasts with an average
diameter ranging from 0.2 –15 cm. Diameters of the largest
clasts range up to 100 cm. The clasts consist of predominantly car-
bonate, chert and quartz, and rare clasts of granite and other volca-
nic or metamorphic rocks. The shapes of the clasts vary from round
to angular. Some clasts show striation marks, rolling cracks or
lacunules on the surface.
Based on the sedimentary features described above, the Biexi-
bastao and Talisayi diamictites have been considered by Wang
et al. (1983) to be marine glaciogenic diamictites.
Correlation
Based on the stratigraphic descriptions of the four type areas,
the successions in the Tielikeli and Aksu-Wusi areas in the
Tarim Block show significant similarities. Similar to the well-
studied Cryogenian and Ediacaran successions of the Yangtze
Block, only two glaciation intervals are developed, and the
uppermost part of the succession is composed of thick carbon-
ate. The C-isotope variation of the upper Sugetblaq and Qige-
blaq formations in the Aksu-Wusi area is similar to that at the
top Doushantuo and Dengying formations in the Yangtze
Block (He et al. 2007a; Zhan et al. 2007). In particular, a
major negative excursion in the lower part of the Upper
Member of the Sugetblaq Fm. can be correlated with the
‘DOUNCE’ excursion at the top of the Doushantuo Fm. (Zhu
et al. 2007). The chemostratigraphy supports the correlation of
the Yulmeinak glaciation with the Nantuo glaciation (635 Ma),
even though there is no distinct cap carbonate recorded atop
the Yulmeinak diamictite.
While the successions in the Quruqtagh and Guozigou-
Keguqingshan areas in the Tianshan Mountains are similar to
each other, they show significant differences from those in the
Tielikeli and Aksu-Wusi areas. At least three glaciogenic diamic-
tites with possible cap carbonates are recorded in the area. The cap
carbonates atop the Altungol and Tereeken diamictites show
characteristic C-isotopic features (Xiao et al. 2004; Kou et al.
2008; Shen et al. 2008). However, C-isotope values from the cap
carbonate atop the Hankalchough diamictite vary dramatically
from section to section (Xiao et al. 2004), demonstrating great
differences when compared with the cap carbonates atop the
Altungol and Tereeken diamictites. The reason for the spatial
 THE TARIM BLOCK, NW CHINA
Fig. 33.9. Composite stratigraphic log of the Cryogenian and Ediacaran in the
Guozigou-Keguqingshan area, western range of Northern Tianshan Mountains
(after Wang et al. 1983; Gao et al. 1985a). Legends as in Figure 33.10.
375
variation of C-isotope values of the cap carbonate atop the Hankal-
chough diamictite remains problematic. Nevertheless, the distinct
association of glaciogenic diamictites and cap carbonates provides
strong evidence for correlation of these diamictites. In particular,
the cap carbonate above the Tereeken diamictite (see Supplemen-
tary Material, Fig. 1S) shows similar sedimentary features to that
of the Nantuo-Marinoan diamictite in other continents, supporting
their correlation. Furthermore, C-isotope chemostratigraphy also
supports a post-Nantuo or post-Cryogenian age of the Hankal-
chough diamictite, because the Dounce-Shuram-Wonaka excur-
sion, which is regarded as a distinct global mid-Ediacaran event,
has been recorded from the Shuiquan Fm. (Xiao et al. 2004).
Detailed integrated correlations of the Cryogenian and Edia-
caran successions of these four areas are given in Figure 33.10.
It should be noted that biostratigraphic correlations using organic-
walled microfossils have been intensively adapted for correlation
of the Neoproterozoic successions in the literature. Because taxo-
nomic problems and restricted occurrences resulted from poor
preservation, biostratigraphic correlation of organic-walled micro-
fossils remains problematic (Gao et al. 1985a; Zhong & Hao
1990). Magnetostratigraphy may support an improved correlation
of the Neoproterozoic successions; however, because of complex
and long-lasting tectonic processes, the original magnetic polarity
can only rarely be discerned. Correlations based on previous
palaeomagnetic studies (see Gao & Zhu 1984; Gao et al. 1985a,
1993) are inconsistent with other correlations and are thus not
yet considered reliable.
Geochronological constraints on the diamictites
Volcanic rocks and tuffaceous beds are well developed within the
Cryogenian and Ediacaran sequences in the Tarim Block. Three
major volcanic episodes are recorded particularly in the Quruqtagh
area, and have been termed (in ascending order) the Beiyixi, Zha-
moketi and Shuiquan episodes by Gao et al. (1985a). These volca-
nic rocks and tuffaceous beds, including basalt, diabase, andesite,
rhyolite and quartz porphyry, provide potential samples for radio-
metric dating. Numerous radiometric ages have been published in
past decades (Lu et al. 1985; Gao et al. 1993; Xu et al. 2005, 2009
and references therein). However, as these ages were based on
different radiometric methods, it is difficult to adapt them for stra-
tigraphic correlation. The most reliable SHRIMP zircon U –Pb
ages from the Qurugtagh area published by Xu et al. (2009)
provide constraints on the four diamictites in the area, suggesting
that the Bayixi diamictite was deposited between 740+7 Ma and
725+10 Ma, the Altungol and Tereeken diamictites between
725+10 Ma and 615+6 Ma, and the Hankalchough diamictite
between 615+6 Ma and c. 542 Ma.
These ages indicate that the oldest (Beiyixi) glaciation in the
Tarim Block is younger than 740 Ma. Although we are not sure
whether the Beiyixi glaciation represents the oldest Neoproterozoic
glaciation on a global scale, however, geochronological data from
the magmatic rocks underlying all the lowest glaciogenic diamic-
tites in the Tarim Block suggest that the oldest glaciation should
be younger than 780 Ma. These ages were reported by Zhang
et al. (2009), including a U–Pb zircon age of 773+3 Ma for the
mafic dyke swarms intruding the c. 820 Ma granite in the Quruqtagh
area, a U–Pb zircon age of 759+7 Ma for the mafic dyke swarms
intruding the blueschist of the Aksu Group of the Aksu-Wusi
region, and a U–Pb zircon age of 815+57 Ma for a mantle-sourced
gneissic granite in the Tielikeli area. Nevertheless, no sedimentary
evidence of glaciation overlying these 760–780 Ma magmatic
rocks and underlying the Beiyixi diamictite rules out glaciations(s)
between 740 Ma and 780 Ma in the Tarim Block.
The reported ages also support the correlation of the Tereeken
glaciation with the c. 635 Ma Nantuo glaciation, and that the
Hankalchough diamictite represents a younger Ediacaran gla-
ciation. However, whether the Hankalchough and Talisayi
376 M. ZHU & H. WANG
diamictites are time-correlative to the 582 Ma Gaskiers glaciation
remains poorly constrained. Based on the C-isotope data from the
Shuiquan Fm. in the Qurugtagh area (Xiao et al. 2004), it is pro-
posed here the Hankalchough diamictite is younger than 551 Ma,
because the negative C-isotopic data from the Shuiquan Fm.
exhibit distinct values of the DOUNCE excursion at the top of
the Doushantuo Fm. of the Yangtze Block. Strata at the top of
this excursion have a high-resolution zircon U – Pb age of
551 Ma (Condon et al. 2005). There is no doubt that Hankal-
chough and Talisayi diamictites represent an end-Ediacaran gla-
ciation rather than a Cambrian one, because Cambrian small
shelly fossils have been reported from the strata overlying the
Hankalchough and Talisayi diamictites (Wang et al. 1983; Qian
1999; Yao et al. 2005).
Fig. 33.10. Stratigraphic correlation of the
Cryogenian and Ediacaran successions in
the Tarim area. Key: 1, dolostone; 2,
limestone; 3, conglomerate; 4,
coarse-grained sandstone; 5, sandstone;
6, siltstone; 7, diamictite; 9, bedded
diamictite; 8, shale and mudstone;
10, silicates or cherts; 11, phosphorite;
12, blueschist; 13, volcanic rocks;
14, dropstones.
Palaeolatitudes and palaeogeography
Systematic palaeomagnetic analyses of the samples from Cryogen-
ian and Ediacaran strata of the Aksu-Wusi, Quruqtagh and
Guozigou-Keguqinshan areas indicate that the palaeolatitudes of
the three areas are ,278 (Table 33.1), which supported the low-
latitude glaciations during the late Neoproterozoic put forward
by the Snowball Earth hypothesis. These palaeomagnetic data,
combined with new data from the mafic dyke swarms (c. 810–
820 Ma) intruding the strata below the Cryogenian successions
in the Tielikeli (328N; Zhang et al. 2004) and Aksu-Wusi
(43 + 68N; Chen et al. 2004) areas, collectively suggest a close
palaeogeographic affinity of the Tarim and Yangtze plates with
Australia and Antarctica during the late Neoproterozoic,
 THE TARIM BLOCK, NW CHINA
Table 33.1. Palaeolatitude data from four areas of the Tarim Block (Gao et al.
1993; Huang et al. 2005; Zhan et al. 2007)
Guozigou-Keguqinshan Quruqtagh Aksu-Wusi
Tarqat Fm. 8.18 Shuiquan Fm. 21.78 Sugetblaq Fm. 24.38
6.68 Zhamoketi Fm. 21.98 c. 278
Kulutieliekti Fm. 10.38 Tereeken Fm. 19.98 Yulmeinak Fm. 24.18
Beiyixi Fm. 0.98
supporting the supercontinental configuration and subsequent
rifting of Rodinia.
This work benefited from discussions with many colleagues, including C. Zhou,
G. Li, B. Xu, S. Xiao and G. Shields. Fieldwork was jointly carried out with
G. Li, F. Zhao, X. Zhao and Z. Yin. This project is supported by the National
Natural Science Foundation of China (nos 40715005 and 40930211). This rep-
resents a contribution of the IUGS- and UNESCO-funded IGCP (International
Geoscience Programme) project #512.
References
Cao, R. 1991. New observation of the Sinian System in the southern
Yardang Mountains, Xinjiang. Regional Geology of China, 1991,
30 – 34 (in Chinese with English abstract).
Chen, Y., Xu, B., Zhan, S. & Li, Y. 2004. First mid-Neoproterozoic
palaeomagnetic results from the Tarim Basin (NW China) and their
geodynamic implications. Precambrian Research, 133, 271– 281.
Condon, D., Zhu, M., Bowring, S., Wang, W., Yang, A. & Jin, Y. 2005.
U –Pb ages from the Neoproterozoic Doushantuo Fm., China.
Science, 308, 95 – 98.
Duan, J. Y., Xia, D. X. & An, S. L. 2005. Deep-water sedimentation and
tectono-palaeogeography of the Neoproterozoic – early Palaeozoic
aulacogen in Kuruktag, Xingjiang, China. Acta Geologica Sinica,
79, 7 – 14 (in Chinese with English abstract).
Gao, Z. & Chen, K. 2003. The Nanhua System of Xinjiang and some
geological issue of Nanhua System in China. Geological Survey
and Research, 26, 8 –14 (in Chinese with English abstract).
Gao, Z. & Qian, J. 1985. Sinian glacial deposits in Xinjiang, Northwest
China. Precambrian Research, 29, 143– 147.
Gao, Z. & Zhu, S. 1984. Precambrian Geology in Xinjiang, China. Xin-
jiang People’s Publishing House, Urumuqi, China (in Chinese with
English summary).
Gao, Z., Wu, S., Li, Y. & Qian, J. 1981. Sinian –Cambrian stratigraphy of
Aksu-Keping area, Xinjiang. Chinese Science Bulletin, 12, 741–743
(in Chinese).
Gao, Z., Wang, W. et al. 1985a. The Sinian System of Xinjiang. Xinjiang
People’s Publishing House, Urumuqi, China (in Chinese with English
abstract).
Gao, Z., Wang, W. et al. 1985b. The Sinian System on Aksu-Wushi
Region, Xinjiang, China. Xinjiang People’s Publishing House,
Urumuqi, China (in Chinese with English summary).
Gao, Z., Chen, J., Lu, S., Peng, C. & Qin, Z. 1993. The Precambrian
Geology in Northern Xinjiang (Precambrian Geology, No. 6). Geo-
logical Publishing House, Beijing (in Chinese with English
summary).
He, J. Y., Xu, B., Meng, X. Y., Kou, X. W., Liu, B., Wang, Y. & Mi, H.
2007a. Neoproterozoic sequence stratigraphy and correlation in
Kurugtagh area, Xinjiang. Acta Petrologica Sinica, 23, 1645–1654
(in Chinese with English abstract).
He, X. B., Xu, B. & Yuan, Z. Y. 2007b. C-isotope composition and
correlation of the Upper Neoproterozoic in Keping area, Xingjiang.
Chinese Science Bulletin, 52, 504– 511.
Huang, B. C., Xu, B., Zhang, C. X., Li, Y. A. & Zhu, R. X. 2005. Palaeo-
magnetism of the Baiyisi volcanic rocks (c. 740 Ma) of Tarim, North-
west China: a continental fragment of Neoproterozoic Western
Australia? Precambrian Research, 142, 83 –92.
377
Kou, X. W., Wang, Y., Wei, W., He, J. Y. & Xu, B. 2008. The Neo-
proterozoic Altungol and Huangyanggou formations in Tarim
plate: recognized newly glaciation and interglaciation? Acta Petrolo-
gica Sinica, 24, 2863– 2868 (in Chinese with English abstract).
Li, H. & Dong, Y. 1991. Sedimentary features of the Sinian Zhamoketi
Formation in the Middle Quruqtagh area of Xinjiang. Xinjiang
Geology, 9, 340–351.
Liou, J. G., Graham, S. A. et al. 1989. Proterozoic blueschist belt in
western China: best documented Precambrian blueschists in the
world. Geology, 17, 1127–1131.
Liu, B., Xu, B., Meng, X. Y., Kou, X. W., He, J. Y., Wei, W. & Mi, H.
2007. Study on the chemical index of alteration of Neoproterozoic
strata in the Tarim plate and its implications. Acta Petrologica
Sinica, 23, 1664–1670 (in Chinese with English abstract).
Lu, S., Ma, G., Gao, Z. & Lin, W. 1985. Sinian ice ages and glacial
sedimentary facies – areas in China. Precambrian Research, 29,
53– 63.
Lu, S., Li, H., Zhang, C. & Niu, G. 2008. Geological and geochronolo-
gical evidence for the Precambrian evolution of the Tarim craton
and surrounding continental fragments. Precambrian Research,
160, 94– 107.
Ma, S., Wang, Y. & Fang, X. 1989. The Sinian at north slope of western
Kunlun Mountains. Xinjiang Geology, 7, 68– 79 (in Chinese with
English abstract).
Ma, S., Wang, Y. & Fang, X. 1991. Basic characteristics of Proterozoic
Eonothem as a table cover on northern slope of western Kunlun
Mountain. Xinjiang Geology, 9, 59– 71 (in Chinese with English
abstract).
Norin, E. 1937. Reports from the Scientific Expedition to the Northwes-
tern Provinces of China under the Leadership of Dr. Sven Hedin, III.
Geology, 1. Geology of Western Quruqtagh, Eastern Tien-Shan.
Bokförlags Aktiebolaget Thule, Stockholm.
Qian, Y. (ed.) 1999. Taxonomy and Biostratigraphy of Small Shelly
Fossils in China. Science Press, Beijing (in Chinese with English
summary).
Qian, J. & Xiao, B. 1984. An Early Cambrian small shelly fauna from
Aksu-Wushi region, Xinjiang. In: Professional Papers of Strati-
graphy and Palaeontology, No. 13. Geological Publishing House,
Beijing, 65– 90 (in Chinese with English abstract).
Shen, B., Xiao, S., Kaufman, A. J., Bao, H., Zhou, C. & Wang, H.
2008. Stratification and mixing of a post-glacial Neoproterozoic
ocean: evidence from carbon and sulfur isotopes in a cap dolostone
from northwest China. Earth and Planetary Science Letters, 265,
209– 228.
Wang, J., Cheng, S., Bai, W. & Wang, L. 1983. Glacigenous strata in
the western part of the northern Tianshan Mountains. In: Precam-
brian Geology, No.1 of the Collected Works of Late Precambrian
Glacigenous Rocks in China. Geological Publishing House,
Beijing, 105– 118 (in Chinese with English summary).
Wang, A. G., Zhang, C. L. & Guo, K. Y. 2004. Depositional types and its
tectonic significance of lower member of Nanhuan System in North
margin of West Kunlun. Journal of Stratigraphy, 28, 248– 256 (in
Chinese with English abstract).
Xiao, S., Bao, H. et al. 2004. The Neoproterozoic Quruqtagh Group in
eastern Chinese Tianshan: evidence for a post-Marinoan glaciation.
Precambrian Research, 130, 1– 26.
Xu, B., Jian, P., Zheng, H., Zou, H., Zhang, L. & Liu, D. 2005. U– Pb
zircon geochronology and geochemistry of Neoproterozoic volcanic
rocks in the Tarim Block of northwest China: implications for the
breakup of Rodinia supercontinent and Neoproterozoic glaciations.
Precambrian Research, 136, 107– 123.
Xu, B., Xiao, S. et al. 2009. SHRIMP zircon U– Pb age constraints on
Neoproterozoic Quruqtagh diamictites in NW China. Precambrian
Research, 168, 247–258.
Yao, J., Xiao, S., Yin, L., Li, G. & Yuan, X. 2005. Basal Cambrian
microfossils from the Yurtus and Xishanblaq formations (Tarim,
north-west China): systematic revision and biostratigraphic corre-
lation of Micrhystridium-like acritarchs from China. Palaeontology,
48, 687– 708.
Zhan, S., Chen, Y., Xu, B., Wang, B. & Faure, M. 2007. Late Neopro-
terozoic palaeomagnetic results from the Sugetbrak Formation of the
378 M. ZHU & H. WANG
Aksu area, Tarim basin (NW China) and their implications to palaeo-
geographic reconstructions and the snowball Earth hypothesis.
Precambrian Research, 154, 143– 158.
Zhang, C. L., Shen, J. L. & Guo, K. Y. 2004. Geochemistry of the Neo-
proterozoic mafic dyke swarm and basalt in south of Tarim Plate and
its tectonic significance. Acta Petrologica Sinica, 20, 473– 482 (in
Chinese with English abstract).
Zhang, C. L., Li, X. H., Li, Z. X., Lu, S. N., Ye, H. M. & Li, H. M. 2007.
Neoproterozoic ultramafic –mafic– carbonatite complex and grani-
toids in Quruqtagh of northeastern Tarim Block, western China: geo-
chronology, geochemistry and tectonic implications. Precambrian
Research, 152, 149– 169.
Zhang, C. L., Li, Z. X., Li, X. H. & Ye, H. M. 2009. Neoproterozoic mafic
dyke swarms at the north margin of the Tarim Block, NW China: age,
geochemistry, petrogenesis and tectonic implications. Journal of Asia
Earth Sciences, 35, 167– 179.
Zhao, W., Li, C., Gao, Z. & Miao, C. 1985. Stromatolites from the Paer-
gangtag Group at the southern slope of Mt. Kuluketag, Xinjiang. Acta
Palaeontologica Sinica, 24, 71 –82.
Zhong, D. & Hao, Y. (eds) 1990. Sinian to Permian Stratigraphy
and Palaeontology of the Tarim Basin, Xinjiang, (I) Kuruktag
Region. Nanjing University Press, Nanjing (in Chinese with
English abstract).
Zhu, M., Strauss, H. & Shields, G. A. 2007. From Snowball Earth to the
Cambrian bioradiation: calibration of Ediacaran– Cambrian Earth
history in South China. Palaeogeography, Palaeoclimatology,
Palaeoecology, 254, 1 –6.
Zhu, W., Zhang, Z., Shu, L., Lu, H., Su, J. & Yang, W. 2008. SHRIMP
U –Pb zircon geochronology of Neoproterozoic Korla mafic dykes in
the northern Tarim Block, NW China: implications for the long
lasting breakup process of Rodinia. Journal of the Geological
Society, London, 165, 887–890.
 Chapter 34

The Hula Hula Diamictite and Katakturuk Dolomite, Arctic Alaska

FRANCIS A. MACDONALD
Department of Earth and Planetary Sciences, Harvard University, Cambridge, MA 02138, USA (e-mail: fmacdon@fas.harvard.edu)

Abstract: The Katakturuk Dolomite is a c. 2-km-thick Neoproterozoic carbonate succession (units K1– K4) exposed in the NE Brooks
Range of Alaska. These strata were deposited on a south-facing (present coordinates), rifted passive margin on the North Slope subterrane
(NSST) of the Arctic Alaska-Chukotka Plate (AACP). The glaciogenic Hula Hula diamictite rests below the Katakturuk Dolomite and
consists of 2– 50 m of diamictite that interfingers with the underlying Mt. Copleston volcanic rocks. Unit K1 of the Katakturuk Dolomite
begins with less than 10 m of dark grey, finely laminated limestone with ‘roll-up’ structures, and continues upwards with nearly 500 m of
recrystallized, ooid-dominated grainstone. The Nularvik dolomite (unit K2 of the Kataktruk Dolomite) rests on unit K1 with a knife-sharp
contact on a heavily silicified surface. The Nularvik dolomite is composed predominantly of laminated micro-peloids hosting tubestone
stromatolites and giant wave ripples, followed by decametres of dolomatized, pseudomorphosed former aragonite crystal fans.
Carbon-isotope chemostratigraphy suggests that the Hula Hula diamictite is an early Cryogenian glacial deposit, and that, despite the
absence of directly underlying glacial deposits, the Nularvik dolomite is a basal Ediacaran cap carbonate. These correlations are sup-
ported by the characteristic sedimentological features in both the carbonate capping the Hula Hula diamictite and the Nularvik dolomite.
Detrital zircon and Palaeozoic fauna provenance studies support the inference that much of the AACP is exotic to Laurentia; however, the
pre-Mississipian relationship between the NSST and the rest of the AACP remains uncertain. Previous palaeomagnetic surveys have been
hampered by pervasive Late Cretaceous overprints. Additional geological mapping, sequence stratigraphy and geochronological data are
needed to correlate Neoproterozoic and Palaeozoic units across the AACP, and constrain relationships between subterranes in the AACP.

The Neoproterozoic Katakturuk Dolomite (units K1 –K4) and the
overlying Cambrian to Ordovician Nanook Limestone form the
backbone of the northernmost ranges of the Arctic National Wild-
life Reserve of Alaska (for a field guide and an introduction to the
general geology of the region, see Molenaar et al. 1987). The pre-
Mississippian stratigraphy of the NE Brooks Range has been
described through geological mapping (Leffingwell 1919; Reed
1968; Reiser 1971; Sable 1977; Robinson et al. 1989), palaeonto-
logical reconnaissance (Dutro 1970) and sequence analysis (Clough
& Goldhammer 2000). Recently, Macdonald et al. (2009b) con-
ducted integrated chemo- and lithostratigraphic studies, identified
two putative glacial horizons and suggested a late Neoproterozoic
age for the Katakturuk Dolomite.
The Katakturuk Dolomite was named by Dutro (1970) for its
exposure in the Katakturuk River canyon in the Sadlerochit Moun-
tains. Katakturuk is an English derivation of the Inupiaq word
Qattaqtuuraq, which translates to ‘a wide open place’ (Clough
1989). The Katakturuk Dolomite has only been positively ident-
ified in three localities: the Sadlerochit Mountains, the Shublik
Mountains and on both sides of the Hula Hula River near
Kikitak Mountain (Fig. 34.1). The succession attains its greatest
thickness in the Sadlerochit Mountains, where it is composed of
c. 2080 m of shallow-water dolomite, and progressively thins to
the south in the Shublik Mountains and at Kikitak Mountain.
Exposures in the NE Brooks Range capture the shelf – slope tran-
sition, facing to the south (present coordinates).
Macdonald et al. (2009b) identified the Hula Hula diamictite at
the base of the Katakturuk Dolomite in the eastern Sadlerochit
Mountains and along the Hula Hula River near Kikitak Mountain
(Fig. 34.1). The type section of the Hula Hula diamictite is on the
east side of the Hula Hula River along Eustik Creek (Figs 34.2 &
34.3a, section F621; 69825.4910 N, 144823.2160 W). At this locality,
the diamictite is overlain by a 10-m-thick, dark, variably dolomi-
tized limestone (unit K1 of the Katakturuk Dolomite) that contains
microbial ‘roll-up’ structures (Hoffman et al. 1998; Pruss et al.
2010). This is one of only two limestone horizons within the
Katakturuk Dolomite.
In the Sadlerochit Mountains, nearly 500 m higher in the
sequence, a micro-peloidal dolostone containing tubestone stroma-
tolites, giant wave ripples and decametres of pseudomorphosed
former aragonite crystal fans rests on a silicified surface. The
type section of the Nularvik dolomite (referred to as the Nularvik
cap carbonate in Macdonald et al. 2009b) is on the west side of the
Nularvik River (Figs 34.2 and 34.3b, section F601; 69837.5880 N,
145805.4510 W). The Nularvik dolomite is succeeded by a major
transgression marked by shale and allodapic carbonate, and then
an additional c. 1200 m of Ediacaran shallow-water dolomite
(units K3 –K4).
Exposure of the Katakturuk Dolomite is limited to the North
Slope subterrane (NSST) of the Arctic Alaska-Chukotka Plate
(AACP). The NSST is tied to the AACP by the Carboniferous
Lisburne Group; however, the pre-Mississippian relationship
between the NSST, the AACP and Laurentia remain uncertain.
Dark-coloured limestones, potentially correlative to the lower-
most unit of the Katakturuk Dolomite (K1), are present in the
Third and Fourth Ranges of the NSST (Fig. 34.1), c. 10 km
and 15 km south of the Shublik Mountains, respectively (Reiser
1971). However, the thickness of this unit is unknown as it is
poorly exposed in the Third Range and structurally duplicated
in the Fourth Range (Macdonald et al. 2009b). Precambrian dolo-
mites dominated by stromatolites and coated grains, possibly cor-
relative with the Katakturuk Dolomite, have been described on
Seward Peninsula and near Snowden Mountain on the southern
subterranes of the AACP (Dumoulin 1988; Dumoulin & Harris
1994), and in the Farewell Terrane (Babcock et al. 1994);
however, no diamictites have been described at these localities.
Terminal Neoproterozoic diamictites occur in the Upper Tindir
Group of East-Central Alaska, but these were definitively depos-
ited on Laurentia (Allison et al. 1981; Young 1982; Macdonald
& Cohen 2011).
Structural framework
Present exposures of the Katakturuk Dolomite in the Sadlerochit,
Shublik and Kikitak Mountains (Fig. 34.1) are the product of
Palaeogene north-vergent thrusting (Wallace & Hanks 1990).
These structures formed during a late-stage reactivation of the
mostly Mesozoic Brookian orogeny (Moore et al. 1997), and are
related to the piecemeal accretion of southern Alaska (Fuis et al.
2008). Pre-Mississippian strata in the NE Brooks Range also pre-
serves Early to Middle Devonian SE-vergent structures associated

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 379– 387. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.34
380 F. A. MACDONALD

Fig. 34.1. Geological map of the Shublik, Sadlerochit and Kikitak Mountains, and the Fourth Range, with a simplified tectonic and location map inset. Post-Ordovician
geology modified from Robinson et al. (1989) and Bader & Bird (1986). Tectonic map modified and simplified from Johnston (2001) and Colpron et al. (2007).
Abbreviations for terranes: AACP, Arctic Alaska-Chukotka Plate; FR, Farewell-Ruby-Hammond Terranes; NAm, Ancestral North America; LA, Late Accreted
Terranes; ?, poorly exposed and of uncertain affinity.

with the Romanzof orogeny (Oldow et al. 1987). These structures
are thought to be distinct from those related to the Late Devonian
Ellesmerian orogeny in the Yukon and Canadian Arctic Islands
(Lane 2007).
Metamorphic grade generally increases from north to south
with exposures of the Katakturuk Dolomite in the Shublik and
Sadlerochit Mountains displaying little folding and simple
block faults. In the NE Sadlerochit Mountains, the underlying
Neoproterozoic rocks are tightly folded and chloritized;
however, this apparent difference in deformation may be due
entirely to lithology and rheology. The overlying Nanook Lime-
stone contains conodonts with a Conodont Alteration Index
(CAI) of 3.5 suggesting that at least these strata have experienced
less than 300 8C (Harris et al. 1990). The Katakturuk Dolomite is
heavily dolomitized and recrystallized; however, the age of the
dolomitzation is unknown and, typically, primary sedimentary
features are preserved. In the Kikitak Mountain area, small-scale
folding is more common, and the basalts are pervasively chlori-
tized. These modifications, coupled with incomplete exposures,
make an exact determination of the thickness of the Hula Hula
diamictite impossible.
Clough & Goldhammer (2000) inferred a NE– SW palaeo-
strandline (present coordinates) from palaeo-current data measured
in tabular cross-bedded grainstones, the orientation of elongated
stromatolites, and from the thickening of outer-ramp to slope
facies to the south. While readily measurable elongated stromato-
lites are not present in the lower kilometre of the Katakturuk Dolo-
mite, facies changes are apparent between the Sadlerochit, Shublik
and Kikitak Mountains, deepening from north to south (Macdonald
et al. 2009b); deepening is not demonstrable in facies changes
from west to east along the ranges. Consequently, for the lower
kilometre of the Katakturuk Dolomite (the Hula Hula Diamictite
and units K1 –K3), Macdonald et al. (2009b) assumed an east –
west palaeo-strandline. Balanced cross-sections from seismic data
reveal a c. 30% Palaeogene north – south shortening in the NE
Brooks Range (Molenaar et al. 1987; Moore et al. 1997). This
results in a restored distance between the Sadlerochit and Shublik
Mountains of c. 11 km, and an additional 10 km north –south and
50 km east –west between the Shublik and Kikitak Mountains.
However, there are few constraints on the pre-Palaeogene tectonic
movement between the two ranges, inhibiting confidence in three-
dimensional basin reconstructions.
 HULA HULA DIAMICTITE AND KATAKTURUK DOLOMITE
Stratigraphy
As much as 3 km of Neoproterozoic strata are exposed in the NE
Brooks Range of Arctic Alaska. The oldest strata in the region are
in map unit ‘O. G.’, which was previously mapped as the Neurokpuk
Formation (Fm.) and consists of tightly folded, mixed siliciclastic
and carbonate rocks (Reiser et al. 1980; Robinson et al. 1989;
Macdonald et al. 2009b). The thickness of this unit is unknown
and its base is not exposed. Map unit O. G. is succeeded by the Mt.
Copleston volcanic rocks. Although this contact is structural in the
Sadlerochit Mountains (Macdonald et al. 2009b), regionally it
has been described as an unconformity (Reiser et al. 1980; Robin-
son et al. 1989). The Mt. Copleston volcanic rocks underlie and
interfinger with the Hula Hula diamictite, which is in turn suc-
ceeded by the c. 2-km-thick Neoproterozoic Katakturuk Dolomite.
Subsidence analysis suggests that these strata were accommodated
by extension on the southern margin of the North Slope subterrane
with units K1 – K3 of the Katakturuk Dolomite deposited on a ther-
mally subsiding passive margin (Macdonald et al. 2009b).
Mt. Copleston volcanic rocks
The Mt. Copleston volcanic rocks are rusty weathering, dark
maroon to black and green tholeitic basalts with 5 mm chlorite,
calcite, zeolite amygdales, and common native copper. In the
western Shublik Mountains, the basalt is up to 450 m thick with
metre-scale individual flows. Along the Hula Hula River, the Mt.
Copleston volcanic rocks are c. 500 m thick with a true thickness
difficult to determine due to structural complexities. The volcanic
rocks are often greenstone, and dominated by volcaniclastic units
in the upper c. 100 m. On the west flank of Kikitak Mountain, the
basalts are over 100 m thick and metamorphosed to greenstone. In
the eastern Sadlerochit Mountains, the basalts appear relatively
low grade with minimal chlorite, 5 –10 mm long plagioclase
lathes, and intact, spherical amygdales. Exposures measure up to
105 m thick with an unconformable, often faulted basal contact
with the underlying unit O. G. The Mt. Copleston volcanic rocks
381
Fig. 34.2. Composite carbon chemo- and
lithostratigraphy of the Katakturuk
Dolomite (K1– K4) and the Nanook
Limestone (N1, N2) in the NE Brooks
Range. Sadlerochit data from measured
sections F601, F602, F607, F501 and F505;
Shublik data from measured sections F513,
F514, F517, F613 and F614; Kikitak data
from measured sections F619 and F624 (see
Figs 34.1 & 34.3 for locations). All
carbonate carbon measurements in ‰
notation. Depositional distance assumes
c. 30% Palaeogene shortening (Molenaar
et al. 1987; Moore et al. 1997). Time scale
is inferred from the detrital zircon ages in
the ‘O.G.’ (Macdonald et al. 2009b),
Cambrian trilobites in the Nanook
Limestone and the similarity of the
chemostratigraphic profile and sedimentary
structures in the intervening carbonates to
other Neoproterozoic age successions (see
text for discussion). HH, Hula Hula
diamictite; V, Mt. Copleston
volcanic rocks.
and the Katakturuk Dolomite are separated by the Hula Hula
diamictite, which ranges in thickness from c. 2 m to 50 m.
Katakturuk Dolomite
In the Sadlerochit Mountains, the Katakturuk Dolomite is over
2000 m thick, whereas in the Shublik Mountains, it is only c.
1000 m thick. This is due both to internal thinning and a basal trun-
cation (Macdonald et al. 2009b). Generally, the Katakturuk Dolo-
mite is composed of massive, light-grey, shallow-water dolomite
with common ooids and cement. However, further to the SE,
along the Hula Hula River, the Katakturuk Dolomite is composed
predominantly of allodapic carbonate of unit K1 and measures
only 400 m thick, with the upper units of the Katakturuk Dolomite
truncated under the sub-Mississippian unconformity (Figs 34.2 &
34.3b). To highlight the major unconformities and disconformities,
and departing from Robinson et al.’s (1989) lithostratigraphic sub-
division, Macdonald et al. (2009b) divided the Katakturuk Dolo-
mite into four informal units (Fig. 34.2): the Cryogenian map
unit K1, the basal Ediacaran Nularvik dolomite (map unit K2),
the early –middle Ediacaran map unit K3 and the late Ediacaran
map unit K4. The uppermost portion of the Katakturuk Dolomite
in the Sadlerochit Mountains, as described by Clough & Goldham-
mer (2000) has been included with the lowermost Nanook Lime-
stone (Macdonald et al. 2009b).
Nanook Limestone
The Katakturuk Dolomite is overlain by the Cambrian to Ordovi-
cian Nanook Limestone. Previous studies suggested an unconfor-
mity at this level (Clough & Goldhammer 2000; Macdonald
et al. 2009b); however, this interpretation may have been compli-
cated by cave breccias from a karstic surface higher in the succes-
sion (personal observations). The lower c. 250 m of the Nanook
Limestone host bed-parallel ichnogenera and C-isotopic profiles
that are consistent with an Early to Middle Cambrian age
382
Fig. 34.3. (a) Geological map of the Kikitak Mountain area near the Hula Hula River. (b) Geological map of the eastern Sadlerochit Mountains along Nularvik Creek. See
Figure 34.1 for locations and keys for both maps.
(Macdonald et al. 2009b). The upper Nanook Limestone contains
Late Cambrian trilobites (Blodgett et al. 1986), and terminates
with c. 160 m of Middle to Late Ordovician strata with the upper-
most beds containing a diverse gastropod assemblage along with
other molluscs, ostracods and brachiopods with a definitive
Late Ordovician age (Blodgett et al. 1986) and a Siberian affinity
(Blodgett et al. 2002).
Glaciogenic deposits and associated strata
Hula Hula diamictite
Reiser et al. (1970) described the orange weathering diamictite
along the Hula Hula River as carbonate debris flows with clasts
of basalt, and included these deposits with the Katakturuk Dolo-
mite. Macdonald (2009) separated the Hula Hula diamictite from
the Katakturuk Dolomite and suggested that sedimentation
occurred under a glacial influence. On the east side of the Hula
Hula River, along Eustik Creek, the Hula Hula diamictite is c.
50 m thick. Although incomplete exposure and structural rep-
etition compromise the measurement of exact thicknesses, discrete
thrust panels allow for confidence in the general stratigraphic
relations. The lower 12 m of the diamictite is composed of cobble-
sized clasts of orange dolomite and green to black basalt, together
with quartzite pebbles, in a green to tan siltstone, and is interfin-
gered with at least four basaltic flows that range in thickness
from 0.2 to 2 m. This lower diamictite is overlain with c. 30 m
of poorly exposed, fine millimetre-laminated siltstone with rare,
gravel- and cobble-sized bedding-piercing outsized clasts and mul-
tiple orange allodapic carbonate beds. The upper 5 m of the Hula
Hula diamictite is a massive, clast-supported diamict with boulders
of dolomite and cobbles of basalt in a calcareous silt matrix. No
striated clasts have been observed.
On the west side of the Hula Hula River near Kikitak Mountain
(Figs 34.2 & 34.3b), the basal Hula Hula diamictite is composed
of c. 10 m of diamictite with cobbles of dolomite and basalt in a
siltstone matrix, and an additional c. 40 m of fine millimetre-
laminated siltstone with rare outsized clasts and multiple beds of
orange allodapic carbonate. The Hula Hula diamictite is also
exposed in the eastern Sadlerochit Mountains along the Nularvik
F. A. MACDONALD
Creek, where the diamictite is only 2 m thick, and is composed
of angular cobbles of dolomite, quartzite and basalt in a coarse,
arkose grit (Fig. 34.4).
Katakturuk Dolomite
Near Kikitak Mountain and on both sides of the Hula Hula River,
the basal unit of the Katakturuk Dolomite (K1) consists of c. 10 m
of dark limestone, and an additional c. 75 m of rhythmite and allo-
dopic carbonate. Near the Hula Hula River, on the north side of
Eustik Creek, the basal 10 m limestone overlying the Hula Hula
diamictite exhibits ‘roll-up’ microbial structures, reminiscent of
the basal Rasthof Formation of northern Namibia (Hoffman
et al. 1998; Pruss et al. 2010), whereas these are not present in
the Sadlerochit Mountains. These shallow upwards to ,500 m
of massively bedded, often silicified, grainstone and packstone
with poorly defined parasequences that are littered with giant
ooids (c. 5 mm in diameter). Unit K1 culminates with c. 30 m of
resistant, silicified grainstone and broken beds of recrystallized
black chert that weather to a distinct black and white (the lower
zebra dolomite of Robinson et al. 1989).
Nularvik dolomite
The Nularvik dolomite (K2) is equivalent to the upper portion of
the zebra dolomite (Robinson et al. 1989; Clough & Goldhammer
2000). In the Sadlerochit Mountains, the Nularvik dolomite (K2)
consists of 20– 45 m of white to buff-coloured, recrystallized,
finely laminated, micro-peloidal dolomite, overlain by tens of
metres of former aragonite crystal fans (Macdonald et al.
2009b). Funnel-shaped calcite and silica cements are common in
the lower 15 m of unit K2. In cross-section, the funnels are less
than 5 cm tall, and up to 2 cm wide, taper downward, and are com-
monly linked at the top along bed parallel cements. The funnels are
filled with isopacous, void-filling cements. In the western Shublik
Mountains, these funnel-shaped cements are laterally equivalent
with tubestone stromatolite bioherms (Corsetti & Grotzinger
2005). The peculiar tubestone stromatolites are distinguished by
evenly distributed, c. 1-cm-diameter, cement-filled cylindrical
tubes, and span as much as 8 m of stratigraphy.
 HULA HULA DIAMICTITE AND KATAKTURUK DOLOMITE
In the central and western Sadlerochit Mountains, giant wave
ripples (Allen & Hoffman 2005) are also present in the Nularvik
dolomite, at the top of the micro-peloidal dolomite. These are suc-
ceeded by breccia, and tens of metres of pseudomorphosed arago-
nite crystal fans, with individual fans measuring as tall as 60 cm
(Clough & Goldhammer 2000; Macdonald et al. 2009b). The
strata hosting the fans are dominated by grainstone and cement
and are often broken, brecciated and recrystallized. Crystal fans
are not present at this horizon in the Shublik Mountains.
In the Sadlerochit Mountains, unit K2 is succeeded by 2 m of
shale and laterally discontinuous allodapic carbonate beds. In the
Shublik Mountains, these deeper water facies expand to over
100 m of shale, rhythmite and allodapic carbonate. These are in
turn overlain by an additional c. 1200 m of dolomite, primarily in
grainstone, biolaminate and stromatolitic facies (units K3 and K4).
Boundary relations with overlying and underlying
non-glacial units
In the eastern Sadlerochit Mountains, along the Nularvik Creek,
the Hula Hula diamictite rests disconformably on pillow basalt
of the Mt. Copleston volcanic rocks, with a basally erosive
contact. In the Kikitak Mountain area, the Hula Hula diamictite
also rests disconformably on the Mt. Copleston volcanic rocks;
yet, the upper 50 m of the Mt. Copleston volcanic rocks consists
of a volcaniclastic diamictite of outsized volcanic gravel and
cobbles in matrix-supported volcanic grit. It is unclear if this
383
Fig. 34.4. C-isotope chemo- and
lithostratigraphy of the Mt. Copleston
volcanic rocks, the Hula Hula diamictite,
and unit K1 along the Nularvik Creek in the
eastern Sadlerochit Mountains (F607), and
along the east side of the Hula Hula River
(F619).
deposit is a debris flow or a glacial diamictite formed of repro-
cessed volcanic rocks. Macdonald et al. (2009b) included this
unit with the Mt. Copleston volcanic rocks rather than the Hula
Hula diamictite because it lacks foreign clasts, shows no obvious
evidence for a glacial origin, and because volcanic breccias
occur at other horizons within the Mt. Copleston volcanic rocks.
Along the Nularvik Creek in the Sadlerochit Mountains the
basal contact of the Katakturuk Dolomite (unit K1) is poorly
exposed, whereas near Kikitak Mountain and on both sides of
the Hula Hula River, the Hula Hula diamictite is overlain with a
knife-sharp contact by c. 10 m of dark limestone. The Nularvik
dolomite (K2) rests above unit K1 on a heavily silicified surface
with a knife-sharp contact (Fig. 34.5).
Chemostratigraphy
High-resolution, carbonate C- and O-isotope chemostratigraphy
through the Katakturuk Dolomite and Nanook Limestone were
reported by Macdonald et al. (2009b). In the Sadlerochit Moun-
tains, above the Hula Hula diamictite, in the basal 20 m of K1,
C-isotope values rise from – 2‰ to þ6‰ (Fig. 34.4) where they
hover between þ3‰ and þ6‰. In the Kikitak Mountain area,
C-isotope values rise from þ1‰ to þ8‰, and also oscillate
around þ5‰ with slightly more variability. In the last para-
sequence of K1, values drop to 0‰.
C-isotope profiles of the Nularvik dolomite display an inverted
S-shaped profile with a nadir at –2‰ (Fig. 34.5). In the Sadlerochit
384
Fig. 34.5. Chemo- and lithostratigraphy of the Nularvik dolomite. See Figure 34.1 for locations of measured sections. All carbonate carbon (filled) and oxygen (hollow)
data are in ‰ notation.
Mountains, above the Nularvik dolomite, C-isotope values are
highly variable through the cement-dominated crystal fans. In
the Shublik Mountains, no crystal fans are present, and instead
the transgressive sequence progresses from grainstone, to ribbo-
nite, to variably dolomitized limestone rhythmite and shale.
C-isotope values bottom out at – 3‰ in these rhythmites, then
jump to þ3‰ above a sharp surface below the overlying allodapic
carbonate (Fig. 34.5).
Palaeolatitude and palaeogeography
The most popular model for the opening of the Arctic Ocean
involves a c. 668 counterclockwise rotation of the AACP away
from the Canadian Arctic islands about a pole in the Mackenzie
Delta region (Carey 1955, 1958; Hamilton 1970; Grantz et al.
1979). Barring any earlier movement relative to Laurentia, this
model would place the Neoproterozoic exposures in the NE
Brooks Range offshore of what is now Banks Island. Lane
(1997) pointed out multiple geological inconsistencies with the
rotation model, including ages of deformation and deposition
and proposed a model pinning Arctic Alaska to near its present
position since Palaeozoic times. However, his ‘fixed’ Alaska
model does not account for growing palaeontological evidence
of Siberian and Baltican Palaeozoic fauna in Alaskan terranes
(Blodgett et al. 2002; Dumoulin et al. 2002).
There are several modified versions of the rotation model that
include differential motion within the AACP (Miller et al. 2006),
and pre-rotation displacement relative to North America
(Sweeney 1982). The Neoproterozoic stratigraphy and Palaeozoic
F. A. MACDONALD
palaeobiogeographic affinities of the AACP do not necessarily
contradict the rotation model for the opening of the Arctic
Ocean, but they do indicate that the pre-Devonian AACP was
exotic to Laurentia (Macdonald et al. 2009b). These models are
difficult to test directly because most palaeomagnetic studies in
northern Alaska have been compromised by a pervasive Late
Cretaceous overprint (Plumley et al. 1989; Stone 1989).
Geochronological constraints
A minimum age constraint on the Katakturuk Dolomite is provided
by Late Cambrian trilobites in the upper portion of the overlying
Nanook Limestone (Blodgett et al. 1986). A lower age constraint
is provided by map unit O. G., which is stratigraphically below
the Mt. Copleston volcanic rocks and the Hula Hula diamictite,
and contains c. 760 Ma (206Pb/207Pb LA-ICPMS) detrital zircon
grains (Macdonald et al. 2009b). A coarse, diabase sill within
map unit O. G., previously assumed to be coeval with the Mt.
Copleston volcanic rocks, yielded a whole rock Rb –Sr isochron
age of 801 + 20 Ma (Moore 1987; Clough & Goldhammer
2000); however, recent U –Pb dates of badellyite in these sills
suggest they are Cretaceous in age (Macdonald 2009).
Discussion
A glacial origin of the Hula Hula diamictite is indicated by
the presence of bed-penetrating outsized clasts interpreted as
dropstones and the association with a geochemically and
 HULA HULA DIAMICTITE AND KATAKTURUK DOLOMITE
sedimentologically distinct overlying dark-coloured limestone.
Although the bulk of the Mt. Copleston volcanic rocks are below
the Hula Hula diamictite with a few small flows inter-fingering
with the lowermost Hula Hula diamictite, it is not certain that
the Mt. Copleston volcanic rocks mark the onset of glaciation, as
there could have been glacial activity prior to the emplacement
of the basalts that failed to leave a record. Nonetheless, a potential
interpretation of the stratigraphy of the Hula Hula diamictite along
Eustik Creek is that the basal c. 10 m of massive diamictite with
interfingering basalts were deposited as glaciomarine deposits
during the encroachment of sea ice; the middle c. 30 m of
millimetre-laminated silts with occasional debris flows and rare
dropstones formed under total ice cover; and the upper c. 5 m of
massive diamictite represents the ice-retreat phase.
Above the Hula Hula diamictite, C-isotope values rise from
–2‰ to þ6‰ in the Sadlerochit Mountains and from þ1‰ to
þ8‰ near Kikitak Mountain (Fig. 34.4). Although it is not clear
why values are more enriched in the deeper-water sections, the
positive, concave trend, and the extremely enriched values are
typical of Cryogenian post-glacial carbonates (Halverson et al.
2005), and the basal negative anomaly is similar in magnitude to
the basal Rasthof Formation (Yoshioka et al. 2003). An early
Cryogenian age of the Hula Hula diamictite is also suggested by
the presence of ‘roll-up’ microbial structures in the overlying
dark limestones, which are reminiscent of the basal Rasthof cap
carbonate of Northern Namibia (Hoffman et al. 1998).
C-isotope profiles of the Nularvik dolomite (K2) display an
inverted S-shaped profile with a nadir at –2‰ (Fig. 34.5). Normal-
ized for thickness (and excluding isotopic values of cements), this
isotopic profile is similar to that of the Ediacaran basal Doushantuo
in South China (Jiang et al. 2003; Zhou & Xiao 2007), which has
been dated at 635.2 + 0.6 Ma (Condon et al. 2005). The C-isotope
profile is also reminiscent of slope sections of the Keilberg Fm. in
northern Namibia, where underlying glacial deposits have been
dated at 635 + 0.5 Ma (Hoffmann et al. 2004), and shelf sections
are 3–4‰ lighter than foreslope sections (Hoffman et al. 2007).
Assuming that carbon is well-mixed in the oceans and on plat-
forms, the relatively enriched values of the Nularvik dolomite
suggest it was deposited early compared to shelf sections in
northern Namibia, and then truncated by exposure surfaces
before seawater reached extremely negative values.
Funnel-shaped calcite and silica cements in unit K2 are similar to
those in the basal Ediacaran Ol cap carbonate of Mongolia (Macdo-
nald et al. 2009a), in the Keilberg Fm. Of Northern Namibia, and in
the basal Doushantuo in South China (Macdonald, unpublished
data). As these are laterally equivalent with tubestone stromatolite
bioherms and are reminiscent in plan view, they may be a related
facies that is characteristic of basal Ediacaran cap carbonates.
The crystal fans in the Nularvik dolomite are formed in grain-
stone with multiple exposure surfaces, pervasive cements, and
broken and brecciated beds (Clough & Goldhammer 2000; Macdo-
nald et al. 2009b). Occasionally, individual fans are tipped over on
their side from the buckling of teepees. C-isotope values are highly
variable through this interval. Together, these data suggest the fans
were formed in a restricted, lagoonal setting with multiple
exposure surfaces (Macdonald et al. 2009b). This is a very differ-
ent depositional environment than the settings for sea-floor pre-
cipitate crystal fan development in other basal Ediacaran cap
carbonates (Peryt et al. 1990; James et al. 2001; Hoffman &
Halverson 2008). Isotopic scatter, evidence of exposure and perva-
sive cementing is also a common feature in other basal Ediacaran
cap carbonates that were deposited in basins lacking active stretch-
ing, such as the upper portion of the Doushantuo Fm. in South
China (Jiang et al. 2003) and the Jbeliat dolostone in Mauritania
(Hoffman & Schrag 2002; Shields et al. 2006). This pre-
transgression shoaling could be a product of isostatic rebound out-
pacing passive subsidence and post-glacial eustatic sea-level rise.
Although no late Cryogenian glacial diamictites have been
identified, the Nularvik dolostone contains an isotopic profile
385
and sedimentary textures in a particular order that are both charac-
teristic of basal Ediacaran cap carbonates globally (Allen &
Hoffman 2005). The lack of glacial deposits can be attributed to
poor preservation potential as the glacio-eustatic sea-level drop
left the carbonate platform exposed or possibly covered with
grounded ice until the post-glacial transgression. Glacial diamic-
tites are also rare under the Keilberg cap carbonate on the Otavi
platform in Northern Namibia (Hoffman & Halverson 2008). In
the Sadlerochit and Shublik Mountains, a late Cryogenian glacia-
tion may be contained in a silicified surface, with the overlying
Nularvik dolomite representing the basal Ediacaran cap carbonate.
The Katakturuk Dolomite has previously been correlated with
the early Neoproterozoic deposits of NW Laurentia (Rainbird
et al. 1996), such as the Lower Tindir Group of the Yukon-Alaska
border area (Young 1982; Macdonald et al. 2010a), the
pre-717.4 Ma Fifteenmile Group in the Ogilvie Mountains (Mac-
donald et al. 2010b), the Little Dal Group in the Mackenzie Moun-
tains (Aitken 1981) and the pre-716.3 Ma Shaler Supergroup of
Victoria Island (Young 1981; Macdonald et al. 2010b). This corre-
lation was based in large part on palaeogeographical reconstruc-
tions that tie the AACP to northwestern Laurentia. However, the
Katakturk Dolomite is younger than c. 760 Ma and likely Cryogen-
ian to Ediacaran in age (Macdonald et al. 2009b). Using the
801 + 20 Ma Rb –Sr date, Clough & Goldhammer (2000)
argued that the Mt. Copleston volcanic rocks represent a rifting
episode coeval with the c. 780 Ma Gunbarrel Event (Park et al.
1995; Harlan et al. 2003), and that the Katakturuk Dolomite was
deposited during the thermal subsidence stage on a passive carbon-
ate ramp. Instead, the Mt. Copleston basalts could be correlative
with the Franklin igneous event. However, NW Laurentia lacks
potentially correlative Cryogenian to Ediacaran successions domi-
nated by platformal carbonate rocks. At present, the origin and
Neoproterozoic position of the NSST and AACP are unknown;
however, the Neoproterozoic and Palaeozoic strata of the AACP
potentially have features in common with peri-Siberian and Balti-
can terranes (Colpron & Nelson 2009; Macdonald et al. 2009b).
Further work is necessary to test these correlations.
I thank my field assistants B. Black, P. Kreycik and W. Macdonald for pushing
through inclement weather and difficult terrain. I thank D. Schrag and G. Eischied
for use of and help in Harvard’s Paleooceanography Laboratory. U. Bold,
K. Knudson, W. Macdonald and K. Wecht are thanked for helping to prepare
samples. I am grateful to P. Hoffman and the NSF Arctic program for providing
financial support, and VECO polar resources for providing logistical support. I
also thank the GSA for a student research grant. I would like to thank
P. Hoffman, D. Jones and J. Clough for helpful discussions and comments
throughout this work. Finally, I thank B. McClelland and E. Arnaud for their
helpful comments that greatly improved the manuscript. This represents a con-
tribution of the IUGS- and UNESCO-funded IGCP (International Geoscience
Programme) project #512.
References
Aitken, J. D. 1981. Stratigraphy and sedimentology of the Upper Proter-
ozoic Little Dal Group, Mackenzie Mountains, Northwest Territories.
In: Campbell, F. H. A. (ed.) Proterozoic Basins of Canada. Geologi-
cal Survey of Canada Paper 81-10.
Allen, P. A. & Hoffman, P. F. 2005. Extreme winds and waves
in the aftermath of a Neoproterozoic glaciation. Nature, 433,
123– 127.
Allison, C. W. A., Young, G. M., Yeo, G. M. & Delaney, G. D. 1981.
Glaciogenic rocks of the Upper Tindir Group, east-central Alaska. In:
Hambrey, M. J. & Harland, W. B. (eds) Earth’s Pre-Pleistocene
Glacial Record. Cambridge University Press, Cambridge.
Babcock, L. E., Blodgett, R. B. & St. John, J. 1994. New Late(?)
Proterozoic-age formations in the vicinity of Lone Mountain,
McGrath Quadrangle, West-Central Alaska. In: Till, A. B. &
Moore, T. E. (eds) Geologic Studies in Alaska by the U.S. Geological
Survey, 1993. United States Government Printing Office,
Washington.
386 F. A. MACDONALD
Bader, J. W. & Bird, K. J. 1986. Geologic map of the Demarcation Point,
Mt. Michelson, Fluxman Island quadrangles, northeastern Alaska.
U.S. Geological Survey Miscellaneous Investigations 1791, 1 sheet,
scale 1:250,000.
Blodgett, R. B., Clough, J. G., Dutro, J. T., Ormiston, A. R., Palmer,
A. R. & Taylor, M. E. 1986. Age revisions of the Nanook Limestone
and Katakturuk Dolomite, northestern Brooks Range, Alaska. In:
Bartsch-Winkler, S. & Reed, K. M. (eds) Geological Studies in
Alaska by the Geological Survey during 1985. US Geological
Survey Circular 978.
Blodgett, R. B., Rohr, D. M. & Boucot, A. J. 2002. Palaeozoic links
among some Alaskan accreted terranes and Siberia based on mega-
fossils. In: Miller, E. L., Grantz, A. & Klemperer, S. L. (eds)
Tectonic Evolution of the Bering Shelf– Chukchi Sea –Arctic
Margin and Adjacent Landmasses. Geological Society of America
Special Paper 360, Boulder, Colorado.
Carey, S. W. 1955. The orocline concept in geotectonics. Royal Society of
Tasmania Proceedings, 89, 255– 288.
Carey, S. W. 1958. Continental drift. In: Carey, S. W. (ed.) Continental
Drift, a Symposium. University of Tasmania, Hobart, Australia.
Clough, J. G. 1989. General stratigraphy of the Katakturuk Dolomite in
the Sadlerochit and Shublik Mountains, Arctic National Wildlife
Refuge, northeastern Alaska. Alaska Division of Geological &
Geophysical Surveys, Public Data File 89-4a, 1– 11.
Clough, J. G. & Goldhammer, R. K. 2000. Evolution of the Neoproter-
ozoic Katakturuk dolomite ramp complex, northeastern Brooks
Range, Alaska. In: Grotzinger, J. P. & James, N. P. (eds) Carbonate
Sedimentation and Diagenesis in the Evolving Precambrian World.
SEPM Special Publication, No. 67. Society of Sedimentary
Geology, Tulsa, Oklahoma.
Colpron, M. & Nelson, J. L. 2009. A Palaeozoic Northwest Passage:
incursion of Caledonian, Baltican and Siberian terranes into eastern
Panthalassa, and the early evolution of the North American Cordil-
lera. In: Cawood, P. & Kroner, A. (eds) Accretionary Orogens
through Space and Time. Geological Society, London, Special
Publications, 318, 273–307.
Colpron, M., Nelson, J. L. & Murphy, D. C. 2007. Northern Cordilleran
terranes and their interactions through time. GSA Today, 17, 1 –7.
Condon, D. J., Zhu, M., Bowring, S. A., Wang, W., Yang, A. & Jin, Y.
2005. U– Pb ages from the Neoproterozoic Doushanto Formation,
China. Science, 308, 95 – 98.
Corsetti, F. A. & Grotzinger, J. P. 2005. Origin and significance of tube
structures in Neoproterozoic post-glacial cap carbonates: example
from Noonday Dolomite, Death Valley, United States. Palaios, 20,
348– 363.
Dumoulin, J. A. 1988. Stromatolite- and coated-grain-bearing carbonate
rocks of the western Brooks Range. In: Galloway, J. P. & Hamil-
ton, T. D. (eds) Geologic Studies in Alaska by the U.S. Geological
Survey during 1987. United States Government Printing Office,
Washington.
Dumoulin, J. A. & Harris, A. G. 1994. Depositional framework and
regional correlation of pre-Carboniferous metacarbonate rocks of
the Snowden Mountain area, Central Brooks Range, Northern
Alaska. US Geological Survey Professional Paper 1545, 1 – 55.
Dumoulin, J. A., Harris, A. G., Gagiev, M., Bradley, D. C. &
Repetski, J. E. 2002. Lithostratigraphic, conodont, and other faunal
links between lower Palaeozoic strata in northern and central
Alaska and northeastern Russia. In: Miller, E. L., Grantz, A. &
Klemperer, S. L. (eds) Tectonic Evolution of the Bering Shelf –
Chukchi Sea– Arctic Margin and Adjacent Landmasses. Geological
Survey of America Special Paper 360, Boulder, Colorado.
Dutro, J. T. 1970. Pre-Carboniferous carbonate rocks, northeastern
Alaska. In: Adkison, W. L. & Brosge, M. M. (eds) Proceed-
ings of the Geological Seminar on the North Slope of Alaska,
American Association of Petroleum Geologists, Pacific Section.
Los Angeles, CA.
Fuis, G. S., Moore, T. E. et al. 2008. Trans-Alaska Crustal Transect and
continental evolution involving subduction underplating and syn-
chronous foreland thrusting. Geology, 36, 267–270.
Grantz, A., Eittreim, S. & Dinter, D. A. 1979. Geology and tectonic
development of the continental margin north of Alaska. Tectonophy-
sics, 59, 263– 291.
Halverson, G. P., Hoffman, P. F., Schrag, D. P., Maloof, A. C. &
Rice, A. H. N. 2005. Toward a Neoproterozoic composite
carbon-isotope record. Geological Society of America Bulletin, 117,
1181– 1207.
Hamilton, W. 1970. The Uralides and the motion of the Russian and
Siberian Platforms. Geological Society of America Bulletin, 81,
2553– 2576.
Harlan, S. S., Heaman, L. M., LeCheminant, A. N. & Premo, W. R.
2003. Gunbarrel mafic magmatic event: a key 780 Ma time marker
for Rodinia plate reconstructions. Geology, 31, 1053– 1056.
Harris, A. G., Lane, R. H. & Tailleur, I. L. 1990. Conodont thermal
maturation patterns in Palaeozoic and Triassic rocks, Northern
Alaska – geological and exploration implications. In: Grantz, A.,
Johnshon, L. & Sweeney, J. F. (eds) The Arctic Ocean Region,
DNAG Series L. Geological Society of America, Boulder, CO.
Hoffman, P. F. & Halverson, G. P. 2008. Otavi Group of the western
Northern Platform, the Eastern Kaoko Zone and the western Northern
Margin Zone. In: Miller, R. M. (ed.) The Geology of Namibia, vol. 2.
Handbook of the Geological Survey of Namibia, Windhoek.
Hoffman, P. F. & Schrag, D. P. 2002. The snowball Earth hypothesis;
testing the limits of global change. Terra Nova, 14, 129–155.
Hoffman, P. F., Kaufman, A. J. & Halverson, G. P. 1998. Comings and
goings of global glaciations on a Neoproterozoic tropical platform in
Namibia. GSA Today, 8, 1– 9.
Hoffman, P. F., Halverson, G. P., Domack, E. W., Husson, J. M.,
Higgins, J. A. & Schrag, D. P. 2007. Are basal Ediacaran
(635 Ma) post-glacial ‘cap dolostones’ diachronous? Earth and Pla-
netary Science Letters, 258, 114–131.
Hoffmann, K. H., Condon, D. J., Bowring, S. A. & Crowley, J. L.
2004. U– Pb zircon date from the Neoproterozoic Ghaub
Formation, Namibia: constraints on Marinoan glaciation. Geology,
32, 817– 820.
James, N. P., Narbonne, G. M. & Kyser, T. K. 2001. Late Neoprotero-
zoic cap carbonates; Mackenzie Mountains, northwestern Canada;
precipitation and global glacial meltdown. Canadian Journal of
Earth Sciences, 38, 1229–1262.
Jiang, G., Kennedy, M. J. & Christie-Blick, N. 2003. Stable isotopic
evidence for methane seeps in Neoproterozoic postglacial cap car-
bonates. Nature, 426, 822– 826.
Johnston, S. T. 2001. The Great Alaskan Terrane Wreck: reconciliation
of palaeomagnetic and geological data in the northern Cordillera.
Earth and Planetary Science Letters, 193, 259–272.
Lane, L. S. 1997. Canada Basin, Arctic Ocean: evidence against a
rotational origin. Tectonics, 16, 363–387.
Lane, L. S. 2007. Devonian –Carboniferous palaeogeography and oroge-
esis, northern Yukon and adjacent Arctic Alaska. Canadian Journal
of Earth Sciences, 44, 679– 694.
Leffingwell, E. 1919. The Canning River region, northern Alaska. US
Geological Survey Professional Paper 109.
Macdonald, F. A. 2009. Neoproterozoic stratigraphy of Alaska and
Mongolia. PhD thesis, Harvard University.
Macdonald, F. A. & Cohen, P. A. 2011. The Tatonduk inlier, Alaska-
Yukon border. In: Arnaud, E., Halverson, G. P. & Shields, G.
(eds) The Geological Record of Neoproterozoic Glaciations.
Geological Society, London, Memoirs, 36, 389–396.
Macdonald, F. A., Jones, D. S. & Schrag, D. P. 2009a. Stratigraphic
and tectonic implications of a new glacial diamictite-cap carbonate
couplet in southwestern Mongolia. Geology, 37, 123–126.
Macdonald, F. A., McClelland, W. C., Schrag, D. P. & Macdonald,
W. P. 2009b. Neoproterozoic glaciation on a carbonate platform
margin in Arctic Alaska and the origin of the North Slope subterrane.
Geological Society of America Bulletin, 121, 448–473.
Macdonald, F. A., Cohen, P. A., Dudás, F. O. & Schrag, D. P. 2010a.
Early Neoproterozoic scale microfossils in the Lower Tindir Group
of Alaska and the Yukon Territory. Geology, 38, 143–146.
Macdonald, F. A., Schmitz, M. D. et al. 2010b. Calibrating the Cryo-
genian. Science, 327, 1241–1243.
Miller, E. L., Toro, J. et al. 2006. New insights into Arctic palaeo-
geography and tectonics from U –Pb detrital zircon geochronology.
Tectonics, 25, 1 –19.
Molenaar, C. M., Mull, C. G. & Swauger, D. A. 1987. Geologic fea-
tures of Ignek Valley and adjacent mountains, northeastern Alaska.
 HULA HULA DIAMICTITE AND KATAKTURUK DOLOMITE
Centennial Field Guide Volume 1: Cordilleran Section of the Geo-
logical Society of America.
Moore, T. E. 1987. Geochemistry and tectonic setting of some volcanic
rocks of the Franklinian assemblage, central and eastern Brooks
Range. In: Tailleur, I. & Weimer, P. (eds) Alaskan North Slope
Geology. SEPM, Pacific Section, California and Alaska Geological
Society, Alaska.
Moore, T. E., Wallace, W. K., Mull, C. G., Adams, K. E., Plafker, G.
& Nokleberg, W. J. 1997. Crustal implications of bedrock geology
along the Trans-Alaska Crustal Transect (TACT) in the Brooks
Range, northern Alaska. Tectonics, 102, 20 645–20 684.
Oldow, J. S., Lallemant, A., Julian, F. E. & Seidensticker, C. M.
1987. Ellesmerian (?) and Brookian deformation in the Franklin
Mountains, northeastern Brooks Range, Alaska, and its bearing on
the origin of the Canada Basin. Geology, 15, 37 – 41.
Park, J. K., Buchan, K. L. & Harlan, S. S. 1995. A proposed giant
radiating dyke swarm fragmented by the separation of Laurentia
and Australia based on palaeomagnetism of ca. 780 Ma mafic intru-
sions in western North America. Earth and Planetary Science
Letters, 132, 129– 139.
Peryt, T. M., Hoppe, A., Bechstadt, T., Koster, J., Pierre, C. &
Richter, D. K. 1990. Late Proterozoic aragonitic cement crusts,
Bambui Group, Minas Gerais, Brazil. Sedimentology, 37, 279–286.
Plumley, P. W., Vance, M. S. & Milazzo, G. 1989. Structural and
palaeomagnetic evidence for Tertiary bending of the Eastern
Brooks Range Flexure, Alaska. In: Hillhouse, J. W. (ed.) Deep
Structure and Past Kinematics of Accreted Terranes, Geophysical
Monograph 50. American Geophysical Union, Washington, DC.
Pruss, S. B., Bosak, T., Macdonald, F. A., McLane, M. & Hoffman,
P. F. 2010. Microbial facies in a Sturtian cap carbonate, the Rasthof
Formation, Otavi Group, northern Namibia. Precambrian Research,
181, 187– 198.
Rainbird, R. H., Jefferson, C. W. & Young, G. M. 1996. The early
Neoproterozoic sedimentary Succession B of Northwestern Lauren-
tia: correlations and palaeogeographic significance. Geological
Society of America Bulletin, 108, 454– 470.
Reed, B. L. 1968. Geology of the Lake Peters area, northeastern Brooks
range, Alaska. US Geological Survey Bulletin 1236, 1 –132.
Reiser, H. N. 1971. Northeastern Brooks Range – a surface expression of
the Prudhoe Bay Section. In: Adkison, W. L. & Brosge, M. M. (eds)
Proceedings of the Geological Seminar on the North Slope of Alaska.
American Association of Petroleum Geologists, Pacific Section,
Los Angeles, CA.
Reiser, H. N., Dutro, J. T., Brosge, W. P., Armstrong, A. K. &
Detterman, R. L. 1970. Progress map, geology of the Sadlerochit
387
and Shublik Mountains, Mt. Michelson C-1, C-2, C-3, C-4 Quadran-
gles, Alaska. US Geological Survey Open-file report, 70-273, scale
1:63,360, 5 sheets.
Reiser, H. N., Brosge, W. P., Dutro, J. T. & Detterman, R. L. 1980.
Geologic map of the Demarcation Point quadrangle, Alaska,
1:250,000, map I-1133. Miscellaneous Investigation Series. U.S.
Geological Survey.
Robinson, M. S., Decker, J., Clough, J. G., Reifenstuhl, R. R.,
Dillon, J. T., Combellick, R. A. & Rawlinson, S. E. 1989.
Geology of the Sadlerochit and Shublik Mountains, Alaska National
Wildlife Reserve, northeast Alaska, Professional Report 100,
1:10,000 scale. State of Alaska, Department of Natural Resources,
Division of Geological and Geophysical Survey.
Sable, E. G. 1977. Geology of the western Romanzof Mountains,
Brooks Range, northeastern Alaska: a comprehensive study of
plutonic, sedimentary, and metamorphic rocks in an eastern Brooks
Range area. Geological Survey, Washington, D.C., Professional
Paper 897.
Shields, G., Deynoux, M., Strauss, H., Paquet, H. & Nahon, D.
2006. Barite-bearing cap dolostones of the Toudeni Basin, north-
west Africa: sedimentary and isotopic evidence for methane
seepage after a Neoproterozoic glaciation. Precambrian Research,
153, 209–235.
Stone, D. B. 1989. Palaeogeography and rotations of Arctic Alaska – an
unresolved problem. In: Kissel, C. & Laj, C. (eds) Palaeomagnetic
Rotations, Continental Deformation. Kluwer Academic, Norwell,
MA.
Sweeney, J. F. 1982. Mid-Palaeozoic travels of Arctic Alaska. Nature,
298, 647–649.
Wallace, W. K. & Hanks, C. R. 1990. Structural provinces of the North-
eastern Brooks Range, Arctic National Wildlife Refuge, Alaska.
AAPG Bulletin, 74, 1100– 1118.
Yoshioka, H., Asahara, Y., Tojo, B. & Kawakami, S. 2003. Systematic
variations in C, O and Sr isotopes and elemental concentrations in
Neoproterozoic carbonates in Namibia: implications for a glacial to
interglacial transition. Precambrian Research, 124, 69 – 85.
Young, G. M. 1981. The Amundsen Embayment, Northwest Territories:
relevance to the Upper Proterozoic evolution of North America. In:
Campbell, F. H. A. (ed.) Proterozoic Basins of Canada. Geological
Survey of Canada Paper 81-10.
Young, G. M. 1982. The late Proterozoic Tindir Group, east-central
Alaska; evolution of a continental margin. Geological Society of
America Bulletin, 93, 759–783.
Zhou, C. & Xiao, S. 2007. Ediacaran d13C chemostatigraphy of South
China. Chemical Geology, 237, 107–126.
 Chapter 35

The Tatonduk inlier, Alaska –Yukon border

FRANCIS A. MACDONALD* & PHOEBE A. COHEN

Department of Earth and Planetary Sciences, Harvard University, Cambridge, MA 02138, USA
*Corresponding author (e-mail: fmacdon@fas.harvard.edu)

Abstract: Glaciogenic deposits of the Rapitan and Hay Creek Groups are exposed in the Tatonduk inlier of east-central Alaska and the
western Yukon. The Rapitan Group ranges in thickness from c. 50 to 700 m with Fe-formation common in the upper 10 m. In the most
distal settings, the Rapitan Group is separated from the diamictite of the Hay Creek Group by over 100 m of sandstone and siltstone;
however, the Hay Creek Group contains large erosive surfaces and cannibalizing breccia, and rarely preserves strata between the two
glaciogenic deposits. The diamictite of the Hay Creek Group is capped by a white- to buff-coloured dolostone with pseudo-teepee struc-
tures, bed-parallel, isopachous sheet-crack cements, and a depleted C-isotope signature. Late Neoproterozoic glacial deposits in the
Tatonduk inlier were formerly assigned to the Tindir Group. To simplify the nomenclature in the northwestern Canadian Cordillera,
the Tindir Group was abandoned and replaced with nomenclature consistent with that of the Windermere Supergroup in the Mackenzie
Mountains. The mixed lithology and anchizone-grade metamorphism distinguish the Rapitan and Hay Creek Groups in the Tatonduk
inlier as attractive future targets for integrated micropalaeontology, geochemistry, palaeomagnetism and geochronology.

Cairnes (1914) referred to Precambrian stratigraphy exposed along
the international border between the United States and Canada
as the Tindir Group. Mertie (1930, 1933) later described the strati-
graphy (and natural history of the region) in remarkable detail and
divided the Tindir Group into seven units. Brabb & Churkin (1969)
produced an excellent map of the geology on the Alaskan side of
the border, while the Tindir Group was mapped on the Yukon
side of the border by Norris (1982); however, exact correlations
of specific units of the Tindir Group across the border remained
ambiguous until more recent mapping and compilation (Dover
1992; Van Kooten et al. 1997) and integrated litho- and chemostra-
tigraphy (Macdonald et al. 2010a, b, 2011; Macdonald & Roots
2010). To improve the consistency of geological maps of the
Yukon and promote the synthesis of geological data, Macdonald
et al. (2011) reassigned the Lower Tindir Group to the Pinguicula
and Fifteenmile Groups of the Mackenzie Mountains Supergroup
and the Upper Tindir Group to the Rapitan, Hay Creek and
‘upper’ groups of the Windermere Supergroup (Figs 35.1 & 2).
The Upper and Lower Tindir Groups were first distinguished by
Payne & Allison (1981), despite the fact that no unconformable
contacts have been observed between the two (Young 1982).
The Lower Tinder Group consists primarily of dolomite and
shale that are commonly cut with north-trending mafic dykes
(Van Kooten et al. 1997). Unlike the Lower Tindir Group, the
units of the Upper Tindir Group are not cut by mafic dykes (Van
Kooten et al. 1997). Young (1982) separated the Upper Tindir
Group into five units in ascending stratigraphic order from unit 1
to unit 5: unit 1 consists primarily of mafic volcanic rocks; unit
2, purple mudstones and diamictite, including Fe-formation; unit
3, an additional massive diamictite; unit 4, platfomal dolomites
and green to light grey shales; unit 5, grey to black shales and lime-
stones. In the updated stratigraphic framework (Macdonald et al.
2011), unit 1 is renamed the Pleasant Creek volcanic rocks, unit
2 is the Rapitan Group, unit 3 and the ‘teepee’ dolomite at the
base of unit 4 constitute the Hay Creek Group, and the rest of
unit 4 and unit 5 are assigned to the ‘upper’ group. The informal
upper group is a provisional name that encompasses the Sheepbed,
Gametrail, Blueflower and Risky formations in the Mackenzie
Mountains (e.g. Aitken 1989; Dalrymple & Narbonne 1996; Mac-
Naughton et al. 2000, 2008) and equivalent strata in the Ogilvie
Mountains (Macdonald et al. 2011).
Exposures of the Windermere Supergroup in the Tatonduk inlier
are typically limited to the walls of creek beds. Allison et al. (1981)
followed Mertie (1933) in designating the exposures along the
Tatonduk River as the type locality of the diamictite-bearing
units. This choice is due in part to the relative ease of access, but
is problematic because of structural complications and the pres-
ence of a major disconformity under the diamictite in the Hay
Creek Group. Owing to lack of exposure, the critical features of
both diamictites are not present at one single locality. Along
Pass Creek (Fig. 35.3), Fe-formation of the Rapitan Group and por-
tions of both diamictites are exposed, but the overlying teepee
dolomite is not present. Along Hard Luck Creek (Fig. 35.3), the
contact between the Fe-formation and underlying diamictite can
be observed, and the teepee dolomite is present, although the dia-
mictite of the Hay Creek Group is not exposed. This lack of
exposure coupled with the presence of multiple disconformities
creates a significant challenge (and many uncertainties) in regional
stratigraphic correlations.
Structural framework
Neoproterozoic diamictites of the Windermere Supergroup are
exposed through an erosional window, referred to as the Tatonduk
inlier, which extends across the Canadian border into Alaska
(Fig. 35.2). Present exposures are the product of a mid-Cretaceous
to Palaeogene foreland fold-and-thrust belt (Norris 1972) associ-
ated with the underplating of the Yakutat Block (Fuis et al.
2008). The Tatonduk inlier protrudes from the elbow of the
Ogilvie deflection, where the orientation of folds and thrusts
transition from north-vergent to west-vergent due to Cenozoic
activity on the Kaltag fault and associated dextral transpressional
structures (Norris 1972). Anchizone-grade metamorphism is
suggested by the presence of economic oil deposits in the overlying
Phanerozoic strata sourced in part from Palaeozoic shales (Van
Kooten et al. 1997).
Late Neoproterozoic strata in the Tatonduk inlier are a north-
western continuation of the Windermere Supergroup in the Mack-
enzie (Aitken 1989), Wernecke (Pyle et al. 2004) and Ogilvie
Mountains (Mustard & Roots 1997), for which deposition com-
menced with rifting of the northwestern margin of Laurentia
(Stewart 1975). Neoproterozoic strata in the Tatonduk inlier
have also been correlated with the Katakturuk Dolomite in the
northeastern Brooks Range of Arctic Alaska (Rainbird et al.
1996); however, the Katakturuk Dolomite was likely deposited

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 389– 396. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.35
390 F. A. MACDONALD & P. A. COHEN

Ogilvie Mountains surfaces and breccias. The Hard Luck Creek Fault (HLF) marks
an abrupt expansion of the stratigraphy, and can be extended to
Tatonduk Inlier Coal Creek Inlier
Macdonald the SE into the Ogilvie Mountains where it has been named the
et al., 2011

Macdonald &
Young, 1992

et al., 2010b;
Van Kooten
et al., 2010a

Roots, 1997
Roots, 2010
Thompson
Mt. Harper Fault, and roughly marks a former rift shoulder of

Macdonald

et al. 1994;
et al., 1997

Macdonald

Mustard &
the Laurentian margin (Mustard & Roots 1997). The HLF and
other Precambrian faults in the region were inverted during mid-
Cretaceous to Palaeogene shortening (Brabb & Churkin 1969;
Jones Ridge & Funnel Ck., Bouvette Bouvette Van Kooten et al. 1997).
Bouvette Fms., Palaeozoic Adams, Jones Ridge Formation Formation
Cambrian undiff. Hillard, Jones Formation
Undiff. Ridge Fms. PH5 PH5

Stratigraphy
Windermere Supergroup

pCtl

Upper Harper Group

5 5 PH4 PH4

Upper Tindir Group

Upper Tindir Group
Upper Tindir Group

“upper” The stratigraphy of the diamictite-bearing units in the Tatonduk

group
Hay Creek
Group
Rapitan Gp.
Pleasant Creek/Mt.
Harper volcanic rocks
1 717.43 ± 0.14 Ma
Fig. 35.1. Nomenclature chart depicting names used by different authors to
describe Neoproterozoic stratigraphy in the Tatonduk and Coal Creek inliers.
Dates are U –Pb CA-IDTIMS zircon ages (Macdonald et al. 2010b).
on a separate margin, as it is doubtful that the pre-Mississippian
Arctic Alaska-Chukotka Plate was part of Laurentia (McClelland,
1997; Blodgett et al. 2002; Dumoulin et al. 2002).
Although unconformities have not been observed in the Wind-
ermere Supergroup of the Tatonduk inlier, multiple disconformi-
ties are present in the Hay Creek Group marked by erosional
4b inlier was reviewed by Allison et al. (1981) and described in
4 PH3 PH3 detail by Young (1982). Additional measured sections are pre-
4a pCtss sented in Macdonald et al. (2010a, supplementary material,
3b
3 PH2 2010b, supplementary material, 2011). Diamictite of the Rapitan
3a
Group is underlain by the Pleasant Creek volcanic rocks, which
2 pCtr 2 PH1 PH1/PH2 are up to 200 m thick and consist chiefly of amygdaloidal pillow
2
1 pCtbs 1 MHVC 1
MHVC basalt and cherty hyaloclastic breccia, with minor tuff, shale and
conglomerate. The Rapitan Group is composed primarily of fine-
2 716.47 ± 0.24 Ma unconformity laminated purple and red mudstone and siltstone that are sprinkled
with dolomite and basaltic lonestones. The upper c. 15 m of the
Rapitan Group commonly hosts a massive diamictite and
Fe-formation. The Rapitan Group is overlain by the Hay Creek
Group, which is composed of up to 150 m of planar laminated
siltstone and sandstone with minor dolomite marl, massive diamic-
tite, and dolomite breccia capped by a white to buff-coloured
dolostone with bed-parallel cements. The uppermost dolostone
of the Hay Creek Group is less than 5 m thick and rests discon-
formably on all of the underlying units of the Windermere Super-
group and the Pleasant Creek volcanic rocks (Fig. 35.4). The upper
group is composed largely of black shale with minor allodapic
carbonate.

Fig. 35.2. Location map of the NW

Cordillera with Cryogenian and Ediacaran
strata in grey. AACP, Arctic Alaska-
Chukotka Plate. Inset shows the
distribution of the Mackenzie Mountains
and Windermere Supergroups in the
Tatonduk Inlier. Teeth on thrust faults.
 THE TATONDUK INLIER, ALASKA– YUKON BORDER
Glaciogenic deposits and associated strata
Rapitan Group
The Rapitan Group (formerly Upper Tindir unit 2) is exposed near
the international border along Pleasant Creek and in outcrops close
to the Tatonduk River (Macdonald et al. 2010b). These strata are
chiefly composed of fine-laminated purple and red mudstone and
siltstone speckled with dolomite gravel lonestones. Scattered
throughout the siltstone are thin diamictite beds (,1 m) dominated
by clasts of dolomite, clastic grit and volcanic cobbles. The
Rapitan Group also contains faceted clasts, boulders with stria-
tions, slumped beds, flame structures, groove casts and flute struc-
tures (Young 1982). The upper c. 15 m of the Rapitan Group
(Fig. 35.5a; section T709) hosts multiple c. 10-cm-thick beds of
Fe-formation, which are interbedded with a laminated diamictite
with bed-penetrating, outsized clasts. The majority of clasts in
the Rapitan Group consist of dolomite derived from the underlying
Fifteenmile Group. Where the upper contact of the Fifteenmile
Group is exposed, it is overlain by a well-sorted, imbricated car-
bonate matrix conglomerate with abundant dolostone clasts of
variable size, interpreted as a debris flow, which is followed by
the parallel-bedded siltstone and sandstone of the Hay Creek
Group. The thickness of the Rapitan Group varies greatly from
,50 m near Pleasant Creek to .700 m c. 20 km to the NW
(Young 1982). Palaeocurrent measurements in the interbedded
siltstone suggest a west-facing margin (present coordinates)
(Young 1982).
391
Fig. 35.3. Geological map of exposures
along the Tatonduk River and Hard Luck
Creek.
Calcareous concretions and jasper lenses are present in the
Fe-formation, while foreign clasts are rare. Chemical analyses
of the Fe-formation yield FeO concentrations up to 50%, high
SiO2/Al2O3, and depletions in most other elements, suggesting
low clastic deposition and a chemical silica contribution
(Young 1982).
Hay Creek Group ( formerly units 3a, 3b and 4a of the Upper
Tindir Group)
In the most western sections along Pass Creek (T710), the Rapitan
Group is overlain by c. 140 m of planar bedded, siltstone, sand-
stone and marl. These beds lack any evidence of tidal influence,
lack lonestones, are commonly stacked in fining-upward Bouma
sequences, and as such, are interpreted as deep-water turbidites.
The turbidite beds are succeeded by an additional diamictite unit,
but the contact between the two is not exposed. In section T710 the
diamictite of the Hay Creek Group is 22 m thick and consists of
10 m of massive diamictite, 10 m of laminated centimetre-beds
of siltstone and an additional 2 m of stratified diamictite (Fig.
35.5b). The massive diamictite is clast-dominated with boulders
of dolomite, and cobbles of Fe-formation, siltstone, conglomerate
and basalt in a marly pink matrix. The bedded diamictite has the
same clast composition as the rock below it, but clasts are slightly
smaller with no boulders, and a matrix of purple silt. Approxi-
mately 20 km to the NW of Pass Creek (section Y.S. 9), still
west of the HLF, the diamictite of the Hay Creek Group is poorly
392 F. A. MACDONALD & P. A. COHEN

Fig. 35.4. Chemo- and lithostratigraphy of the ‘upper’ group in east-central Alaska. Y.S. indicates section from Young (1982). Sections have been projected along the
Hard Luck Fault (i.e. to the NW) to an east– west line, with distances between sections estimated from the projection. Note the change in scale for two sections SW of the
Hard Luck fault, which are approximately four times as thick as sections NE of the fault. See Figure 35.3 for section locations.

exposed, but lonestones c. 2 m in diameter are present, and the dia-
mictite is capped by a buff-coloured dolostone (Young 1982).
Along the Tatonduk River, Young (1982) described a massive,
over 250-m-thick, crudely stratified, purple diamictite interbedded
with minor amounts of contorted purple mudstone and lenses
of chert. We assign this diamictite to the Hay Creek Group.
Clasts are up to 0.6 m across, and consist predominantly of
dolomite from the Fifteenmile Group, with minor limestone,
basalt and chert. Faceted and striated clasts have been described
at this locality (Allison et al. 1981), but have not been observed
elsewhere in the Tatonduk inlier. The diamictite of the Hay
Creek Group is also very thick c. 20 km due north, NE of the
HLF (Van Kooten et al. 1997), where it cuts down into the
underlying stratigraphy.
Northeast of the HLF, the diamictite of the Hay Creek Group
is either absent, or represented by a dolomite (matrix and clast)
breccia. No foreign clasts have been identified with the exception
of some clasts of the Pleasant Creek volcanic rocks near the base.
This breccia cuts out the underlying units of the Windermere
Group and rests on the Pleasant Creek volcanic rocks (Fig. 35.4).
The overlying dolomite, ,5 m thick, rests disconformably on
underlying units of the Hay Creek Group as well as the Pleasant
Creek volcanic rocks and Rapitan Group strata. It is white to
buff-coloured dolostone with bed-parallel cements (pseudo-
teepee structures of Young 1982). These pseudo-teepees do not
show a polygonal plan-form or a concentration of cements
along the broken pieces, as is typical of teepees that are of a sub-
aerial exposure origin (Kendall & Warren 1987). Instead,
cements are isopachous and bed-parallel, and beds are contorted
and irregularly buckled, suggesting intraformational detachment
during deposition. Similar ‘sheet-crack’ cements are present in
several basal Ediacaran cap carbonates globally (Hoffman &
Macdonald 2010).
The ‘Upper’ Group
The Hay Creek Group is overlain by the ‘upper’ group (formerly
units 4b and 5 of the Upper Tindir Group), which consists of as
much as 50 m of planar laminated siltstone, sandstone and dolomi-
tic marl, and an additional sequence of black shale interbedded
with minor organic-rich limestone (Fig. 35.4; section T707).
Like the underlying units, the ‘upper’ group displays a major stra-
tigraphic expansion to the SW ranging from 40–75 m thick in the
Yukon to c. 700 m thick along the Tatonduk River in Alaska (Mac-
donald et al. 2010a). Both Hay Creek and the ‘upper’ group strata
are consistent with a SW-facing margin (present coordinates).
These ‘upper’ group units form the final clastic-carbonate cycle
prior to deposition of the Cambrian sandstone (Backbone Ranges
Formation) and the commencement of the miogeocline after
break-up of Rodinia. This group completes Windermere Super-
group sedimentation and episodic extension.
Boundary relations with overlying and underlying
non-glacial units
The basal contact of the Rapitan Group was not seen; however,
volcanic fragments similar in composition to the underlying Plea-
sant Creek volcanic rocks are common in the lower half of the
massive diamictite. The uppermost exposures of the Rapitan
Group commonly consist of Fe-formation. At some localities the
upper contact is exposed, and overlain by a well-sorted, dolomite
 THE TATONDUK INLIER, ALASKA– YUKON BORDER 393

(a) (b) erosional disconformity. These interpretations are consistent

with the erosional disconformity and breccia in the Hay Creek
Upper Unit 2 Unit 3b Group NE of the HLF where much of the underlying stratigraphy
Hard Luck Creek Pass Creek is missing. Although the top contact is not exposed SW of the HLF,
to the NE the breccia is sharply overlain by the buff-coloured
T709 T710 dolostone that defines the top of the Hay Creek Group.

v
Chemostratigraphy
10 20
Carbon-isotope values of the dolomite at the top of the Hay Creek
Group are extremely depleted, although somewhat variable
(Fig. 35.4; Macdonald et al. 2010a, supplementary data). Near
8 16 the international border (section T708), values are consistently
between –3‰ and –3.5‰, whereas near Hard Luck Creek (sec-
tions T701 and T709), values are more scattered, ranging from
–6‰ to þ2‰. In the ‘upper’ group, C-isotope values tend to be
(metres)

6 12 enriched (Fig. 35.4).

Carbon and Sr isotopes were reported from unit 5 of the Upper
Tindir Group in Canada (Kaufman et al. 1992); however, these
sections are actually correlative with the Fifteenmile Group
v
(Macdonald et al. 2010a, b, 2011). The lowest Sr-isotope values
4 8 from these sections are near 0.7064 (Macdonald et al. 2010a), a
value that is typical for pre-Sturtian carbonate rocks (Halverson
et al. 2007).

2 4 Palaeolatitude and palaeogeography

v No palaeomagnetic studies have been reported from the Tatonduk

inlier. Although the lithologies of Pleasant Creek volcanic rocks
0 0 and the Rapitan Group (basalts and Fe-rich clastic sediments,
respectively) are ideal targets, most palaeomagnetic studies in
northern Alaska are compromised by a pervasive Late Cretaceous
overprint (Plumley et al. 1989). However, palaeopoles on the
diamictite clast size Rapitan Group in the Mackenzie Mountains yield a palaeolatitude
lithologies of 6 + 48 (Park 1997). This pole is consistent with the grand mean
massive boulder pole on the contemporaneous Franklin LIP (Denyszyn et al. 2009),
unsorted and demonstrates that Laurentia straddled the equator during depo-
cobble sition of the Rapitan Group (Macdonald et al. 2010b). Although
stratified Laurentia remained at low latitudes until 615 Ma according to
gravel the controversial Long Range Dyke pole (Murthy et al. 1992;
imbricated Hodych et al. 2004), Laurentia appears to have migrated to high
conglomerate latitudes by 590 Ma (Murthy 1971). McCausland et al. (2007)
clast provide an excellent review of the Ediacaran palaeomagnetic
matrix compostion data from Laurentia.
composition v volcanic
iron- Geochronological constraints
formation clastic
dolomite Radiometric data
marl carbonate
The NNW-trending mafic dykes that intrude the Fifteenmile Group
siltstone iron- in the Tatonduk inlier have yielded a wide range of K –Ar biotite
formation
and whole-rock ages: 532 + 11 Ma, 572 + 16 Ma, 588 + 14 Ma
and 644 + 18 Ma (Van Kooten et al. 1997). These dykes have
Fig. 35.5. (a) Stratigraphy of the Rapitan Group exposed along Hard Luck not been observed to intrude any of the Windermere Supergroup
Creek, showing diamictite and its relationship to the Fe-formation. (b) units and are possibly feeder dykes for the Pleasant Creek
Stratigraphy of the Hay Creek diamictite along Pass Creek, including diamictite volcanic rocks. Because alteration can lead to argon loss (Westcott
and associated lithofacies. See Figure 35.3 for section locations. 1966), these dates are minimum age constraints for the Fifteen-
mile Group.
Along strike to the east in the Coal Creek inlier (Fig. 35.3),
clast, dolomite matrix diamictite, interpreted as a debris flow, Rapitan Group correlatives rest above the Mt. Harper volcanic
which is followed by the parallel-bedded siltstone and sandstone complex (MHVC). Zircons extracted from rhyolite in member D
with Bouma sequences of the Hay Creek Group. of the MHVC were dated at 717.43 + 0.14 Ma, and zircons
Young (1982) interpreted the diamictite of the Hay Creek Group from a tuff within the Rapitan Group correlatives were dated at
along the Tatonduk River as allochthonous in the sedimentary 716.47 + 0.24 Ma (U – Pb ID-TIMS; Macdonald et al. 2010b).
sense, having been derived from a more proximal setting to the In the Rapitan Group of the Mackenzie Mountains, a clast of leu-
east of the HLF in a massive slope failure, with an unconformable cogranite in the Rapitan Group has a U –Pb TIMS bulk zircon age
basal contact. Allison et al. (1981) interpreted this contact as an of 755 + 18 Ma (Ross & Villeneuve 1997).
394 F. A. MACDONALD & P. A. COHEN
Micropalaeontology
Fossils interpreted as microscopic flatworm impressions by
Allison (1975), but possibly sponge spicules (Andy Knoll
pers. comm.), have been described in the shale interbedded
with basalt in the lower portion of the Windermere Supergroup.
It is not clear if these samples were collected from the Pleasant
Creek volcanic rocks, Rapitan Group or the Hay Creek Group, as
they are reported in the ‘basalt and red beds’ unit of Brabb &
Churkin (1969), which incorporates all three. Recent biomarker
work suggests that the presence of sponge spicules in the Wind-
ermere Supergroup is not inconsistent with a Cryogenian age
(Love et al. 2009). Microfossils have also previously been
described in chert of Fifteenmile Group (in strata previously mis-
mapped as unit 5 of the Upper Tindir Group; Macdonald et al.
2010a), including cyanobacterial coccoids, acritarchs such as
Trachyhystrichosphaera, vase-shaped microfossils and unique,
enigmatic siliceous scales (Allison 1980; Allison & Hilgert
1986; Allison & Awramik 1989). More recent studies have
extracted these microfossils from the surrounding carbonate
rock, and have demonstrated that they are composed of phos-
phate rather than silica, suggesting a green algae taxonomic affi-
nity (Cohen et al. 2011). A pre-717 Ma for these fossils is
supported by C- and Sr-isotope correlations (Macdonald et al.
2010a), the presence of dykes cutting the sections that are
co-magmatic with the Pleasant Creek volcanic rocks, and
geochronology of correlative rocks in the Coal Creek inlier
(Macdonald et al. 2010b, 2011).
Discussion
Depositional setting
The Windermere Supergroup contains two Cryogenian glaciogenic
deposits (Rapitan Group and the diamictite of the Hay Creek
Group) separated by c. 140 m of non-glacial strata (Fig. 35.4). A gla-
ciomarine depositional setting for the Rapitan Group is suggested
by the presence of faceted and striated clasts, bed-penetrating
dropstones, and common outsized and exotic clasts, along with evi-
dence for subaqueous slumping in the form of graded grain flows
and debris flows. Young (1988) ascribed the Fe-formations of the
Rapitan Group, and equivalent strata in the Tatonduk inlier, to
rift-related hydrothermal activity. However, these Fe-formations
are intimately associated with well-developed dropstones. Presum-
ing these diamictites were deposited during the terminal ice retreat,
and presuming an increased solubility of iron in the ocean due
to low oxygen levels under long-lived sea ice (Martin 1965),
Fe-formation can be attributed to an influx of oxygenated fresh
water concentrated at the termini of ice streams (Kirschvink 1992).
The parallel bedded sandstones and siltstones in the lower
portion of the Hay Creek Group are interpreted as turbidites as
Fig. 35.6. Neoproterozoic stratigraphy in
the Tatonduk inlier and correlations with
the Windermere Supergroup exposed c.
500 km east in the Mackenzie Mountains of
Canada. The schematic stratigraphy of the
Windermere Supergroup is modified from
Halverson et al. (2008). IRD mudstones are
laminated fine-grained sediments with
lonestones interpreted as ice rafted debris.
 THE TATONDUK INLIER, ALASKA– YUKON BORDER
they lack any evidence of wave action or traction currents, and
beds are stacked in fining-upward Bouma cycles. These green-grey
turbidites are distinguished from the red and purple grain flows in
the Rapitan Group by the lack of lonestones.
Young (1982) interpreted the diamictite of the Hay Creek Group
SW of the HLF as the product of a massive slope failure, and ident-
ified an unconformity at the base. Nonetheless, there is evidence of
a glaciomarine influence on deposition with the presence of
striated clasts (Allison et al. 1981), and a well-bedded diamictite
with dropstones near the top of Hay Creek Group along Pass
Creek (Fig. 35.5b). A depositional setting along the upper slope,
with diamictite being the product of syn-glacial slope failure, is
consistent with the interpretation that the expansion of the strati-
graphy of all of the Windermere Supergroup across the HLF
roughly approximates the slope – shelf transition.
From outcrops along the Tatonduk River, Allison et al. (1981)
cited deformation in Rapitan Group sediments near the overlying
disconformity to suggest that the underlying stratigraphy was
unconsolidated at the time of deposition of the upper diamictite.
However, slumps and syn-sedimentary folds are common through-
out the Rapitan Group (Young 1982), so this deformation in the
Rapitan Group may be syn-sedimentary or the product of grounded
ice during the Rapitan glaciation.
Regional correlations
Another line of evidence that the upper diamictite represents a
second, later glacial event is that it is overlain with a buff-coloured
dolostone. The latter is considered a basal Ediacaran cap carbonate
as it contains bed-parallel, isopachous sheet-crack cements
(Hoffman & Macdonald 2010) and a C-isotope profile similar to
that of thin basal Ediacaran cap carbonates in China (Zhou &
Xiao 2007), Australia (Kennedy 1996) and Namibia (Halverson
et al. 2005).
Lithologically, the Rapitan Group in the Tatonduk inlier is very
similar to the clast poor siltstone and Fe-formation of the Sayunei
Formation in the Mackenzie Mountains (Young 1976; Yeo 1984).
The diamictite of the Hay Creek Group can be correlated with the
Ice Brook Formation (Fig. 35.6). This correlation is particularly
attractive because the carbonate at the top of the Hay Creek
Group shares sedimentological and isotopic characteristics with
the Ravensthroat cap dolomite in the Mackenzie Mountains
(Aitken 1991; James et al. 1999).
Regional correlations of Neoproterozoic strata in NW Canada
are particularly important in light of the recent geochronological
constraints provided by volcanic tuffs interbedded with the Fif-
teenmile and Rapitan Groups in the Ogilvie Mountains (Macdo-
nald et al. 2010b). Coupled with the robust palaeomagnetic poles
in NW Canada (Evans 2000; Denyszyn et al. 2009) and the rich
micropalaeontological record in the Tatonduk inlier (Allison &
Arwimik 1989; Macdonald et al. 2010b), the inliers of the
Ogilvie Mountains hold great promise of a calibrated record of
tectonics, climate, chemistry and life in Cryogenian oceans.
We are grateful to the Yukon Geological Survey, the Department of Earth and
Planetary Sciences, Harvard University, P. Hoffman, A. Knoll and the NSF for
support. We thank R. B. Blodgett and D. Jones for helpful discussions. We also
thank the YUCH National Park office in Eagle for allowing access to the
field area. This represents a contribution of the IUGS- and UNESCO-funded
IGCP (International Geoscience Programme) project #512.
References
Aitken, J. D. 1989. Uppermost Proterozoic formations in central Macken-
zie Mountains, Northwest Territories. Geological Survey of Canada
Bulletin, 368, 1 –26.
Aitken, J. D. 1991. The Ice Brook Formation and Post-Rapitan, Late
Proterozoic glaciation, Mackenzie Mountains, Northwest Territories.
Geological Survey of Canada Bulletin, 404, 1 – 43.
395
Allison, C. W. A. 1975. Primitive fossil flatworm from Alaska: new
evidence bearing on ancestry of the Metazoa. Geology, 3, 649– 653.
Allison, C. W. A. 1980. Siliceous microfossils from the lower Cambrian
of northwest Canada: possible source for biogenic chert. Science,
211, 53– 55.
Allison, C. W. A. & Hilgert, J. W. 1986. Scale micro-fossils from the
Early Cambrian of northwest Canada. Journal of Paleontology, 60,
973– 1015.
Allison, C. W. A. & Awramik, S. M. 1989. Organic-walled microfossils
from earliest Cambrian or latest Proterozoic Tindir Group rocks,
northwest Canada. Precambrian Research, 43, 253–294.
Allison, C. W. A., Young, G. M., Yeo, G. M. & Delaney, G. D. 1981.
Glaciogenic rocks of the Upper Tindir Group, east-central Alaska. In:
Hambrey, M. J. & Harland, W. B. (eds) Earth’s Pre-Pleistocene
Glacial Record. Cambridge University Press, Cambridge, 720–723.
Blodgett, R. B., Rohr, D. M. & Boucot, A. J. 2002. Paleozoic links
among some Alaskan accreted terranes and Siberia based on mega-
fossils. In: Miller, E. L., Grantz, A. & Klemperer, S. L. (eds) Tec-
tonic Evolution of the Bering Shelf – Chukchi Sea –Arctic Margin and
Adjacent Landmasses. Geological Society of America Special Paper
360, Boulder, Colorado, 273– 290.
Brabb, E. E. & Churkin, M. J. 1969. Geologic Map of the Charley River
Quadrangle, East-Central Alaska. U.S. Geological Survey Map
I-573.
Cairnes, D. D. 1914. The Yukon-Alaska International Boundary Between
Porcupine and Yukon Rivers. Geological Survey of Canada Memoir
67, Map 140-A.
Cohen, P. A., Schopf, J. W., Butterfield, N. J., Kudryaytsev, A. &
Macdonald, F. A. 2011. Phosphate biomineralization in mid-
Neoproterozoic protists. Geology, 39, 539– 542.
Dalrymple, R. W. & Narbonne, G. M. 1996. Continental slope sedi-
mentation in the Sheepbed Formation (Neoproterozoic, Windermere
Supergroup), Mackenzie Mountains, N.W.T. Canadian Journal of
Earth Sciences, 33, 848–862.
Denyszyn, S. W., Halls, H. C., Davis, D. W. & Evans, D. A. D. 2009.
Paleomagnetism and U– Pb geochronology of Franklin dykes in High
Arctic Canada and Greenland: a revised age and paleomagnetic pole
for constraining block rotations in the Nares Strait region. Canadian
Journal of Earth Sciences, 46, 689– 705.
Dover, J. H. 1992. Geologic map and fold and thrust belt interpretation of
the south eastern part of the Charley River Quadrangle, East Central
Alaska. US Geological Survey Miscellaneous Investigations, Map
I-1942, scale 1:100,000, 2 sheets.
Dumoulin, J. A., Harris, A. G., Gagiev, M., Bradley, D. C. &
Repetski, J. E. 2002. Lithostratigraphic, conodont, and other
faunal links between lower Paleozoic strata in northern and central
Alaska and northeastern Russia. In: Miller, E. L., Grantz, A. &
Klemperer, S. L. (eds) Tectonic Evolution of the Bering Shelf –
Chukchi Sea –Arctic Margin and Adjacent Landmasses. Geological
Survey of America Special Paper 360, Boulder, Colorado, 291– 312.
Evans, D. A. D. 2000. Stratigraphic, geochronological and paleomagnetic
constraints upon the Neoproterozoic climatic paradox. American
Journal of Science, 300, 347–433.
Fuis, G. S., Moore, T. E. et al. 2008. Trans-Alaska Crustal Transect
and continental evolution involving subduction underplating and
synchronous foreland thrusting. Geology, 36, 267–270.
Halverson, G. P., Hoffman, P. F., Schrag, D. P., Maloof, A. C. &
Rice, A. H. N. 2005. Toward a Neoproterozoic composite carbon-
isotope record. Geological Society of America Bulletin, 117,
1181–1207.
Halverson, G. P., Dudás, F. O., Maloof, A. C. & Bowring, S. A. 2007.
Evolution of the 87Sr/86Sr composition of Neoproterozoic Seawater.
Palaeogeography, Palaeoclimatology, Palaeoecology, 256, 103–129.
Hodych, J. P., Cox, R. A. & Kosler, J. 2004. An equatorial Laurentia at
550 Ma confirmed by Grenvillian inherited zircons dated by LAM
ICP-MS in the Skinner Cove volcanics of western Newfoundland:
implications for inertial interchange true polar wander. Precambrian
Research, 129, 93– 113.
Hoffman, P. F. & Macdonald, F. A. 2010. Sheet-crack cements and
early regression in Marinoan (635 Ma) cap dolostones: regional
benchmarks of vanishing ice-sheets? Earth and Planetary Science
Letters, 300, 374– 384.
396 F. A. MACDONALD & P. A. COHEN
James, N. P., Narbonne, G. M. & Kyser, K. T. 1999. Neoproterozoic Cap
Carbonate Facies; Mackenzie Mountains, NW Canada; Abiotic Pre-
cipitation and Global Glacial Meltdown. Geological Society of
America Abstracts with Programs, 31, 487.
Kaufman, A. J., Knoll, A. H. & Awramik, S. M. 1992. Biostratigraphic
and chemostratigraphic correlation of Neoproterozoic sedimentary
successions: Upper Tindir Group, northwestern Canada, as a test
case. Geology, 20, 181– 185.
Kendall, C. G. S. C. & Warren, J. 1987. A review of the origin and
setting of teepees and their associated fabrics. Sedimentology, 34,
1007– 1027.
Kennedy, M. J. 1996. Stratigraphy, sedimentology and isotope geo-
chemistry of Australian Neoproterozoic postglacial cap dolostones:
deglaciation, d13C excursions and carbonate precipitation. Journal
of Sedimentary Research, 66, 1050–1064.
Kirschvink, J. L. 1992. Late Proterozoic low-latitude global glaciation:
the snowball earth. In: Schopf, J. W. & Klein, C. (eds) The Proter-
ozoic Biosphere. Cambridge University Press, Cambridge, 51 –52.
Love, G. D., Fike, D. A. et al. 2009. Fossil steroids record the appearance
of Demospongiae during the Cryogenian period. Nature, 457,
718– 722.
Macdonald, F. A. & Roots, C. F. 2010. Upper Fifteenmile Group in the
Ogilvie Mountains and correlations of early Neoproterozoic strata in
the northern Cordillera. In: MacFarlane, K. E., Weston, L. H. &
Blackburn, L. R. (eds) Yukon Exploration and Geology 2009.
Yukon Geological Survey, Whitehorse, YT, 237–252.
Macdonald, F. A., Cohen, P. A., Dudás, F. O. & Schrag, D. P. 2010a.
Early Neoproterozoic scale microfossils in the Lower Tindir Group
of Alaska and the Yukon Territory. Geology, 38, 143– 146.
Macdonald, F. A., Schmitz, M. D. et al. 2010b. Calibrating the
Cryogenian. Science, 327, 1241– 1243.
Macdonald, F. A., Smith, E. F., Strauss, J. V., Cox, G. M., Halverson,
G. P. & Roots, C. F. 2011. Neoproterozoic and early Paleozoic cor-
relations in the western Ogilvie Mountains, Yukon. In: MacFar-
lane, K. E., Weston, L. H. & Blackburn, L. R. (eds) Yukon
Exploration and Geology 2010. Yukon Geological Survey, White-
horse, 161– 182.
MacNaughton, R. B., Narbonne, G. M. & Dalrymple, R. W. 2000.
Neoproterozoic slope deposits, Mackenzie Mountains, northwestern
Canada: implications for passive-margin development and Ediacaran
faunal ecology. Canadian Journal of Earth Sciences, 37, 997–1020.
MacNaughton, R. B., Roots, C. F. & Martel, E. 2008. Neoprotero-
zoic– (?)Cambrian lithostratigraphy, northeast Sekwi Mountain map
area, Mackenzie Mountains, Northwest Territories: new data from
measured sections. Geological Society of Canada, Current Research
2008, 16, 1– 17.
Martin, H. 1965. Beobachtungen zum Problem der jung-präkambrischen
Glazialen Ablagerungen in Südwestafrika. (Observations concerning
the problem of the late Precambrian glacial deposits in South West
Africa.) Geologische Rundschau, 54, 115– 127.
McCausland, P. J. A., Van der Voo, R. & Hall, C. M. 2007.
Circum-Iapetus paleogeography of the Precambrian – Cambrian
transition with a new paleomagnetic constraint from Laurentia.
Precambrian Research, 156, 125– 152.
McClelland, W. C. 1997. Detrital zircon studies of the Proterozoic
Neruokpuk Formation, Sadlerochit and Franklin Mountains, northern
Alaska. Geological Society of America Abstracts with Programs, 25, 28.
Meertie, J. B. 1930. Geology of the Eagle-Circle district, Alaska. US
Geological Survey Bulletin, 816, 121–122.
Meertie, J. B. 1933. The Tatonduk-Nation district, Alaska. US Geologi-
cal Survey Bulletin, 836-E, 345–454.
Murthy, G. S. 1971. The paleomagnetism of diabase dykes from the
Grenville Province. Canadian Journal of Earth Sciences, 8,
802– 812.
Murthy, G. S., Gower, C. F., Tubrett, M. & Patzold, R. 1992.
Paleomagnetism of Eocambrian Long Range dykes and Double
Mer Mormation from Labrador, Canada. Canadian Journal of
Earth Sciences, 29, 1224–1234.
Mustard, P. S. & Roots, C. F. 1997. Rift-related volcanism, sedimen-
tation and tectonic setting of the Mount Harper Group, Ogilvie
Mountains, Yukon Territory. Geological Survey of Canada Bulletin,
492.
Norris, D. K. 1972. En echelon folding in the northern Coldillera of
Canada. Bulletin of Canadian Petroleum Geology, 20, 634– 642.
Norris, D. K. 1982. Geology, Ogilvie River, Yukon Territory, Geological
Survey of Canada, Map 1526A, 1:250,000 scale.
Park, J. K. 1997. Paleomagnetic evidence for low-latitude glaciation
during deposition of the Neoproterozoic Rapitan Group, Mackenzie
Mountains, N.W.T., Canada. Canadian Journal of Earth Sciences,
34, 34 – 49.
Payne, M. W. & Allison, C. W. A. 1981. Paleozoic continental-margin
sedimentation in east-central Alaska. Geology, 9, 274–279.
Plumley, P. W., Vance, M. S. & Milazzo, G. 1989. Structural and
paleomagnetic evidence for Tertiary bending of the Eastern
Brooks Range Flexure, Alaska. In: Hillhouse, J. W. (ed.) Deep
Structure and Past Kinematics of Accreted Terranes, Geophysical
Monograph 50. American Geophysical Union, Washington, DC,
127– 150.
Pyle, L. J., Narbonne, G. M., James, N. P., Dalrymple, R. W. &
Kaufman, A. J. 2004. Integrated Ediacaran chronostratigraphy,
Wernecke Mountains, northwestern Canada. Precambrian Research,
132, 1– 27.
Rainbird, R. H., Jefferson, C. W. & Young, G. M. 1996. The early Neo-
proterozoic sedimentary Succession B of Northwestern Laurentia:
correlations and paleogeographic significance. Geological Society
of America Bulletin, 108, 454– 470.
Ross, G. M. & Villeneuve, M. E. 1997. U –Pb geochronology of stranger
stones in Neoproterozoic diamictites, Canadian Cordillera: impli-
cations for provenance and ages of deposition. Radiogenic age and
isotopic studies, Report 10, Geological Survey of Canada, Current
Research 1997-F, 141– 155.
Stewart, J. H. 1975. Initial deposits in the Cordilleran geosyncline:
evidence of a late Precambrian (,850 m.y.) continental seperation.
Geological Society of America Bulletin, 83, 1345–1360.
Van Kooten, G. K., Watts, A. B. et al. 1997. Alaska Division of Geo-
logical and Geophysical Surveys, Report of Investigations 96-6A, 3
sheets, scale 1:125,000, Fairbanks.
Westcott, M. R. 1966. Loss of argon from biotite in a thermal meta-
morphism. Nature, 210, 83– 84.
Yeo, G. M. 1984. The Rapitan group: relevance to the global association
of Late Proterozoic glaciation and iron-formation. PhD thesis,
University of Western Ontario, London, Ontario.
Young, G. M. 1976. Iron-formation and glaciogenic rocks of the Rapitan
Group, Northwest Territories, Canada. Precambrian Research, 3,
137– 158.
Young, G. M. 1982. The late Proterozoic Tindir Group, east-central
Alaska; evolution of a continental margin. Geological Society of
America Bulletin, 93, 759–783.
Young, G. M. 1988. Proterozoic plate tectonics, glaciation and
iron-formations. Sedimentary Geology, 58, 127– 144.
Zhou, C. & Xiao, S. 2007. Ediacaran d13C chemostatigraphy of South
China. Chemical Geology, 237, 107– 126.
 Chapter 36

Neoproterozoic glacial record in the Mackenzie Mountains, northern Canadian Cordillera

P. F. HOFFMAN1,2* & G. P. HALVERSON3,4

1
Department of Earth and Planetary Sciences, Harvard University, Cambridge, MA 02138, USA
2
School of Earth and Ocean Sciences, University of Victoria, Victoria, BC V8W 2Y2, Canada
3
School of Earth and Environmental Sciences, The University of Adelaide, North Terrace, Adelaide, SA 5005, Australia
4
Present address: Department of Earth and Plantary Sciences, McGill University, Montreal, QC H3A 2A7, Canada
*Corresponding author (e-mail: paulfhoffman@yahoo.com)

Abstract: In the Mackenzie Mountains, an arcuate foreland thrust-fold belt of Late Cretaceous–Paleocene age in the northern Canadian
Cordillera, two discrete glacial– periglacial sequences of Cryogenian age (the Rapitan Group and the Stelfox Member of the Ice Brook
Fm.) are separated by c. 1.0 km of non-glacial strata. The older Rapitan diamictite occurs in an amagmatic rift basin; the younger Stelfox
diamictite occurs on a passive-margin continental slope.
The Rapitan Group consists of three formations. The lower Mount Berg Fm. is a complex of diamictites and conglomerates of limited
extent. The middle Sayunei Fm. is a thick sequence of maroon-coloured mudrocks hosting innumerable graded layers of silt- and fine-
grained sandstone. It lacks wave- or traction current-generated bedforms, and is lightly sprinkled with granule aggregates (‘till pellets’)
and lonestones of dolostone and rare extrabasinal granitoids. It is capped by a hematitic Fe-formation that was reworked into the discon-
formably overlying Shezal diamictite. The Shezal Fm. is a complex of olive-green coloured boulder diamictites with subordinate, dark-
grey shales, siltstones and parallel-sided sandstones. Some of the boulders are faceted and striated, and include dolostone, quartzite, silt-
stone and gabbro in declining order of abundance. Diamictite terminates abruptly at the top of the Shezal Fm., which is sharply overlain
by dark shales or by ,52 m of fetid, dark-grey, 13C-depleted limestone with graded bedding.
The Stelfox Member is dominated by non-stratified, carbonate-clast diamictite with faceted and striated clasts, locally associated with sub-
ordinate, well-laminated shales containing till pellets and ice-rafted dropstones. It is thin or absent on the palaeocontinental shelf, but thickens
seaward (southwestward) on the palaeocontinental slope. A thin clay drape separates it from a laterally continuous post-glacial ‘cap’ dolo-
stone, which is a very pale coloured, micro- to macropeloidal dolostone with low-angle cross-laminae, giant wave ripples and local bioherms
of corrugated stomatolites. In the NW, the dolostone is followed by reddish and greenish marls, followed by black shale of the Sheepbed Fm.
In the SE, the dolostone is overlain by pink or grey limestones with well-developed sea-floor cements pseudomorphic after aragonite. In this
area, the top of the dolostone is ferruginous and contains digitate rosettes of sea-floor barite cement, variably calcitized. The dolostone–
limestone contact is perfectly conformable, and synclinal structures previously intepreted as karst features are tectonic in origin.
The grand mean palaeomagnetic pole for the well-studied Franklin Large Igneous Province (c. 718 Ma) of Arctic Laurentia, coeval with
the basal Rapitan Group in the Mount Harper area, Yukon Territory, places the Mackenzie Mountains firmly in the tropics, at 18 + 38N
palaeolatitude, at the onset of the Rapitan glaciation.
Carbon (d13C), oxygen (d18O) and strontium (87Sr/86Sr) isotopes have been measured in carbonates bracketing the Rapitan and Stelfox
diamictites. Sulphur isotope data (d34S) have been obtained from carbonate-associated sulphate and barite above the younger diamictite, and
calcium isotope data (d44Ca) from the younger carbonate itself. The results are broadly consistent with data from other areas. Iron isotope
(d57Fe) and cerium anomaly (Ce/Ce*) values increase systematically upwards through the Sayunei Fe-formation, supporting an interpret-
ation that deposition occurred within a redox chemocline through which the basin floor descended as a consequence of isostatic loading by the
advancing Shezal ice sheet.

Supplementary material: Data are available at http://www.geolsoc.org.uk/SUP18470.

Glacial marine diamictites (‘tillites’) were recognized in the
northwestern Mackenzie Mountains (Fig. 36.1) and to the west
in the Yukon Territory by Shell Oil Company geologists in
1958 (Ziegler 1959), during the course of economic assessment
of associated Fe-formations. At the time, they were thought to
be early Palaeozoic in age (Ziegler 1959). They were referred to
as the Rapitan Group by Green & Godwin (1963). Correlative
strata were studied in the Hayhook Lake area (Fig. 36.1), 300 km
to the SE, by Upitis (1966), Gabrielse et al. (1973), Young
(1976) and Eisbacher (1978), where they were recognized as
being of Neoproterozoic (Hadrynian) age. Gabrielse et al. (1973)
erected a type section for the Rapitan Group near Hayhook
Lake at 63834’03”N, 127802’41”W. It includes the Sayunei,
Shezal and Twitya formations (Fig. 36.2). Eisbacher (1978)
enlarged the Rapitan Group to include the Keele Fm. because
of its gradational relationship with the Twitya Fm. However,
Yeo (1981) and most subsequent workers follow the original
definition (Green & Godwin 1963) and limit the Rapitan
Group to glaciogenic strata beneath the Twitya Fm. Additional
stratigraphic and sedimentological studies of the Rapitan
Group along the 400 km arc of the Mackenzie Mountains were
carried out by Eisbacher (1981a, b, 1985) and Yeo (1981, 1984,
1986).
A second, thinner, glaciogenic unit was discovered .1000 m
stratigraphically above the Rapitan Group by Aitken (1991a, b).
He defined it as the Stelfox Member (,272 m) of the Ice Brook
Fm., the type section of which is located in the Sayunei Range at
648080 0700 N, 1298000 3800 W (Aitken 1991b). It is overlain by a dis-
tinctive carbonate unit, the ‘Tepee dolostone’ of Eisbacher (1978,
1981a) and Aitken (1991b), subdivided into the Ravensthroat and
Hayhook formations by James et al. (2001). The dolomite features
unusual sedimentary structures, including sheet-crack cements,
reverse-graded peloids, giant wave ripples, corrugated stromato-
lites and sea-floor cements (crystal fans) originally composed of
aragonite and barite (Eisbacher 1981a; Aitken 1991b; James
et al. 2001; Hoffman & Schrag 2002; Allen & Hoffman 2005;
Hoffman & Macdonald 2010). It has many similarities with post-
glacial cap dolostones on other palaeocontinents (Hoffman et al.
2011), which allow the base of the Ediacaran Period to be recog-
nized globally (Knoll et al. 2006).
Correlatives of the Mackenzie Mountains and Windermere super-
groups, including both glacial horizons, are discontinuously exposed
in Wernecke and Ogilvie in central Yukon Territory (Macdonald &
Roots 2009; Macdonald et al. 2010b), and in the Tatonduk inlier
of east-central Alaska (Young 1982; Macdonald et al. 2010a;
Macdonald & Cohen 2011).

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 397– 411. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.36
398 P. F. HOFFMAN & G. P. HALVERSON

134 oW 132 oW 128oW 126o W

130oW
M Norman
platfor a
ck Wells
Iron SNAKE RIVER
Ck
m en
mou zi
Cranswick R.
ntain e
Ri
v 65oN

e
Corn Gayna

r
Creek R. P
la

fr
Arctic te
Red R. a u

on
Stone Knife R. t

Th
Shale Lake

ru
st
Twitya R. 64oN
Boomerang L.
Moose Horn R.

Stelfox Mtn Hayhook

June Lake Lake
REDSTONE
Ravens-
This Sekwi throat R.
figure Brook
63oN

PLATEAU Coates L.
Windermere Supergroup Thunder-
cloud R. Fig. 36.1. Outcrop distribution of the
Sheepbed Rapitan Group and the Windermere
Rapitan Group Supergroup in the Mackenzie Mountains,
Creek
measured sections showing locations of measured sections
forming the basis of this report.

This chapter is a synopsis of previous work, augmented by River (648490 2300 N, 1308270 1100 W: 22 m), Stone Knife River
seven new sections of the Rapitan Group, measured by P.F.H. (648410 5300 N, 1298530 1900 W: 114 m), Shale Lake (648320 4100 N,
in the Iron Creek (658030 1500 N, 1338070 3100 W: .775 m), Gayna 1298220 3700 W: 995 m), Boomerang Lake (638460 4600 N, 1278280 0500 :
.236 m), Hayhook Lake (638340 3800 N, 1278050 3400 W: 997 m)
and Ravensthroat River (638130 5700 N, 1278030 1800 W: 230 m)
Group Formation Member Reference areas; and 13 sections of the Stelfox diamictite (all ,35 m)
Gabrielse and its post-glacial carbonate sequence at Cranswick River
Sheepbed (658050 5400 N, 1328260 1700 W), Arctic Red River (648560 0500 N,
et al. (1973) 1318030 3000 W), Gayna River (648490 2200 N, 1308280 2700 W), Stone-
Hayhook James et al. knife River (648400 3600 N, 1298530 4800 W), Shale Lake (648310 1200 N,
(2001) 1298290 0900 W), Twitya River (648130 0700 N, 1288380 1100 W),
Ravensthroat
Moose Horn River (638570 1300 N, 1278300 3900 W), Hayhook
Stelfox Lake (638340 1700 N, 1278110 4400 W), Stelfox Mountain (638360 0100 N,
Ice Brook Aitken 1278500 4700 W; 638350 4600 N, 1278530 0400 W; 638350 6000 N,
Delthore
(1991b) 1278530 5700 W; 638350 3700 N, 1278550 1100 W) and Ravensthroat
Durkan River (638170 0500 N, 1278080 4000 W). The named locations are
Keele indicated in Figure 36.1. The coordinates refer to the base of the
section.
Twitya
Eisbacher
Shezal
(1978)
Rapitan Sayunei Structural framework
Mt Berg
The Mackenzie Mountains are the physiographic expression of an
Copper Cap Gabrielse arcuate, NE-vergent, foreland thrust –fold belt of Late Cretac-
Coates Lake Redstone River et al. (1973) eous –Paleocene age in the northern Canadian Cordillera (Aitken
& Long 1978; Aitken 1982; Narbonne & Aitken 1995). Late
Thundercloud Jefferson (1983) Neoproterozoic strata of the Windermere Supergroup (Fig. 36.3)
Little Dal Aitken (1981) are principally exposed in the hanging wall of the Plateau Thrust
system (Fig. 36.1). Their subhorizontal (Redstone Plateau) or
Fig. 36.2. Stratigraphic nomenclature and defining references for units SW-dipping attitude depends on their position above a flat or
associated with Cryogenian glaciogenic strata (black with inverted triangles) in ramp in the thrust plane. Outcrop-scale strain is heterogeneous;
the Mackenzie Mountains. structurally intact, low-strain sections can be located by
 THE MACKENZIE MOUNTAINS, NORTHERN CANADIAN CORDILLERA 399

(a) (b) (c) (d)

Mackenzie Mountains composite Hayhook Lake (type sec.) Hayhook Lake (ref. sec.) Iron Creek
63o34’03”N, 127o02’41”W 63o34’38”N, 127o05’34”W 65o03’15”N, 133o07’31”W
0 0.0
Sekwi (km)
CAMBRIAN

(m) (m)
(km) -0.1

Shezal
Vampire
-50 -50
Backbone -0.2
Ranges

Keele
Ingta -0.3
Risky -100 -100
-2 -0.4
EDIACARAN

Blueflower
-0.5 fs BIF
-150 -150
WINDEREMERE SUPERGROUP

Gametrail

olive green (Little Dal-Katherine derivation)

-0.6

Sayunei
fs
Sheepbed -0.7 -200 -200
fs
Ravensthroat BIF
-4 Ice Stelfox -0.8 fs
Brook BIF
Twitya

Keele -0.9 -250 -250

Shezal
fs
CRYOGENIAN

-1.0 BIF

maroon
Twitya -300 -300
-1.1

Mount Berg
unnamed cap
Shezal -1.2
-6 Rapitan -350 -350
Group Sayunei -1.3 fs

Mt Berg
-1.4
Shezal

Coates Lake -400 -400

Coates?
Group -1.5

-1.6 -450 -450

-8 BIF
MACKENZIE MTNS SUPERGROUP

Little -1.7
LEGEND
Dal
-1.8 -500 mud-matrix
Group R diamictite
BIF
Sayunei

R -1.9
wacke-matrix
-2.0 -550 diamictite
TONIAN(?)

-10 Katherine
maroon

Group -2.1 scaly-matrix

Sayunei

-600 diamictite
-2.2

Tsezotene 780 Ma mudstone with

(U-Pb) -2.3
silty turbidites,
Coates

-650
-2.4 debrites, sparse
'Unit H1' ice-rafted debris
-12
R fs

carbonate reefal sulphate mudstone siltstone sandstone basalt- haematitic haematite- flooding
carbonate evaporite gabbro mudstone jaspilite surface
Fig. 36.3. (a) Composite columnar section of Neoproterozoic strata in the Mackenzie Mountains (modified after Narbonne & Aitken 1995). (b) Type section of the
Rapitan Group (Gabrielse et al. 1973) in the Hayhook Lake area (see Fig. 36.1 for location) as remeasured by P.F.H. (c) Detailed section of the upper Sayunei and Shezal
formations 1.6 km NW of the Rapitan Group type section. Note Fe-formation (‘BIF’) at the top of the Sayunei Formation, representing ‘basin-facies’ Fe-formation
(Eisbacher 1985). (d) Section of the lower Rapitan Group near Iron Creek, representative of ‘transitional-facies’ Fe-formation (Eisbacher 1985).

mapping. Long slopes are unattractive because of scree. Published An unconformity at the base of the Backbone Ranges Fm. (Early
1:250 000-scale map coverage is incomplete and stratigraphically Cambrian) cuts progressively down-section from SW to NE, with
inconsistent. the result that the Windermere Supergroup is missing to the NE of
400 P. F. HOFFMAN & G. P. HALVERSON
the Plateau Thrust system. To the SW, the glaciogenic units are
buried by younger strata. Although the outcrop belt of Windermere
strata is less than 30 km wide, it fortuitously preserves the outer
shelf-edge and upper slope of a continental terrace developed in
late Cryogenian (Keele Fm.) to early Ediacaran (Sheepbed Fm.)
time (Ross 1991; Narbonne & Aitken 1995; Dalrymple & Nar-
bonne 1996; Day et al. 2004). Continental rifting, leading to the
formation of the continental terrace, controlled sedimentation
during the Coates Lake and Rapitan groups (Eisbacher 1981a,
1985; Jefferson 1983), as well as during the preceding Mackenzie
Mountains Supergroup (Turner & Long 2008). The Coates Lake
and Rapitan basins are interpreted by Jefferson & Ruelle (1986)
and Yeo (1981), respectively, as rhombochasms (‘pull-apart’
basins) associated with hypothetical strike –slip systems, systems
that are not mutually compatible in orientation.
Stratigraphy
Mackenzie Mountains Supergroup
The Mackenzie Mountains Supergroup is a broadly conformable
succession composed of sandstone, siltstone, carbonate and eva-
porite of mostly shallow-marine origin (Fig. 36.3a). Its upper
part, the Little Dal Group, is 2 km thick in its type section (Gab-
rielse et al. 1973; Aitken 1981; Halverson 2006) and is dominated
by carbonate with a recessive interval of gypsiferous siltstone. The
carbonates are platformal in the SE and basinal in the NW, with
reefal build-ups in the lower part (Aitken 1981; Turner et al.
1997). The carbonate is conformably overlain by and locally inter-
stratified with pillow basalt, the Little Dal lavas of Aitken (1982).
These are possibly the extrusive equivalents of the Tsezotene
dykes and sills (Aitken 1982), which belong to the 780 Ma Gun-
barrel large igneous province (Harlan et al. 2003).
Coates Lake and Rapitan groups
The Coates Lake Group is an assemblage of sandstone, carbonate-
clast conglomerate, gypsiferous siltstone and carbonate, including
fetid basinal limestone with turbidites and ‘debrites’ (i.e. coarse-
grained mass-flow deposits) (Ruelle 1982; Jefferson & Ruelle
1986). The Rapitan Group (Eisbacher 1978, 1981a) includes the
Sayunei (sigh-YOU-knee) Fm., composed of maroon coloured,
subaqueously deposited, fine-grained clastics with subordinate
debrites and lonestones, and the overlying Shezal (shiz-ALL)
Fm., a stack of mostly olive-coloured, polymictic ‘diamictites’
(i.e. massive, foliated or bedded wackestone with randomly dis-
persed, matrix-supported pebbles and boulders, characteristically
faceted and striated) with thin interbeds of dark shale and sand-
stone. Locally, an older diamictite complex (Mount Berg Fm.)
occurs below the Sayunei Fm. (Yeo 1981). The Rapitan Group
onlaps tilted Coates Lake and Little Dal group strata unconform-
ably (Eisbacher 1978, 1981a, 1985). Basinal facies of the Coates
Lake Group are black, pyritic and organic-rich; those of the
Rapitan Group are maroon, hematitic and organic-poor.
Twitya, Keele and Ice Brook formations
A major marine transgression followed the Rapitan glaciation, pro-
viding accommodation for 330–765 m of dark grey shale, siltstone
and fine-grained sandstone of the Twitya Fm. (Eisbacher 1978,
1981a; Aitken 1982). Resting sharply upon boulder diamictite of
the upper Shezal Fm., at the base of the Twitya Fm., are 0– 52 m
of dark-grey, thin-bedded limestone (Eisbacher 1978, 1981a;
Aitken 1982), which are absent in the Rapitan Group type
section (Hayhook Lake) and thickest at Shale Lake (Fig. 36.1).
Above the Twitya Fm. lie 220 –600 m of cyclic, shallow-marine
sandstone, siltstone, limestone and dolomitized limestone of the
Keele Fm. (Day et al. 2004). A thin, jasper-pebble conglomerate
occurs widely near its base. Glendonites, interpreted as pseudo-
morphs after ikaite (CaCO3.6H2O), occur close to the same
horizon (James et al. 2005).
The outer edge of the Keele platform is exposed near Shale Lake
and on Stelfox Mountain (Fig. 36.1), where a SW-dipping ‘break-
away scarp’ (Aitken 1991b, p. 26) separates shallow-marine Keele
strata on the footwall from hotel-size megaclast breccia (Durkan
Member), turbiditic siltstone (Delthore Member) and glaciogenic
diamictite (Stelfox Member) of the Ice Brook Fm. on the
hanging wall (Aitken 1991b; Shen et al. 2008). The Stelfox dia-
mictite is buttressed against the palaeoscarp and its feather-edge
steps across the trace of the palaeoscarp on Stelfox Mountain (at
638350 5800 N, 1278520 0200 W) onto the Keele Formation of the foot-
wall (Fig. 36.4b). The Ravensthroat Fm. passes across the fault line
without displacement, proving that movement on the palaeoscarp
ended before the glacial termination. Eisbacher (1981a) inferred
that the Durkan megabreccia was triggered by glacioeustatic fall,
but this interpretation was rejected by Aitken (1991b), who
argued that pore-fluid overpressures would not develop in uncon-
solidated sediments unless the base-level fall was catastrophically
rapid. He (Aitken 1991b) insisted that only the Stelfox Member
was glacial in origin. P.F.H. therefore prefers the name Stelfox gla-
ciation over Ice Brook glaciation, despite the latter’s priority in the
literature (Kaufman et al. 1997).
Ravensthroat, Hayhook and Sheepbed formations
As with the Rapitan glaciation, major marine flooding followed by
Stelfox diamictite deposition, providing accommodation on the
Keele platform for the formation of a transgressive, post-glacial
carbonate (Raventhroat/Hayhook formations) and 446 –562 m of
black shale and minor siltstone of the Sheepbed Fm. (Pyle et al.
2004; Shen et al. 2008). On the continental slope SW of the
Keele platform, the Sheepbed Fm. reaches 1050 m in thickness
at Sekwi Brook (Fig. 36.1) and includes significantly more silt-
stone, deposited by both turbidity flows and unidirectional
(NW-directed) contour currents (Dalrymple & Narbonne 1996).
The contourites are estimated to have formed in water depths of
1.0– 1.5 km and imply that the slope was open to the world
ocean in a gulf .100 km wide (Dalrymple & Narbonne 1996).
Glaciogenic and associated strata
Rapitan Group
The remeasured type section of the Rapitan Group (Gabrielse
et al. 1973) is shown in Figure 36.3b. The Sayunei Fm.
(487 m) consists of maroon-coloured mudstone with innumerable
graded beds of siltstone and/or fine-grained sandstone, some
with outsized clasts of dolostone. Granule- to small pebble-sized
lonestones are sprinkled throughout the mudstone. The clasts are
commonly rimmed by chlorite, formed by reaction between the
dolostone and the host mudstone. Some clasts are aggregates of
sand or granules, interpreted as ‘till pellets’ (Ovenshine 1970) by
Young (1976) and Eisbacher (1981a). Bedforms attributable to
wave action are absent, as is evidence of bottom traction cur-
rents. Greenish mudstone beds occur, but none are dark grey
or black, in stark contrast to the basinal facies of the underlying
Coates Lake Group.
In the Hayhook Lake area (Fig. 36.1), the top of the Sayunei
Fm. is marked by 10–16 m of brick-red, hematitic mudstone and
hematite jaspilite (Fig. 36.3c). This interval is parallel-laminated
and lacks evidence for bottom traction currents. Within the
Fe-rich formation are lonestones and thin dolostone-clast debrites
(Young 1976). The former include rounded cobbles and small
boulders of porphyritic quartz monzonite (i.e. two-feldspar
 THE MACKENZIE MOUNTAINS, NORTHERN CANADIAN CORDILLERA 401

(a)
o
155-210

NW SE
50 50

Sheepbed Fm
40 40
G Sheepbed Formation
o G
200-220
R
30 G 30

Hayhook fm
G
o R
198-210 R
20 20
o 190-200o
215
G

Ravensthroat fm
240o
10 190o
R R 190-210o 10
195o
190o 210-230o
230o 195o
0 0

Ice Brook Fm (Stelfox Mb)

-10 -10
aeolian(?)
-20 -20
Keele Formation

-30 -30

Keele Fm
o
150-225 -40
-40 o
200

o
200
-50 -50
Cranswick R. Arctic Red R. Gayna R. Stone Knife R. Shale Lake Twitya R. Moose Horn R. Ravensthroat R.
65o05’54”N, 64o56’05”N, 64o49’22”N, 64o40’36”N, 64o31’12”N, 64o13’07”N, 63o57’13”N, 63o17’05”N,
132o26’17”W 131o03’30”W 130o28’27”W 129o53’48”W 129o29’09”W 128o38’11”W 127o30’39”W 127o08’40”W

(b)
W RG red / green marly shale
LEGEND E
50 black shale 50
allodapic limestone current ripples (flow azimuth)
peloidal dolostone (primary) sea-floor aragonite cement
40 carbonate-clast diamictite sea-floor barite cement 40
limestone giant wave ripples (crestline)
dolostone (secondary) corrugate stromatolite
30 sheet-crack cement
30
quartz sandstone
quartz-sericite siltstone crossbedding (flow azimuth) Hayhook
formation
20 20

Sheepbed Formation
10 Ravensthroat 10
o o 295-300o
formation
170-205 275-280 235o
0 0

-10 -10
en, 1991b)

Stelfox Member Fig. 36.4. Stratigraphic relations based on

Ice Brook Formation
columnar sections spanning the Stelfox
-20 -20 Member and its post-glacial carbonate
scarp” (Aitk

sequence (Raventhroat and Hayhook

Keele Formation
-30 -30 formations) in lines oriented parallel (a) and
transverse (b) to the depositional strike of the
“breakaway

Neoproterozoic passive margin. The

-40 Delthore Member -40 ‘breakaway scarp’ (Aitken 1991b) marks the
Ice Brook Formation
outer edge of the Keele Formation shelf.
Megaclast breccia of the Durkan Member
-50 -50 (not shown) conformably underlies the
Stelfox Mt. Stelfox Mt. Stelfox Mt. Stelfox Mt. Stelfox Mt. Moose Horn R. Hayhook Lake
63o36’08”N, 63o35’37”N, 63o36’00”N, 63o35’46”N, 63o36’01”N, 63o57’13”N, 63o34’17”N, Delthore Member and both abut the
127o57’43”W 127o55’11”W 127o53’57”W 127o53’04”W 127o50’47”W 127o30’39”W 127o11’44”W ‘breakaway scarp’ at a buttress unconformity.

granitoid with 5–15% quartz). The Fe-formation is overlain dis- In the Iron Creek area (Fig. 36.3d), an aggregate thickness of 100–
conformably by maroon-coloured diamictite (basal Shezal Fm.), 120 m of hematite jaspilite with pea-sized jasper nodules (mini-
which contains rounded and angular clasts from the lower unit. concretions) is sandwiched between units of poorly stratified,
In some sections, the Fe-formation is missing and was probably boulder diamictite with subordinate beds and lenses of tabular cross-
removed by erosion at the base of the Shezal Fm. bedded, medium-grained, well-rounded, monocrystalline, quartz
402 P. F. HOFFMAN & G. P. HALVERSON
sandstone (Yeo 1981, 1986; Klein & Beukes 1993). Outsized clasts
and debrites occur within the jaspilite intervals, as do thin seams and
lenses of Fe-carbonate (siderite). The Hayhook Lake area and Iron
Creek sections represent respectively the ‘basinal’ and ‘transitional’
(more proximal) facies of Fe-formation distinguished in the Rapitan
Group by Eisbacher (1985).
The Shezal Fm. diamictite is mostly olive-green or grey in
colour, except for the basal part, which is maroon and clearly
derived from the underlying Sayunei Fm. (Fig. 36.3c). The clasts
consist of dolomite (Little Dal Group), commonly stromatolitic,
quartzite (Katherine Group), siltstone (Coates Lake Group and
Tsezotene Fm.), basalt and gabbro. Extrabasinal clasts are rare.
Faceted and striated clasts, particularly siltstone, are abundant
and unequivocally glacial in origin (Young 1976; Eisbacher
1981a). Diamictite occurs in crudely tabular, poorly stratified,
matrix-supported bodies up to 90 m thick. The matrix consists
variably of wackestone, structureless mudstone, or mudstone
with anastomosing (‘scaly‘) foliation (Fig. 36.3c) that is subparal-
lel to bedding but unrelated to the axis of maximum tectonic com-
pression. The Shezal Fm. (320 m thick) is unusually well-exposed
in a river canyon 1.6 km NW of the type area (Fig. 36.3c), where
diamictite units are separated by subordinate, recessive intervals of
well-stratified siltstone, graded sandstone and dark-grey shale. A
6-m-thick duplex structure with thrusting directed toward the NE
occurs in the reference section between –320 and –330 m
(Fig. 36.3c). In this section, the tops of the diamictite bodies
tend to be more sharply defined than their bases.
No post-glacial carbonate is present in the type section of the
Rapitan Group (Fig. 36.3b), but elsewhere the basal Twitya Fm.
consists of thin graded beds of dark-grey, allodapic limestone (Eis-
bacher 1978, 1981a; Aitken 1982). The carbonate reaches a thick-
ness of 52.5 m at Shale Lake. At Stone Knife River (Fig. 36.1),
where 40 m of limestone is exposed, its lower part is hummocky
cross-stratified. Near Corn Creek (Fig. 36.1), bordering the Wer-
necke Mountains, up to 300 m of massive dolomite (Mount
Profeit Fm.) is laterally equivalent to the lower Twitya post-glacial
limestone (Eisbacher 1981a).
Stelfox Member (Ice Brook Fm.)
Whereas the Rapitan Group is up to 1500 m thick (Yeo 1981), the
Stelfox Member is less than 40 m thick in most sections (Fig. 36.4)
and the thickest, 308 m (Atiken 1991a), is also the deepest-water
section of Stelfox diamictite in the Mackenzie Mountains. Basin-
ward thickening could account for the modest thickness of the
Stelfox diamictites, which are only preserved in shelf and upper-
slope settings, in contrast to the deep rift-basin setting of the
Rapitan Group.
At Shale Lake (Fig. 36.4a), the Stelfox diamictite is directly
underlain by 5–9 m of well-sorted, well-rounded, medium-
grained, quartz sandstone of possible aeolian origin. Large-scale
tabular cross-bedding indicates transport to the SSW (c. 2058
azimuth, n ¼ 2), consistent with easterly palaeowinds (see
below). The presence of aeolianite beneath the diamictite would
suggest that glacioeustasy preceded or outranked glacioisostasy.
The Stelfox Member is typically recessive and masked by scree
from the overlying Ravensthroat Fm., which along with its modest
thickness accounts for its delayed recognition. The dominant
lithology is massive to weakly stratified diamictite, in which
rounded clasts of tan dolostone and grey limestone are supported
by a matrix of brownish-tan carbonate-rich mudrock with dis-
persed grains of quartz sand and granules. The clasts are derived
from the underlying Keele Fm. and the absence of quartzite
clasts suggests that carbonates were the only components of the
Keele Fm. that were strongly lithified at the time of glaciation.
Beds and lenses of quartz-chert sandstone, micaceous wacke and
polymictic debrite are subordinate components of the Stelfox
Member. At Cranswick River section (Fig. 36.4a), the diamictite
has a terrigenous matrix and the carbonate clasts, freed in float,
are visibly faceted and striated. Ice-rafted dropstones and ‘till
pellets’ (Ovenshine 1970) are well-developed in laminated facies
(Aitken 1991a, b).
Ravensthroat– Hayhook post-glacial carbonate couplet
The Stelfox diamictite or its equivalent erosion surface is every-
where sharply overlain by a transgressive, 10–15-m-thick, cap
dolostone (Fig. 36.4), previously known informally as the ‘Tepee
dolostone’ (Eisbacher 1981a, 1985; Aitken 1991b). The pale, incan-
descent, yellowish-grey scree profusely generated by the informally
defined Ravensthroat Fm. (James et al. 2001) could not be more
unlike the jet-black feathers of its avian namesake. East of 1308W
longitude, the dolostone is overlain by up to 15 m of ledge-forming
grey or pink limestone (Fig. 36.4), informally defined as the
Hayhook Fm. (James et al. 2001). West of 1308W longitude, it is
overlain by an ascending sequence of red, green and grey shales
with thin beds of current-rippled limestone (Fig. 36.4). This tri-
coloured sequence is remarkably similar to that overlying the
homologous postglacial cap dolostones (Hambrey & Spencer
1987; Halverson et al. 2004) in East Greenland (e.g. Kap Weber,
Andrée Land) and East Svalbard (e.g. Ditlovtoppen, Ny Friesland).
The Ravensthroat Fm. consists of micro- and macropeloidal
dolostone with ubiquitous, low-angle, cross-lamination defined
by normal and reverse-graded laminae (James et al. 2001). The
peloids are structureless, sub-spherical aggregates of dolomicrite
up to 3 mm in diameter. They commonly rest on abraded facets.
Giant wave ripples (cf. Allen & Hoffman 2005) are well developed
in many sections (Fig. 36.4). They are trochoidal in profile, with
sharp crests and lobate troughs, and individual ripple trains
aggrade for up to 1.4 m through bidirectional accretion of peloidal
laminae. Their synoptic relief, crest to trough, is up to 40 cm. The
structures are different in form and origin from peritidal ‘tepees’
(Assereto & Kendall 1977; Kendall & Warren 1987); they are
not brecciated, lack syndepositional cements, and their crestlines
are linear and parallel, not polygonal, in plan view (Eisbacher
1981a; Aitken 1991b; James et al. 2001). Their crestlines generally
trend NE, at a high angle to the Keele shelf edge (Fig. 36.5).
Decametric bioherms composed of corrugated stromatolites
(James et al. 2001) oriented subparallel to the crests of giant
wave ripples occur in some sections (Fig. 36.4). The corrugated
1 130o 128o
65o
3
1 Keele
4
shelf
slope 5
o
64
50 km 5
1
132o 130o
Fig. 36.5. Mean orientations of crestlines (double bars) of giant wave ripples in
the Ravensthroat Fm. at different locations. Numerals are the numbers of
independent ripple trains measured at each location.
 THE MACKENZIE MOUNTAINS, NORTHERN CANADIAN CORDILLERA
stromatolites are closely similar to those that are intergradational
with ‘tubestone’ stromatolites in the homologous Keilberg cap
dolostone in northern Namibia. Evidence of subaerial exposure
is absent, but there can be no doubt that the Ravensthroat Fm.
was deposited above wave-base and within the euphotic zone.
In virtually all sections from Shale Lake to Ravensthroat River
(Fig. 36.1), a distance of nearly 200 km, the uppermost 4– 10 cm of
the Ravensthroat Fm. contains a remarkable digitate crust com-
posed of sea-floor barite (BaSO4) cement, variably pseudomor-
phosed by calcite. At Shale Lake, the barite crust is developed
directly on a train of giant wave ripples. Because of pseudo-
morphic replacement in some sections, the barite cement was for-
merly misinterpreted as an aragonite cement (Aitken 1991b, fig. 8)
or as microdigitate stromatolites (James et al. 2001, fig. 10). The
digits consist of rosettes or bundles of bladed crystals whose mor-
phology is quite distinct from the prismatic (pseudohexagonal)
habit of the former aragonite cements in the overlying Hayhook
Fm. (James et al. 2001). The digits preserve growth lamellae, 1–
2 mm thick, and typically tilt to the SW due to preferential
growth in the seaward direction. The spaces between the digits
are filled by laminated, peloidal, ferroan dolomite, the dark
brown colour of which contrasts with the pale colouration of the
rest of the Ravensthroat Fm. The borders of the barite digits are
ragged due to the growth of vertically standing barite crystals
that lap out onto the upper surfaces of individual laminae of infill-
ing peloidal dolomite. This provides textural proof that the barite
digits formed simultaneously with the peloidal infills. This
interpretation agrees with that of barite cements at the homologous
post-glacial dolostone– limestone transition in central Australia
(Kennedy 1996).
The tips of the barite digits, marking the Ravensthroat –
Hayhook transition, are typically ‘colonized’ by slender prismatic
fans of pseudomorphosed aragonite, while at the same level the
peloidal ferroan dolostone changes to calcimicrite, initially with
some reworked dolostone. Locally, the barite crust was dissolved
away instead of pseudomorphically replaced, and the barite
cement layer is represented by a collapse microbreccia of
ferroan dolostone clasts, overlain by calcimicrite of the basal
Hayhook Fm.. The local occurrence of this microbreccia contribu-
ted to the misinterpretation of the contact as a karstic unconformity
(James et al. 2001, fig. 9B).
Numerous mesoscopic folds at the Ravensthroat –Hayhook
contact, previously interpreted as karst channels (James et al.
2001, fig. 9A), are consistently oriented parallel to tectonic clea-
vage and regional folds. They are not karst channels because the
primary stratification in both units is everywhere parallel to the
contact. We interpret them as lobate-cuspate folds (Ramsay &
Huber 1987) of tectonic origin and Late Cretaceous –Paleocene
age, related to bedding-parallel shortening at the rheological inter-
face between the stiff Ravensthroat dolostone and the weak
Hayhook limestone. Contrary to James et al. (2001), we find no
evidence of karstic unconformity, subaerial exposure or isostatic
adjustment at the Ravensthroat –Hayhook contact.
The Hayhook Fm. is 0 –15 m thick (Fig. 36.4) and consists of
flaggy micrite with pseudomorphosed crystal fans of aragonite
cement (James et al. 2001). The cement fraction generally
increases upward and reaches nearly 100% in some sections,
notably Shale Lake (Fig. 36.1). The detailed interplay between
cement growth, burial by micrite and renewed cement develop-
ment indicates precipitation at a free-face on the sea-floor.
Boundary relations with overlying and underlying
non-glacial units
Rapitan Group
A low-angle unconformity separates the glaciogenic Rapitan
Group from various units of the underlying Coates Lake and
403
Little Dal groups (Gabrielse et al. 1973; Eisbacher 1978, 1981a,
1985; Aitken 1982; Jefferson & Ruelle 1986). In the Hayhook
Lake area (Fig. 36.1), the Coates Lake Group is onlapped by the
Sayunei Fm. along the unconformity (Eisbacher 1978, 1981a).
The presence of submarine carbonate-clast talus breccias in the
lower Rapitan Group (beneath and within the lower Sayunei
Fm.) on the northeastern margin of the outcrop belt near Shale
Lake (648320 4100 N, 1298220 3700 W) suggests a master normal fault
on that side of the basin. It may have been structurally inverted
during orogenic contraction to form a part of the Plateau Thrust
system. The maroon colour of the Sayunei stems in part from its
derivation from the Coates Lake Group, which is dominated by
reddish siltstones of the Redstone River Fm. Similarly, the olive-
green colouration of the overlying Shezal diamictites reflects its
derivation from the grey carbonates and greenish basalts of the
Little Dal Group.
A knife-sharp, marine, flooding surface separates the Shezal Fm.
from fetid, dark-grey, flaggy limestone and shale of the lower
Twitya Fm. (Eisbacher 1978, 1981a). Where the Rapitan Group
is absent, the Twitya transgresses older units disconformably (Nar-
bonne & Aitken 1995).
Stelfox Member
A low-angle unconformity (Fig. 36.4a) separates the Stelfox dia-
mictite from shelf carbonates and clastics of the underlying
Keele Fm. (Day et al. 2004). West of the palaeoscarp at Stelfox
Mountain (Fig. 36.4b), a sharp disconformity separates the
Stelfox diamictite from parallel-laminated, greenish-grey siltstone
of the underlying Delthore Member (Aitken 1991b). As the
Delthore Member was deposited below storm wave-base, glacial
erosion may be required to account for the sharp disconformity.
A major marine flooding surface separates the Stelfox Member,
or the Keele Fm. where the Stelfox is absent, from the overlying
Ravensthroat Fm. Where the basal contact of the dolostone is
not covered by its own scree (e.g. Cranswick River, Arctic Red
River, Stoneknife River), 10–20 cm of clay separate it from the
underlying diamictite.
The top of the Hayhook limestone is a conformable marine
flooding surface overlain by organic-rich black shale of the
Sheepbed Fm. (Pyle et al. 2004; Shen et al. 2008).
Chemostratigraphy
Carbon isotopes
Carbon-isotopic records have been published for the Little Dal and
Coates Lake groups (Halverson 2006), the Twitya post-glacial
limestone and Keele Fm. (Narbonne et al. 1994; Kaufman et al.
1997; Hoffman & Schrag 2002), and the Ravensthroat –Hayhook
post-glacial carbonate (James et al. 2001). The Little Dal Group
is enriched in 13C relative to the PDB standard, except for a
depleted interval in the Upper Carbonate Fm. The overlying
Coates Lake Group begins with a deep negative excursion to
–8‰ in d13C, followed by a positive excursion to þ8‰ before
settling back to modestly enriched values.
Values of d13C in the Twitya post-glacial limestone increase
up-section from – 2.5‰ at the base. Carbonates of the Keele Fm.
are strongly enriched,  þ 8‰, but plunge to – 8‰ in the last
30 m of section beneath the Stelfox diamictite.
The Ravensthroat Fm. is also depleted in 13C (Fig. 36.6) and values
decline up-section. The Ravensthroat–Hayhook contact coincides
with a –2‰ step function. This is broadly consistent with low-
temperature equilibrium fractionation between dolomite and calcite
(Friedman & O’Neil 1977). Within the Hayhook Fm., d13C rises
by c. 1‰ up-section, with no systematic difference in values
between micro-drilled micrite and sea-floor cement (Fig. 36.6).
404 P. F. HOFFMAN & G. P. HALVERSON

(a) (per mil V-PDB) An alternative, less consequential, interpretation of the d18O
-20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 25 data is that the micrite was preferentially altered by diagenetic
fluids on account of its finer grain size compared with the
20 cements, and therefore its greater surface-to-volume ratio. The
d18O data are more depleted in 18O and display more scatter in
15
the second section (Fig. 36.6b), at Ravensthroat River. This
section illustrates the ease with which d18O, but not d13C, was
diagenetically altered.
10
Sheepbed Fm
Hayhook fm 5 Strontium isotopes
Ravensthroat fm
Stelfox Mb 0 New and previously published (Kaufman et al. 1993, 1997)
87
(m) Sr/86Sr data from Sr-rich (300 –3000 ppm), low-Mn (Mn/
Sr , 0.1) limestones bracketing the Rapitan and Stelfox glacia-
(b) tions in the Mackenzie Mountains are given in Halverson et al.
30 (2007). Non-radiogenic values of 0.70550 –0.70622 (n ¼ 10) are
observed in the Little Dal Group, compared with 0.70644–
25 0.70669 (n ¼ 5) in the Coates Lake Group. No data are currently
available for the post-Rapitan limestone (basal Twitya Fm.), but
20
limestones directly beneath the Stelfox diamictite (uppermost
Keele Fm.) have ratios of 0.70718–0.70720 (n ¼ 3). Sea-floor
δ O δ C
18 13
cements from the Hayhook post-glacial limestone are marginally
15
calcite (micrite) less radiogenic at 0.70714– 0.70716 (n ¼ 4) than the Coates Lake
calcite (cement) Group, whereas the coexisting micrites are more radiogenic, with
dolopelmicrite 10 minimum values of 0.70751–0.70792 (n ¼ 5), which is consistent
James et al. (2001) with preferential alteration. All reported Hayhook values are from
5 samples with Sr concentrations .600 ppm and Mn/Sr , 0.1.

-20 -18 -16 -14 -12 -10 -8 -6 -4 -2
(per mil V-PDB)
Fig. 36.6. C- and O-isotope data from dolomite of the Ravensthroat Fm.
and from coexisting micrite and sea-floor cement of the Hayhook formation
at (a) Shale Lake and (b) Ravensthroat River.
Oxygen isotopes
In order to test the possibility of low-temperature dolomite – calcite
equilibrium fractionation, which would imply a sea-floor origin
for the Ravensthroat dolomite, and also for the possibility that
the Hayhook sea-floor cements and associated micrites might
retain a record of bottom and surface waters, respectively, we
plot unpublished O-isotope data for two sections that we
sampled in detail (Fig. 36.6). Exceptional petrographic preser-
vation of the sea-floor cements motivated this effort. At Shale
Lake, the Ravensthroat-Hayhook contact coincides with a –6‰
step function in d18O, taking the micritic component to be repre-
sentative of the surface waters, from which the Ravensthroat was
also derived (Fig. 36.6a). The magnitude of the step function in
d18O, roughly 3 that in d13C, is compatible with a low-
temperature equilibrium fractionation between dolomite and
calcite (Friedman & O’Neil 1977).
The sea-floor cements are consistently more enriched in 18O
than the micritic component of the Hayhook (Fig. 36.6a), consist-
ent with lower temperatures and/or higher salinities of the bottom
waters, from which the cements were precipitated, relative to the
surface waters. The cements become more enriched in 18O
up-section (Fig. 36.6a), which could indicate an increase in the
temperature and/or salinity gradient, or alternatively an increase
in water depth, with time. The cements exhibit considerably
more scatter in d18O than does the micrite, which could reflect
the difficulty in micro-drilling cement without contamination
from the micrite that fills in between the slender prisms of
former aragonite and even their originally hollow interiors.
Accordingly, the reduction in scatter towards the top of the
Hayhook reflects the more massive and continuous nature of the
uppermost cements.
0
0 (m) Sulphur isotopes and iron speciation
Sulphur isotopes have been measured from carbonate-associated
sulphate in the Ravensthroat and Hayhook formations, as well as
from the barite sea-floor cement at the top of the Ravensthroat
Fm. (M.T. Hurtgen, pers. comm. 2008). The d34SCAS and
d34Sbarite values are mutually consistent and agree with values
from correlative strata in Namibia (Hurtgen et al. 2006), but the
data have yet to be published.
Sulphide S isotopes and Fe-speciation (ratio of highly reactive
Fe to total Fe) have been studied in organic-rich black shales of
the Sheepbed Fm. (Fig. 36.3) in order to infer changes in the oxy-
genation of the atmosphere and the deep ocean, respectively, in
the aftermath of the Stelfox glaciation (Shen et al. 2008). Large
variability (.35‰) in d34Ssulphide is observed from bottom to
top, indicating that sulphate was not limiting (i.e. .2 mM). This
implies higher rates of oxidative weathering (the major source
of marine sulphate) than before the glaciation (Hurtgen et al.
2002). There is a dramatic shift in the Fe-speciation ratio
(FeHR/FeT), however, between 140 and 160 m above the base
of the lithologically homogeneous black shale sequence that is
420 m thick in total. The average value below 140 m is 0.43
(n ¼ 32), indicating bottom water anoxia, whereas above 160 m
it is 0.11 (n ¼ 27), which is consistent with oxygenated bottom
waters (Raiswell & Canfield 1998; Lyons & Severmann 2006;
Canfield et al. 2007).
Calcium isotopes
Calcium isotopes (d44Ca) were measured in limestone of the lower
Twitya Fm., and in dolostone and limestone of the Ravensthroat
and Hayhook formations, respectively (Silva-Tamayo et al. 2010).
The data from the first are consistent with mid-Cryogenian post-
glacial carbonates in Brazil, and the latter with basal Ediacaran
post-glacial sequences in Brazil and Namibia (Silva-Tamayo
et al. 2010). Compared with the younger sequences, the older
d44Ca profiles appear basally truncated, similar to their sequence
stratigraphies and d13C profiles (Hoffman & Schrag 2002). The
 THE MACKENZIE MOUNTAINS, NORTHERN CANADIAN CORDILLERA
younger sequences exhibit paired negative and positive d44Ca
excursions that span twice the entire Phanerozoic range of variabil-
ity (Farkas et al. 2007). The paired anomaly in the lower Hayhook
limestone is preceeded (Ravensthroat Fm.) and followed (upper
Hayhook Fm.) by relatively stable values lying well within the
Phanerozoic range (Silva-Tamayo et al. 2010).
Equilibrium isotope fractionation during carbonate precipitation
maintains seawater roughly 1.0‰ heavier than the sources and
sinks of Ca in the ocean. Thus, imbalance between Ca input and
output will cause deviations in the d44Ca of seawater, and therefore
in marine carbonate. The magnitude of the equilibrium fraction-
ation is expected to vary with temperature and carbonate precipi-
tation rate, but relatively modest variability in d44Ca (n ¼ 10)
within the Ravensthroat Fm. (Silva-Tamayo et al. 2010), deposited
during global deglaciation when changes in temperature and pre-
cipitation rate may have been large, implies that those dependen-
cies were either small or counteracting at that time. The paired
negative and positive excursions of 1.0‰ apiece in the lower
Hayhook limestone and correlatives (n ¼ 31) represents first, a
large excess of Ca input over output, followed by the reverse, a
large excess of Ca output over input (Silva-Tamayo et al. 2010).
Synglacial hematite jaspilite (Fe-formation), Rapitan Group
REE geochemistry of the Sayunei jaspilite (Klein & Buekes, 1993)
is described in Hoffman et al. (2011). Planavsky et al. (2010)
present elemental data (P, Fe, Mn and Al) for jaspilite samples
from the Sayunei Fm. (n ¼ 41) and correlative strata (Upper
Tindir Group) in the Tatonduk inlier. In common with other
Cryogenian glaciogenic jaspilites, the samples have much higher
phosphate contents (average P/Fe ¼ 1.44%, n ¼ 44) than Palaeo-
proterozoic or Archaean Fe-formations (Planavsky et al. 2010).
Iron isotopes (d57Fe) and accompanying rare-earth elements
(REE), including redox-sensitive cerium anomalies (Ce/Ce*),
have been measured through the 16.4-m-thick interval of hematitic
mudstone and hematite-jaspilite (banded Fe-formation) at the top
of the Sayunei Fm. in the Hayhook Lake area (Fig. 36.7). The
405
d57Fe values increase systematically upsection from – 0.7‰ to
1.2‰. The largest rise coincides with the transition from hematitic
mudstone to hematite-jaspilite, which also corresponds to the
highest Fe concentration (c. 58 wt% Fe2O3). The increase in
d57Fe is accompanied by a rise in Ce/Ce* (Fig. 36.7).
Palaeontology
Neoproterozoic palaeontology of the Mackenzie Mountains is
summarized in Narbonne & Aitken (1995). When combined
with contemporary work in the Wernecke Mountains (Narbonne
& Hofmann 1987) and ongoing stratigraphic and palaeontological
studies in correlative strata of the Tatonduk inlier, a pattern of
change over time emerges that is broadly consistent with other
areas (Macdonald et al. 2010a, b; Macdonald & Cohen 2011). A
number of eukaryotic crown groups, both algal and protistan,
had evolved and diversified before the Rapitan glaciation. The
record of these groups is cryptic from shortly before the Rapitan
glaciation until long after the Stelfox glaciation.
Soft-bodied, Ediacara-type macrofossils, representing benthic
polypoid and frond-like organisms, occur in the upper Ediacaran
Blueflower Fm. (Fig. 36.1), where they coexist with infaunal
burrows (simple, meandering and subordinate patterned) that
extend into the overlying Risky Fm. (Narbonne & Aitken 1990;
Narbonne 1994). In the same area, near Sekwi Brook (Fig. 36.1),
simple radially symmetric body fossils and rare trace fossils
occur in the middle Sheepbed Fm. (Narbonne & Aitken 1990;
Narbonne 1994). The Stelfox glaciation is not recognized at
Sekwi Brook, but at Shale Lake (Fig. 36.1), Fe-speciation data
indicate that bottom waters became oxygenated just below the
middle Sheepbed Fm. (Shen et al. 2008).
An assemblage of simple, centimetric annuli and discs, thought
to be biogenic, occur in the upper Twitya Fm. (Fig. 36.3a) of the
Sayunei Range (Hofmann et al. 1990), south of Twitya River
(Fig. 36.1). As possible sponge-grade metazoan fossils pre-dating
the terminal Cryogenian glaciation, the ‘Twitya-discs’ have
recently been joined by calcareous structures at different horizons
Fig. 36.7. Fe isotopes (d57Fe), wt% Fe2O3
and Ce anomaly data (Ce/Ce*) from the
Sayunei Fe-formation in the Hayhook Lake
reference section. Data tabulated in
Halverson et al. (2011).
406 P. F. HOFFMAN & G. P. HALVERSON
in South Australia (Wallace & Woon 2008; Maloof et al. 2010),
and a sponge biomarker in the Oman Salt Basin (Love et al. 2009).
Other characteristics
The Crest deposit at Iron Creek (Fig. 36.3d), near Snake River, is a
hematite –jaspilite interval 120 m thick within glaciogenic dia-
mictites of the Rapitan Group (Yeo 1986; Klein & Beukes
1993). The deposit contains 5.6 billion tonnes of ‘ore’ averaging
47.2% Fe and regional reserves are estimated to exceed 18.6
billion tonnes (Yeo 1986). Because of remoteness and difficult
terrain, the deposit has never been economic.
Palaeolatitude and palaeogeography
Palaeomagnetic data from the Mackenzie Mountains relating to
the Rapitan and Stelfox glaciations have been reviewed (Evans
2000; Evans & Raub 2011). Morris (1977) recognized three
natural remanent magnetic components in the Rapitan Group, of
which the one inferred to be the youngest (Cretaceous?) gives a
high palaeolatitude and the other two give low palaeolatitudes. A
larger data set obtained by Park (1997) gives broadly similar
results, but only after anomalous declinations are corrected for
large vertical-axis rotations of presumed structural origin. The
high-inclination poles were acquired during tectonic folding and
the low-inclination poles before that time, although Morris
(1977) and Park (1997) disagreed on which of the two clusters
of low-inclination poles was older. The palaeolatitudes of their
inferred oldest components are 08 + 28 (Morris 1977) and
06 + 78 (Park 1997). Neither result is definitively syndepositional,
but the near-equatorial palaeolatitudes agree with more recent
palaeomagnetic results from mid-Neoproterozoic (800 –740 Ma)
sedimentary rocks in western Laurentia (Weil et al. 2004, 2006).
Shallow mafic igneous rocks provide more reliable palaeomag-
netic results than sedimentary rocks because they are strongly
magnetized, resist low-temperature remagnetization, have not
been compacted (no flattening of palaeomagnetic inclination),
and provide baked-contact tests of the primary age of remanent
magnetization, which can be directly dated as the crystallization
age of the rock. The Franklin Large Igneous Province (LIP) com-
prises comagmatic mafic dykes, sills and lavas extending collec-
tively for 2500 km across Arctic Laurentia. The grand mean
palaeopole (8.48N, 163.88E, A95 ¼ 2.88, n ¼ 78 sites) for the
Franklin LIP (Denyszyn et al. 2009) places the Mount Harper
area and the Mackenzie Mountains, respectively, at 21+3 and
18 + 38N palaeolatitude at the time of magmatism, which
spanned the onset of the Rapitan glaciation (see below). These
results for the Rapitan glaciation are more reliable than those
from the Rapitan Group itself.
There are no direct palaeomagnetic constraints on the Stelfox
glaciation. McMechan (2000b) cites Park (1994) as giving a
palaeolatitude of c. 358 for the Stelfox glaciation, but this was
based on palaeopoles from the Risky Fm., which is c. 2 km strati-
graphically above the Stelfox Member (Fig. 36.3a). Park (1994)
estimated the Risky Fm. to be younger than the 615 Ma Long
Range Dykes of Newfoundland and Labrador (Kamo et al. 1989;
Kamo & Gower 1994). Although more results are needed for
better statistics, the preliminary palaeopole for the Long Range
Dykes (19.08N, 355.38E, A95 ¼ 17.48, n ¼ 5 dykes) places the
Mackenzie Mountains near the palaeoequator at 615 Ma (McCaus-
land et al. 2007).
Geochronological constraints
In the Mount Harper area of the Ogilvie Mountains, 310 km west
of Corn Creek (Fig. 36.1), up to 120 m of massive and bedded
diamictite, correlative with the lower Rapitan Group, was depos-
ited during the waning stages of bimodal volcanism in an active
rift basin (Mustard & Roots 1997; Macdonald et al. 2010b). The
presence of faceted and striated clasts, bed-penetrating
dropstones, common outsized and exotic clasts, and glacial push
structures (Hart River inlier) within a marine succession support
a glacial marine origin for the diamictite at an ice-sheet grounding-
line. A quartz-phyric rhyolite within the volcanic pile, stratigraphi-
cally beneath the diamictite, and a felsic tuff within diamictite,
c. 60 m above the base of the glacial –periglacial sequence, give
U –Pb zircon dates of 717.43 + 0.14 and 716.47 + 0.24 Ma
(2s), respectively (Macdonald et al. 2010b). These dates are indis-
tinguishable in age from mafic dykes and sills of the Franklin
Large Igneous Province (Heaman et al. 1992; Denyszyn et al.
2009; Macdonald et al. 2010b), which straddled the palaeoequator
at the time of their emplacement. The beginning of glacial marine
sedimentation in the Mount Harper rift basin at 717 Ma is the best
indicator of the onset of Rapitan glaciation in the northern
Canadian Cordillera (Macdonald et al. 2010b). These data are
consistent with maximum age constraints obtained from the
Mackenzie Mountains and Windermere Supergroup, including
a weighted mean 207Pb/206Pb date of 779 + 2.3 Ma (Harlan
et al. 2003) obtained for two baddeleyite analyses from a
gabbro sill (Carcajou Canyon Gabbro) intruding the Tsezotene
Fm. (Fig. 36.3), and a weighted average 207Pb/206Pb date of
755 + 18 Ma obtained for two concordant but low-Pb zircon
grains from a single leucogranite ‘dropstone’ within turbidites
close to the base of the Sayunei Fm. near Shale Lake (Ross &
Villeneuve 1997).
The Stelfox glaciation has no independent radiometric age con-
straint. Mafic tuffs in the upper Keele Fm. (Fig. 36.3a) contain only
xenocrystic zircons. The Stelfox glaciation is assumed, largely on
the basis of its post-glacial carbonate sequence, to be correlative
with the terminal Cryogenian glaciation (Knoll et al. 2006),
which terminated in 635 Ma.
Discussion
Palaeoenvironmental interpretation of the Rapitan Group
The Rapitan Group is widely interpreted as a glacial marine suc-
cession because of the occurrence of faceted and striated clasts,
sand aggregates (‘till pellets‘), erratic dropstones with impact-
related structures, and soft-sediment deformation horizons (e.g.
push structures, scaly matrix in particular diamictite sheets) of
apparent glacitectonic origin (Young 1976; Eisbacher 1981a,
1985; Yeo 1981; Aitken 1982; Macdonald et al. 2010b). Eisbacher
(1985) presents a comprehensive facies model ranging from fluc-
tuating grounding-line diamictites (Shezal Fm.) through transi-
tional facies to distal proglacial turbidites and fine-grained
suspension fallout accumulated in structurally controlled, deep-
water basins (Sayunei Fm.). The distal nature of the Sayunei turbi-
dites, characterized by Bouma C-D sequences, is emphasized (Yeo
1981; Eisbacher 1985) and the source of suspended sediment is
inferred to be turbid meltwater plumes emanating from discharge
sites at ice grounding-lines at the basin margin (Young 1976; Yeo
1981; Klein & Beukes 1993). Eisbacher (1985, p. 242, fig. 8)
suggests that the paucity of dropstones and diamictite, respect-
ively, in the Sayunei Fm. reflect a suppression of icebergs and a
buttressing of ice grounding-lines by a ‘stable sea-ice cover’
(Dowdeswell et al. 2000). Such a permanent ice cover would
account for the low organic content and deep maroon colour of
the Sayunei mudrocks, in contrast to the basinal shale and lime-
stone of the underlying non-glacial Copper Cap Fm., which are
mostly fetid and black in colour. Removal of the permanent ice
cover by climate warming would have allowed icebergs and ice
grounding-lines to advance basinward at the time of the
Sayunei-Shezal transition (Eisbacher 1985).
 THE MACKENZIE MOUNTAINS, NORTHERN CANADIAN CORDILLERA
Palaeomagnetic data place the Rapitan rift basins in palaeolati-
tudes comparable to the present southern Red Sea Rift basin
(Evans & Raub 2011). At the Last Glacial Maximum (LGM,
c. 20 ka), arguably as severe a glaciation as any in the Phanerozoic
eon, the lowest nearby glacial moraines, in the Ethiopian Highlands,
are 3750 m above sea level (Umer et al. 2004). Air temperatures at
that altitude are on average 26 8C colder than the same latitude at sea
level, reached by Rapitan glaciers. This gives some sense of the con-
trast between the Rapitan glaciation and the LGM.
Origin of ‘transitional-facies’ Fe-formation
A local hydrothermal source of Fe for transitional-facies
Fe-formation (Fig. 36.3d) was favoured by Yeo (1981) and
Young (1988, 2002). However, when compared with Neoarchaean
and Palaeoproterozoic Fe-formations, the rare-earth element
(REE) chemistry of the Rapitan Fe-formation is ‘much less dis-
tinctly influenced by hydrothermal input’, suggesting that such
input was ‘highly diluted by ocean waters at Rapitan time’
(Klein & Beukes 1993). Moreover, where submarine volcanism
was demonstratively active during the Rapitan glaciation (Mount
Harper area), Fe-formation is absent. Where Fe-formation is
present, in the Tatonduk inlier (Macdonald & Cohen 2011) and
the Mackenzie Mountains, interbedded volcanic rocks are non-
existent (Macdonald et al. 2010b).
Global Fe-speciation data suggest that Cryogenian and Edia-
caran deep waters were ferruginous, rather than euxinic (Canfield
et al. 2008). This raises the question why Fe-formation (hematite
jaspilite) was only deposited during glacial times, and possibly
only during the ‘Sturtian’ glaciation (Hoffman et al. 2011). If the
Rapitan rift-basins were covered by thick permanent marine ice
during fine-grained sedimentation of the Sayunei Fm. (Eisbacher
1985), primary productivity would have crashed, limiting bacterial
sulphate reduction and allowing high concentrations of Fe in acidic
waters in the absence of hydrothermal sources (Mikucki et al.
2009). The Crest Fe deposit at Iron Creek (Fig. 36.3d) is situated
within a sequence of transitional-facies diamictites (Eisbacher
1985), where local oxidizing power capable of titrating dissolved
Fe would have been provided by subglacial meltwater discharges
at basin-margin ice grounding-lines, assuming air bubbles in the
meteoric ice (compressed snow) contained oxygen.
Low primary productivity related to the permanent ice cover as
well as low seawater pH would account for elevated P/Fe ratios in
the Fe-formation at Iron Creek and perhaps other synglacial Cryo-
genian Fe-formations (Planavsky et al. 2010). The explanation for
the high P/Fe ratio favoured by Planavsky et al. (2010) involves
‘unprecedented continental P fluxes during postglacial and inter-
glacial time periods, given the extraordinary extent and duration
of Cryogenian ice cover and the high levels of P delivery expected
from glaciated catchments’. The problem with their explanation is
that six of the seven Cryogenian Fe-formations in their database
were deposited before the termination of the first major
Cryogenian glaciation.
Origin and stratigraphic localization of ‘basin-facies’
Fe-formation
Why does Fe-formation in the basin facies (Eisbacher 1985) occur
at the top of the Sayunei Fm., disconformably beneath the Shezal
diamictite (Fig. 36.3b,c)? If we accept Eisbacher’s (1985) expla-
nation that the advance of the Shezal grounding-line was triggered
by the removal of permanent sea ice over the basin, two factors
would have contributed to the deposition of Fe-formation. First,
as the ice cover thinned and finally disappeared, anoxic and oxy-
genic photosynthesis could have precipitated Fe2O3-precursor
from anoxic Fe(II)-rich basin waters. Second, after the ice cover
was removed, air –sea gas exchange and wind-driven mixing
407
would have quickly oxygenated the surface waters, leading to
abiotic Fe(III) precipitation.
After the Sayunei basin became ice-free, the water vapour
content of air blowing across the basin would have increased due
to evaporation. This would likely have caused a positive change
in the mass-balance of downwind ice sheets, due to increased pre-
cipitation and decreased melting on account of fog and clouds. On
a ‘modern Snowball Earth’ (i.e. present geography with complete
sea-ice cover, Voigt & Marotzke 2009), the Red Sea Rift would be
isolated from the Indian Ocean because of the roughly 1.0 km fall
in global mean sea level (glacioeustasy). As tropical sea-surface
temperatures rose to melting point, sea ice on the Red Sea basin
would disappear, while the Indian Ocean would remain ice-bound
because of sea – glacial flow from the south (Warren et al. 2002;
Goodman & Pierrehumbert 2003). Accordingly, the Shezal dia-
mictite would represent the time interval between the opening of
the Sayunei basin ‘oasis’ and overall glacial termination. Once
illuminated, primary productivity in the basin, now crowded
with icebergs, would be stimulated.
Fe-isotope and Ce-anomaly data (Fig. 36.7) shed additional light
on the basin-facies Fe-formation near Hayhook Lake (Fig. 36.3c).
Controls on Fe-isotope variation in natural systems are not well
known, and data from ancient rocks are still quite limited. Labora-
tory experiments indicate that broadly similar equilibrium and
kinetic isotope fractionations are associated with both abiotic
and biological Fe-oxidation pathways (Johnson & Beard 2006).
Consider the d57Fe trend in the Fe-formation (Fig. 36.7) not in
terms of a secular change in seawater Fe-isotope composition,
but as a change in net Fe-isotope fractionation (D57Fehem-Fe2þ)
between hematite (or ferric-oxyhydroxide precursor) and dis-
solved Fe(II) across a redox chemocline in the water column
(Halverson et al. 2011). This could come about either because
the Fe-isotope fractionation varied as a function of the Fe(II)
concentration, assumed to increase with depth across the chemo-
cline, or because of progressive oxidation of Fe(II) as it upwelled
across the chemocline. Progressive oxidation would drive the iso-
topic composition of the dissolved Fe(II) to more 57Fe-depleted
values, lowering the d57Fe of the hematite produced. In either
case, the d57Fe of hematite should decrease from the base of the
chemocline to the top. Accordingly, the upward increase in
d57Fe observed within the Fe-formation (Fig. 36.7) suggests an
increase in water depth with time, which is consistent with more
reducing conditions up-section inferred from the Ce/Ce* data
and with transgression inferred on sedimentological grounds
over the same stratigraphic interval (Klein & Beukes 1993). As
the Sayunei Fe-formation is directly overlain by the Shezal diamic-
tite (Fig. 36.3), the rise in relative sea level is logically attributed to
lithospheric downwarping under the load of the advancing Shezal
ice sheet (Halverson et al. 2011).
Palaeoenvironmental interpretation of the Stelfox (Ice Brook)
glaciation
The Stelfox Member of the Ice Brook Fm. is a subaqueous unit
interpreted as glaciomarine because of the presence of faceted
and striated clasts in diamictite, sand and granule aggregates
(‘till pellets’), dropstones with impact-related structures in finely
laminated mudrocks, extremely angular quartz grains derived
from well-rounded quartz arenites of the Keele Fm., and diamictite
in a shallow-shelf setting where large-scale mass-flows are less
likely to occur (Aitken 1991a, b).
In discussing the Stelfox-equivalent Vreeland Diamictites in
northeastern British Columbia, McMechan (2000a, b) described
mud-rich diamictites and related fine-grained terrigenous facies
as resembling deposits associated with temperate, wet-based gla-
ciers in southern Alaska, in contrast to those of cold-based glaciers
in Antarctica. She argued that they are inconsistent with the
existence of a Cryogenian ‘Snowball’ Earth, assuming that its
408 P. F. HOFFMAN & G. P. HALVERSON
mean-annual tropical surface temperature was close to that of
present-day Antarctica. She noted the prevalence of biogenic
facies (e.g. diatomaceous ooze) beneath Antarctic ice shelves, on
the Antarctic continental shelf and adjacent to the outlets of
fast-flowing ice streams, implying a near-absence of fine-grained
terrigenous input from meltwater plumes (Domack 1988; Ander-
son et al. 1991). Citing the same authors, she posited that signifi-
cant meltwater flow requires surface melting and therefore
summer air temperatures above 0 8C, which do not occur on Ant-
arctica or on a ‘Snowball’ Earth (McMechan 2000b).
Much has been learned about the Antarctic glacial regime since
the studies cited by McMechan (2000a, b). More than 150 subgla-
cial lakes have been found, some at the origins of fast-flowing ice
streams, and they are not stagnant but are subject to active hydro-
dynamic exchange (Siegert et al. 2005; Wingham et al. 2006; Bell
et al. 2007). Modelling suggests that basal melting occurs under
most of the interior of the East Antarctic Ice Sheet and all of its
outlet ice streams (Pattyn 2010). Recent collapse of the Larsen B
ice shelf revealed the former existence of both terrigenous and bio-
genic sedimentation beneath the ice shelf (Domack et al. 2005;
Damiani & Giorgetti 2008). Springtime diatom blooms on the
East Antarctic margin may be triggered, not inhibited, by Fe-rich
sub-glacial meltwater disharges, from which terrigenous mud sub-
sequently settles (Leventer et al. 2006). There is no doubt that
accumulation rates for muds are lower in polar than in temperate
settings, but sedimentation rates in the Cryogenian are as yet
unknown in the absence of biogenic facies, varve chronologies,
or data on meteoritic dust content. We agree that ‘any model for
Neoproterozoic glaciation should include a significant period of
more moderate glacial conditions prior to the end of glaciation’
(McMechan 2000b), but we maintain that even the most extreme
glaciation must experience more moderate conditions at its termin-
ation, especially in tropical marine environments, and that this is
fully consistent with the hysteresis loop predicted by the Snowball
Earth hypothesis (Hoffman 2009).
Calcium isotope record of the Stelfox glaciation and
its aftermath
The Snowball Earth hypothesis (Hoffman & Schrag 2002) pro-
vides a potential explanation for the paired negative and positive
d44Ca excursions and for their stratigraphic location in the lower
Hayhook Fm. (Silva-Tamayo et al. 2010). During a snowball gla-
ciation, a steady rise in CO2 drives carbonate dissolution in the
ocean (a source of Ca) and precludes carbonate deposition
(Hoffman & Schrag 2002, fig. 10). This is consistent with the
absence of primary carbonate in the relevant synglacial strata.
The d44Ca of dissolved Ca in snowball brine should therefore
fall towards the value of the Ca input. The first carbonate that pre-
cipitates from the brine will be depleted in 44Ca by c. 1.0‰ com-
pared with normal carbonates. In the snowball aftermath, Ca input
to the ocean from erosion is high, but rapid ocean warming and the
more leisurely drawdown of CO2, through silicate weathering,
drive excess Ca output (carbonate deposition) over input, raising
d44Ca. During deglaciation, however, ocean surface waters are
flooded by glacial meltwater, marine and terrestrial, and are thus
isolated from the deep brine by an exceptionally stable density
stratification (Shields 2005). Consequently, the 44Ca-depleted
snowball signature is not observed in the Ravensthroat Fm.,
which was deposited in surface waters during the snowball melt-
down (James et al. 2001). As melting neared completion, the
flux of meltwater waned and ocean mixing by winds and tides
resumed. The negative d44Ca excursion in the lower Hayhook Fm.
(Silva-Tamayo et al. 2010) records the mixing of 44Ca-depleted
snowball brine and the meltwater lid. It does not record the gener-
ation of the isotope anomaly in dissolved Ca of the brine, but
rather its delayed expression as precipitated carbonate when the
brine was mixed into the meltwater lid, which was saturated with
carbonate from dissolution of carbonate rock ‘flour’, exposed by
deglaciation (Fairchild 1993). Once the ocean was mixed, the post-
glacial excess of Ca output over input drove the positive d44Ca excur-
sion. Silva-Tamayo et al. (2010), who assume that the stratigraphic
expression of the negative d44Ca excursion corresponds to its time
of formation, attribute the negative excursion to Ca influx from
weathering immediately after deglaciation. Their model requires
that Ca input greatly exceeded output, which appears unlikely at a
time of ocean warming (raising saturation) and mixing (facilitating
degassing), when the surface ocean is known to have been critically
oversaturated from the widespread occurrence of syndeglacial cap
dolostones (Hoffman et al. 2007).
Correlation and palaeogeography
The post-glacial carbonate sequences following the older and
younger Cryogenian glaciations are stratigraphically, lithologi-
cally and isotopically distinct (Kennedy et al. 1998; Hoffman &
Schrag 2002). They provide a surprisingly successful basis for dis-
tinguishing the glaciations globally (Knoll et al. 2006; Hoffman
et al. 2011). As cosmopolitans, however, they are less well
suited for palaeogeographic reconstruction. In a wide orogen like
the North American Cordillera, the distinction between Neoproter-
ozoic successions that are indigenous to Laurentia and those orig-
inating elsewhere is fundamental (Johnston 2008; Hildebrand
2009). All exposed Neoproterozoic rocks in the Mackenzie Moun-
tains (Fig. 36.1) were thrust northeastward relative to cratonic
Laurentia in the Upper Cretaceous –Paleocene. Sequence and
chemostratigraphic comparison of successions deposited before,
between and after the Cryogenian glaciations implies that the
Tatonduk, Coal Creek, Hart River and Wernecke inliers of the
Yukon Territory expose a Neoproterozoic succession continuous
with that of the Mackenzie Mountains (Macdonald & Roots
2009; Macdonald et al. 2010a, b; Macdonald & Cohen 2011).
The co-occurrence of 718– 716 Ma mafic magmatism in the
Mount Harper area (Coal Creek inlier) and on cratonic Laurentia
(Franklin LIP) therefore implies that the Neoproterozoic succes-
sion of the Ogilvie and Mackenzie Mountains formed on the
margin of Laurentia.
In contrast, the lithostratigraphically alien Katakturuk Dolomite
(Ediacaran) in the northeastern Brooks Range, Arctic Alaska, is
unlikely to have had a Laurentian affinity before Siluro-Devonian
time, at the earliest (Macdonald et al. 2009). Accordingly, the stra-
tigraphically underlying Hula Hula diamictite represents a Cryo-
genian glaciation of Chukotka-Arctic Alaska, not Laurentia, and
the Mount Copleston volcanics within and beneath the diamictite
are predicted to be asynchronous with the Franklin LIP.
What about Neoproterozoic successions in the rest of the North
American Cordillera? They host named glacial horizons in British
Columbia (Toby and Vreeland), Idaho (Scout Mountain), Utah
(Dutch Peak and Mineral Fork), California (Surprise and Wild-
rose) and Sonora (Mina el Mezquite). Conventionally, all have
been assumed to originate at the rifted margin of Laurentia
(Stewart 1972). In the Cordilleran collisional model (Johnston
2008; Hildebrand 2009), all except possibly the Mineral Fork ori-
ginated elsewhere. In that model, they collided with the Laurentian
margin in the Late Cretaceous as part of a composite ribbon con-
tinent named ‘Rubia’ (Hildebrand 2009) or ‘Saybia’ (Johnston
2008). Targeted detrital zircon dating of the various glacial hor-
izons, and their host successions, should be an effective means
of testing the Rubia/Saybia hypothesis. From the perspective of
the Mackenzie Mountains, the sequence stratigraphy and lithology
of Neoproterozoic successions assigned to the Windermere Super-
group in British Columbia (Ross et al. 1989) and to the South are
not similar. The differences may reflect facies changes, as conven-
tionally assumed, but it seems fair to conclude that Neoproterozoic
strata of the North American Cordillera provide no grounds at
present for rejecting the Rubia/Saybia hypothesis. To date, all
 THE MACKENZIE MOUNTAINS, NORTHERN CANADIAN CORDILLERA
that we do know is that palaeomagnetic data from the Franklin LIP
place the then northward-facing rifted margin of Laurentia in the
Mackenzie Mountains at 18 + 38N latitude at the onset of the
Rapitan glaciation, and possibly closer to the palaeo-equator
during the Stelfox glaciation.
Fieldwork in the Mackenzie Mountains was licensed by the Aurora Research
Institute and supported by a grant from the Astrobiology Institute of the US
National Aeronautics and Space Administration (NASA), and by grants
EAR-9905495 and EAR-0417422 from the US National Science Foundation
(NSF) to P.F.H. C. A. Ferguson, M. T. Hurtgen, F. A. Macdonald, A. C.
Maloof, Y. Shen and A. V. Turchyn participated in the fieldwork. C and O iso-
topes were analysed at the Laboratory for Geochemical Oceanography at
Harvard University, directed by D. P. Schrag. Sr isotope data were obtained at
the Geochronology Laboratory at the Massachusetts Institute of Technology,
directed by S. A. Bowring. Fe isotopes were measured at the Laboratoire des
Mécanismes et Transferts en Géologie (LMTG) at the Université P. Sabatier,
Toulouse, France, directed by F. Poitrasson. P.F.H. acknowledges discussions
with J. D. Aitken, R. W. Dalrymple, E. W. Domack, G. H. Eisbacher,
H. Gabrielse, L. M. Heaman, H. J. Hofmann, N. P. James, C. W. Jefferson,
A. J. Kaufman, T. Kurt Kyser, M. E. McMechan, W. A. Morris, D. C. Murphy,
G. M. Narbonne, J. K. Park, G. M. Ross, G. M. Yeo and G. M. Young. Comments
by E. Arnaud, G. Narbonne, E. C. Turner and G. Yeo substantially improved
the manuscript. This represents a contribution of the IUGS- and UNESCO-
funded IGCP (International Geoscience Programme) project #512.
References
Aitken, J. D. 1981. Stratigraphy and sedimentology of the Upper Proter-
ozoic Little Dal Group, Mackenzie Mountains, Northwest Territories.
In: Campbell, F. H. A. (ed.) Proterozoic Basins of Canada. Geologi-
cal Survey of Canada Paper, 81 – 10, 47– 71.
Aitken, J. D. 1982. Precambrian of the Mackenzie fold belt — a strati-
graphic and tectonic overview. In: Hutchinson, R. W., Spence, C.
D. & Franklin, J. M. (eds) Precambrian Sulfide Deposits. Geologi-
cal Association of Canada Special Paper, 25, 149– 161.
Aitken, J. D. 1991a. Two late Proterozoic glaciations, Mackenzie Moun-
tains, northwestern Canada. Geology, 19, 445– 448.
Aitken, J. D. 1991b. The Ice Brook Formation and post-Raptian, Late
Proterozoic glaciation, Mackenzie Mountains, Northwest Territories.
Geological Survey of Canada Bulletin, 404.
Aitken, J. D. & Long, D. G. F. 1978. Mackenzie tectonic arc – reflection
of early basin configuration? Geology, 6, 626–629.
Allen, P. A. & Hoffman, P. F. 2005. Extreme winds and waves in the
aftermath of a Neoproterozoic glaciation. Nature, 433, 123–127.
Anderson, J. B., Kennedy, D. S., Smith, M. J. & Domack, E. W. 1991.
Sedimentary facies associated with Antarctica’s floating ice masses.
In: Anderson, J. B. & Ashley, G. M. (eds) Glacial Marine Sedimen-
tation: Paleoclimatic Significance. Geological Society of America
Special Paper, 261, 1– 25.
Assereto, R. L. A. M. & Kendall, C. G. St. C. 1977. Nature, origin and
classification of peritidal tepee structures and related breccas. Sedi-
mentology, 24, 153– 210.
Bell, R. E., Studinger, M., Shuman, C. A., Fahnestock, M. A. &
Joughin, I. 2007. Large subglacial lakes in East Antarctica at the
onset of fast-flowing ice streams. Nature, 445, 904– 907.
Canfield, D. E., Poulton, S. W. & Narbonne, G. M. 2007. Late-
Neoproterozoic deep-ocean oxygenation and the rise of animal life.
Science, 315, 92 – 95.
Canfield, D. E., Poulton, S. W., Knoll, A. H., Narbonne, G. M.,
Ross, G., Goldberg, T. & Strauss, H. 2008. Ferruginous conditions
dominated later Neoproterozoic deep-water chemistry. Science, 321,
949– 952.
Dalrymple, R. W. & Narbonne, G. M. 1996. Continental slope sedi-
mentation in the Sheepbed Formation (Neoproterozoic, Windermere
Supergroup), Mackenzie Mountains, N.W.T. Canadian Journal of
Earth Sciences, 33, 848–862.
Damiani, D. & Giorgetti, G. 2008. Provenance of glacial-marine sedi-
ments under the McMurdo/Ross Ice Shelf (Windless Bight, Antarc-
tica): Heavy minerals and geochemical data. Palaeogeography,
Palaeoclimatology, Palaeoecology, 260, 262– 283.
409
Day, E. S., James, N. P., Narbonne, G. M. & Dalrymple, R. W.
2004. A sedimentary prelude to Marinoan glaciation, Cryogenian
(Middle Neoproterozoic) Keele Formation, Mackenzie
Mountains, northwestern Canada. Precambrian Research, 133,
223– 247.
Denyszyn, S. W., Halls, H. C., Davis, D. W. & Evans, D. A. D. 2009.
Paleomagnetism and U– Pb geochronology of Franklin dykes in High
Arctic Canada and Greenland: a revised age and paleomagnetic pole
constraining block rotations in the Nares Strait region. Canadian
Journal of Earth Sciences, 46, 689– 705.
Domack, E. W. 1988. Biogenic facies in the Antarctic glacimarine
environment: basis for a polar glacimarine summary. Palaeogeogra-
phy, Palaeoclimatology, Palaeoecology, 63, 357– 372.
Domack, E. W., Duran, D. et al. 2005. Stability of the Larsen B ice shelf
on the Antarctic Peninsula during the Holocene epoch. Nature, 436,
681– 685.
Dowdeswell, J. A., Whittington, J. A., Jennings, A. E., Andrews, J.
T., Mackensen, A. & Marienfield, P. 2000. An origin for lami-
nated glacimarine sediments through sea-ice build-up and suppressed
iceberg rafting. Sedimentology, 47, 557– 576.
Eisbacher, G. H. 1978. Re-definition and Subdivision of the Rapitan
Group, Mackenzie Mountains. Geological Survey of Canada Paper
77– 35.
Eisbacher, G. H. 1981a. Sedimentary tectonics and glacial record in the
Windermere Supergroup, Mackenzie Mountains, northwestern
Canada. Geological Survey of Canada Paper, 80– 27.
Eisbacher, G. H. 1981b. Late Precambrian tillites of the northern Yukon-
Northwest Territories region, Canada. In: Hambrey, M. J. &
Harland, W. B. (eds) Earth’s Pre-Pleistocene Glacial Record.
Cambridge University Press, Cambridge, 724–727.
Eisbacher, G. H. 1985. Late Proterozoic rifting, glacial sedimentation and
sedimentary cycles in the light of Windermere deposition, western
Canada. Palaeogeography, Palaeoclimatology, Palaeoecology, 51,
231– 254.
Evans, D. A. D. 2000. Stratigraphic, geochronological and Paleomagnetic
constraints upon the Neoproterozoic climatic paradox. American
Journal of Science, 300, 347–433.
Evans, D. A. D. & Raub, T. D. 2011. Neoproterozoic glacial Palaeolati-
tudes: a global update. In: Arnaud, E., Halverson, G. P. & Shields,
G. (eds) The Geological Record of Neoproterozoic Glaciations. Geo-
logical Society, London, Memoirs, 36, 93 –112.
Fairchild, I. J. 1993. Balmy shores and ice wastes: the paradox of carbon-
ates associated with glacial deposits in Neoproterozoic times. Sedi-
mentology Review, 1, 1 –16.
Farkas, J., Böhm, F. et al. 2007. Calcium isotope record of Phanerozoic
oceans: implications for chemical evolution of seawater and its
causative mechanisms. Geochimica et Cosmochimica Acta, 71,
5117–5134.
Friedman, I. & O’Neil, J. R. 1977. Chapter KK. Compilation of stable
isotope fractionation factors of geochemical interest. In: Fleischer,
M. (ed.) Data of Geochemistry 6th edn. United States Geological
Survey Professional Paper 440– KK, Washington, DC.
Gabrielse, H., Blusson, S. L. & Roddick, J. A. 1973. Geology of the
Flat River, Glacier Lake and Wrigley Lake map-areas, District of
Mackenzie and Yukon Territory. Geological Survey of Canada
Memoir 366.
Goodman, J. & Pierrehumbert, R. T. 2003. Glacial flow of floating
marine ice in ‘Snowball Earth’. Journal of Geophysical Research,
108, 3308, doi: 10.1029/2002JC001471.
Green, L. H. & Godwin, C. I. 1963. Snake River area: mineral industry
of Yukon Territory and southwestern District of Mackenzie, 1962.
Geological Survey of Canada Paper 63 –38, 15 – 18.
Halverson, G. P. 2006. A Neoproterozoic chronology. In: Xiao, S. &
Kaufman, A. J. (eds) Neoproterozoic Geobiology and Paleobiology.
Springer, Dordrecht, 231– 271.
Halverson, G. P., Maloof, A. C. & Hoffman, P. F. 2004. The Marinoan
glaciation (Neoproterozoic) in northeast Svalbard. Basin Research,
16, 297– 324.
Halverson, G. P., Dudás, F. Ö., Maloof, A. C. & Bowring, S. A. 2007.
Evolution of the 87Sr/86Sr composition of Neoproterozoic seawater.
Palaeogeography, Palaeoclimatology, Palaeoecology, 256,
103– 129.
410 P. F. HOFFMAN & G. P. HALVERSON
Halverson, G. P., Poitrasson, F., Hoffman, P. F., Nedelec, A.,
Montel, J.-M. & Kirby, J. 2011. Fe isotope and trace element geo-
chemistry of the Neoproterozoic syn-glacial Raptian iron formation.
Earth and Planetary Science Letters, doi: 10.1016/j.epsl.2011.06.21.
Hambrey, M. J. & Spencer, A. M. 1987. Late Precambrian Glaciation of
Central East Greenland. Meddelelser om Grønland, Geoscience 19.
Harlan, S. S., Heaman, L., LeCheminant, A. N. & Premo, W. R. 2003.
Gunbarrel mafic magmatic event: a key 780 Ma time marker for
Rodinia plate reconstructions. Geology, 31, 1053–1056.
Heaman, L. M., LeCheminant, A. N. & Rainbird, R. H. 1992. Nature
and timing of Franklin igneous events, Canada: implications for a
Late Proterozoic mantle plume and the break-up of Laurentia.
Earth and Planetary Science Letters, 109, 117–131.
Hildebrand, R. S. 2009. Did westward subduction cause Cretaceous-
Tertiary orogeny in the North American Cordillera? Geological
Society of America, Special Paper 457.
Hoffman, P. F. 2009. Pan-glacial – a third state in the climate system.
Geology Today, 25, 107– 114.
Hoffman, P. F. & Schrag, D. P. 2002. The snowball Earth hypothesis:
testing the limits of global change. Terra Nova, 14, 129–155.
Hoffman, P. F. & Macdonald, F. A. 2010. Sheet-crack cements and
early regression in Marinoan (635 Ma) cap dolostones: regional
benchmarks of vanishing ice-sheets? Earth and Planetary Science
Letters, 300, 374–384.
Hoffman, P. F., Halverson, G. P., Domack, E. W., Husson, J. M.,
Higgins, J. A. & Schrag, D. P. 2007. Are basal Ediacaran
(635 Ma) post-glacial ‘cap dolostones’ diachronous? Earth and Pla-
netary Science Letters, 258, 114–131.
Hoffman, P. F., Macdonald, F. A. & Halverson, G. P. 2011. Chemical
sediments associated with Neoproterozoic glaciation: iron formation,
cap carbonate, barite and phosphorite. In: Arnaud, E., Halverson,
G. P. & Shields, G. (eds) The Geological Record of Neoproterozoic
Glaciations. Geological Society, London, Memoirs, 36, 67 –80.
Hofmann, H. J., Narbonne, G. M. & Aitken, J. D. 1990. Ediacaran
remains from intertillite beds in northwestern Canada. Geology, 18,
1199– 1202.
Hurtgen, M. T., Arthur, M. A., Suits, N. S. & Kaufman, A. J. 2002.
The sulfur isotopic composition of Neoproterozoic seawater sulfate:
implications for a snowball Earth? Earth and Planetary Science
Letters, 203, 413–429.
Hurtgen, M. T., Halverson, G. P., Arthur, M. A. & Hoffman, P. F.
2006. Sulfur cycling in the aftermath of a 635-Ma snowball
glaciation: evidence for a syn-glacial sulfidic deep ocean. Earth
and Planetary Science Letters, 245, 551–570.
James, N. P., Narbonne, G. M. & Kyser, T. K. 2001. Late Neoprotero-
zoic cap carbonates: Mackenzie Mountains, northwestern Canada:
precipitation and global glacial meltdown. Canadian Journal of
Earth Science, 38, 1229– 1262.
James, N. P., Narbonne, G. M., Dalrymple, R. W. & Kyser, T. K. 2005.
Glendonites in Neoproterozoic low-latitude, interglacial, sedimentary
rocks, northwest Canada: insights into Cryogenian ocean and
Precambrian cold-water carbonates. Geology, 33, 9 – 12.
Jefferson, C. W. 1983. The Upper Proterozoic Redstone Copper Belt,
Mackenzie Mountains, Northwest Territories. PhD thesis, University
of Western Ontario, London, Ontario, 445.
Jefferson, C. W. & Ruelle, J. C. L. 1986. The Late Proterozoic
Redstone Copper Belt, Mackenzie Mountains, Northwest Territories.
In: Morin, J. A. (ed.) Mineral Deposits of Northern Cordillera.
Canadian Institute of Mining and Metallurgy Special Volume, 37,
154– 168.
Johnson, C. M. & Beard, B. L. 2006. Fe isotopes: an emerging technique
for understanding modern and ancient biogeochemical cycles. GSA
Today, 16, 4– 10.
Johnston, S. T. 2008. The Cordilleran ribbon continent of North America.
Annual Reviews of Earth and Planetary Sciences, 36, 495– 530.
Kamo, S. L. & Gower, C. F. 1994. Note: U– Pb baddeleyite
dating clarifies age of characteristic paleomagnetic remanence of
Long Range dykes, southeastern Labrador. Atlantic Geology, 30,
259– 262.
Kamo, S. L., Gower, C. F. & Krogh, T. E. 1989. Birthplace for the
Iapetus Ocean? A precise U –Pb zircon and baddeleyite age for the
Long Range dikes, southeast Labrador. Geology, 17, 602–605.
Kaufman, A. J., Jacobsen, S. B. & Knoll, A. H. 1993. The Vendian
record of Sr and C isotopic variations in seawater: implications for
tectonics and paleoclimate. Earth and Planetary Science Letters,
120, 409– 430.
Kaufman, A. J., Knoll, A. H. & Narbonne, G. M. 1997. Isotopes,
ice ages and terminal Proterozoic earth history. Proceedings of
the National Academy of Sciences (USA), 94, 6600– 6605.
Kendall, C. G. St. C. & Warren, J. 1987. A review of the origin and
setting of tepees and their associated fabrics. Sedimentology, 34,
1007– 1027.
Kennedy, M. J. 1996. Stratigraphy, sedimentology and isotopic geochem-
istry of Australian Neoproterozoic postglacial cap dolostones: degla-
ciation, d13C excursions and carbonate precipitation. Journal of
Sedimentary Research, 66, 1050– 1064.
Kennedy, M. J., Runnegar, B., Prave, A. R., Hoffmann, K.-H. &
Arthur, M. A. 1998. Two or four Neoproterozoic glaciations?
Geology, 26, 1059–1063.
Klein, C. & Beukes, N. J. 1993. Sedimentology and geochemistry of the
glaciogenic Late Proterozoic Rapitan Fe-formation in Canada.
Economic Geology, 88, 542–565.
Knoll, A. H., Walter, M. R., Narbonne, G. M. & Christie-Blick, N.
2006. The Ediacaran Period: a new addition to the geologic time
scale. Lethaia, 39, 13 – 30.
Leventer, A., Domack, E. et al. 2006. Marine sediment record from
the East Antarctic margin reveals dynamics of ice sheet recession.
GSA Today, 16, 4– 10.
Love, G. D., Grosjean, E. et al. 2009. Fossil steroids record the appear-
ance of Demospongiae during the Cryogenian period. Nature, 457,
718– 721, doi: 10.1038/nature07673.
Lyons, T. W. & Severmann, S. 2006. A critical look at iron paleoredox
proxies based on new insights from modern euxinic marine basins.
Geochimica et Cosmochimica Acta, 70, 5698– 5722.
Macdonald, F. A. & Roots, C. F. 2009. Upper Fifteenmile Group in the
Ogilvie Mountains and correlations of early Neoproterozoic strata in
the northern Cordillera. In: MacFarlane, K. E., Weston, L. H. &
Blackburn, L. R. (eds) Yukon Exploration and Geology 2009,
Yukon Geological Survey, Whitehorse, 237–252.
Macdonald, F. A. & Cohen, P. A. 2011. The Tatonduk inlier,
Alaska-Yukon border. In: Arnaud, E., Halverson, G. P. &
Shields, G. (eds) The Geological Record of Neoproterozoic
Glaciations. Geological Society, London, Memoirs, 36, 389–396.
Macdonald, F. A., Mcclelland, W. C., Schrag, D. P. & Macdonald,
W. P. 2009. Neoproterozoic glaciation on a carbonate platform
margin in Arctic Alaska and the origin of the North Slope subterrane.
Geological Society of America Bulletin, 121, 448–473.
Macdonald, F. A., Cohen, P. A., Dudás, F. Ö. & Schrag, D. P. 2010a.
Early Neoproterozoic siliceous scale microfossils in the Lower Tindir
Group of Alaska and the Yukon Territory. Geology, 38, 143– 146.
Macdonald, F. A., Schmitz, M. D. et al. 2010b. Calibrating the
Cryogenian. Science, 327, 1241– 1243.
Maloof, A. C., Rose, C. V. et al. 2010. Possible animal-body fossils in
pre-Marinoan limestones from South Australia. Nature Geoscience,
3, 653– 659.
McCausland, P. J. A., Van der Voo, R. & Hall, C. M. 2007.
Circum-Iapetus paleogeography of the Precambrian– Cambrian
transition with a new paleomagnetic constraint from Laurentia.
Precambrian Research, 156, 125– 152.
McMechan, M. E. 2000a. Vreeland Diamictites – Neoproterozoic
glaciogenic slope deposits, Rocky Mountains, northeast British
Columbia. Bulletin of Canadian Petroleum Geology, 48, 246– 261.
McMechan, M. E. 2000b. Reply to discussion. Vreeland Diamictites –
Neoproterozoic glaciogenic slope deposits, Rocky Mountains, north-
east British Columbia. Bulletin of Canadian Petroleum Geology, 48,
364– 366.
Mikucki, J. A., Pearson, A. et al. 2009. A contemporary microbially
maintained subglacial ferrous ‘ocean’. Science, 324, 397– 400.
Morris, W. A. 1977. Paleolatitude of upper Precambrian glaciogenic
Rapitan Group and the use of tillites as chronostratigraphic marker
horizons. Geology, 5, 85 – 88.
Mustard, P. S. & Roots, C. F. 1997. Rift-related volcanism, sedimen-
tation and tectonic setting of the Mount Harper Group, Ogilvie Moun-
tains, Yukon Territory. Geological Survey of Canada Bulletin, 492.
 THE MACKENZIE MOUNTAINS, NORTHERN CANADIAN CORDILLERA
Narbonne, G. M. 1994. New Ediacaran fossils from the Mackenzie
Mountains, northwestern Canada. Journal of Palaeontology, 68,
411– 416.
Narbonne, G. M. & Hofmann, H. J. 1987. Ediacaran biota of the
Wernecke Mountains, Yukon, Canada. Palaeontology, 30, 647– 676.
Narbonne, G. M. & Aitken, J. D. 1990. Ediacaran fossils from the Sekwi
Brook area, Mackenzie Mountains, northwestern Canada. Palaeon-
tology, 33, 945– 980.
Narbonne, G. M. & Aitken, J. D. 1995. Neoproterozoic of the Macken-
zie Mountains, northwestern Canada. Precambrian Research, 73,
101– 121.
Narbonne, G. M., Kaufman, A. J. & Knoll, A. H. 1994. Integrated che-
mostratigraphy and biostratigraphy of the Windermere Supergroup,
northwestern Canada: implications for Neoproterozoic correlations
and early evolution of animals. Geological Society of America Bulle-
tin, 106, 1281– 1292.
Ovenshine, T. A. 1970. Observations of ice rafting in Glacier Bay,
Alaska, and the identification of ancient ice-rafted deposits. Geologi-
cal Society of America Bulletin, 81, 891–894.
Park, J. K. 1994. Palaeomagnetic constraints on the position of Laurentia
from middle Neoproterozoic to Early Cambrian times. Precambrian
Research, 69, 95 – 112.
Park, J. K. 1997. Paleomagnetic evidence for low-latitude glaciation
during deposition of the Neoproterozoic Rapitan Group, Mackenzie
Mountains, N.W.T., Canada. Canadian Journal of Earth Sciences,
34, 34 – 49.
Pattyn, F. 2010. Antarctic subglacial conditions inferred from a hybrid
ice sheet/ice stream model. Earth and Planetary Science Letters,
295, 451– 461.
Planavksy, N. J., Rouxel, O. J., Bekker, A., Lalonde, S. V., Konhau-
ser, K. O., Reinhard, C. T. & Lyons, T. W. 2010. The evolution of
the marine phosphate reservoir. Nature, 467, 1088– 1090.
Pyle, L. J., Narbonne, G. M., James, N. P., Dalrymple, R. W. &
Kaufman, A. J. 2004. Integrated Ediacaran chronostratigraphy,
Wernecke Mountains, northwestern Canada. Precambrian Research,
132, 1– 27.
Raiswell, R. & Canfield, D. E. 1998. Sources of iron for pyrite
formation in marine sediments. American Journal of Science, 298,
219– 245.
Ramsay, J. G. & Huber, M. I. 1987. The Techniques of Modern Structural
Geology, Vol. 2. Academic Press, New York.
Ross, G. M. 1991. Tectonic setting of the Windermere Supergroup
revisited. Geology, 19, 1125– 1128.
Ross, G. M., Villeneuve, M. E. 1997. U– Pb geochronology of stranger
stones in Neoproterozoic diamictites, Canadian Cordillera: impli-
cations for provenance and ages of deposition. In: Radiogenic Age
and Isotopic Studies: Report 10. Geological Survey of Canada
Current Research, 1997-F, 141– 155.
Ross, G. M., McMechan, M. E. & Hein, F. J. 1989. Proterozoic history:
the birth of the miogeocline. In: Rickets, B. D. (ed.) Western Canada
Sedimentary Basin: A Case History. Canadian Society of Petroleum
Geologists, Calgary, 79 – 104.
Ruelle, J. C. L. 1982. Depositional environments and genesis of strati-
form copper deposits of the Redstone Copper Belt, Mackenzie Moun-
tains, N.W.T. In: Hutchinson, R. W., Spence, C. D. & Franklin,
J. M. (eds) Precambrian Sulfide Deposits. Geological Association
of Canada Special Paper, 25, 701–737.
Shen, Y., Zhang, T. & Hoffman, P. F. 2008. On the co-evolution of
Ediacaran oceans and animals. Proceedings of the National
Academy of Sciences USA, 105, 7376–7381.
Shields, G. A. 2005. Neoproterozoic cap carbonates: a critical appraisal
of existing models and the plumeworld hypothesis. Terra Nova, 17,
299 – 310.
Siegert, M. J., Carter, S., Tabacco, I., Popov, S. & Blankenship, D.
2005. A revised inventory of Antarctic sub-glacial lakes. Antarctic
Science, 17, 453– 460.
411
Silva-Tamayo, J. C., Nägler, T. F. et al. 2010. Global Ca isotope
variations in c. 0.7 Ga old post-glacial carbonate successions. Terra
Nova, 22, 188– 194.
Stewart, J. H. 1972. Initial deposits in the Cordilleran Geosyncline: evi-
dence of a Late Precambrian (,850 m.y.) continental separation.
Geological Society of America Bulletin, 83, 1345– 1360.
Turner, E. C. & Long, D. G. F. 2008. Basin architecture and syndeposi-
tional fault activity during deposition of the Neoproterozoic Macken-
zie Mountains supergroup, Northwestern Territories, Canada.
Canadian Journal of Earth Sciences, 45, 1159– 1184.
Turner, E. C., James, N. P. & Narbonne, G. M. 1997. Growth
dynamics of Neoproterozoic calcimicrobial reefs, Mackenzie
Mountains, northwest Canada. Journal of Sedimentary Research,
67, 437– 450.
Umer, M., Kebede, S. & Osmaston, H. 2004. Quaternary glacial activity
on the Ethiopian mountains. In: Ehlers, J. & Gibbard, P. L. (eds)
Quaternary Glaciations – Extent and Chronology, Part III: South
America, Asia, Africa, Australia, Antarctica. Elsevier, Amsterdam,
171–174.
Upitis, U. 1966. The Rapitan Group, southwestern Mackenzie Mountains,
Northwest Territories. MSc thesis, McGill University, Montreal.
Voigt, A. & Marotzke, J. 2009. The transition from the present-day
climate to a modern Snowball Earth. Climate Dynamics, doi:
10.1007/s00382-009-0633-5.
Wallace, M. W. & Woon, E. 2008. Giant Cryopgenian reefs as windows
into pre-Ediacaran life. In: Gallagher, S. J. & Wallace, M. W.
(eds) Selwyn Symposium 2008: Neoproterozoic Extreme Climates
and the Origin of Early Metazoan Life. Geological Society of Austra-
lia, Extended Abstracts, 91, 17 – 22.
Warren, S. G., Brandt, R. E., Grenfell, T. C. & McKay, C. P. 2002.
Snowball Earth: ice thickness on the tropical ocean. Journal of Geo-
physical Research, 107(C10), 3167, doi: 10.1029/2001JC001123.
Weil, A. B., Geismann, J. W. & Van der Voo, R. 2004. Paleomagnetism
of the Neoproterozoic Chuar Group, Grand Canyon Supergroup,
Arizona: implications for Laurentia’s Neoproterozoic APWP and
Rodinia break-up. Precambrian Research, 129, 71 –92.
Weil, A. B., Geismann, J. W. & Ashby, J. M. 2006. A new paleomagnetic
pole for the Uinta Mountain supergroup, Central Rocky Mountain
States, USA. Precambrian Research, 147, 234– 259.
Wingham, D. J., Siegert, M. J., Shepherd, A. & Muir, A. S. 2006.
Rapid discharge connects Antarctic subglacial lakes. Nature, 440,
1033–1036.
Yeo, G. M. 1981. The Late Proterozoic Rapitan glaciation in the northern
Cordillera. In: Campbell, F. H. A. (ed.) Proterozoic Basins of
Canada. Geological Survey of Canada Paper, 81 –10, 25 –46.
Yeo, G. M. 1984. The Rapitan Group: relevance to the global association
of Late Proterozoic glaciation and Fe-formation. PhD thesis, Univer-
sity of Western Ontario, London (Ontario).
Yeo, G. M. 1986. Fe-formation in the late Proterozoic Rapitan Group,
Yukon and Northwest Territories. In: Morin, J. A. (ed.) Mineral
Deposits of Northern Cordillera. Canadian Institute of Mining and
Metallurgy, Special Volume, 37, 142– 153.
Young, G. M. 1976. Fe-formation and glaciogenic rocks of the Rapitan
Group, Northwest Territories, Canada. Precambrian Research, 3,
137– 158.
Young, G. M. 1982. The late Proterozoic Tindir Group, east-central
Alaska: evolution of a continental margin. Geological Society of
America Bulletin, 93, 759–783.
Young, G. M. 1988. Proterozoic plate tectonics, glaciation and
Fe-formations. Sedimentary Geology, 58, 127–144.
Young, G. M. 2002. Stratigraphic and tectonic settings of Proterozoic
glaciogenic rocks and banded Fe-formations: relevance to the
snowball Earth debate. Journal of African Earth Sciences, 35,
451– 466.
Ziegler, P. 1959. Frühpaläozoische Tillite im östlichen Yukon-
Territorium (Kanada). Eclogae Geologicae Helvetiae, 52, 735– 741.
 Chapter 37

The record of Neoproterozoic glaciations in the Windermere Supergroup,

southern Canadian Cordillera

MARK D. SMITH1, EMMANUELLE ARNAUD2*, R.W.C. ARNOTT1 & GERALD M. ROSS3

1
Department of Earth Sciences, University of Ottawa, Ottawa, Ontario, K1N 6N5, Canada
2
School of Environmental Sciences, University of Guelph, Guelph, Ontario, N1G 2W1, Canada
3
Kupa’a Farm, P.O.Box 458, Kula, Hawai’i, 96790, USA
*Corresponding author (e-mail: earnaud@uoguelph.ca)

Abstract: The Neoproterozoic Windermere Supergroup (WSG) is exposed over an area of 35 000 km2 in the southern Canadian
Cordillera, and consists primarily of deep-marine meta-sedimentary rocks interpreted to have been deposited during rifting and sub-
sequent post-rift thermal relaxation. The main exposures of the WSG occur within thrust panels and structural culminations of the
eastern Cordilleran orogen. Within the thick stratigraphic succession (c. 9 km) are three units of glaciogenic affinity: Toby, Vreeland
and Old Fort Point (OFP) formations. The Toby Formation (Fm.) is composed of up to 2500 m of diamictite, interbedded with conglom-
erate, sandstone, mudstone, carbonate and mafic volcanic rocks. The Vreeland Formation ranges from 350 m to 2000 m in thickness and
consists of diamictite, interbedded with mudstone, sandstone and conglomerate. The OFP ranges from 60 to 450 m in thickness and con-
sists of a distinctive threefold stratigraphic package of basal siltstone grading upward into limestone –siltstone rhythmite, organic-rich
mudstone and an overlying heterolithic unit of diamictite, breccia, conglomerate, sandstone, siltstone to mudstone and limestone. A
locally prominent unconformity marks the base of the OFP upper member. Both the Toby and Vreeland formations represent remobilized
glacially derived marine sediments deposited by sediment-gravity flows. Deposition of the Toby Fm. was fault-controlled during an
active tectonic phase (rifting), whereas the Vreeland Fm. accumulated during the subsequent quiescent phase (post-rift) with limited
structural control. The OFP is interpreted to be a regionally extensive deep-marine post-glacial marker temporally associated with the
glaciogenic Vreeland Fm. Although direct geochronological ages for WSG units in southwestern Canada are generally absent, high-
precision radiometric ages of underlying and overlying igneous events constrain the relative maximum and minimum timing of depo-
sition from c. 740– 728 Ma to c. 569 Ma. At the base of the WSG succession, the Toby Fm. may be as young as c. 685 Ma based on
ages obtained from potential stratigraphic correlatives in the USA. There is no direct age constraint for the deposition of the Vreeland
Fm.; its minimum timing is based on its stratigraphic relationship with the post-glacial OFP. The middle member of the OFP was pre-
cisely dated at 607.8 + 4.7 Ma using the Re–Os method, placing it in the Ediacaran Period. Published geochemical and stable isotopic
data are similarly limited for all three units with only some d34Spy values available from one section of the OFP. Recent work has focused
on detailed sedimentological and stratigraphic studies of the Toby and OFP formations with future efforts being directed towards inte-
grated geochemical and isotopic research. Additional geochronological constraints are needed to refine palaeogeographical models and
strengthen regional correlations with other North American Cordilleran glaciogenic units.

The record of Neoproterozoic glaciogenic sedimentation in the
southern Canadian Cordillera (Fig. 37.1) is preserved in three
units within the Windermere Supergroup (WSG): a glacial interval
at the base of the succession is inferred from the diamictite-bearing
Toby Fm., with another event higher in the succession inferred
from the diamictite-bearing Vreeland Fm. (glacial) and Old Fort
Point Fm. (post-glacial) (Fig. 37.2) (Ross et al. 1989, 1995).
including mapping of the Toby Fm. and related structures in the
Paradise Mine (Atkinson 1975), Delphine Creek (Root 1987),
Mount Forster (Bennett 1986), Creston/Salmo (Glover & Price
1976) and Crawford Bay/Columbia Point areas (Lis & Price
1976). Ongoing investigations are focusing on the relationship
between Toby Fm. facies variability and structural trends as well
as isotopic analyses of associated strata.

Toby Fm. Vreeland Fm.

The Toby Fm. (also known as the Toby Conglomerate) is best
exposed in the Purcell Mountains of southeastern British Colum-
bia. The variability of its lithofacies has precluded the identifi-
cation of a type section; the formation was named by Walker
(1926) after well-exposed outcrops located along Toby Creek,
west of Invermere, British Columbia (508300 N, 1168020 W)
(Fig. 37.1, A). This area has been extensively modified by Proter-
ozoic, Devonian and Mesozoic deformation and the structural
context of these rocks is often complicated (e.g. Root 1987).
Previous studies of the Toby Fm. are relatively few, with
primary emphasis on regional mapping relative to Proterozoic-
and Mesozoic-age structures (Walker 1926; Leech 1954; Aalto
1971; Reesor 1973; Root 1987; Pope 1990; Warren 1997). The
glacial origin of the Toby Fm. was first proposed based on a
more detailed study of facies types and variability, petrography,
as well as trends in clast characteristics in the Toby Fm. itself
(Aalto 1971). Additional mapping of specific regions has followed,
The Vreeland Fm. has received little study (due largely to inac-
cessibility) despite its impressive exposures and thickness in the
Pine and Monkman Pass regions, central British Columbia (from
548050 N to 578000 N) (Fig. 37.1). It was originally mapped by
Slind & Perkins (1966) as an unnamed Precambrian conglomer-
ate –schist. Later, it was briefly examined by Stelck et al. (1978),
but was not really described until regional mapping conducted
by McMechan (1987) and McMechan & Thompson (1985,
1995a, b). It was initially referred to as a diamictite within the
Misinchinka Group or the Mount Vreeland –Paksumo Pass dia-
mictites (McMechan 1990), until Hein & McMechan (1994)
referred to it as the Vreeland Fm. McMechan (2000) provided
the first detailed stratigraphic description and palaeogeographical
interpretation in the Monkman Pass area. No type section has
been defined for the Vreeland Fm., although the Mount Vreeland
sections (548350 N, 1218320 W) in the Snake Indian thrust sheet
(Fig. 37.1,B) and the Paksumo Pass section (548200 N,

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 413– 423. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.37
414 M. D. SMITH ET AL.
1208490 W) in the Wapiti Pass thrust sheet (Fig. 37.1,C) of the
Monkman Pass area are thought to be representative exposures
(McMechan 2000).
Old Fort Point Fm.
The Old Fort Point Fm. (OFP) is the preferred name to describe the
unique stratigraphic marker in the WSG that is exposed locally
over the entire outcrop region of southwestern Canada
(Fig. 37.1,D – K). Walcott (1910) was the first to describe the
OFP from the Lake Louise area of Alberta using its distinctive
purple and green colours and lithology (siltstone, limestone,
breccia) to separate the Precambrian Corral Creek (lower con-
glomerate to sandstone) and Hector (upper slate) formations.
The name and type section (528520 1600 N, 1188030 4000 W) come
from outcrops at a prominent landmark near Jasper, Alberta
Fig. 37.1. Generalized geological map
showing outcrop distribution of
Windermere Supergroup in western Canada
and specific outcrop localities (A– K)
discussed in the text (modified from Ross
et al. 1995). A comprehensive list of
outcrop locations for the Toby Fm. can be
found in Aalto (1971) and Reesor (1973).
Outcrop localities for the Vreeland Fm. are
from McMechan (2000) and for the OFP
from Ross & Murphy (1988) and Smith
(2009). Mtns, Mountains.
(Charlesworth et al. 1967). A more representative and better
exposed section crops out c. 15 km west of Jasper along the
major highway (528520 2100 N, 1188170 3500 W). The regionally
extensive nature of the OFP has led to various names: (i) basal
Hector Fm. (Walcott 1910) or Mount Temple and Taylor Lake
members (e.g. Gussow 1957) in the Lake Louise area, Alberta;
(ii) Old Fort Point Fm. (e.g. Weiner 1966; Charlesworth et al.
1967) in the Jasper area, Alberta; (iii) Middle Miette Marker
(e.g. McDonough 1989) in the Mount Robson area, British Colum-
bia; (iv) Kaza Group marker (e.g. Ross & Murphy 1988) in the
Cariboo Mountains, British Columbia; (v) Baird Brook Division
(e.g. Kubli 1990) in the Dogtooth Range of the Purcell Mountains,
British Columbia; (vi) Comedy Creek unit (Grasby & Brown
1993), Selkirk Mountains, British Columbia; (vii) upper and
lower markers (e.g. Warren 1997) in the Purcell Mountains,
British Columbia (Fig. 37.1). Although there were early attempts
to correlate between certain locations (Charlesworth et al. 1967;
 WINDERMERE SUPERGROUP
Aitken 1969), it was Ross & Murphy (1988) who first recognized
the basin-wide correlation of the OFP (and equivalents), its palaeo-
geographical significance and importance in Neoproterozoic event
stratigraphy. Subsequent work by Ross et al. (1995) examined the
S-isotopic evolution of the OFP and WSG strata in the Cariboo
Mountains, British Columbia (Fig. 37.1). A regionally comprehen-
sive study of the OFP integrating sedimentology, stratigraphy,
geochemistry and stable isotopes has recently been completed
(Smith 2009).
Structural framework and basin setting
The WSG forms part of a long arcuate belt of semicontinuous out-
crops in western North America that extends from the Yukon –
Alaska border region to northwestern Mexico (e.g. Ross et al.
1989). In southwestern Canada, exposures crop out over an area
of 35 000 km2 in a series of thrust sheets in the Main Ranges of
the western Fold-and-Thrust Belt and the Omineca Belt of the
southern Canadian Cordillera (Fig. 37.1). Major structures affect-
ing WSG strata in the region include the Southern Rocky Mountain
Trench (Fig. 37.1) and a broad gently north-plunging structure in
the Purcell Mountains (the Purcell Anticlinorium). Proterozoic-
age structures that affected WSG deposition include regionally sig-
nificant transfer faults (e.g Mount Forster and Moyie) and associ-
ated uplifted blocks (e.g. Windermere High and Montania; Reesor
1973; Lis & Price 1976; Root 1987; Warren 1997).
All WSG strata were subjected to Mesozoic orogenic defor-
mation that resulted in a wide range of structural complexity and
metamorphic grade between different structural panels. Extensive
tracts of sub-greenschist to greenschist grade exist in the western
Main Ranges of the Rocky Mountains (Lake Louise, Jasper, north-
ern Cariboo Mountains and eastern Purcells Mountains), whereas
higher-grade metamorphic rocks (biotite to upper amphibolite)
415
Fig. 37.2. Comparative stratigraphic
columns for the Windermere Supergroup
from the Monkman Pass area and the
Cariboo-Purcell-Rocky Mountains
(modified from Ross et al. 1995).
Nomenclature of rift and post-rift strata
(Kaza-Cariboo, Horsethief Creek and
Miette groups refer to similar strata,
depending on location within the region).
Relative stratigraphic position of the
glaciogenic Toby, Vreeland and Old Fort
Point formations and proposed correlation
between the basal Framstead and Old Fort
Point formations are shown. Ages shown
are discussed in the text and, in
stratigraphic order, are Malton Gneiss
(McDonough & Parrish 1991); rift-related
volcanic rocks from Idaho (Lund et al.
2003; Fanning & Link 2004); Old Fort
Point Fm. (Kendall et al. 2004); and Hamill
Group volcanic rocks (Colpron et al. 2002).
Bracketed ages indicate that the ages are
not from rocks of the southern Canadian
Cordillera, but rather interpreted
correlatives of the Irene Fm.
are found in the southern Cariboo and western Purcell Mountains,
as well as the Shuswap Complex (e.g. Simony et al. 1980;
Ross et al. 1995). Areas of low metamorphic grade and relatively
uncomplicated tectonic deformation show exceptional preservation
of primary sedimentary textures and structures. This allowed the
development of a consistent internal stratigraphy of the WSG and
reconstruction of the Windermere basin in southwestern Canada
(Ross & Murphy 1988; Ross et al. 1989, 1995; Ross 2000).
The WSG is thought to represent deposition in two tectonic
phases (e.g. Stewart 1972; Ross 1991). The first phase was charac-
terized by rifting during the break-up of the supercontinent
Rodinia, whereas the second phase occurred during post-rift
thermal relaxation. At the base of the WSG, the Toby Fm. together
with the Irene Fm. volcanic rocks are interpreted to have accumu-
lated during the rifting phase (Aalto 1971; Glover & Price 1976;
Root 1987). The variable thickness of the Toby Fm. is attributed
to syn-sedimentary fault-controlled deposition (Aalto 1971; Root
1987). Based on the nature of Proterozoic-age faults, Root
(1987) suggested that the basin underwent two episodes of exten-
sion; one trending approximately east –west, resulting in a series of
NNE and NNW faults and another trending NW –SE, resulting in a
series of NE– SW normal faults.
The Vreeland Fm. and OFP both accumulated during the post-
rift phase with possible local structural control (McMechan
2000; Smith 2009). During this later phase, the basin is thought
to be an elongate NW-flowing turbidite system (Ross et al. 1989,
1995; Ross 1991), based on consistent palaeocurrent data and
bimodal pattern (.2.6 Ga and 1.9–1.75 Ga) of U –Pb detrital
zircon provenance (Ross & Bowring 1990; Ross & Parrish 1991).
Basement clasts from the Vreeland Fm. yielded ages between
1865 to 1842 Ma (Ross & Villeneuve 1997), suggesting derivation
from a source area to the north and NE (McMechan 2000).
Although there is broad agreement about these two tectonic
phases, the exact tectonic setting of the Windermere basin and
416 M. D. SMITH ET AL.
its evolution remains controversial. Some argue that the upper part
of the WSG in the southern Canadian Cordillera accumulated on a
continental passive margin with a substantial Proto-Pacific Ocean
(Ross 1991; Ross et al. 1995; Dalrymple & Narbonne 1996).
Others prefer an intracontinental rift and restricted ocean basin
setting for the southern WSG, and a late Neoproterozoic –early
Cambrian inception for the Proto-Pacific Ocean (Colpron et al.
2002). The controversy and uncertainty is in part due to the pres-
ence of two intervals of rift-related igneous rocks within the
southern Canadian Cordillera (Bond & Kominz 1984; Ross 1991;
Colpron et al. 2002), diachronous dates for the timing of rifting
along the whole of Laurentia’s Pacific margin (see discussion in
Lund et al. 2003), and tectonic deformation that currently precludes
certain identification of a western Windermere basin margin.
Stratigraphy
In the southern Canadian Cordillera, the WSG is a thick (c. 9 km),
unconformity-bounded succession of predominantly coarse-
grained deep-marine siliciclastic rocks with subordinate carbonate
and mafic volcanic rocks (Fig. 37.2). Older units associated with
the rift phase (e.g. Toby and Irene formations) are more limited
in areal extent and tend to show greater lateral facies variations
compared with units associated with the post-rift phase (e.g.
Ross et al. 1989, 1995; Ross 1991; Warren 1997). At the base of
the WSG, the Toby Fm. unconformably overlies shallow-water
deposits of the Mount Nelson Fm. (Neoproterozoic/Mesoprotero-
zoic?, Root 1987; Ross & Villeneuve 2003) or deep-marine sedi-
mentary rocks of the Mesoproterozoic Purcell Supergroup (Aalto
1971; Root 1987). The Toby Fm. is associated with localized
rift-related mafic volcanic rocks of the Irene Fm. and a laterally
discontinuous carbonate horizon at its top (e.g. Root 1987;
Warren 1997). These basal units of the WSG are largely restricted
to the Purcell Mountains region (Fig. 37.1). Potential high-grade
metamorphic equivalents of the Toby (e.g. Simony et al. 1980;
Murphy et al. 1991) and Irene volcanic rocks (e.g. Simony et al.
1980; Sevigny 1988) may exist in the Shuswap Complex and
northern Cariboo Mountains, although the stratigraphic control is
poor (Fig. 37.1).
The nomenclature of WSG succession that overlies the Toby
Fm. in the southern Canadian Cordillera is rather complex: a
direct result of correlation challenges across thrust-bound panels
with varying levels of metamorphism, general absence of reliable
stratigraphic markers and poor geochronological constraints. The
Horsethief Creek Group (Purcell Mountains), Kaza and lower/
middle Cariboo groups (Cariboo Mountains), Miette Group
(Rocky Mountains) and Misinchinka Group (Rocky Mountains) all
refer to parts of the same c. 5 km deep-marine turbidite system
exposed over 35 000 km2 (e.g. Ross et al. 1989) (Figs 37.1 &
37.2). Lithologically, the thick package is dominated by coarse-
grained arkosic sandstone and granule conglomerate interbedded
with mudstone-dominated intervals or minor carbonate.
The Vreeland Fm. of the Misinchinka Group is a local excep-
tion, being composed of up to 2000 m of diamictite interbedded
with mudstone, sandstone and conglomerate (McMechan 2000).
Mapping of the Vreeland Fm. has shown that the diamictite units
undergo a lateral facies change into typical deep-marine coarse-
grained turbidites of the WSG in both a westward (McMechan
1987) and southward direction (McMechan & Thompson 1995a).
Throughout much of the basin, the overall monotony of the
thick-bedded, immature turbidites of the Horsethief Creek,
Miette and Kaza groups is interrupted by the fine-grained, region-
ally widespread OFP (and correlative units) (Ross & Murphy
1988; Ross et al. 1995). Nowhere is the OFP observed to be in
contact with diamictite of the Vreeland Fm.; rather, the OFP over-
lies WSG turbiditic strata interpreted as deep-marine lateral
equivalents. The OFP is overlain by coarse-grained basin-floor tur-
bidite deposits that form a kilometre-scale shoaling upwards trend
to mudstone-dominated upper slope units and shelf platform car-
bonates (Ross et al. 1989). A significant regional unconformity
marks the end of WSG deposition and the beginning of a latest
Neoproterozoic to Cambrian rift– drift succession with the
clastic and volcanic sediments of the upper Cariboo, Hamill and
Gog groups.
Glaciogenic deposits and associated strata
Toby Fm.
The Toby Fm. consists primarily of diamictite interbedded with
conglomerate, sandstone, dolomitized or recrystallized limestone
and mudstone (Fig. 37.3) (Aalto 1971). Its thickness is highly vari-
able, ranging from 0 to 2500 m. Lithofacies, in general, and sedi-
mentary characteristics of diamictite, in particular (including clast
concentration, clast size, matrix and clast lithology, lateral conti-
nuity and bed thickness) are highly variable over short distances,
a fact that has been stressed by all workers in this region. The
following descriptions rely heavily upon the regional study of
Aalto (1971), with additional observations based on more recent
fieldwork by E. Arnaud and K. Root.
Diamictite is predominantly massive, matrix- or clast-supported
with sub-angular to sub-rounded clasts up to 2 m, floating in a
muddy sandstone or sandy mudstone matrix. Clast lithology
includes quartzite, volcanic greenstone, slate, dolomite and chert
derived from the underlying Mount Nelson Fm. as well as rare
extrabasinal granite (Leech 1954; Reesor 1973; Loveridge et al.
1981). The diamictite matrix exhibits a similar lithological
variability. Tectonic overprinting typically obscures clast fabric
in many outcrops. Where relatively unaltered, clasts show no pre-
ferred orientation. Diamictite units vary in thickness from several
metres to tens of metres. In outcrops with good lateral exposure,
basal contacts are relatively planar, and conformable or erosional.
Conglomerate occurs in lenticular or planar units (centimetre to
metre scale) interbedded with diamictite and sandstone. It is distin-
guished from the diamictite by its coarse sandstone matrix, and
consistently high clast concentration (65 –80%). It is generally
unorganized and massive with occasional subtle coarse-tail
grading. Sandstone varies from coarse- to fine-grained, moderately
to poorly sorted, and occurs in planar beds or lenticular units
(centimetre to metre scale). It is most commonly massive or lami-
nated, although some sandstone units exhibit ripple cross-
lamination, climbing ripples or normal grading. Dolomitized or
recrystallized limestone (centimetre to metre scale thickness)
with varying amounts of quartz grains is also observed within
and at the top of the Toby Fm. at various sites within the region.
These carbonate strata are planar –tabular and can be laterally
persistent over hundreds of metres, although their overall distri-
bution is patchy. Finally, mudstone is relatively common through-
out the Toby Fm. It is massive or laminated and occurs in relatively
planar tabular units (centimetre to metre scale thickness). Outsized
clasts are found within these mudstone units, and some are seen to
depress or puncture underlying laminations (Aalto 1971).
In terms of associated strata, the underlying Mount Nelson and
Dutch Creek formations consist of thick packages of carbonate and
siliciclastic strata (kilometre scale; Walker 1926; Reesor 1973;
Root 1987). Slate predominates in the Dutch Creek Fm.,
whereas thick packages of white quartzite and dolomite inter-
bedded with intervals of argillite, siltstone and conglomerate
characterize the Mount Nelson Fm. In the Purcell Mountains, the
Toby Fm. is overlain by the Horsethief Creek Group, which con-
sists of a thick package (,100–2000 þ m) of mudstone, sand-
stone, conglomerate, calcareous mudstone and dolostone
(Walker 1926; Root 1987; Kubli 1990; Warren 1997). Further to
the south, the Toby Fm. is interbedded and overlain by volcanic
greenstone of the Irene Fm., and coarse- to fine-grained clastic
and carbonate lithofacies of the Monk Fm. (Aalto 1971).
 WINDERMERE SUPERGROUP 417

Vreeland Fm.

Description of the Vreeland Fm. is based entirely on regional

mapping studies in the Pine and Monkman Pass areas (McMechan
& Thompson 1985, 1995a, b; McMechan 1987) and geological
descriptions from the Snake Indian and Wapiti Pass thrust sheets
(Fig. 37.1) (McMechan 1990, 2000). The Vreeland Fm. comprises
a thick succession of diamictite interbedded with mudstone, sand-
stone and conglomerate (Fig. 37.4). Diamictite units are generally
massive (c. 98%) with a sandstone or siltstone to mudstone matrix

Fig. 37.3. Stratigraphic columns of the Toby Fm. from the Mount Brewer
(c. 508230 N, 1168140 W) and Paradise Mine localities (c. 508280 N, 1168180 W)
with lithofacies representative of other outcrops; however, their lateral
continuity, stratigraphic thickness and stratigraphic distribution vary widely
across the Purcell Mountains. Note the different scales used in each log. Facies Fig. 37.4. Stratigraphic columns of the Vreeland Fm. near Mount Vreeland in
codes used: (first letters) D, diamictite; G, conglomerate; S, sandstone; F, the Snake Indian thrust sheet, and Paksumo Pass in the Wapiti thrust sheet,
fine-grained facies; (second letters in diamictite and conglomerate) m, matrix modified from McMechan (2000). St, cross-trough bedded sandstone; Fld,
supported; c, clast-supported; (third letter in any lithofacies) m, massive; l, laminated fine-grained facies with lonestones. See Figure 37.3 caption for other
laminated; d, deformed; r, rippled; h, horizontally bedded; s, stratified. facies codes.
418 M. D. SMITH ET AL.

and form laterally extensive (up to 2 km) tabular sheets that range
from a few metres to .40 m thick. The basal contact of massive dia-
mictite units is sharp and locally erosive. Subtle normal grading is
observed locally in diamictite in addition to rare stratified diamictite
defined by clast layers, or sandstone/siltstone stringers and lenses.
Clasts are lithologically diverse and consist of both intra-basinal
(mudstone, sandstone, carbonate) and extra-basinal (felsic or mafic
plutonic, volcanic, quartzite) varieties. Differing provenance trends
are apparent in the two main areas based on clast lithology.
Interstratified mudstone units are up to 38 m thick with rare sand-
stone interbeds, and common lonestones. Sandstone units are
sharp-based and range from very fine- to very coarse-grained and
very thin- to thick-bedded. Sedimentary structures include scours,
cross-stratification, graded bedding and intraclast rip-ups. Con-
glomerate and sandstone units commonly form lenticular channel-
like deposits. Conglomerate is matrix- and clast-supported with a
sandstone matrix, which contrasts with the typical siltstone to mud-
stone matrix of the diamictite. Clast lithologies are similarly
diverse, rounded to angular, and range up to boulder size.
In the Snake Indian thrust sheet, the top of the Vreeland Fm. is
marked by a pyrite-rich zone up to 30 cm thick and locally the
diamictites are overlain by a thin (few metres), parallel laminated
limestone. The conformably overlying Framstead Fm. is
mudstone-dominated, but locally is composed of sandstone to con-
glomerate or limestone. Locally, large (up to 650 m long) carbon-
ate olistoliths occur on, or just above the basal contact with the
Vreeland Fm., and also near its upper contact. These olisoliths
occur as discrete, randomly oriented blocks that lack internal
deformation. Lithologies are predominantly tan-weathered
shallow-water carbonates and include laminated dolomite with
rare layers of teepee structures and bladed calcite, stromatolitic
dolomite, sandy to pebbly dolomite, dolomite conglomerate, and
quartzose fenestral and vuggy dolomite.

Old Fort Point Fm.

The following sedimentological and stratigraphic description

stems mostly from a recent detailed regional study (Smith 2009),
but also incorporates work of earlier authors (Fig. 37.1,D –K)
(Walcott 1910; Weiner 1966; Charlesworth et al. 1967; Aitken
1969; Murphy 1986; Pell & Simony 1987; Ross & Murphy
1988; McDonough 1989; Deschesne 1990; Kubli 1990; Grasby
& Brown 1993; Ross et al. 1995; Warren 1997; Ross & Ferguson
2003a, b). The OFP comprises three distinctive lithological
members, which despite variable metamorphic grade, form a
consistent stratigraphic relationship throughout the southern
Canadian Cordillera (Fig. 37.5).
The lower member is a purple, green, grey or red-brown fine-
grained unit that ranges from 50 to 125 m thick. The basal
portion is composed of siltstone to mudstone that grades upward
into rhythmic couplets of limestone –siltstone. Beds typically
range from ,1 cm to 10 cm in thickness and exhibit common trac- Fig. 37.5. Stratigraphic columns of the Old Fort Point Fm., simplified from
tional sedimentary structures including planar lamination, lenticu- Smith (2009). At the Geikie Siding section in Jasper National Park, the
regionally widespread lower two members are exposed and overlain sharply by
lar bedding, ripple cross-lamination, normal grading and minor
a thinly developed upper member. The Re– Os isochron age was obtained from
scours. Subordinate lithofacies include very fine- to fine-grained
the middle member at this section (Kendall et al. 2004). At the Upper
sandstone interstratified with uncommon limestone-clast breccia
Boomerang section on the Lake Louise Ski Resort, only a thickly developed
beds. Palaeocurrents measured from 3D current-ripples generally Upper Member is exposed, with complete erosion of the lower two members.
indicate flow toward the SW to NW. Str, stratified (ripple cross-laminated and cross-bedded). See Figure 37.3 for
The middle member ranges from 2 to 15 m in thickness and other facies codes.
consists of a dark grey organic-rich mudstone/pelite. The basal
contact with the underlying lower member is usually gradational
over a few metres. The unit is characterized by alternating siltstone In contrast to the lower two members, the upper member is litho-
and mudstone laminae with subtle normal grading. Isolated, thin- logically diverse and highly variable in thickness (,0.5–165 m).
bedded dark-grey massive or planar-laminated limestone and The basal contact with underlying strata is always sharp and,
very fine ripple-stratified sandstone occur locally. This member locally, it completely erodes the lower two members. Lithologies
is regionally extensive. Its fissile nature typically results in poor include diamictite, breccia, conglomerate, mudstone, siltstone,
or covered exposures, although notable exceptions occur (e.g. sandstone, quartz arenite, calcareous arenite, arenaceous limestone
Ross et al. 1995; Kendall et al. 2004). and limestone. Diamictite and breccia to conglomerate are
 WINDERMERE SUPERGROUP
generally sharp, commonly erosively based, massive beds with
mudstone/siltstone or well-sorted, coarse quartz-rich sandstone
matrix. Bed thickness (,0.5 to .10 m), clast size (centimetre to
metre) and clast lithologies are variable. Clasts include fragments
of the OFP members and shallow-marine carbonates, some with
rare bladed calcite crystals. Mudstone and siltstone exhibit
planar laminations, micro-scours, subtle normal grading, large
single chlorite flakes and rare scours. This lithofacies is locally
thick (c. 100 m) and monotonous with rare interbeds of diamictite,
breccia, conglomerate or sandstone. Sandstone in the upper
member consists of a range of textures: fine- to very coarse-
grained, poor- to well-sorted and mineralogical maturity (imma-
ture to supermature). The well-sorted, mature quartz arenites are
interbedded with dark, organic-rich limestone beds and these
units can exhibit a range of compositions between the two end
members (e.g. calcareous arenites or arenaceous limestones). Sedi-
mentary structures include massive beds, planar lamination and
cross-stratification. Rare palaeocurrent measurements from the
upper member are consistently towards the SW to NW. Soft-
sediment deformation features such as load structures, convolute
or contorted bedding and ductile folding are common in both
fine- and coarse-grained facies of the upper member.
Boundary relations with overlying and underlying
non-glacial units
Toby Fm.
The Toby Fm. rests unconformably on various members of the
Mount Nelson Fm. or the underlying Dutch Creek Fm. In
outcrop, the unconformity is commonly subtle (,108), although
a distinctive angular unconformity with tilted underlying Mount
Nelson strata is evident in some places. The upper contact of the
Toby Fm. with the Horsethief Creek Group is conformable and
gradational throughout the Purcell Mountains and typically estab-
lished based on the loss of diamictite. To the west and south of the
Purcell Mountains, the Toby Fm. is conformably overlain by
pillow lavas of the Irene Fm. as well as conglomerate and diamic-
tite facies of the lower Monk Fm. (Aalto 1971). Considering there
are localized lenses of conglomerate and diamictite in the Irene
Fm., and that the upper contact of the Toby Fm. is defined in the
Purcell Mountains with the loss of diamictite facies, the upper
contact of the Toby Fm. may actually occur several tens of
metres above the Irene volcanic rocks (Aalto 1971).
Vreeland Fm.
In the western Snake Indian thrust sheet, the basal contact of the
Vreeland Fm. is not exposed (McMechan 2000). In the eastern
Wapiti Pass thrust sheet, the basal diamictite appears to interfinger
with the upper 150 m of the fine-grained facies of the underlying
Paksumo Fm. (McMechan & Thompson 1995b). The Vreeland
Fm. is conformably overlain by the Framstead Fm., and in the
western exposures by a carbonate olistolith-bearing unit, whereas
in the eastern exposures it is overlain by a sandstone unit (McMe-
chan 2000). The carbonate olistolith-bearing unit correlates from
west to east where it overlies the lowermost Framstead Fm. sand-
stone. This suggests at the western Snake Indian thrust sheet
locality that either local erosion of the sandstone unit occurred
or it is actually a lateral equivalent of diamictite of the Vreeland
Fm. (McMechan 1990).
Old Fort Point Fm.
The OFP conformably and gradationally overlies deep-marine tur-
biditic strata of the WSG interpreted to be lateral facies equivalent
419
of the Vreeland Fm. diamictite. The contact is commonly marked
by a distinctive change in colour (e.g. grey to purple) and lithology
(appearance of lower member siltstones). However, at some
locations the lower and middle members have been eroded and
the upper member forms a sharp unconformity over older WSG
strata. The top of the OFP is conformably overlain by younger
WSG strata. In many cases the contact is identified by a sharp
change in grain size and/or framework mineralogy (e.g. OFP
quartz arenite to WSG arkosic sandstone). The challenge arises
where OFP mudstone units are overlain directly by a younger mud-
stone unit of the WSG. The contact is taken to coincide with the
termination of organic-rich limestone beds or the appearance of
common sandstone interbeds. Field observations can sometimes
be confirmed by a decrease in gamma-ray counts or correlative
geochemical trends (e.g. decrease in total organic carbon (TOC),
Mo) (Smith 2009).
Chemostratigraphy
The only published geochemical or isotopic data from the three
units is the S-isotopic study of Ross et al. (1995) on pyrite in the
post-rift strata of the WSG, Cariboo Mountains, British Columbia
(Fig. 37.1,E). A total of 170 samples were analysed for d34Spy and
these showed a broad range of isotopic values (c. 50‰) (Ross et al.
1995). As part of that study, a suite of 36 samples were collected
from a measured section that includes the OFP. A strong corre-
lation between lithology (and inferred sedimentation rate) and
pyrite isotopic composition was reported (Ross et al. 1995). For
example, the most negative d34Spy values ( –31.9‰ to –25.3‰)
corresponded to siltstone and mudstone of the OFP, whereas
the interval in the underlying Kaza Group with the highest
sandstone –mudstone ratio was the most positive (around
þ11.9‰ to þ14.5‰) (Ross et al. 1995).
New isotopic data (d18Ocarb, d13Ccarb, d13Corg, d34Spy) from OFP
sections and lithogeochemical (major, trace and rare-earth-
element analyses) data from a regional study of WSG mudstones
have recently been obtained (Smith 2009). Preliminary analysis
shows lower member limestone units exhibiting negative
d13Ccarb isotopic values and the middle member characterized by
distinctive chemical (e.g. TOC, Mo, V/Cr) and isotopic (e.g.
d13Corg) signatures.
Other characteristics
Economic deposits associated with the glaciogenic strata are
limited. Local copper mineralization has been observed below,
or in the basal parts of the OFP (Ross 2000; Smith 2009).
The only description of possible Precambrian fossils from gla-
ciogenic strata are some problematic discoid structures at the
base of the OFP near Arnica Lake, west of Banff, Alberta
(Fig. 37.1). The structures are c. 3 mm in size, lentil-shaped with
either hypo- or epirelief, have concentric wrinkles and a five-point
star-like figure inside the concentric pattern (Hofmann 1971;
Gussow 1973; Smith 2009). The specimens are attributed to
Chuaria circularis (Hofmann 1971; Gussow 1973), but they lack
any carbonaceous material that would strengthen a biogenic
origin and certainly require further detailed work. Ediacaran
fauna (including Namacalathus and Cloudina assemblages
similar to those found in the Nama Group of Namibia) have other-
wise been reported from the uppermost part of the WSG (Hofmann
& Mountjoy 2001 and references therein).
Palaeolatitude and palaeogeography
There are no published palaeolatitute data for the WSG in south-
western Canada. A near-equatorial palaeogeographical position
420 M. D. SMITH ET AL.
is based on palaeomagnetic data obtained from broadly correlative
WSG strata in the Mackenzie Mountains, northern Canadian Cor-
dillera (Park 1997).
Geochronological constraints
The predominantly siliciclastic nature of the WSG in the southern
Canadian Cordillera has led to a paucity of high-precision geochro-
nological constraints. Currently, the local maximum depositional
age of the WSG in the region is constrained by a U– Pb zircon
date of 736 þ 23/ –17 Ma [mean square weighted deviation
(MSWD) 2.36, four-point TIMS regression] from orthogneiss
basement rocks of the Malton Gneiss Complex (McDonough &
Parrish 1991). This age corresponds well with U –Pb zircon
dates obtained from WSG basement rocks elsewhere in the
Canadian Cordillera, such as 740 + 36 Ma (Parrish & Scammell
1988) and 728 þ 8/– 7 Ma (Evenchick et al. 1984; see Ross
et al. 1995 for a full review). A U –Pb zircon age of 569.6 +
5.3 Ma (three-point 207Pb/206Pb weighted average) from the
syn-rift felsic volcanic rocks of the Hamill Group that unconform-
ably overlie the WSG constrains the minimum timing of deposition
(Colpron et al. 2002).
No direct ages have been obtained from the Toby Fm. or Irene
volcanic rocks in southern British Columbia. Mafic metavolcanic
rocks of the Huckleberry Fm. in northeastern Washington, inter-
preted to be Irene Fm. correlatives were imprecisely dated with a
preliminary age of 762 + 44 Ma (MSWD 0.06) from a three-
point Sm –Nd isochron of one whole-rock and two pyroxene sep-
arates (Devlin et al. 1988). Other possible treatments of the Sm –
Nd data produced even larger error estimates: 795 + 115 Ma
(MSWD 6.71) from all the data excluding one pyroxene separ-
ate; 674 + 212 Ma (MSWD 1.0) from a seven-point whole-rock
isochron; 719 + 200 Ma (MSWD 1.0) from two whole-rock
and two pyroxene separates (Devlin et al. 1988). More recent
work on possible correlative rift-related volcanic rock units in
Idaho has obtained precise SHRIMP U –Pb zircon ages of
685 + 7 Ma, 684 + 4 Ma (Lund et al. 2003) and 709 + 5 Ma
(Fanning & Link 2004). A comparable U – Pb zircon age of
688.9 þ 9.5/ –6.2 Ma was obtained from syn-rift felsic volcani-
clastics of the Gataga Volcanics, northern British Columbia
(Ferri et al. 1999). In contrast, high-precision U –Pb (ID-TIMS)
dates were obtained from possible correlative strata in the
Central Ogilvie Mountains, NW Canada (716.47 + 0.24 Ma;
Macdonald et al. 2010).
No absolute age constraints are available for the Vreeland Fm.
Its relative age is based on regional mapping and lithostratigraphic
correlations within the WSG (McMechan 1987, 1990, 2000).
Kendall et al. (2004) obtained two comparable five-point Re –Os
isochron ages from organic-rich black mudstone of the OFP
middle member in the Jasper area, Alberta (Fig. 37.1): an impre-
cise 634 + 57 Ma (MSWD 65) using the conventional inverse
aqua regia digestion method, and a more precise 607.8 + 4.7 Ma
(MSWD 1.2) using the CrO3-H2SO4 digestion technique that is
thought to best reflect depositional age of the OFP.
Discussion
Toby Fm.
The variable level of erosion below the sub-Toby unconformity
suggests a period of significant erosion and uplift preceded WSG
sedimentation. The depositional setting of the Toby Fm. itself
has been controversial, specifically regarding the extent of
glacial influence on deposition. Most lithologies record remobili-
zation of sediment by sediment-gravity flows. Some authors
have favoured a glacial setting for these reworked deposits based
on the regional extent of the Toby Fm., the presence (albeit
localized) of extrabasinal clasts, striated clasts, clast clusters
within diamictites and outsized clasts interpreted as far-travelled
and ice-rafted glacial debris (Aalto 1971; Warren 1997). Others
favoured the resedimentation of locally derived material formed
along a fault-bounded margin undergoing extension with limited
or no glacial influence (Walker 1926; Reesor 1973; Root 1987).
The predominance of intrabasinal clasts, the mounting evidence
for structural controls on facies type and variability (ongoing
research), in addition to the localized and rare occurrence of
glacial indicators, suggest that tectonic activity imparted a
primary control on deposition with localized glacial influence.
Carbonate strata within and at the top of the Toby Fm. have
received limited attention to date. Preliminary results from
ongoing work suggest that primary carbonate was accumulating
on topographic highs while coarse-grained clastics accumulated
in fault-bounded basins. Stable isotopic analyses of these
carbonates have so far yielded limited information, with current
research focused on assessing the effects of metamorphism on
isotopic signatures.
Vreeland Fm.
The Vreeland Fm. is interpreted to have been deposited in a mid-
slope, glacially influenced marine setting (McMechan 2000). The
diamictite units are interpreted to be resedimented sediment-
gravity flow deposits (e.g. unconfined debris flows) of glaciogenic
debris or the result of rainout processes from fine-grained sus-
pended sediment plumes and ice-rafted debris (McMechan 2000).
The argument for a glacial origin is based on large, sub-angular
extrabasinal basement clasts in the diamictite and the occurrence
of lonestones interpreted as dropstones in the mudstone units of
the Snake Indian thrust sheet (McMechan 2000). The absence of
dropstones in the Wapiti Pass thrust sheet Vreeland sediments
suggests that there was little to no ice-rafted material at this palaeo-
geographical location (McMechan 2000). The considerable strike
length (c. 400 km) of diamictite suggests glacial erosion over a
large area, whereas its thickness (c. 2000 m) suggests active subsi-
dence accompanied by local faulting (McMechan 2000).
In the Snake Indian thrust sheet, diamictite units are overlain
locally by a discontinuous thin-laminated grey limestone (cap car-
bonate?) deposited during post-glacial sea-level rise and concomi-
tant shutdown of coarse clastic sediment flux to the basin
(McMechan 1990, 2000). The large shallow-marine carbonate
olistoliths in the Framstead Fm. suggest continued uplift or fault-
ing causing instability and downslope movement of parts of the
carbonate shelf edge (McMechan 2000), although simple high-
stand shedding during post-glacial eustastic rise could also be a
potential delivery mechanism.
Old Fort Point Fm.
The OFP was deposited along a deep-marine basin-slope to basin-
floor transect and overlies coarse-grained turbidites of the WSG.
The OFP is interpreted to be related to an Ediacaran-aged post-
glacial eustatic rise and a shutdown of the supply of coarse, imma-
ture siliciclastic sediment into the basin (Ross & Murphy 1988).
Fine-grained, mostly turbiditic strata of the lower and middle
members are regionally uniform in thickness and lithofacies,
suggesting synchronous, basin-wide deep-marine deposition
during transgression and highstand (Ross & Murphy 1988; Ross
et al. 1995; Smith 2009).
Conversely, the variable lithology and thickness of the upper
member appears to be controlled by a combination of more local
factors, including palaeogeographical location, topography along
the slope, erosional mass wasting, structural activity and/or rela-
tive sea-level changes (Smith 2009). Deposition of the upper
member is interpreted to have coincided with a fall of a relative
 WINDERMERE SUPERGROUP
sea level caused, at least in part, by regional uplift (Smith 2009). In
western North America, other Ediacaran-aged uplift or erosion
features have been observed, suggesting some type of renewed
extensional activity during the post-rift phase (e.g. Warren 1997;
Fedo & Cooper 2001; Clapham & Corsetti 2005). Where observa-
ble, the locally sharp, basal contact is interpreted as a sequence
boundary (possibly overprinting the basal surface of forced
regression) within the OFP (Smith 2009). Related erosional mass
wasting, submarine canyon incision and/or syn-sedimentary
growth faults controlled initial sediment transport fairways
(lowstand) that were subsequently filled during a renewed relative
sea-level rise (transgression and highstand) (Smith 2009). The
quartz-rich lithologies of the upper member compared to the
arkosic sandstone of other WSG turbidites reflect winnowing and
maturation from residence on the shelf during the earlier transgres-
sion and highstand (Ross 2000; Smith 2009).
Regional correlations
The Toby, Vreeland and OFP formations have been correlated
with other units along the North American Cordillera based on
lithostratigraphic similarities (see Lund et al. 2003 or Colpron
et al. 2002 for most recent reviews). The Toby Fm. has been cor-
related with the Kingston Peak Fm. (Death Valley), the Edwards-
burg and Pocatello formations (Idaho), the Shedroof Conglomerate
(NW Washington) and the Sayunei and Shezal formations (Mack-
enzie Mountains, NW Canada) (Gabrielse & Campbell 1991; Link
et al. 1993; Ross et al. 1995). The Vreeland Fm. has been corre-
lated to the similar diamictite-bearing Toobally (Pigage & Mac-
Naughton 2004) and Icebrook (Aitken 1991; James et al. 2001)
formations in the northern Canadian Cordillera. The Mount
Lloyd George diamictite in north-central British Columbia (Eisba-
cher 1981a, b) is of uncertain age and may correlate with either the
Toby or the Vreeland formations.
The OFP overlies turbidites of the WSG that are interpreted to
be correlative with the Vreeland Fm. (McMechan 1990, 2000;
Ross et al. 1995). The OFP lower member rhythmic limestone –
siltstone couplets are interpreted as a deep-marine equivalent of
the thin limestone overlying the Vreeland Fm. (McMechan
2000) and of the cap-carbonate succession overlying the Ice
Brook Fm. (Aitken 1991; James et al. 2001). The organic-rich
OFP middle member potentially correlates with the pyrite-rich
horizon at the top of the Vreeland Fm. (McMechan 2000) and rep-
resents maximum flooding conditions and anoxic (?) bottom-water
conditions during earliest post-glacial highstand. The correlative
link between the glaciogenic Ice Brook/Vreeland formations and
the post-glacial OFP is based primarily on stratigraphic relation-
ships and on the consistent stratigraphic presence of associated
carbonates and carbonate olistoliths possessing distinctive bladed
calcite crystals (Ross et al. 1995).
The basal glaciogenic event in the southern Canadian Cordillera
is represented by deposition of the Toby Fm., which is thought to
be younger than c. 740– 728 Ma (see Ross et al. 1995 for full
review). Various data treatments of the Sm –Nd ages (c. 795–
674 Ma) from correlative volcanic rocks (Devlin et al. 1988) are
considered too imprecise to be useful. Current precise ages from
potentially equivalent strata in Idaho and NW Canada provide
better absolute timing constraints on the glacial episodes because
they were obtained from volcanic rocks intercalated with diamic-
tite units, as opposed to unconformably underlying basement
rocks. However, further geochronological studies are needed
to determine the timing of glacial conditions in the southern
Canadian Cordillera relative to the timing of glacial conditions
in Idaho and NW Canada. The disparity in radiometric age con-
straints for rift-related magmatic rocks along the Cordilleran
margin suggests a diachronous, protracted history of crustal exten-
sion and magmatism and thus underscores the need for caution
when correlating Cordilleran glacial deposits without precise
421
geochronological constraints (Lund et al. 2003; Fanning & Link
2004). The timing of the younger glaciogenic event is
constrained by the c. 608 Ma OFP (Kendall et al. 2004), which
provides a relative minimum age for Vreeland diamictite depo-
sition based on current regional stratigraphic correlations.
M. D. S.’s PhD research on the OFP is a product of the Windermere Consortium,
initiated by R. W. C. A. and G. M. R. and jointly funded by the Natural Sciences
and Engineering Research Council of Canada and industry partners (Anadarko
Petroleum, Canadian Natural Resources Ltd, Devon Canada Ltd, Encana Corp.,
Husky Energy Corp., Nexen Inc. and Shell). M. D. S. has received additional
funding from an Ontario Graduate Scholarship (OGS), a Canadian Society of
Petroleum Geologists Scholarship and two Grants-In-Aid from the American
Association of Petroleum Geologists. E. A.’s work on the Toby Fm. is funded
by the Natural Sciences and Engineering Research Council of Canada. Parks
Canada and B. C. Parks are thanked for granting research and collections
permits to M. D. S. Critical reviews and detailed comments by Brian Kendall
and Paul Link improved the content and clarity of the manuscript. This represents
a contribution of the IUGS- and UNESCO-funded IGCP (International
Geoscience Programme) project #512.
References
Aalto, K. R. 1971. Glacial marine sedimentation and stratigraphy of the
Toby conglomerate (upper Proterozoic), southeastern British Colum-
bia, northwestern Idaho and northeastern Washington. Canadian
Journal of Earth Sciences, 8, 753– 787.
Aitken, J. D. 1969. Documentation of the sub-Cambrian unconformity,
Rocky Mountains main ranges, Alberta. Canadian Journal of Earth
Sciences, 6, 193–200.
Aitken, J. D. 1991. Two late Proterozoic glaciations, Mackenzie Moun-
tains, northwestern Canada. Geology, 19, 445– 448.
Atkinson, S. J. 1975. Surface geology of the Paradise Basin (82K/8W).
In: Geology in British Columbia. British Columbia Ministry of Mines
and Petroleum Resources 1975: G7– G12.
Bennett, S. 1986. Geology of the Mount Forster map area. British
Columbia Ministry of Energy, Mines and Petroleum Resources,
Preliminary Map 62.
Bond, G. C. & Kominz, M. A. 1984. Construction of tectonic subsidence
curves for the early Paleozoic miogeocline, southern Canadian Rocky
Mountains-implications for subsidence mechanisms, age of breakup,
and crustal thinning. Geological Society of America Bulletin, 95,
155– 173.
Charlesworth, H. A. K., Weiner, J. L. et al. 1967. Precambrian Geo-
logy of the Jasper Region. Research Council of Alberta Bulletin 23.
Clapham, M. E. & Corsetti, F. A. 2005. Deep valley incision in the term-
inal Neoproterozoic (Ediacaran) Johnnie Fm., eastern California,
USA: tectonically or glacially driven? Precambrian Research, 141,
154– 164.
Colpron, M., Logan, J. M. & Mortensen, J. K. 2002. U– Pb zircon age
constraint for late Neoproterozoic rifting and initiation of the lower
Paleozoic passive margin of western Laurentia. Canadian Journal
of Earth Sciences, 39, 133– 143.
Dalrymple, R. W. & Narbonne, G. M. 1996. Continental slope sedi-
mentation in the Sheepbed Fm. (Neoproterozoic, Windermere Super-
group), Mackenzie Mountains, N.W.T. Canadian Journal of Earth
Sciences, 33, 848–862.
Dechesne, R. G. 1990. Geology of the Ptarmigan Creek map area (east
half) and adjacent regions, Main Ranges, Rocky Mountains, British
Columbia. Geological Survey of Canada Paper 90-1D, 81 – 89.
Devlin, W. J., Brueckner, H. K. & Bond, G. C. 1988. New isotopic data
and a preliminary age for volcanics near the base of the Windermere
Supergroup, northeastern Washington, U.S.A. Canadian Journal of
Earth Sciences, 25, 1906–1911.
Eisbacher, G. H. 1981a. Sedimentary tectonics and glacial record in
the Windermere Supergroup, Mackenzie Mountains, northwestern
Canada. Geological Survey of Canada Paper 80-27.
Eisbacher, G. H. 1981b. The Late Precambrian Mount Lloyd George dia-
mictites, northern British Columbia. In: Hambrey, M. J. & Harland,
W. B. (eds) Earth’s Pre-Pleistocene Glacial Record. Cambridge
University Press, Cambridge, 728–730.
422 M. D. SMITH ET AL.
Evenchick, C. A., Parrish, R. R. & Gabrielse, H. 1984. Precambrian
gneiss and Late Proterozoic sedimentation in north-central British
Columbia. Geology, 12, 233– 237.
Fanning, C. M. & Link, P. K. 2004. U –Pb SHRIMP ages of Neoproter-
ozoic (Sturtian) glaciogenic Pocatello Fm., southeastern Idaho.
Geology, 32, 881–884.
Fedo, C. M. & Cooper, J. D. 2001. Sedimentology and sequence stratigra-
phy of Neoproterozoic and Cambrian units across a craton-margin
hinge zone, southeastern California, and implications for the early
evolution of the Cordilleran margin. Sedimentary Geology, 141–
142, 501– 522.
Ferri, F., Rees, C. J., Nelson, J. L. & Legun, A. S. 1999. Geology and
mineral deposits of the northern Kechika Trough between Gataga
River and the 60th parallel. British Columbia Ministry of Energy
and Mines Bulletin 107.
Gabrielse, H. & Campbell, R. B. 1991. Upper Proterozoic assemblages.
In: Gabrielse, H. & Yorath, C. J. (eds) Geology of the Cordilleran
Orogen. Geological Survey of Canada, Geology of Canada, 4,
125– 150.
Glover, J. K. & Price, R. A. 1976. Stratigraphy and structure of the
Windermere Supergroup, southern Kootenay Arc, British Columbia.
Geological Survey of Canada Paper 76-1B, 21 –23.
Grasby, S. E. & Brown, R. L. 1993. New correlations of the
Hadrynian Windermere Supergroup in the northern Selkirk Moun-
tains, British Columbia. Geological Survey of Canada Paper 93-1A,
199– 206.
Gussow, W. C. 1957. Cambrian and Precambrian geology of southern
Alberta. Alberta Society of Petroleum Geologists Guidebook,
Seventh Annual Field Conference, 7 – 19.
Gussow, W. C. 1973. Chuaria sp. cf. C. circularis Walcott from the
Precambrian Hector Fm., Banff National Park, Alberta, Canada.
Journal of Paleontology, 47, 1108– 1112.
Hein, F. J. & McMechan, M. E. 1994. Proterozoic – Lower Cambrian
strata of the Western Canada Sedimentary Basin. In: Mossop, G. D.
& Shetsen, I. (comps) Geological Atlas of the Western Canada
Sedimentary Basin. Canadian Society of Petroleum Geologists,
Calgary, 57 –68.
Hofmann, H. J. 1971. Precambrian fossils, pseudofossils and Problema-
tica in Canada. Geological Survey of Canada Bulletin 189.
Hofmann, H. J. & Mountjoy, E. W. 2001. Namacalathus-Cloudina
assemblage in Neoproterozoic Miette Group (Byng Fm.), British
Columbia; Canada’s oldest shelly fossils. Geology, 13, 819– 821.
James, N. P., Narbonne, G. M. & Kyser, T. K. 2001. Late Neoprotero-
zoic cap carbonates; Mackenzie Mountains, northwestern Canada;
precipitation and global glacial meltdown. Canadian Journal of
Earth Sciences, 38, 1229– 1262.
Kendall, B. S., Creaser, R. A., Ross, G. M. & Selby, D. 2004. Con-
straints on the timing of Marinoan ‘Snowball Earth’ glaciation by
187
Re– 187Os dating of a Neoproterozoic, post-glacial black shale in
Western Canada. Earth and Planetary Science Letters, 222,
729– 740.
Kubli, T. E. 1990. Geology of the Dogtooth Range, northern Purcell
Mountains, British Columbia. PhD thesis, University of Calgary.
Leech, G. B. 1954. Canal Flats, British Columbia. Geological Survey of
Canada Paper 54-7.
Link, P. K., Christie-Blick, N. et al. 1993. Middle and Late Proterozoic
stratified rocks of the western United States Cordillera, Colorado
Plateau, and Basin and Range Province. In: Reed, J., Sims, P.,
Houston, R. S., Rankin, D. W., Link, P. K., Van Schmus, W. R.
& Bickford, M. E. (eds) Precambrian: Conterminous United
States. Geological Society of America Decade of North American
Geology Series, C-3, 474–690.
Lis, M. G. & Price, R. A. 1976. Large-scale block faulting during depo-
sition of the Windermere Supergroup (Hadrynian) in southeastern
British Columbia. Geological Survey of Canada Paper 76-1A,
135– 136.
Loveridge, W. D., Leech, G. B., Stevens, R. D. & Sullivan, R. W.
1981. Zircon and monazite age of a granitic clast in Toby Conglom-
erate (Windermere Supergroup), Canal Flats, British Columbia.
Geological Survey of Canada Paper 81-1C, 131 – 134.
Lund, K., Aleinikoff, J. N., Evans, K. V. & Fanning, C. M. 2003.
SHRIMP U –Pb geochronology of Neoproterozoic Windermere
Supergroup, central Idaho; implications for rifting of western Lauren-
tia and synchroneity of Sturtian glacial deposits. Geological Society
of America Bulletin, 115, 349– 372.
Macdonald, F. A., Schmitz, M. D. et al. 2010. Calibrating the
Cryogenian. Science, 327, 1241– 1243.
McDonough, M. R. 1989. The structural geology and strain history of
the northern Selwyn Range, Rocky Mountains, near Valemount,
British Columbia. PhD thesis, University of Calgary.
McDonough, M. R. & Parrish, R. R. 1991. Proterozoic gneisses of the
Malton Complex, near Valemount, British Columbia; U– Pb ages
and Nd isotopic signatures. Canadian Journal of Earth Sciences,
28, 1202– 1216.
McMechan, M. E. 1987. Stratigraphy and structure of the Mount Selwyn
area, Rocky Mountains, northeastern British Columbia. Geological
Survey of Canada Paper 85-28.
McMechan, M. E. 1990. Upper Proterozoic to Middle Cambrian history
of the Peace River Arch; evidence from the Rocky Mountains. Bulle-
tin of Canadian Petroleum Geology, 38A, 36 – 44.
McMechan, M. E. 2000. Vreeland diamictites; Neoproterozoic glacio-
genic slope deposits, Rocky Mountains, Northeast British Columbia.
Bulletin of Canadian Petroleum Geology, 48, 246– 261.
McMechan, M. E. & Thompson, R. I. 1985. Geology of southeast
Monkman Pass area (93I/SW), British Columbia. Geological
Survey of Canada Open File 1150.
McMechan, M. E. & Thompson, R. I. 1995a. Geology, Ovington Creek,
west of sixth meridian, British Columbia. Geological Survey of
Canada Map 1873A.
McMechan, M. E. & Thompson, R. I. 1995b. Geology, Wapiti Pass, west
of sixth meridian, British Columbia. Geological Survey of Canada
Map 1872A.
Murphy, D. C. 1986. Stratigraphy and structure of the east-central
Cariboo Mountains, British Columbia, and implications for the geo-
logical evolution of the southeastern Canadian Cordillera. PhD
thesis, Carleton University.
Murphy, D. C., Walker, R. T. & Parrish, R. R. 1991. Age and geo-
logical setting of Gold Creek Gneiss, crystalline basement of the
Windermere Supergroup, Cariboo Mountains, British Columbia.
Canadian Journal of Earth Sciences, 28, 1217– 1231.
Park, J. K. 1997. Paleomagnetic evidence for low-latitude glaciation
during deposition of the Neoproterozoic Rapitan Group, Mackenzie
Mountains, N.W.T., Canada. Canadian Journal of Earth Sciences,
34, 34 – 49.
Parrish, R. R. & Scammell, R. J. 1988. The age of the Mount Copeland
syenite gneiss and its metamorphic zircons, Monashee Complex,
southeastern British Columbia. Geological Survey of Canada Paper
88-2, 21 –28.
Pell, J. & Simony, P. S. 1987. New correlations of Hadrynian strata,
south-central British Columbia. Canadian Journal of Earth Sciences,
24, 302– 313.
Pigage, L. C. & MacNaughton, R. B. 2004. Reconnaissance geology of
northern Toobally Lake (95D/8), southeast Yukon. In: Emond, D. S.
& Lewis, L. L. (eds) Yukon Exploration and Geology 2003. Yukon
Geological Survey, 199–219.
Pope, A. 1990. Geology and mineral deposits of the Toby-Horsethief
Creek map area, northern Purcell Mountains, southeast British
Columbia (82K). Ministry of Energy, Mines and Petroleum
Resources, Mineral Resources Division Open File Report 1990-26,
1 –53.
Reesor, J. E. 1973. Geology of the Lardeau Map-area, east-half, British
Columbia. Geological Survey of Canada Memoir 369.
Root, K. G. 1987. Geology of the Delphine Creek area, southeastern
British Columbia: implications for the Proterozoic and Paleozoic
development of the Cordilleran divergent margin. PhD thesis, Uni-
versity of Calgary.
Ross, G. M. 1991. Tectonic setting of the Windermere Supergroup
revisited. Geology, 19, 1125– 1128.
Ross, G. M. 2000. The Neoproterozoic Windermere Supergroup: an
on-land continental margin turbidite system. Geological Survey of
Canada Open File 3932.
Ross, G. M. & Murphy, D. C. 1988. Transgressive stratigraphy, anoxia
and regional correlations within the late Precambrian Windermere
grit of the southern Canadian Cordillera. Geology, 16, 139– 143.
 WINDERMERE SUPERGROUP
Ross, G. M. & Bowring, S. A. 1990. Detrital zircon geochronology of the
Windermere Supergroup and the tectonic assembly of the southern
Canadian Cordillera. Journal of Geology, 98, 879– 893.
Ross, G. M. & Parrish, R. R. 1991. Detrital zircon geochronology
of metasedimentary rocks in the southern Omineca Belt,
Canadian Cordillera. Canadian Journal of Earth Sciences, 28,
1254– 1270.
Ross, G. M. & Villeneuve, M. E. 1997. U– Pb geochronology of stran-
ger stones in Neoproterozoic diamictites, Canadian Cordillera;
implications for provenance and ages of deposition. Geological
Survey of Canada Paper 1997-F, 141– 155.
Ross, G. M. & Villeneuve, M. E. 2003. Provenance of the Mesoproter-
ozoic (1.45 Ga) Belt basin (western North America): another piece in
the pre-Rodinia paleogeographic puzzle. Geological Society of
America Bulletin, 115, 1191– 1217.
Ross, G. M. & Ferguson, C. R. 2003a. Geology and structure cross-
sections, Eddy, British Columbia. Geological Survey of Canada,
‘A’ series map 1967A.
Ross, G. M. & Ferguson, C. R. 2003b. Geology and structure cross-
sections, Lanezi Lake, British Columbia. Geological Survey of
Canada, ‘A’ series map 2001A.
Ross, G. M., McMechan, M. E. & Hein, F. J. 1989. Proterozoic History:
the Birth of the Miogeocline. In: Ricketts, B. D. (ed.) The Western
Canadian Sedimentary Basin. Canadian Society of Petroleum Geo-
logists, Calgary, 79 – 104.
Ross, G. M., Bloch, J. D. & Krouse, H. R. 1995. Neoproterozoic strata
of the southern Canadian Cordillera and the isotopic evolution of
seawater sulfate. Precambrian Research, 73, 71– 99.
Sevigny, J. H. 1988. Geochemistry of late Proterozoic amphibolites and
ultramafic rocks, southeastern Canadian Cordillera. Canadian
Journal of Earth Sciences, 25, 1323– 1337.
423
Simony, P. S., Ghent, E. D., Craw, D., Mitchell, W. & Robbins, D. B.
1980. Structural and metamorphic evolution of the northeast flank of
the Shuswap Complex, southern Canoe River area, British Columbia.
In: Coney, P. J. (ed.) Cordilleran Core Complexes. Geological
Society of America Memoir 153.
Slind, O. L. & Perkins, G. D. 1966. Lower Paleozoic and Proterozoic
sediments of the Rocky Mountains between Jasper, Alberta and
Pine River, British Columbia. Bulletin of Canadian Petroleum
Geology, 14, 442– 468.
Smith, M. D. 2009. Stratigraphic and geochemical evolution of the Old
Fort Point Fm., southern Canadian Cordillera: the deep-marine
perspective of Ediacaran post-glacial environmental change. PhD
thesis, University of Ottawa.
Stelck, C. R., Burwash, R. A. & Stelck, D. R. 1978. The Vreeland
High; a Cordilleran expression of the Peace River Arch. Bulletin of
Canadian Petroleum Geology, 26, 87 –104.
Stewart, J. H. 1972. Initial deposits in the Cordilleran geosyncline;
evidence of a Late Precambrian (,850 m.y.) continental separation.
Geological Society of America Bulletin, 83, 1345– 1360.
Walcott, C. D. 1910. Pre-Cambrian rocks of the Bow River Valley, Alberta,
Canada. Smithsonian Miscellaneous Collections, 53, 423–431.
Walker, J. F. 1926. Geology and Mineral Deposits of the Windermere
Map-area, British Columbia. Geological Survey of Canada Memoir
148.
Warren, M. J. 1997. Crustal extension and subsequent crustal thickening
along the Cordilleran rifted margin of ancestral North America, west-
ern Purcell Mountains, southeastern British Columbia. PhD thesis,
Queen’s University. Available at: http://www.collectionscanada.
ca/obj/s4/f2/dsk3/ftp04/nq22501.pdf.
Weiner, J. L. 1966. The Old Fort Point Fm., Jasper, Alberta. PhD thesis,
University of Alberta.
 Chapter 38

Neoproterozoic strata of southeastern Idaho and Utah: record of Cryogenian

rifting and glaciation

PAUL KARL LINK1* & NICHOLAS CHRISTIE-BLICK2

1
Department of Geosciences, Idaho State University, Pocatello, ID 83209-8072, USA
2
Department of Earth and Environmental Sciences and Lamont-Doherty Earth Observatory of Columbia University,
Palisades, NY 10964-8000, USA
*Corresponding author (e-mail: linkpaul@isu.edu)

Abstract: Neoproterozoic strata in southeastern Idaho and Utah include the ,766 Ma Uinta Mountain Group and Big Cottonwood
Formation (Fm.) deposited in an east-trending rift basin and, to the west, the lower part of a westward-thickening rift to passive-
margin succession that initiated c. 720 Ma. The latter contains a lower diamictite and volcanic succession, with a complex stratigraphic
interval of Cryogenian marine glacial deposits (Pocatello and Mineral Fork formations and correlatives). This is overlain by a mostly
terrigenous succession of ,667 Ma strata assigned to the upper member of the Pocatello Fm. and Brigham Group in southeastern
Idaho, to the Kelley Canyon Fm. and Brigham Group in northern and western Utah, and to the McCoy Creek Group and Prospect Moun-
tain Quartzite in adjacent Nevada. Although the Brigham Group and correlative deposits contain no direct evidence for glaciation, widely
developed, though stratigraphically restricted, incised valleys, with erosional relief from a few metres to as much as 160 m, are inferred to
represent subsequent times of Cryogenian glacially lowered sea level. Overall interpretations of the stratigraphy and sedimentology of
these rocks have changed little in the past 10–15 years. The most important recent advances relate to U– Pb geochronology. In strata that
lie unconformably below demonstrable glacial deposits, the lower Uinta Mountain Group (formerly thought to be c. 900 Ma) contains
populations of detrital zircons as young as 766 + 5 Ma. Cryogenian magmatism north of the Snake River Plain in central Idaho is recog-
nized near House Mountain, east of Boise at c. 725 + 5 Ma, in the Pioneer Mountains Core Complex at about 695 Ma, and in central and
east-central Idaho at 685–650 Ma. Clasts interpreted to be from the rift-related Bannock Volcanic Member of the Pocatello Fm. are dated
at 717 + 4 Ma and 701 + 4 Ma. The overlying diamictite-bearing Scout Mountain Member contains a mafic lapilli tuff near the base
(686 + 4 Ma) and a reworked fallout tuff near the top (667 + 5 Ma). Strongly negative C-isotope data have been obtained from
some of the carbonate rocks, although the latter constitute only a small fraction of the succession. Palaeomagnetic data are available
only for the Uinta Mountain Group, and suggest an equatorial palaeolatitude.

Neoproterozoic glaciogenic rocks, locally in excess of 1 km thick,
are exposed widely but discontinuously in southeastern Idaho, and
northern and western Utah (Fig. 38.1). The deposits are assigned to
a plethora of local formal and informal stratigraphic units, for his-
torical reasons and because of their varied expression. Names used
for commonly correlated diamictite-bearing units include Poca-
tello Fm. in Idaho, and Mineral Fork Fm., Sheeprock Group, for-
mation of Perry Canyon, Trout Creek sequence (units 3 and 5),
and Horse Canyon Fm. in Utah (Fig. 38.2). Crittenden et al.
(1971, 1983) and Link et al. (1993, 1994) provide regional
reviews of diamictites and their interpretation. Important papers
for specific locations include Blackwelder (1932), Crittenden
et al. (1952), Ojakangas & Matsch (1980), Blick (1981), Christie-
Blick (1982, 1983a, 1983b, 1985, 1997), Christie-Blick & Link
(1988), Christie-Blick & Levy (1989) and Yonkee et al. (2000a)
for northern and west-central Utah; Ludlum (1942), Trimble
(1976), Link (1981, 1983, 1987; Link et al. 2005) for the Pocatello
area of Idaho; and Misch & Hazzard (1962), Bick (1966) and
Rodgers (1994) for westernmost Utah and eastern Nevada. Type
localities are described by Misch & Hazzard (1962; Trout Creek
sequence and McCoy Creek Group), Bick (1966; Horse Canyon
Fm.), Crittenden et al. (1971; Kelley Canyon Fm., Brigham
Group and Pocatello Fm.), Christie-Blick (1982, 1983a; Mineral
Fork Fm. and Sheeprock Group), Crittenden et al. (1983; Pocatello
Fm.), Link (1983, 1987; Pocatello Fm.) and Link et al. (1985;
Brigham Group).
The stratigraphy and sedimentology of these strata were worked
out primarily in the 1970s and early 1980s. The correlations first
synthesized by Crittenden et al. (1971) were based on regional
field mapping, and have largely stood the test of time as summar-
ized in Link et al. (1993). More recent research has focused on U –
Pb geochronology (Fanning & Link 2004, 2008) and geochemistry
(Smith et al. 1994; Young 2002; Lorentz et al. 2004; Corsetti et al.
2007; Dehler et al. 2007).
Structural framework
The rocks are thought to have accumulated in rift-related basins
associated with the development of a passive continental margin
in western North America between c. 665 Ma and c. 520 Ma,
and in part atop erosional topography with as much as 900 m of
local relief (Stewart 1972; Stewart & Suczek 1977; Bond et al.
1983, 1985; Christie-Blick & Levy 1989; Levy & Christie-Blick
1991a; Ross 1991; Christie-Blick 1983a, 1997). They crop out
today within and along the eastern flank of the late Jurassic to
early Cenozoic Cordilleran thrust-and-fold belt, and across the
eastern edge of the late Cenozoic Basin and Range extensional
province (Fig. 38.1; Armstrong & Oriel 1965; Armstrong 1968;
Levy & Christie-Blick 1989; Allmendinger 1992; Wernicke
1992; DeCelles 2004; DeCelles & Coogan 2006). The structurally
lowest thrust sheets encompassing thick glacial and associated
deposits at the present level of exposure (generally on the
eastern side of the thrust belt) belong to the Willard –Paris –
Putnam system in northern Utah and southeastern Idaho, and to
the Tintic – Sheeprock –Canyon Range system of west-central
Utah (Levy & Christie-Blick 1989; DeCelles 2004; Fig. 38.1).
For example, the Pocatello Fm. in the Bannock Range of southeast-
ern Idaho and the formation of Perry Canyon near Ogden in the
northern Wasatch Range of Utah represent a 200 km strike-parallel
outcrop band within the Putnam-Paris thrust sheet. Although the
rocks are generally foliated, with pervasive development of chlor-
ite and locally biotite in the greenschist facies, detrital zircon
geochronology and C-isotope studies have been successfully

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 425– 436. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.38
426 P. K. LINK & N. CHRISTIE-BLICK

Fig. 38.1. Map showing areas of outcrop of Neoproterozoic rocks of Idaho and Utah, with key locations shown (after Link et al. 1994). Inset map is present geography.
Main map is palinspastic reconstruction after Levy & Christie-Blick (1989).

conducted (Fanning & Link 2004, 2008; Lorentz et al. 2004; complex, the entire Neoproterozoic succession is overturned
Corsetti et al. 2007). One or more generations of Neogene exten- beneath the Pole Canyon thrust in the southern Sheeprock Moun-
sional faults cut all of these ranges. In many cases, therefore, tains, over a lateral distance of c. 10 km (Christie-Blick 1983b). In
rocks transported eastward during Cretaceous thrusting have the northern Wasatch Range, the formation of Perry Canyon crops
been translated westward during Neogene extension (Levy & out in an east-vergent overturned fold above the Willard thrust
Christie-Blick 1989). fault. At Portneuf Narrows, SE of Pocatello, the type section of
Detailed field studies suggest that primary sedimentary features the Pocatello Fm. is exposed in an overturned fold, cut by a Creta-
and correlations are discernable through the deformation. In the ceous tear fault.
structurally highest hinterland thrust sheet of the Deep Creek Neoproterozoic rocks of the central Wasatch Range and Uinta
Range, close to the Utah – Nevada state line, garnet and staurolite Mountains are parautochthonous with respect to cratonic North
grade rocks are present in the Trout Creek sequence (Nelson America, having been displaced eastward no more than a few kilo-
1966; Rodgers 1994). East of there, but in the same thrust metres by mostly blind structures (Bruhn et al. 1986). In the Uinta
Mountain Group some strata are basically unmetamorphosed and
retain organic carbon (Dehler & Sprinkel 2005). The Big Cottonwood
(Ma)

Allochthon Parautochthon Fm. in the Wasatch Range is at greenschist facies. The least deformed
C Age

Deep Creek Central

Range SE Idaho N. Wasatch Sheeprock Wasatch Uinta Mtns and metamorphosed glacial deposits are found in the Mineral Fork
Prospect Camelback Geertsen Prospect Tintic Fm Tintic
Fm. of the central Wasatch parautochthon, except within the
Windermere Supergroup

Mountain Mountain Canyon Mountain

v ? aureole adjacent to Oligocene stocks (Christie-Blick 1983a).
Ediacaran

McCoy Browns Hole Fm 580 v

600 Creek Gp Mutual Fm
Inkom Fm ‘Mutual’ Fm
?
Neoproterozoic

Caddy Canyon Fm
fault? v 667 Stratigraphy
d Poca- d d d Mineral
d Trout d tello v Fm of v Sheep-
700 Creek v 686 Fm Perry rock Fork Fm
d sq d Canyon d Group
<717 Neoproterozoic and lower Cambrian, predominantly siliciclastic
fault? not exposed thrust ? ? rocks in southeastern Idaho and adjacent Utah are divisible into
Cryogenian

Big Uinta Mtn.

Cottonwood Gp three intervals. Pre-glacial deposits are best represented by
Fm < 766
800 locally conglomeratic sandstone and siltstone of the Uinta Moun-
Palaeoproterozoic to tain Group in the Uinta Mountains (Figs 38.2 & 38.3) and by
Archaean basement
Brigham Group 580 radiometric age (Ma) comparable quartzite and argillite of the Big Cottonwood Fm. in
and correlatives v volcanics the central Wasatch Range. Glacial and associated deposits are
Diamictite-bearing interval d glacial diamictite
represented by the Pocatello Fm. and correlatives (Figs 38.2 &
Uinta Mountain Group unconformity
and correlatives
38.4; Crittenden et al. 1971, 1983). Terminal Neoproterozoic to
lower Cambrian quartzite, minor siltstone and minor carbonate
Fig. 38.2. Utah and Idaho Neoproterozoic correlation chart showing are ‘post-glacial’ in terms of preserved facies. However, they are
stratigraphic names, ages, and locations of diamictites, volcanic rocks and relevant to the theme of this volume because they contain an indir-
carbonate strata. ect record of sea-level change (incised valleys) that may reflect the
 NEOPROTEROZOIC IDAHO AND UTAH
waxing and waning of ice sheets elsewhere on the planet. The rocks
are assigned in most places to the Brigham Group (Figs 38.2 & 38.5;
Crittenden et al. 1971; Christie-Blick 1982; Link et al. 1985). Cor-
relative strata in the Deep Creek Range are the McCoy Creek Group
and overlying Prospect Mountain Quartzite (Fig. 38.2; Misch &
Hazzard 1962). All of these Neoproterozoic and lower Cambrian
strata are broadly equivalent to the Windermere Supergroup of
Washington and western Canada and to the Kingston Peak Fm.
and overlying rocks in Death Valley (Link et al. 1993; Lund
et al. 2003, 2011; Hoffman & Halverson 2011; Mrofka &
Kennedy 2011; Petterson et al. 2011; Smith et al. 2011).
Uinta Mountain Group and Big Cottonwood Fm.
The Cryogenian but pre-glacial Uinta Mountain Group consists of
4–7 km of pervasively cross-bedded arkosic and quartzose sand-
stone, with subordinate medial and upper intervals of mudrock
(Fig. 38.3; Hansen 1965; Sanderson & Wiley 1986; Stone 1993;
Link et al. 1993; Dehler & Sprinkel 2005; Dehler et al. 2010).
The Big Cottonwood Fm. is of similar thickness (5 km) and lithol-
ogy (interstratified quartzite and argillite), but is more strongly
427
Fig. 38.3. Stratigraphy of the Uinta
Mountain Group in the central Uinta
Mountains (Kings Peak quadrangle) and in
the eastern range near Brown’s Park (after
Dehler et al. 2007, and E. M. Kingsbury,
Idaho State University, pers. comm. 2008).
SB, sequence boundary.
folded and less easily studied than the Uinta Mountain Group in
the steep-sided Wasatch Range canyons in which it is exposed.
The most distinctive attribute of these apparently non-descript
rocks is the presence of tidal rhythmites (Chan et al. 1994;
Ehlers et al. 1997; Ehlers & Chan 1999). Direct correlation
between the Uinta Mountain Group and Big Cottonwood Fm. is
not possible because younger rocks intervene between available
exposures. However, the successions are thought to be of broadly
the same age on the basis of gross lithostratigraphy and provenance
(Condie et al. 2001; Dehler et al. 2007), a stratigraphic position
below glacial deposits in the case of the latter, and the east –west
alignment of the sedimentary basins in which the successions are
thought to have accumulated (Christie-Blick 1997). Further, detri-
tal zircons in both units show the same populations, with the
youngest population in the Uinta Mountain Group at ,766 Ma
(Dehler et al. 2007).
Pocatello Fm. and correlative units
The Pocatello Fm. of southeastern Idaho consists of a lower inter-
val of mafic volcanic flows, fragmental volcanic rocks and minor
428 P. K. LINK & N. CHRISTIE-BLICK
Fig. 38.4. Stratigraphy of diamictite and volcanic succession, southeastern Idaho and northern and western Utah (after Link et al. 1994). Available radiometric ages are
from Fanning & Link (2004, 2008).
intrusives (Bannock Volcanic Member); a heterolithic medial unit
(Scout Mountain Member) that includes two intervals of diamic-
tite, cobble conglomerate, ferruginous sandstone and at least two
zircon-bearing tuff beds; and an unnamed upper member consist-
ing primarily of laminated shale or phyllite (Fig. 38.4; Ludlum
1942; Crittenden et al. 1971, 1983; Trimble 1976; Link 1981,
1983, 1987; Link et al. 2005; Rodgers et al. 2006). A thin lami-
nated carbonate, locally resedimented as multiple layers of breccia,
is present at the base of the upper member. The lower diamictite
contains abundant intrabasinal volcanic clasts, whereas the upper
contains clasts of quartzite (locally striated), granitic rocks and
felsic volcanic rocks interpreted to have been derived from the sub-
jacent Bannock Volcanic Member.
Lying unconformably on the Big Cottonwood Fm. in the vicin-
ity of Mineral Fork in the central Wasatch Range, the Mineral Fork
Fm. consists of as much as 800 m of diamictite, siltstone, sand-
stone and conglomerate (Fig. 38.4; Crittenden et al. 1952; Ojakan-
gas & Matsch 1980; Christie-Blick 1983a, 1997; Link et al. 1994).
The diamictite, which consists primarily of rounded clasts of quart-
zite, carbonate, other sedimentary rocks and minor igneous rocks
in a sandy matrix, contains striated stones, lonestones in laminated
host rock, and sheets and lenses of texturally mature sandstone.
These strata are unconformably overlain in turn by quartzite of
Mutual Fm. and Tintic Quartzite (Ediacaran to lower Cambrian).
Comparable deposits, also assigned to the Mineral Fork Fm., are
present in the Charleston-Nebo thrust sheet of the southern
Wasatch Range (small outcrops not shown in Fig. 38.1), and in
the parautochthon of Antelope Island. In the case of the latter, the
Mineral Fork Fm. rests not on the Big Cottonwood Fm., but on
gneisses of Archaean to Palaeoproterozoic age, and it is overlain
more or less conformably by a well-developed laminated carbonate
and argillite correlative with the upper member of the Pocatello Fm.
at Pocatello (Kelley Canyon Fm.; Christie-Blick 1983a; Crittenden
et al. 1983; Bryant 1988; Yonkee et al. 2000a). Diamictite at this
location is coarse-grained, with clasts composed largely of gneiss.
The formation of Perry Canyon of northern Utah consists of
more than 2 km of diamictite, greywacke, argillite, sandstone
and mafic volcanic rocks (Crittenden et al. 1983). These strata
crop out widely though discontinuously in the Willard thrust
sheet (Ogden area, Huntsville, Little Mountain and Fremont
Island in Fig. 38.1). Neither the base nor the top is exposed at
most locations. North and east of Ogden, in the northern
Wasatch Range, however, the Perry Canyon locally overlies the
Palaeoproterozoic Facer Fm., and underlies a poorly developed
laminated carbonate. The latter passes upward into several
hundred metres of siltstone (Kelley Canyon Fm.; Crittenden
et al. 1971). Diamictite clasts consist primarily of granitic rocks
and quartzite (Crittenden et al. 1983).
More than 2 km of diamictite, greywacke, conglomerate, quart-
zite and argillite is present in the Otts Canyon Fm. and overlying
Dutch Peak Fm. of the Sheeprock Group in the Sheeprock Moun-
tains in west-central Utah (Christie-Blick 1982, 1997; Crittenden
et al. 1983). The lower part of the Otts Canyon Fm., stratigraphi-
cally below all exposed diamictite, consists of phyllitic argillite.
The uppermost part of the Otts Canyon Fm. is intruded by mafic
sills. The Dutch Peak Fm. is overlain directly by shale of the
Kelley Canyon Fm. without an intervening carbonate. Diamictite
in the Otts Canyon Fm. consists predominantly of quartzite
clasts in a phyllitic matrix (Christie-Blick 1982). Substantially
thicker diamictite and greywacke of the Dutch Peak Fm. consist
primarily of clasts of granitic rocks, carbonate and quartzite in a
phyllitic to sandy matrix, with notable lateral variations in the rela-
tive abundance of clast types.
Diamictite is present in units 3 and 5 of the lower part of the
Trout Creek sequence of Misch & Hazzard (1962; Horse Canyon
Fm. of Bick 1966) in the Deep Creek Range on the Utah –
Nevada state line (Rodgers 1994). Details of the stratigraphic
succession are less well established than at other locations
owing to structural complexity and higher metamorphic grade
(Christie-Blick 1982).
 NEOPROTEROZOIC IDAHO AND UTAH
Fig. 38.5. Pocatello area (Pocatello Fm., Blackrock Canyon Limestone and
Brigham Group) stratigraphy showing available C-isotope data (after Lorentz
et al. 2004). In the C-isotope curve, interval ‘a’ is the cap carbonate that overlies
the diamictite of the Scout Mountain Member, interval ‘b’ is the Blackrock
Canyon Limestone, and interval ‘c’ is from thin dolomites within the marine
Caddy Canyon Quartzite. Qz, quartzite.
Brigham Group
In most areas of southeastern Idaho and northern Utah, many hun-
dreds of metres of shale and argillite directly overlie glacial strata
(upper member of the Pocatello Fm., Kelley Canyon Fm., and
Trout Creek sequence unit 6). These pass upwards into a quartzose
succession up to several kilometres thick of late Neoproterozoic
(late Cryogenian and Ediacaran) to early Cambrian age. At most
locations in southeastern Idaho and northern Utah, these strata
are assigned to the Brigham Group (Crittenden et al. 1971;
Christie-Blick 1982; Stewart 1982; Link et al. 1985, 1987, 1993;
Levy & Christie-Blick 1991b). In westernmost Utah and eastern
Nevada, the rocks are placed in the upper Trout Creek sequence,
McCoy Creek Group and Prospect Mountain Quartzite (Misch &
Hazzard 1962; Fig. 38.2).
The stratigraphy of the Brigham Group varies in detail, particu-
larly in the distribution of carbonate rocks (e.g. Blackrock Canyon
Limestone in Idaho) and volcanic rocks (e.g. Browns Hole Fm. in
the northern Wasatch Range), and in the abundance of siltstone at
the level of the Caddy Canyon Quartzite (Christie-Blick 1982).
The interval that encompasses the upper part of the Caddy
Canyon Quartzite (as much as 2 km of orthoquartzite), Inkom
Fm. (up to 150 m of olive drab to greyish red or liver-coloured silt-
stone) and Mutual Fm. (several hundred metres of greyish red
pebbly quartzite) contains incised valley-fill conglomerates, corre-
lated with confidence from range to range over two states
(Fig. 38.2; Crittenden et al. 1971; Christie-Blick 1982). The suc-
cession at Pocatello is representative (Fig. 38.5).
429
A striking feature of the Brigham Group, in addition to its
overall lithostratigraphy, is the presence at many localities of an
unconformity (sequence boundary) with prominent conglomerate-
filled incised valleys at or near the top of the Caddy Canyon Quart-
zite from Pocatello south and west to the Sheeprock Mountains
(Christie-Blick et al. 1988; Christie-Blick & Levy 1989; Levy &
Christie-Blick 1991b; Levy et al. 1994; Christie-Blick 1997).
The valleys are typically as much as several tens of metres deep
(up to 160 m deep in the southern Sheeprock Mountains). A
second unconformity, at or near the base of the Mutual Fm., is
recognized on the basis of a regionally persistent abrupt upward
coarsening of facies, and the presence of incised valleys in the
southern Sheeprock Mountains and central Wasatch Range
(Christie-Blick 1997).
Glaciogenic deposits and associated strata
Glaciogenic deposits in southeastern Idaho and Utah consist of a
heterogeneous assemblage of sedimentary facies associations
(see Link et al. 1994 for the most recent review) including
massive diamictite, stratified diamictite and graded sandstone, dia-
mictite and laminated mature sandstone, and carbonate, shale and
sandstone. The massive diamictite association is characterized by
diamictite as much as hundreds of metres thick with little or no
stratification. Where present, bedding is indistinct, and defined
by subtle variations in clast abundance, inverse and normal
grading, and lenses and interbeds of stratified sandstone or argil-
lite. This association is best developed in the upper Scout Moun-
tain Member, lower part of the Mineral Fork Fm. and in the
formation of Perry Canyon (Fig. 38.4). The stratified diamictite
and graded sandstone association includes bedded diamictite,
disorganized clast-supported conglomerate, massive to graded
sandstone with parallel lamination and cross-lamination, and
rhythmically bedded and laminated argillite. The association is
best expressed in the lower part of the Scout Mountain Member
and formation of Perry Canyon and in western locations (Shee-
prock Mountains and Deep Creek Range). The diamictite and
laminated mature sandstone association is characterized by strati-
fied diamictite with contorted lenses of conglomerate and finer-
grained deposits, and parallel-laminated medium- to coarse-
grained sandstone. Much of the Mineral Fork Fm. and parts of
the Dutch Peak Fm. and Trout Creek sequence is composed of
this association. The carbonate, shale and sandstone association
consists of generally laminated limestone and dolomite as much
as several metres thick, laminated, cross-laminated and graded silt-
stone, and fine- to coarse-grained sandstone. It is present in the
upper Scout Mountain Member, and the upper formation of
Perry Canyon.
Striated clasts and outsized clasts in laminated facies are
present widely in the massive and stratified diamictite facies
associations, though they are not common. The best examples
are found in the Mineral Fork Fm. of the central Wasatch
Range (Christie-Blick 1983a), particularly at the Mineral Fork
locality (striated clasts) and on the north side of Little Cotton-
wood Canyon (outsized clasts in laminated facies). Such isolated
stones are present locally in the Dutch Peak Fm., and in the for-
mation of Perry Canyon on Fremont Island. Striated clasts have
been observed also in the Dutch Peak Fm. (Sheeprock Moun-
tains) and in the upper diamictite of the Scout Mountain
Member in Idaho.
Boundary relations with overlying and underlying
non-glacial units
The lower contacts of diamictite-bearing strata range from uncon-
formable to concordant and perhaps conformable. Diamictite rests
upon an unconformity at most exposures of the Mineral Fork Fm.
430 P. K. LINK & N. CHRISTIE-BLICK
(Fig. 38.4) and locally at the base of the formation of Perry
Canyon. A striated and grooved surface with roches moutonnées
is preserved beneath the Mineral Fork Fm. on the north side of
Big Cottonwood Canyon (Christie-Blick 1983a, 1997). The age
of the Big Cottonwood Fm., which underlies the Mineral Fork at
that locality and all others except Antelope Island, is uncertain.
By correlation with the ,766 Ma Uinta Mountain Group, the
Big Cottonwood is perhaps no more than a few tens of millions
of years older than the glacial deposits (Fig. 38.3). We know that
the Big Cottonwood was sufficiently lithified to maintain a palaeo-
gradient of ,408 at valley walls without deformation, to preserve
glacial grooves, and to provide a source for well-rounded clasts
in the Mineral Fork Fm.
Elsewhere (Antelope Island, and locally in the northern Wasatch
Range, Yonkee et al. 2000b), diamictite overlies igneous and
metasedimentary rocks of Archaean to Palaeoproterozoic age. At
most locations, the lower contacts of the Pocatello Fm., the for-
mation of Perry Canyon and Sheeprock Group are not exposed.
Diamictite-bearing intervals in those sections are concordant
with underlying non-diamictic deposits (Fig. 38.4). The base of
the Trout Creek sequence (Horse Canyon Fm.) in the Deep
Creek Range is faulted.
Upper contacts with non-glacial deposits are gradational to thick
siltstone or shale in the case of the Scout Mountain Member
(Idaho), the Mineral Fork Fm. on Antelope Island, the formation
of Perry Canyon (northern Wasatch Mountains), the Dutch Peak
Fm. (Sheeprock Mountains) and the Trout Creek sequence (Deep
Creek Range; Fig. 38.4). The upper contact is unconformable at
most other exposures of the Mineral Fork Fm., and not exposed
at Little Mountain or on Fremont Island.
A laminated dolostone (in excess of 10 m thick) overlies diamic-
tite of the Mineral Fork Fm. on Antelope Island (Christie-Blick
1983a; Yonkee et al. 2000a). A correlative considerably thinner
carbonate is exposed in the northern Wasatch Range, where it over-
lies a variety of glacial and non-glacial facies (Crittenden et al.
1983). A comparably thin laminated carbonate, which overlies the
upper diamictite of the Scout Mountain Member east of Pocatello
(Link 1983, 1987; Corsetti et al. 2007), is present both in situ and
as a sedimentary breccia. The laminated dolomite at Pocatello
passes up into an upward-fining succession containing several beds
of limestone and an epiclastic tuff (Fanning & Link 2004, 2008;
Lorentz et al. 2004). Laminated carbonates above diamictite are
not present in the Sheeprock Mountains or in the Deep Creek Range.
Chemostratigraphy
Cryogenian and Ediacaran strata in Idaho and Utah have been the
subject of several chemostratigraphic and other geochemical and
provenance studies. C-isotope data have been obtained from
three relatively thin carbonate-bearing intervals in the post-glacial
part of the succession (Smith et al. 1994; Lorentz et al. 2004; Cor-
setti et al. 2007), and from organic matter from the upper part of the
pre-glacial Uinta Mountain Group (Dehler et al. 2007). Chemical
Index of Alteration (CIA) data are available for the glaciogenic
Mineral Fork Fm. (Young 2002), and for the Uinta Mountain
Group and Big Cottonwood Fm. (Condie et al. 2001). Provenance
has been studied widely in both pre-glacial and post-glacial depos-
its (Farmer & Ball 1997; Ball & Farmer 1998; Condie et al. 2001;
Mueller et al. 2007).
C-isotope ratios range from –2.9 to –6.9‰ within carbonate
beds in the uppermost part of the Scout Mountain Member, includ-
ing the thin laminated dolostone (level ‘a’ and ‘cap’ in Fig. 38.5;
Smith et al. 1994; Lorentz et al. 2004; Corsetti et al. 2007). The
data cluster around –4.5 to –5.5‰ in two of three sections (sec-
tions 2 and 3) sampled by Lorentz et al. (2004), without a well-
defined stratigraphic trend. Data are more scattered in a third
section (their section 1). O-isotope values of –13.6 to – 22.5‰
and a positive correlation with d13C in sections 1 and 2 suggest
diagenetic alteration. Lorentz et al. (2004) nevertheless took the
general consistency of C-isotope data and the absence of a diage-
netic trend in the isotopic cross-plot for section 3 to indicate that
measured d13C values provide a reasonable approximation for
the isotopic composition of seawater at the time of deposition.
C-isotope values cluster between þ1.0 and –1.5‰ in the Black-
rock Canyon Limestone in Idaho (level ‘b’ in Fig. 38.5; Corsetti
et al. 2007). Thin layers of dolomite in the middle of the Caddy
Canyon Quartzite in Idaho are characterized by d13C values
between þ3.9 and þ8.8‰ (three measurements, level ‘c’ in
Fig. 38.5; Smith et al. 1994; Corsetti et al. 2007).
C-isotope data have been obtained also from organic-rich shales
in the Red Pine Shale in the upper part of the Cryogenian Uinta
Mountain Group, stratigraphically below the glacial deposits
(Dehler et al. 2007). Whole-rock d13C values for organic matter
range from –16.9 to –30.8‰, and are comparable to values
obtained from other Cryogenian marine successions, including
the Chuar Group of the Grand Canyon Supergroup (Dehler et al.
2005a). Total organic carbon for the Red Pine Shale varies from
0.07 to 5.9% (Dehler et al. 2007).
Mudstones in the upper part of the Mineral Fork Fm. are unu-
sually Fe-rich (c. 15% Fe2O3), and comparable with Fe-rich gla-
ciogenic deposits of the Rapitan Group in the northern Canadian
Cordillera (Young 2002). The carbonate-corrected CIA for dia-
mictite in the Mineral Fork Fm. ranges from 65 to 70, values
that are significantly higher than those reported for the glaciogenic
Palaeoproterozoic Gowganda Fm. of Ontario, Canada (CIA , 60).
The Gowganda data were taken to indicate reduced chemical
weathering under frigid conditions. Data from the Mineral Fork
Fm. are consistent with the incorporation of a high proportion of
weathered silicates (sedimentary rocks). About 94% of clasts
larger than 1 cm are sedimentary (mean of 22 counts excluding
the Antelope Island locality; Christie-Blick 1983a). CIA values
for shales and argillites in the Uinta Mountain Group and Big
Cottonwood Fm. are higher than for the Mineral Fork Fm.
(mostly 75–85), reflecting significant chemical weathering of
their sources (Condie et al. 2001).
Provenance data are available for both pre-glacial and post-
glacial strata. Neodymium (Nd) and Sr-isotope data from shales
of the Uinta Mountain Group are consistent with a Laurentian pro-
venance, with both Archaean and mixed Proterozoic components
(Ball & Farmer 1998; Condie et al. 2001; Mueller et al. 2007).
A Nd-isotope study of the Trout Creek and McCoy Creek succes-
sions, above the glacial interval, suggest a Palaeoproterozoic
source within the Mojave and Yavapai provinces (Farmer & Ball
1997). Detrital zircons from Brigham Group sandstone are of
Grenvillian age (1250 – 1000 Ma), with smaller populations of
Mesoproterozoic, Palaeoproterozoic and Archaean age. These
are interpreted to represent a trans-continental provenance
(Stewart et al. 2001).
Other characteristics
Organic-walled spheres and aggregates c. 5– 20 mm in diameter
[Bavlinella faveolata (Shepeleva) Vidal] have been described
from the Red Pine Shale and older formations of the Uinta Moun-
tain Group, along with locally abundant leiosphaerid acritarchs
and filaments (Vidal & Ford 1985; Nagy & Porter 2005; Sprinkel
& Waanders 2005; Dehler et al. 2007). Similar unicells, dyads and
aggregates, also referred to as Bavlinella, were recognized by
Knoll et al. (1981) in mudrocks of the Mineral Fork Fm. In
some cases, organic material appears to have been pyritized or to
be composed of abiotic framboidal pyrite (Dehler et al. 2007;
N. J. Butterfield, pers. comm. 2009). However, the existence of
at least some microfossils is supported by a restricted stratigraphic
range, the presence of transitional morphologies, and in the case
of the Mineral Fork, the absence of pyrite (Knoll et al. 1981;
S. Porter, pers. comm. 2009). Trace fossils, including Skolithos,
 NEOPROTEROZOIC IDAHO AND UTAH
are present widely in the upper part of the Brigham Group (Camel-
back Mountain Quartzite and correlatives; Fig. 38.4), consistent
with a Cambrian age.
Palaeolatitude and palaeogeography
Hematite-cemented sedimentary rocks in the Uinta Mountain
Group yield a well-defined palaeomagnetic pole with a mean of
0.88N, 161.38E (a95 ¼ 4.68; n ¼ 9 sampling localities consisting
of 79 sites; Weil et al. 2006). The characteristic remanent magne-
tization (ChRM) is east-directed (or antipode), of low positive or
negative inclination, and typically unblocked over a narrow
range of high laboratory temperatures between 660 8C and
680 8C. A second magnetization is north- to NE-directed with
moderate to steep inclination. The presence of dual polarities
suggests that the ChRM was acquired close to the time of depo-
sition, and over a span long enough to average out secular variation
of the geomagnetic field. The second magnetization is inferred to
be a recent or modern overprint. The data are consistent with a
low palaeolatitude for Laurentia during deposition of Uinta Moun-
tain Group sediment, with a tight counterclockwise apparent polar
wander path in the south Pacific, and with deposition over an inter-
val that was sufficiently short (tens of millions of years) for the
palaeolatitude not to have changed significantly during deposition.
Sampling for this most recent study was more comprehensive than
that undertaken by Bressler (1981), but inferences about pole
location and palaeolatitude are not statistically distinguishable.
Palaeomagnetic data are not available for the glacial interval or
for the later Cryogenian and Ediacaran in the area of interest. Suit-
able rocks are not present in the glaciogenic section, which in most
places is sufficiently deformed and metamorphosed (greenschist)
to be problematic. Nor has it been possible to achieve a positive
fold test, though that was tried twice in the Brigham Group near
Huntsville, Utah (as reported in Link et al. 1994). Based on a
global assessment of the reliability of the best available palaeo-
magnetic data from Cryogenian strata and interpreted positions
of continents, Evans et al. (2000) concluded that glacial deposits
in southeastern Idaho and Utah accumulated at a palaeolatitude
of ,58, and that western North America remained at low palaeo-
latitude through the end of the Ediacaran (cf. Torsvik et al. 1996).
Geochronological constraints
A population of detrital zircons from the formation of Outlaw
Trail, in the lower part of the eastern Uinta Mountain Group,
yields a concordia age of 766 + 5 Ma, suggesting the rocks must
be younger (Dehler et al. 2007). This is substantially younger
than previous estimates of c. 900 Ma for the upper part of the
Uinta Mountain Group (Link et al. 1993; Stone 1993), and it
places a very conservative upper bound on the age of pre-glacial
sedimentation in Utah. Based on similarities in microfossils to
Chuar Group strata in the Grand Canyon that contain a 740 Ma
tuff bed, the Uinta Mountain Group is estimated to span 766–
740 Ma, and to be Cryogenian in age. The Big Cottonwood Fm.
is estimated to span the same time interval.
The timing of Neoproterozoic glaciation in the Pocatello Fm. of
southeastern Idaho is bracketed by SHRIMP U – Pb dating of
volcanic clasts within diamictite and two intervals of tuff in the
Scout Mountain Member. Felsic volcanic clasts in diamictite
near Pocatello are dated as 717 + 4 Ma (Fanning & Link 2004)
and 701 + 4 Ma (Fanning & Link 2008). These cobble- to
boulder-sized clasts are interpreted as having been eroded from
uplifted exposures of the underlying Bannock Volcanic Member
of the Pocatello Fm. (Link 1983). On Oxford Mountain near the
Idaho –Utah state line (Fig. 38.1), an epiclastic plagioclase-phyric
mafic tuff breccia interbedded with diamictite-bearing rocks near
the base of the Scout Mountain Member yielded a SHRIMP U –
431
Pb concordia age of 686 + 4 Ma (Fanning & Link 2008; a separate
older population of zircons (709 + 5 Ma) was reported by Fanning
& Link 2004). Much or all of the Scout Mountain Member is thus
younger than 686 Ma. Above the upper Scout Mountain diamictite
and the overlying laminated carbonate, an epiclastic tuff contains
zircons dated as 667 + 5 Ma (Fanning & Link 2004).
The more stratigraphically complete sections of northern and
west-central Utah remain to be dated. Attempts to separate zircons
from a clast of rhyolite obtained from conglomerate near the base
of the Dutch Peak Fm. (Sheeprock Mountains) and from a suite of
intermediate-composition volcanic clasts collected from the for-
mation of Perry Canyon at Little Mountain were not successful.
Cryogenian magmatism is also recognized north of the Snake
River Plain in central Idaho. Three samples of felsic orthogneiss
exposed in Wildhorse Creek of the Pioneer Mountains yield
SHRIMP U –Pb zircon upper-concordia intercept ages of
692.3 + 5.2, 695.7 + 8.0 and 696.5 + 9.0 Ma (location in
Fig. 38.1; K.M. Durk-Autenrieth, Idaho State University, pers.
comm. 2007). The age is thus close to 695 Ma. A somewhat
older U –Pb age (725 + 5 Ma) was obtained for the House Moun-
tain orthogneiss near the southern edge of the Atlanta lobe of the
Idaho batholith (location shown on Fig. 38.l; M. Schmitz, Boise
State University, pers. comm. 2006). In central western Idaho,
684 Ma volcanic rocks overlie diamictite near Edwardsburg
(Lund et al. 2003).
Lund et al. (2010) report ages of 665–650 Ma for Cryogenian
alkalic plutonic rocks of the Beaverhead –Big Creek belt in
central Idaho. As these are intrusive rocks, within exposures of
Proterozoic basement, their relations to Neoproterozoic glacio-
genic successions are not clear. However, these new ages extend
the locations of Cryogenian magmatism into a NW-trending
swath across much of Idaho.
The only other direct age constraint on the Neoproterozoic of
Utah and Idaho is a 580 +7 Ma age (Fig. 38.2; 40Ar/39Ar on horn-
blende from trachyte from the upper Browns Hole Fm.; Christie-
Blick & Levy 1989). This unit overlies strata that contain incised
valleys interpreted to represent sea-level drawdown associated
with a younger Cryogenian glaciation (Christie-Blick & Levy 1989).
Discussion
The Neoproterozoic and early Cambrian palaeoenvironmental and
palaeogeographic evolution of Idaho and Utah reflect a combi-
nation of climatic and tectonic controls. Mostly marine glacial
deposits are dated in southeastern Idaho as c. 686 + 4 Ma to
c. 667 + 5 Ma (Cryogenian). Thicker, more complete sections in
northern and west-central Utah may include older glacial deposits,
but age data are not yet available for those rocks. Direct evidence
for a younger Cryogenian glaciation is not preserved in this region.
However, stratigraphically restricted incised valleys as much as
160 m deep are inferred to be at least in part of glacial-eustatic
origin. Available evidence suggests that all of the Neoproterozoic
deposits accumulated at low palaeolatitude.
Depositional settings and climatic controls
The Uinta Mountain Group accumulated in braided fluvial and
shallow marine environments (Wallace & Crittenden 1969; Crit-
tenden & Wallace 1973; Sanderson 1984; Dehler et al. 2005b,
2007, 2010). Beginning at the base, marine intervals have been
documented in the Jesse Ewing Canyon Fm. and in the informal
formations of Red Castle, Mount Agassiz, Dead Horse Pass, Moo-
sehorn Lake and Red Pine Shale (Fig. 38.3; Dehler et al. 2007;
E. M. Kingsbury, pers. comm. 2009). Tidal rhythmites preserved
in the coeval Big Cottonwood Fm. provide some of the best evi-
dence for a marine connection in spite of an intracratonic setting
that must have been many hundreds of kilometres from the
432 P. K. LINK & N. CHRISTIE-BLICK
nearest oceanic crust (Chan et al. 1994; Ehlers et al. 1997; Ehlers
& Chan 1999; Dehler et al. 2005b).
Glaciogenic and associated deposits in southeastern Idaho and
Utah accumulated for the most part in a deep marine setting
(water depths as great as hundreds of metres; Christie-Blick 1983a,
1997; Crittenden et al. 1983; Link 1983; Link et al. 1994). This is
indicated by the overall character of the facies, by their association
with turbidite sandstones and subaqueous extrusive volcanic rocks,
by the presence of ice-rafted dropstones and till clots (albeit
sporadically), and in the Sheeprock Mountains by a progradational
stratigraphic architecture. The Mineral Fork Fm. is thought to have
accumulated close to the grounding line of a partially buoyant ice
sheet, on the basis of glacial grooves and probable roches mouton-
nées at the basal contact, syndepositional deformation, and abun-
dant well-stratified sandstone interpreted as subaqueous outwash.
Direct evidence for grounding is not present elsewhere, although
quartzite in the Otts Canyon and Dutch Peak formations of the
Sheeprock Mountains may have accumulated in a glacial–fluvial
to braid-delta setting (Link et al. 1994). The appreciably greater
water depths implied by most of the deposits, in comparison with
‘syn-rift’ strata at lower and higher stratigraphic levels, is thought
to indicate an isostatic response to the weight of the nearby ice
sheet as well as tectonically driven subsidence.
Up to several hundreds of metres of siltstone that in most places
directly overlie glaciogenic deposits, locally with multiple beds of
intervening carbonate, are attributed to glacial –eustatic sea-level
rise combined with thermally driven tectonic subsidence. C-
isotope data from the thin laminated carbonate beds that overlie
Scout Mountain Member diamictite are comparable with the
strongly depleted d13C values of cap carbonates elsewhere
(Corsetti et al. 2007). Near-zero to positive d13C signatures in stra-
tigraphically higher carbonate layers are comparable with data
from the Johnnie Fm. of Death Valley (Corsetti & Kaufman
2003) and are consistent with a late Cryogenian age.
The Brigham Group and correlatives are dominated by braided
fluvial, braid delta and shallow marine deposits (Link et al. 1987;
Christie-Blick & Levy 1989; Levy & Christie-Blick 1991b; Levy
et al. 1994; Christie-Blick 1997). Incised valleys at several
scales and horizons, at or near the top of the Caddy Canyon Quart-
zite, represent approximately the same stratigraphic level as the
glaciogenic Ice Brook Fm. in the Mackenzie Mountains of north-
western Canada (Aitken 1991; Levy & Christie-Blick 1991b; Levy
et al. 1994; Ross et al. 1995; Christie-Blick 1997). Taken together,
these observations are consistent with a glacial– eustatic origin and
a late Cryogenian age (655 –635 Ma; Hoffman & Li 2009).
The origin of the sequence boundary at or near the base of the
Mutual Fm. is less clear (Levy & Christie-Blick 1991b; Christie-
Blick 1997). Eustatic and tectonic explanations are both permitted
by the only available age constraint (an 40Ar/39Ar age of
580 + 7 Ma from the overlying Browns Hole Fm. at Huntsville,
Utah). A glacial –eustatic origin is consistent with age estimates
for the Gaskiers glaciation of Newfoundland. (ID-TIMS dates on
ash beds below, within and above the glaciogenic Gaskiers Fm.
constrain its age to c. 584–582 Ma; Bowring et al. 2003.) A diffi-
culty with the glacial –eustatic hypothesis is that while sea-level
rise after an initial drawdown accounts for the thickness of the
Mutual (hundreds of metres) over a broad area, it does not
explain why fluvial sedimentation continued. Christie-Blick
(1997) suggested that this might have to do with an increase in
sediment supply. A difficulty with a tectonic interpretation for
the unconformity is that regional uplift reduces available sedimen-
tary accommodation. So renewed subsidence is then required to
account for the considerable thickness of overlying quartzite.
Tectonic setting
The rocks of southeastern Idaho and Utah are inferred to have accu-
mulated in rift-related basins associated with the development of a
passive continental margin, remnants of which are preserved today
from eastern Alaska to eastern California (Stewart 1972; Stewart &
Suczek 1977; Bond et al. 1983, 1985; Christie-Blick & Levy 1989;
Ross 1991; Levy & Christie-Blick 1991a). Details of the tectonic
history are unresolved, particularly with respect to the timing of
continental separation. Evidence for crustal extension is necessarily
indirect owing to the difficulty in documenting stratigraphic growth
in available outcrop and to the tendency for originally faulted basin
margins to be offset by younger structures. The pre-glacial Uinta
Mountain Group and Big Cottonwood Fm. are thought to have
been deposited in an east-trending ,766 Ma rift basin, based
upon abrupt thinning of these deposits towards the north across a
fault-controlled boundary (Sears et al. 1982; Christie-Blick 1997;
Mueller et al. 2007). Stone’s (1993) alternative interpretation of
the Uinta Mountain Group in terms of a seaway that encroached
on an area of low topography fails to account for a stratigraphic
thickness as great as 7 km. Abrupt thickness changes in the Shee-
prock Group of the Sheeprock Mountains are attributed to a combi-
nation of facies change, progradation into a deep depocentre, and
tectonic tilting towards the SE at a hinged basin margin (Christie-
Blick 1982, 1997; Crittenden et al. 1983). Inferred stratigraphic
growth encompasses the interval of glaciation, although it may
have begun earlier. Magmatism, assumed to be rift-related, is wide-
spread at this same stratigraphic level, ranging in age from c. 720
to c. 685 Ma, and as young as 650 Ma in the case of intrusive
rocks in central Idaho. The Bannock Volcanic Member of the Poca-
tello Fm. includes mafic volcanic rocks with a rift-related trace
element signature (Harper & Link 1986).
Facies and thickness variation at the level of the upper member
of the Pocatello Fm. and Brigham Group (post-glacial) are less
pronounced, consistent with the onset of thermally driven subsi-
dence of a passive margin as early as c. 665 Ma. However, quan-
titative analysis of tectonic subsidence in Cambro-Ordovician
strata indicates that a second pulse of thermal subsidence began
as late as c. 520 Ma, if account is taken of recent changes to the
Cambrian timescale (Bond et al. 1983; Christie-Blick & Levy
1989; Levy & Christie-Blick 1991a; Link et al. 1994). Early Cam-
brian timing for this second pulse is supported in the central
Wasatch Range by angular discordance of up to 108 between the
Big Cottonwood Fm. and overlying Cambrian quartzite, which
Christie-Blick & Levy (1989) and Christie-Blick (1997) took to
indicate rift-related reactivation of the northern bounding fault of
the Big Cottonwood basin. The mismatch between the magnitude
of early Palaeozoic post-rift thermal subsidence and the minimal
evidence for crustal extension after c. 665 Ma is best explained
in terms of inhomogeneous extension of the lithosphere in latest
Neoproterozoic time (Christie-Blick & Levy 1989). Stratigraphic
evidence for multiple rifting events and for one or more times of
passive margin formation at or after c. 665 Ma is inconsistent
with palaeomagnetically based interpretations of a rift-to-drift
transition prior to 750 Ma (Wingate & Giddings 2000; Torsvik
2003; Weil et al. 2006). The latter requires assumptions about
the identity of counterpart blocks, and about the positioning of
those blocks at the time of continental break-up.
Regional correlations
Available stratigraphic and geochronological data support regional
correlation. The pre-glacial Uinta Mountain Group and Big Cot-
tonwood Fm. are comparable in terms of age and tectonic setting
to the Chuar Group of the Grand Canyon region (Karlstrom
et al. 2000). Marine strata within the three successions are approxi-
mately coeval (c. 766 –740 Ma), suggesting that they may have
accumulated within the same epicontinental ‘ChUMP seaway’,
which Dehler et al. (2005b, 2007) regarded as encompassing at
least part of the Pahrump Group in the Death Valley area of
eastern California. Cryogenian glacial deposits of Idaho and
Utah are inferred to correlate approximately with the Kingston
 NEOPROTEROZOIC IDAHO AND UTAH
Peak Fm. (upper Pahrump Group; Stewart & Suczek 1977; Link
et al. 1993; Corsetti et al. 2007; Mrofka & Kennedy 2011; Petter-
son et al. 2011) and the Edwardsburg Fm. in central Idaho (Lund
et al. 2003, 2011), although a glacial origin for the diamictite-
bearing Edwardsburg Fm. has not been demonstrated. The succes-
sions in southeastern Idaho and Utah have also traditionally been
broadly correlated with the Windermere Supergroup in Canada
(Link et al. 1993; Lund et al. 2003), although age control is not
yet adequate to explore diachrony in the timing of either glaciation
or tectonism in western North America, as predicted by the
‘zipper-rift’ hypothesis of Eyles & Januszczak (2004).
Over the years, our research in the Neoproterozoic of the western United States
has been supported by the National Science Foundation (most recently NSF
08-19884 to Link), U.S. Geological Survey, and by the Donors of the Petroleum
Research Fund, administered by the American Chemical Society. We acknowl-
edge mentors (J. C. Crowell and Max D. Crittenden, Jr) plus colleagues and
students too numerous to mention individually, who have contributed to our emer-
ging understanding of these rocks. This represents a contribution of the IUGS- and
UNESCO-funded IGCP (International Geoscience Programme) project #512.
References
Aitken, J. D. 1991. The Ice Brook Formation and post-Rapitan, Late
Proterozoic glaciation, Mackenzie Mountains, Northwest Territories.
Geological Survey of Canada Bulletin, 404.
Allmendinger, R. W. 1992. Fold and thrust tectonics of the western
United States exclusive of the accreted terranes. In: Burchfiel,
B. C., Lipman, P. W. & Zoback, M. L. (eds) The Cordilleran
Orogen: Conterminous US, The Geology of North America. Geologi-
cal Society of America, Boulder, Colorado, G-3, 583– 607.
Armstrong, F. C. & Oriel, S. S. 1965. Tectonic development of Idaho-
Wyoming thrust belt. American Association of Petroleum Geologists
Bulletin, 49, 1847–1866.
Armstrong, R. L. 1968. Sevier orogenic belt in Nevada and Utah.
Geological Society of America Bulletin, 79, 429–458.
Ball, T. T. & Farmer, G. L. 1998. Infilling history of a Neoproterozoic
intracratonic basin: Nd isotope provenance studies of the Uinta
Mountain Group, western United States. Precambrian Research,
87, 1 – 18.
Bick, K. F. 1966. Geology of the Deep Creek Mountains, Tooele and Juab
Counties, Utah. Utah Geological and Mineralogical Survey, Bulletin,
77, 120.
Blackwelder, E. 1932. An ancient glacial formation in Utah. Journal
of Geology, 40, 289–304.
Blick, N. 1981. Late Precambrian glaciation in Utah. In: Hambrey, M. J.
& Harland, W. B. (eds) Earth’s Pre-Pleistocene Glacial Record.
Cambridge University Press, Cambridge, 740– 744.
Bond, G. C., Kominz, M. A. & Devlin, W. J. 1983. Thermal subsidence
and eustasy in the Lower Paleozoic miogeocline of western North
America. Nature, 306, 775– 779.
Bond, G. C., Christie-Blick, N., Kominz, M. A. & Devlin, W. J. 1985.
An Early Cambrian rift to post-rift transition in the Cordillera of
western North America. Nature, 316, 742– 745.
Bowring, S. A., Myrow, P., Landing, E., Ramezani, J. & Grotzinger,
J. P. 2003. Geochronological constraints on terminal Neoproterozoic
events and the rise of metazoans. European Geophysical Union
Annual Meeting, Nice, 2003. Geophysical Research Abstracts, 5,
13219.
Bressler, S. L. 1981. Preliminary paleomagnetic poles and correlation
of the Proterozoic Uinta Mountain Group, Utah and Colorado.
Earth & Planetary Science Letters, 55, 53 –64.
Bruhn, R. L., Picard, M. D. & Isby, J. S. 1986. Tectonics and sedimen-
tology of Uinta Arch, western Uinta Mountains and Uinta Basin. In:
Peterson, J. A. (ed.) Paleotectonics, Sedimentation in the Rocky
Mountain Region United States. American Association of Petroleum
Geologists Memoir, 41, 333–352.
Bryant, B. 1988. Evolution and early Proterozoic history of the margin of
the Archean continent in Utah. In: Ernst, W. G. (ed.) Metamorphism
433
and Crustal Evolution of the Western United States. Prentice-Hall,
Englewood Cliffs, New Jersey, 432– 445.
Chan, M. A., Kvale, E. P., Archer, A. W. & Sonett, C. 1994. Oldest
direct evidence of lunar– solar tidal forcing encoded in sedimentary
rhythmites, Proterozoic Big Cottonwood Formation, Central Utah.
Geology, 22, 791– 794.
Christie-Blick, N. 1982. Upper Proterozoic and Lower Cambrian
rocks of the Sheeprock Mountains, Utah: regional correlation
and significance. Geological Society of America Bulletin, 93,
735– 750.
Christie-Blick, N. 1983a. Glacial-marine and subglacial sedimentation,
Upper Proterozoic Mineral Fork Formation, Utah. In: Molnia, B. F.
(ed.) Glacial Marine Sedimentation. Plenum Press, New York,
703– 776.
Christie-Blick, N. 1983b. Structural geology of the southern Sheeprock
Mountains, Utah: regional significance. In: Miller, D. M., Todd, V.
R. & Howard, K. A. (eds) Tectonic and Stratigraphic Studies in the
Eastern Great Basin. Geological Society of America Memoir, 157,
101– 124.
Christie-Blick, N. 1985. Upper Proterozoic glacial-marine and subgla-
cial deposits at Little Mountain, Utah. Brigham Young University
Geology Studies, 32, 9 –18.
Christie-Blick, N. 1997. Neoproterozoic sedimentation and tectonics in
west-central Utah. Brigham Young University Geology Studies, 42,
1– 30.
Christie-Blick, N. & Link, P. K. 1988. Glacial-marine sedimentation,
Mineral Fork Formation (Late Proterozoic) Utah. In: Holden,
G. S. (ed.) Geological Society of America Annual Meeting Field
Trip Guidebook 1988. Colorado School of Mines – Professional
Contributions, 12, 259– 274.
Christie-Blick, N. & Levy, M. (eds) 1989. Late Proterozoic and Cam-
brian tectonics, sedimentation, and record of Metazoan radiation in
the western United States. 28th International Geological Congress
Field Trip Guidebook, American Geophysical Union, Washington
DC, T331.
Christie-Blick, N., Grotzinger, J. P. & von der Borch, C. C. 1988.
Sequence stratigraphy in Proterozoic successions. Geology, 16,
100– 104.
Condie, K. C., Lee, D. & Farmer, G. L. 2001. Tectonic setting and
provenance of the Neoproterozoic Uinta Mountain and Big
Cottonwood groups, northern Utah: constraints from geochemistry,
Nd isotopes, and detrital modes. Sedimentary Geology, 141– 142,
443– 464.
Corsetti, F. A. & Kaufman, A. J. 2003. Stratigraphic investigations of
carbon isotope anomalies and Neoproterozoic ice ages in Death
Valley, California. Geological Society of America Bulletin, 115,
916– 932.
Corsetti, F. A., Link, P. K. & Lorentz, N. J. 2007. d13C chemostratigra-
phy of the Neoproterozoic succession near Pocatello, Idaho: impli-
cations for glacial chronology and regional correlations: In: Link,
P. K. & Lewis, R. S. (eds) Proterozoic Geology of Western North
America and Siberia: SEPM (Society for Sedimentary Geology)
Special Publication, 86, 193– 208.
Crittenden, M. D., Jr. & Wallace, C. A. 1973. Possible equivalents of
the Belt Supergroup in Utah. Idaho Bureau of Mines and Geology,
Moscow ID, Belt Symposium, 1, 116–138.
Crittenden, M. D., Sharp, B. J. & Calkins, F. C. 1952. Geology of
the Wasatch Mountains east of Salt Lake City, Parleys Canyon to
Traverse Range In: Marsell, R. E. (ed.) Geology of the Central
Wasatch Mountains. Utah Geological and Mineralogical Survey
Guidebook to the Geology of Utah, 8, 1 –37.
Crittenden, M. D., Jr., Schaeffer, F. E., Trimble, D. E. & Woodward,
L. A. 1971. Nomenclature and correlation of some upper
Precambrian and basal Cambrian sequences in western Utah and
southeastern Idaho. Geological Society of America Bulletin, 82,
581– 602.
Crittenden, M. D., Jr, Christie-Blick, N. & Link, P. K. 1983. Evi-
dence for two pulses of glaciation during the Late Proterozoic in
northern Utah. Geological Society of America Bulletin, 94, 437– 450.
DeCelles, P. G. 2004. Jurassic to Eocene evolution of the Cordilleran
thrust belt and foreland basin system, western U.S.A. American
Journal of Science, 304, 105–168.
434 P. K. LINK & N. CHRISTIE-BLICK
DeCelles, P. G. & Coogan, J. C. 2006. Regional structure and
kinematic history of the Sevier fold-and-thrust belt, central Utah.
Geological Society of America Bulletin, 118, 841–864,
doi: 10.1130/B25759.1.
Dehler, C. M. & Sprinkel, D. A. 2005. Revised stratigraphy and corre-
lation of the Neoproterozoic Uinta Mountain Group, northeastern
Utah. In: Dehler, C. M., Pederson, J. L., Sprinkel, D. A. &
Kowallis, B. J. (eds) Uinta Mountain Geology. Utah Geological
Association Publication, 33, 17 –30.
Dehler, C. M., Elrick, M., Bloch, J., Crossey, L., Karlstrom, K. &
Des Marais, D. 2005a. High-resolution d13C stratigraphy of the
Chuar Group (c. 770– 742 Ma), Grand Canyon: implications for
mid-Neoproterozoic climate change. Geological Society of America
Bulletin, 117, 32– 45.
Dehler, C. M., Sprinkel, D. A. & Porter, S. M. 2005b. Neoproterozoic
Uinta Mountain Group of northeastern Utah: Pre-Sturtian geographic,
tectonic, and biologic evolution. In: Pederson, J. & Dehler, C. M.
(eds) Interior Western United States. Geological Society of America
Field Guide 6, 1 –25.
Dehler, C. M., Porter, S. M., De Grey, L. D., Sprinkel, D. A. &
Brehm, A. 2007. The Neoproterozoic Uinta Mountain Group
revisited: a synthesis of recent work on the Red Pine Shale and
related undivided clastic strata, Northeastern Utah. In: Link, P. K.
& Lewis, R. S. (eds) Proterozoic Geology of Western North
America and Siberia, SEPM Special Publication, 86, 151–166.
Dehler, C. M., Fanning, C. M., Link, P. K., Kingsbury, E. M. & Rybc-
zynski, D. 2010. Incipient Rodinia breakup, marine transgression,
and peri-Gondwanan sediment source in western Laurentia at
,766– 742 Ma: new SHRIMP data from the Uinta Mountain
Group and Big Cottonwood Formation, northern Utah. Geological
Society of America Bulletin, 122, 1686– 1699, doi: 10.1130/B30094.
Ehlers, T. & Chan, M. 1999. Tidal sedimentology and estuarine depo-
sition of the Proterozoic Big Cottonwood Formation, Utah. Journal
of Sedimentary Research, 69, 1169–1180.
Ehlers, T. A., Chan, M. A. & Link, P. K. 1997. Proterozoic tidal, glacial
and fluvial sedimentation in Big Cottonwood Canyon, Utah. Brigham
Young University Geology Studies, 42, 31 –58.
Evans, D. A. D., Li, Z. X., Kirschvink, J. L. & Wingate, M. T. D. 2000.
A high-quality mid-Neoproterozoic paleomagnetic pole from South
China, with implications for ice ages and the breakup configuration
of Rodinia. Precambrian Research, 100, 313–334.
Eyles, N. & Januszczak, N. 2004. ‘Zipper-rift’: a tectonic model for
Neoproterozoic glaciations during the breakup of Rodinia after
750 Ma. Earth-Science Reviews, 65, 1– 73.
Fanning, C. M. & Link, P. K. 2004. U –Pb SHRIMP ages of Neoproter-
ozoic (Sturtian) glaciogenic Pocatello Formation, southeastern Idaho.
Geology, 32, 881–884.
Fanning, C. M. & Link, P. K. 2008. Age constraints for the Sturtian
glaciation; data from the Adelaide Geosyncline, South Australia
and Pocatello Formation, Idaho, USA. In: Gallagher, S. J. &
Wallace, M. W. (eds) Neoproterozoic extreme climates and the
origin of early metazoan life. Geological Society of Australia
Extended Abstracts, 91, 57 – 62.
Farmer, G. L. & Ball, T. T. 1997. Sources of Middle Proterozoic to Early
Cambrian siliciclastic sediments in the Great Basin: a Nd isotope
study. Geological Society of America Bulletin, 109, 1193–1205.
Hansen, W. R. 1965. Geology of the Flaming Gorge Area Utah-
Colorado-Wyoming. US Geological Survey Professional Paper,
490, 196.
Harper, G. D. & Link, P. K. 1986. Geochemistry of Upper Proterozoic
rift-related volcanics, northern Utah and southeastern Idaho.
Geology, 14, 864–867.
Hoffman, P. F. & Li, Z.-X. 2009. A palaeogeographic context for Neo-
proterozoic glaciation. Palaeogeography, Palaeoclimatology,
Palaeoecology, 277, 158–172.
Hoffman, P. F. & Halverson, G. P. 2011. Neoproterozoic glacial record
in the Mackenzie Mountains, northern Canadian Cordillera. In:
Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The
Geological Record of Neoproterozoic Glaciations. Geological
Society, London, Memoirs, 36, 397 –411.
Karlstrom, K. E., Bowring, S. A. et al. 2000. Chuar Group of the
Grand Canyon: Record of breakup of Rodinia, associated change in
the global carbon cycle, and ecosystem expansion by 740 Ma.
Geology, 28, 619–622.
Knoll, A. H., Blick, N. & Awramik, A. M. 1981. Stratigraphic and
ecologic implications of late Precambrian microfossils from Utah.
American Journal of Science, 281, 247–263.
Levy, M. & Christie Blick, N. 1989. Pre-Mesozoic palinspastic recon-
struction of the eastern Great Basin (western United States).
Science, 245, 1454– 1462.
Levy, M. & Christie-Blick, N. 1991a. Tectonic subsidence of the early
Paleozoic passive continental margin in eastern California and
southern Nevada. Geological Society of America Bulletin, 103,
1590– 1606.
Levy, M. & Christie-Blick, N. 1991b. Late Proterozoic paleogeography
of the eastern Great Basin. In: Cooper, J. D. & Stevens, C. H. (eds)
Paleozoic Paleogeography of the Western United States – II. Society
of Economic Paleontologists and Mineralogists, Los Angeles,
Pacific Section, 1, 371– 386.
Levy, M., Christie-Blick, N. & Link, P. K. 1994. Neoproterozoic
incised valleys of the eastern Great Basin, Utah and Idaho: fluvial
response to changes in depositional base level. In: Dalrymple,
R. W., Boyd, R. & Zaitlin, B. A. (eds) Incised Valley Systems:
Origin and Sedimentary Sequences. SEPM (Society for Sedimentary
Geology) Special Publication, 51, 369– 382.
Link, P. K. 1981. Upper Proterozoic diamictites in southeastern Idaho.
In: Hambrey, M. J. & Harland, W. B. (eds) Earth’s Pre-Pleistocene
Glacial Record. Cambridge University Press, Cambridge, 736– 739.
Link, P. K. 1983. Glacial and tectonically influenced sedimentation in the
Upper Proterozoic Pocatello Formation, southeastern Idaho. In:
Miller, D. M., Todd, V. R. & Howard, K. A. (eds) Tectonic and
Stratigraphic Studies in the Eastern Great Basin. Geological
Society of America Memoir, 157, 165– 181.
Link, P. K. 1987. The Late Proterozoic Pocatello Formation; A record of
continental rifting and glacial marine sedimentation, Portneuf
Narrows, southeastern Idaho. In: Beus, S. S. (ed.) Centennial Field
Guide Vol. 2. Rocky Mountain Section of the Geological Society of
America, Decade of North American Geology, Boulder, CO,
139– 142.
Link, P. K., LeFebre, G. B., Pogue, K. R. & Burgel, W. D. 1985. Struc-
tural geology between the Putnam Thrust and the Snake River Plain.
In: Kerns, G. & Kerns, R. (eds) Orogenic Patterns and Stratigraphy
of North-Central Utah and Southeastern Idaho. Utah Geological
Association Publication, 14, 67 –74.
Link, P. K., Jansen, S. T., Halimdihardja, P., Lande, A. C. & Zahn, P.
D. 1987. Stratigraphy of the Brigham Group (Late Proterozoic –
Cambrian), Bannock, Portneuf, and Bear River Ranges, southeastern
Idaho. In: Miller, W. R. (ed.) The Thrust Belt Revisited. Wyoming
Geological Association Guidebook, 38, 133–148.
Link, P. K., Christie-Blick, N. et al. 1993. Middle and Late Proterozoic
stratified rocks of the western United States Cordillera, Colorado
Plateau, and Basin and Range Province. In: Reed, J., Sims, P.,
Houston, R. S., Rankin, D. W., Link, P. K., Van Schmus, W. R.
& Bickford, M. E. (eds) Precambrian: Conterminous United
States. Geological Society of America Decade of North American
Geology Series, C-3, 474–690.
Link, P. K., Miller, J. M. G. & Christie-Blick, N. 1994. Glacial-marine
facies in a continental rift environment: Neoproterozoic rocks of the
western United States Cordillera. In: Deynoux, M., Miller,
J. M. G., Domack, E. W., Eyles, N., Fairchild, I .J. & Young,
G. M. (eds) Earth’s Glacial Record. International Geological
Correlation Project 260. Cambridge University Press, Cambridge,
29 – 59.
Link, P. K., Corsetti, F. A. & Lorentz, N. J. 2005. Pocatello Formation
and overlying strata, southeastern Idaho: snowball Earth diamictites,
cap carbonates, and Neoproterozoic isotopic profiles. In: Pederson,
J. & Dehler, C. M. (eds) Interior Western United States. Geological
Society of America Field Guide, 6, 251– 259, doi: 10.1130/
2005.fld006(12).
Lorentz, N. J., Corsetti, F. A. & Link, P. K. 2004. Seafloor precipitates
and C-isotope stratigraphy from the Neoproterozoic Scout Mountain
Member of the Pocatello Formation, southeast Idaho: implications
for Neoproterozoic earth system behavior. Precambrian Research,
130, 57– 70.
 NEOPROTEROZOIC IDAHO AND UTAH
Ludlum, J. C. 1942. Pre-Cambrian formations at Pocatello, Idaho.
Journal of Geology, 50, 85 – 95.
Lund, K., Aleinikoff, J. N., Evans, K. V. & Fanning, C. M. 2003.
SHRIMP U–Pb geochronology of Neoproterozoic Windermere
Supergroup, central Idaho: implications for rifting of western Lauren-
tia and synchroneity of Sturtian glacial deposits. Geological Society
of America Bulletin, 115, 349–372.
Lund, K., Aleinikoff, J. N., Evans, K. V., duBray, E. A., Dewitt, E. H.
& Unruh, D. M. 2010. SHRIMP U–Pb dating of recurrent Cryogen-
ian and Late Cambrian – Early Ordovician alkalic magmatism in
central Idaho: implications for Rodinian rift tectonics. Geological
Society of American Bulletin, 122, 430– 453.
Lund, K., Aleinikoff, J. N. & Evans, K. V. 2011. The Edwardsburg
Formation and related rocks, Windermere Supergroup, central
Idaho, U.S.A. In: Arnaud, E., Halverson, G. P. & Shields-Zhou,
G. (eds) The Geological Record of Neoproterozoic Glaciations. Geo-
logical Society, London, Memoirs, 36, 437–447.
Misch, P. & Hazzard, J. C. 1962. Stratigraphy and metamorphism of
late Precambrian rocks of central-east Nevada and adjacent Utah.
American Association of Petroleum Geologists Bulletin, 46,
310– 316.
Mueller, P. A., Foster, D. A., Mogk, D. W., Wooden, J. L., Kamenov,
G. D. & Vogl, J. J. 2007. Detrital mineral chronology of the Uinta
Mountain Group: implications for the Grenville flood in southwestern
Laurentia. Geology, 35, 431– 434.
Mrofka, D. & Kennedy, M. 2011. The Kingston Peak Formation in the
eastern Death Valley Region. In: Arnaud, E., Halverson, G. P. &
Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic
Glaciations. Geological Society, London, Memoirs, 36, 449– 458.
Nagy, R. M. & Porter, S. M. 2005. Paleontology of the Neoproterozoic
Uinta Mountain Group. In: Dehler, C. M., Pederson, J. L., Sprin-
kel, D. A. & Kowallis, B. J. (eds) Uinta Mountain Geology. Utah
Geological Association Publication, 33, 49 –62.
Nelson, R. B. 1966. Structural development of northernmost Snake
Range, Kern Mountains, and Deep Creek Range, Nevada and Utah.
American Association of Petroleum Geologists Bulletin, 50,
921– 951.
Ojakangas, R. W. & Matsch, C. L. 1980. Upper Precambrian (Eocam-
brian) Mineral Fork Tillite of Utah: a continental glacial and glacio-
marine sequence. Geological Society of America Bulletin, 91,
495– 501.
Petterson, R., Prave, A. R. & Wernicke, B. P. 2011 Glaciogenic and
related strata of the Neoproterozoic Kingston Peak Formation in the
Panamint Range, Death Valley region, California. In: Arnaud, E.,
Halverson, G. P. & Shields-Zhou, G. (eds) The Geological
Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 459–466.
Rodgers, D. W. 1994. Stratigraphy, correlation, and depositional environ-
ments of Upper Proterozoic and Lower Cambrian rocks of the
southern Deep Creek Range, Utah. In: Kerns, G. J. & Kerns, R.
L., Jr. (eds) Geology of Northwest Utah, Southern Idaho and
Northeast Nevada. Utah Geological Association Publication, 13,
79 – 91.
Rodgers, D. W., Long, S. P., McQuarrie, N., Burgel, W. D. &
Hersely, C. F. 2006. Geologic Map of the Inkom Quadrangle,
Bannock County, Idaho, Idaho Geological Survey Map, T-06-2,
scale 1:24,000.
Ross, G. M. 1991. Tectonic setting of the Windermere Supergroup
revisited. Geology, 19, 1125– 1128.
Ross, G. M., Block, J. D. & Krouse, H. R. 1995. Neoproterozoic strata
of the southern Canadian Cordillera and the isotopic evolution of
seawater sulfate. Precambrian Research, 73, 71– 99.
Sanderson, I. D. 1984. The Mount Watson Formation, an interpreted
braided-fluvial deposit in the Uinta Mountain Group (upper Precam-
brian), Utah. The Mountain Geologist, 21, 157– 164.
Sanderson, I. D. & Wiley, M. T. 1986. The Jesse Ewing Canyon
Formation, an interpreted alluvial fan deposit in the basal Uinta
Mountain Group (Middle Proterozoic), Utah. The Mountain Geol-
ogist, 23, 77– 89.
Sears, J. W., Graff, P. J. & Holden, G. S. 1982. Tectonic evolution of
lower Proterozoic rocks, Uinta Mountains Utah and Colorado. Geo-
logical Society of America Bulletin, 93, 990–997.
435
Smith, L. H., Kaufman, A. J., Knoll, A. H. & Link, P. K. 1994.
Chemostratigraphy of predominantly siliciclastic Neoproterozoic
successions: a case study of the Pocatello Formation and
Lower Brigham Group, Idaho, USA. Geological Magazine, 131,
301– 314.
Smith, M. D. S., Arnaud, E., Arnott, R. W. C. & Ross, G. M. 2011.
The record of Neoproterozoic glaciation in the Windermere Super-
group, southern Canadian Cordillera. In: Arnaud, E., Halverson,
G. P. & Shields-Zhou, G. (eds) The Geological Record of Neopro-
terozoic Glaciations. Geological Society, London, Memoirs, 36,
413– 423.
Sprinkel, D. A. & Waanders, G. 2005. Stratigraphy, organic microfos-
sils, and thermal maturation of the Neoproterozoic Uinta Mountain
Group in eastern Uinta Mountains, northeastern Utah. In: Dehler,
C. M., Pederson, J. L., Sprinkel, D. A. & Kowallis, B. J. (eds)
Uinta Mountain Geology. Utah Geological Association Publication,
33, 63– 73.
Stewart, J. H. 1972. Initial deposits in the Cordilleran geosyncline:
evidence of a Late Precambrian (,850 m.y.) continental separation.
Geological Society of America Bulletin, 83, 1345– 1360.
Stewart, J. H. 1982. Regional relations of Proterozoic Z and Lower Cam-
brian rocks in the western United States. In: Cooper, J. D., Troxel,
B. W. & Wright, L. A. (eds) Geology of Selected Areas in the San
Bernardino Mountains, Western Mojave Desert, and Southern
Great Basin. Guidebook for Fieldtrip No. 9, Cordilleran Section
Geological Society of America. Death Valley Publishing Company,
Shoshone Ca.,171– 186.
Stewart, J. H. & Suczek, C. A. 1977. Cambrian and latest Precambrian
paleogeography and tectonics in the western United States. In:
Stewart, J. H., Stevens, C. H. & Fritsche, A. E. (eds) Paleozoic
Paleogeography of the Western United States. Society of Economic
Paleontologists and Mineralogists, Pacific Section, Pacific Coast
Paleogeography Symposium, I, 1 – 18.
Stewart, J. H., Gehrels, G. E., Barth, A. P., Link, P. K., Christie-
Blick, N. & Wrucke, C. T. 2001. Detrital zircon provenance of
Mesoproterozoic to Cambrian arenites in the western United States
and northwestern Mexico. Geological Society of America Bulletin,
113, 1343–1356.
Stone, D. E. 1993. Tectonic evolution of the Uinta Mountains:
palinspastic restoration of a structural cross section along longitude
1098150 , Utah. Utah Geological Survey Miscellaneous Publication,
93-8.
Torsvik, T. H. 2003. The Rodinia jigsaw puzzle. Science, 300,
1379–1381.
Torsvik, T. H., Smethurst, M. A. et al. 1996. Continental break-up and
collision in the Neoproterozoic and Palaeozoic – a tale of Baltica and
Laurentia. Earth-Science Reviews, 40, 229–258.
Trimble, D. E. 1976. Geology of the Michaud and Pocatello quadrangles,
Bannock and Power Counties, Idaho. US Geological Survey Bulletin,
1400, 88.
Vidal, G. & Ford, T. D. 1985. Microbiotas from the Late Proterozoic
Chuar Group (northern Arizona) and Uinta Mountain Group (Utah)
and their chronostratigraphic implications. Precambrian Research,
28, 349– 389.
Wallace, C. A. & Crittenden, M. D. 1969. The stratigraphy, deposi-
tional environment and correlation of the Precambrian Uinta Moun-
tain Group, western Uinta Mountains, Utah. In: Lindsey, J. B. (ed.)
Geologic Guidebook of the Uinta Mountains. Intermountain Associ-
ation of Geologists Field Conference, 16, 127–142.
Weil, A. B., Geissman, J. W. & Ashby, J. M. 2006. A new palaeomag-
netic pole for the Neoproterozoic Uinta Mountain Group, Central
Rocky Mountain States, USA. Precambrian Research, 147,
234– 259.
Wernicke, B. 1992. Cenozoic extensional tectonics of the U.S. Cordillera.
In: Burchfiel, B. C., Lipman, P. W. & Zoback, M. L. (eds)
The Cordilleran Orogen: Conterminous U.S. Geological Society of
America, Boulder, Colorado, The Geology of North America, G-3,
553– 581.
Wingate, M. T. D. & Giddings, J. W. 2000. Age and paleomagnetism
of the Mundine Well dyke swarm, Western Australia. Implications
for an Australia-Laurentia connection at 755 Ma. Precambrian
Research, 100, 335–357.
436 P. K. LINK & N. CHRISTIE-BLICK

Yonkee, W. A., Willis, G. C. & Doelling, H. H. 2000a. Proterozoic and
Cambrian sedimentary and low-grade metasedimentary rocks on
Antelope Island, Utah. In: King, J. K. & Willis, G. C. (eds) The
Geology of Antelope Island. Utah Geological Survey Publication,
00-1, 37– 47.
Yonkee, W. A., Willis, G. C. & Doelling, H. H. 2000b. Petrology and
geologic history of the Precambrian Farmington Canyon Complex,
Antelope Island, Utah. In: King, J. K. & Willis, G. C. (eds) The
Geology of Antelope Island. Utah Geological Survey Publication,
00-1, 5– 36.
Young, G. M. 2002. Geochemical investigation of a Neopro-
terozoic glacial unit; the Mineral Fork Formation in the
Wasatch Range, Utah. Geological Society of America Bulletin, 114,
387– 399.
 Chapter 39

The Edwardsburg Formation and related rocks, Windermere Supergroup,

central Idaho, USA

KAREN LUND1 *, JOHN N. ALEINIKOFF2 & KARL V. EVANS1

1
US Geological Survey, MS 973, Federal Center Box 25046, Denver, CO 80225, USA
2
US Geological Survey, MS 964, Federal Center Box 25046, Denver, CO 80225, USA
*Corresponding author (e-mail: klund@usgs.gov)

Abstract: In central Idaho, Neoproterozoic stratified rocks are engulfed by the Late Cretaceous Idaho batholith and by Eocene volcanic
and plutonic rocks of the Challis event. Studied sections in the Gospel Peaks and Big Creek areas of west-central Idaho are in roof pen-
dants of the Idaho batholith. A drill core section studied from near Challis, east-central Idaho, lies beneath the Challis Volcanic Group and
is not exposed at the surface. Metamorphic and deformational overprinting, as well as widespread dismembering by the younger igneous
rocks, conceals many primary details. Despite this, these rocks provide important links for regional correlations and have produced
critical geochronological data for two Neoproterozoic glacial periods in the North American Cordillera.
At the base of the section, the more than 700-m-thick Edwardsburg Formation (Fm.) contains interlayered diamictite and volcanic
rocks. There are two diamictite-bearing members in the Edwardsburg Fm. that are closely related in time. Each of the diamictites is
associated with intermediate composition tuff or flow rocks and the diamictites are separated by mafic volcanic rocks. SHRIMP U–
Pb dating indicates that the lower diamictite is about 685 + 7 Ma, whereas the upper diamictite is 684 + 4 Ma. The diamictite units
are part of a cycle of rocks from coarse clastic, to fine clastic, to carbonate rocks that, by correlation to better preserved sections, are
thought to record an older Cryogenian glacial to interglacial period in the northern US Cordillera.
The more than 75-m-thick diamictite of Daugherty Gulch is dated at 664 + 6 Ma. This unit is preserved only in drill core and the
palaeoenvironmental interpretation and local stratigraphic relations are non-unique. Thus, the date for this diamictite may provide a
date for a newly recognized glaciogenic horizon or may be a minimum age for the diamictite in the Edwardsburg Fm.
The c. 1000-m-thick Moores Lake Fm. is an amphibolite facies diamictite in which glacial features have not been observed. However,
it is part of a sedimentary cycle from unsorted siliclastic deposits to mud and carbonate deposits. Using lithostratigraphy and available
geochronology, the Moores Lake Fm. is correlated with a younger succession of Cryogenian glaciogenic rocks in southeastern Idaho.
Traditional correlations of Neoproterozoic rocks in the Cordillera recognize two levels of Cryogenian diamictites. The Edwardsburg
and Moores Lake diamictites along the middle Cordillera fit well into the scenario of two glacial events. Because of the correlations, dates
that provide ages for the diamictites in central Idaho (and corroborated in southeastern Idaho, Link & Fanning 2008) could constrain the
age of correlated glaciogenic deposits elsewhere in the Cordillera. However, in the absence of dates for the glaciogenic diamictites in
Canadian and southern US Cordilleran sections, the correlations are considered possible but uncertain.

Two belts of Neoproterozoic continental margin facies rocks are
recognized in the western North American Cordillera, one along
the Canadian Cordillera (Hoffman & Halverson 2011; Smith
et al. 2011) and another along the southern US Cordillera in the
Great Basin (Link & Christie-Blick 2011; Mrofka & Kennedy
2011; Petterson et al. 2011) (Fig. 39.1). Within those belts, two
horizons of glaciogenic strata are widely correlated (Crittenden
et al. 1972; Gabrielse 1972; Stewart 1972). Using the nomencla-
ture for the more completely preserved succession in the Canadian
Cordillera, the preferred stratigraphic terminology for upper Cryo-
genian through Ediacaran continental margin strata (including gla-
ciogenic rocks) of the western North American Cordillera is
‘Windermere Supergroup’ (Ross 1991; Link et al. 1993). In the
well-known compilations of the Windermere Supergroup, there
was a major gap in the belts of the northern US Cordillera, from
northeastern Washington to southeastern Idaho. However,
mapping and direct dating of volcanic rocks in a NW-striking
belt of metamorphosed Neoproterozoic rocks that cross central
Idaho confirm that this ‘gap’ region contains thick, widely distrib-
uted, but previously unrecognized exposures of the Windermere
Supergroup, including diamictite-bearing units within Edwards-
burg and Moores Lake (Fig. 39.1). The metamorphic grade and
(or) restricted exposure of some of the central Idaho Neoprotero-
zoic rocks limit sedimentological, isotopic, palaeoclimatic and
stratigraphic interpretations. However, these studies, combined
with limited dating and lithostratigraphic evaluations in the
Canadian and Great Basin Cordillera segments, resulted in
continental-scale correlation of units (Lund et al. 2003; Lund
2008). Further, these rocks contribute to the reconstruction of
stratigraphic trends of the Cryogenian –Ediacaran Margin in the
northern US Cordillera (Lund et al. 2010) and produce important
geochronological data that potentially constrain the ages of other
Cryogenian glaciogenic units.
The best-preserved exposures of Neoproterozoic rocks in this
region are in roof pendants within the Idaho batholith, in the
Gospel Peaks and near the seasonal community of Big Creek,
west-central Idaho (Fig. 39.2). There are other smaller or less com-
plete exposures in east-central Idaho.
In west-central Idaho, the first mapping of the Gospel Peaks roof
pendant resulted in a description of lithological map units and in
the interpretation that they were Neoproterozoic to early Palaeo-
zoic in age, but did not provide an interpretation of the sedimentary
environment (Lund 1984). Early study of the section near Big
Creek resulted in the description of volcanic rocks, diamictite
and associated carbonate and quartzite units as local lithological
units within thick Mesoproterozoic formations (Leonard 1962).
Further studies of the Big Creek section included geological
mapping and a successful dating study (Fig. 39.3). These studies
resulted in a description of regionally distributed formations,
interpretation of depositional settings, documentation of Neopro-
terozoic ages and correlations to other Cryogenian glaciogenic
units in the Cordillera (Lund et al. 2003; Lund 2004).
In east-central Idaho, coarse lithic tuff (Daugherty Gulch,
Fig. 39.2) was described from drill core and suggested to be
Neoproterozoic in age (Jacob 1990; Oliver & Blackwell 2002).
This drill hole is in a region dominated by a series of
fragmentary exposures of structurally juxtaposed, poorly dated,
Neoproterozoic(?) to Middle Ordovician(?) miogeoclinal

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 437– 447. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.39
438 K. LUND ET AL.
successions (Hobbs & Hays 1990; Hobbs et al. 1991). Geologi-
cal and geochronological study confirmed that this is a Cryogen-
ian diamictite, providing another link between similar strata in
central Idaho and additional data for correlations with Cryogen-
ian glaciogenic units in the Cordillera (Lund et al. 2010).
Structural framework
In central Idaho, rocks of the Windermere Supergroup were
involved in intense Cretaceous thrust faulting and folding prior
to, and during emplacement of, the engulfing Cretaceous Idaho
batholith and prior to cover by, and intrusion of, Eocene Challis
Fig. 39.1. Map and stratigraphic column
showing Windermere Supergroup and
related rocks in central Idaho, as well as
distribution of equivalent rocks along the
North American Cordillera. Grey pattern,
layered rocks of Windermere Supergroup
or equivalent (see Lund et al. 2003 for
sources); black dots, Neoproterozoic
syenite-diorite suites in central Idaho (665–
650 Ma); black star, diamictite of
Daugherty Gulch (664 Ma).
volcanic-plutonic rocks (Fisher et al. 1992; Evans & Green
2003; Lund 2004). Because of this, the Windermere Supergroup
is now exposed in roof pendants and thrust– fault disrupted sec-
tions that strike NW across central Idaho within regional-scale
thrust-bounded panels (Fig. 39.2). Metamorphism and ductile
deformation overprint and disguise much of the direct evidence
needed for primary sedimentological and stratigraphic interpret-
ations of strata as well as for reconstructions of the tectonic
setting of the Neoproterozoic depositional basin.
Rocks of the Windermere Supergroup underwent regional
metamorphism at upper greenschist to amphibolite facies con-
ditions. In most places, layering is compositional due to a combi-
nation of metamorphism and ductile deformation. However,
 THE EDWARDSBURG FORMATION
116° 115°
suture
45°30’ Gospel
Peaks
River
Salmon
Fig. 39.3
Big Creek
45°
McCall
Quaternary-Miocene sedimentary and volcanic rocks
Eocene -Cretaceous intrusive and volcanic rocks
Jurassic-Permian volcanic and sedimentary rocks
Palaeozoic sedimentary rocks
Fig. 39.2. Generalized geological map showing location of Neoproterozoic rocks in roof pendants, central Idaho. Roof pendants discussed in the text and the location for
Daugherty Gulch drill hole are shown.
primary sedimentary structures are preserved locally such that stra-
tigraphic younging can be determined. The preserved younging
indicators allow reconstruction of stratigraphy and mesoscopic
structures. In the Gospel Peaks roof pendant, Windermere Super-
group rocks are repeated in two thrust sheets, one as part of an
upright section and the other in a map-scale (kilometre-scale wave-
length) syncline. In the Big Creek roof pendant, graded volcani-
clastic sandstone beds indicate that the exposures are steeply
overturned, top-to-the-SW (Fig. 39.3). At Daugherty Gulch, the
drill hole intersected upright, moderately dipping rocks that are
at the toe of a thrust sheet not exposed at the surface because of
younger normal fault reactivation of thrust faults and cover by
younger units.
Across central Idaho, igneous rocks are interlayered with or geo-
graphically associated with diamictite-bearing units. As well as
providing datable rocks, these units provide information on the
tectonic setting of the local Windermere basin. Bimodal volcanic
rock interlayered with diamictite in the Edwardsburg Fm. and
with the diamictite of Daugherty Gulch, as well as a series of
bimodal alkalic plutonic suites, are aligned along the same NW
trend (Lund et al. 2010). These aligned alkalic igneous rocks
provide evidence for significant, long-lived, intracontinental
extension related to the NW-trending sedimentary basin into
which the central Idaho Windermere rocks were deposited.
Stratigraphy
The stratigraphy of the Windermere Supergroup in central Idaho
(Fig. 39.1) is based on composite sections from several roof pen-
dants and isolated exposures. The older units include two possibly
glaciogenic diamictite-bearing members in the Edwardsburg Fm.
of west-central Idaho and the diamictite of Daugherty Gulch in
439
114°
Mo o
Ida
nta
h
na
Salmon
113o
Daugherty
Gulch
Challis
0 50 km
Ordovician-Cambrian syenite-diorite suite
Cryogenian syenite-diorite suite
Neoproterozoic volcanic and metasedimentary rocks
Mesoproterozoic metasedimentary rocks
east-central Idaho (west and east of about longitude 1158, respect-
ively, Fig. 39.2). The diamictite in the Moores Lake Fm. in west-
central Idaho, is part of the younger diamictite-bearing package.
The following descriptions are from Lund et al. (2003, 2010)
and Lund (2004).
The basal unit of the Windermere Supergroup in central Idaho is
the upper greenschist to lower amphibolite facies Edwardsburg
Fm. (Fig. 39.1). Four lithological members are described in the for-
mation from the most complete exposure near Big Creek
(Fig. 39.3). The three lower members are also exposed in two
thrust sheets in the Gospel Peaks roof pendant. At the base of the
formation, the Wind River Meadows Member is composed of dia-
mictite and local intertongued rhyodacite lithic tuff. The succeed-
ing Golden Cup Member is mafic volcanic and volcaniclastic
rocks. The Placer Creek Member is composed of two distinct
types of diamictite. Intertongued with, and at the top of the
Placer Creek Member, is the Hogback Rhyolite. This extrusive
rhyolite is locally preserved in the Big Creek roof pendant
(Fig. 39.3).
A younger, cleaved, lower-greenschist facies volcaniclastic dia-
mictite is present in a drill hole near Daugherty Gulch, 10 km west
of Challis, east-central Idaho (Fig. 39.2). In the lower part of the
core, the 1143-m-deep drill hole intersected 75 m of diamictite
without reaching the base of the unit. The upper part of the dia-
mictite is an unsorted mixture of quartz-rich matrix and
sedimentary-rock clasts. Below that, most of the unit is unsorted
rock with volcanic and sedimentary clasts in a matrix of mixed
sedimentary and tuffaceous lithic grains and sericitized clay to
silt. The deepest part of the core contains rock with mostly tuffa-
ceous matrix. Because the units cut by the drill hole do not crop
out, total unit thickness is not known.
Above the volcanic and diamictite rocks of the Edwardsburg
Fm., the Moores Station Fm. is composed of thick carbonaceous
440 K. LUND ET AL.

Fig. 39.3. Geological map of Big Creek roof pendant, central Idaho, showing locations of dated samples from Edwardsburg Fm., modified from Lund et al. (2003).
00KL040, location of dated rhyodacite tuff of Wind River Meadows Member; 97KE074, location of dated Hogback Rhyolite Member. Members (M) of the Edwardsburg
Fm. shown in inset: Hogback Rhyolite, Placer Creek, Golden Cup, Wind River Meadows.
 THE EDWARDSBURG FORMATION
phyllite with lesser interlayered marble lenses. A section of
similar, but lower metamorphic grade, phyllitic carbonate and car-
bonaceous siltite was intersected in the drill hole above the volca-
niclastic diamictite of Daugherty Gulch. Because the upper units of
the Daugherty Gulch drill core were incompletely retained and
archived, a reported upper unit has not been examined. This
upper unit is described as grey, sandy and phyllitic dolostone
(Jacob 1990). Below this dolostone lie two units, (i) dark grey,
pyrite-bearing, carbonaceous phyllite, and calcareous and dolomi-
tic marble and (ii) white, muscovite-bearing, medium-grained,
dolomitic marble (Jacob 1990). These rocks were interpreted to
be equivalent to three nearby Cambrian (?) units (Jacob 1990).
However, the Cambrian correlations are problematic because the
rocks in the core are more cleaved, are of a slightly higher meta-
morphic grade, and have different preserved bedding character-
istics than nearby exposed Cambrian rocks. At the time the hole
was drilled, there were no published descriptions of Neoprotero-
zoic carbonate or phyllite rocks in central Idaho but, based on
evaluation of the Neoproterozoic units, correlation between
rocks in the upper part of the Daugherty Gulch core and the
Moores Station Fm. is preferred (Fig. 39.1; Lund et al. 2010).
An upper level of possible glaciogenic and post-glacial rocks is
preserved in the Moores Lake and Missouri Ridge Formations
(Fig. 39.1). The Moores Lake Fm. is best preserved in the Gospel
Peaks roof pendant where it comprises coarse quartzite cobble dia-
mictite with a micaceous schist and metasandstone matrix. The
diamictite of the Moores Lake Fm. is succeeded by the Missouri
Ridge Fm., a fine-grained siliciclastic and carbonate unit.
Glaciogenic deposits and associated strata
Edwardsburg Fm.
The Edwardsburg Fm. is estimated to be c. 700 m thick on the basis
of mapped exposures, but aggregate thickness of its members is as
much as 1200 m thick.
At the base of the formation (Fig. 39.1), the Wind River
Meadows Member is about 200 m thick. The lower part of the
Wind River Meadows Member is composed of minor fine-grained
quartzite and thin mafic volcanic layers. The upper part of the
member is composed of diamictite that is laterally gradational
with rhyodacite flows and volcaniclastic conglomerate.
The diamictite in the Wind River Meadows Member is massive
to weakly layered metagreywacke with cobbles and pebbles of
quartzite and calc-silicate compositions. The clasts are mostly
,3 cm long, but can be as much as 20 cm long, sub-angular to
rounded, and stretched parallel to the overprinting cleavage or
foliation. As measured from map exposure, the diamictite is as
much as 200 m thick. The unit is massive and there is no clear
remnant compositional layering. Clasts in the diamictite are
similar to the underlying Neoproterozoic(?) and Mesoproterozoic
metasedimentary rocks.
The diamictite of the Wind River Meadows Member may inter-
tongue with, or be regionally gradational into, rhyodacite flow
breccia, tuff, graded volcaniclastic sandstone and volcaniclastic
conglomerate of heterogeneous composition. Where present, rhyo-
dacite flow and flow-breccia rocks locally dominate the base of the
Wind River Meadows Member. Volcaniclastic conglomerate con-
tains a heterogeneous mixture of rhyodacite clasts and minor quart-
zite clasts in water-worked tuffaceous matrix. The composition of
the clasts indicates that both the local rhyodacite and older quart-
zite units were being eroded and redeposited. Graded volcaniclas-
tic sandstone beds associated with the volcaniclastic conglomerate
provide younging indicators, suggesting that the exposures NW of
Big Creek are steeply overturned with top to the SW (Fig. 39.3).
Metamorphosed mafic volcanic rocks of the Golden Cup
Member form the middle of the Edwardsburg Fm. and are c.
400 m thick. The mafic volcanic rocks are massive medium- to
441
coarse-grained amphibolite containing relict pyroxene (now plagi-
oclase and chlorite) and plagioclase phenocrysts (Leonard 1962)
and fine-grained, indistinctly layered amphibolite containing horn-
blende, biotite, plagioclase and quartz. The massive amphibolite
contains zones of stretched, calcite-filled cavities suggesting
amygdaloidal texture. Possible fragmental and pillow textures
are found in the coarser-grained porphyritic amphibolites.
Above the mafic volcanic rocks of the Golden Cup Member are
two diamictite units of the Placer Creek Member that in total are
c. 100 m thick. The lower diamictite contains a heterogeneous
mixture of intrabasinal mafic and felsic volcanic clasts, indicating
that the deposits are probably the result of local reworking of vol-
canic flows. The upper diamictite contains a mixture of extrabasinal
quartzite and pebble-sized clasts of calc-silicate composition
(similar in composition to underlying older sedimentary rocks),
together with intrabasinal mafic and rhyodacitic volcanic pebbles.
The pebbles are stretched parallel to cleavage and most are
c. 2 cm long.
The Hogback Rhyolite is an c. 200-m-thick local unit that is located
at the same stratigraphic level as diamictite of the Placer Creek
Member elsewhere. The Hogback Rhyolite has relict potassium feld-
spar, plagioclase and dipyramidal quartz phenocrysts; groundmass
crystals of aligned biotite and muscovite define the metamorphic foli-
ation. Possible flow and cooling textures, including relict fiamme and
pumice fragments as well as flow breccia and water-worked layers,
are preserved. These features indicate that, although exposure of
this rhyolite is restricted, it is an extrusive flow unit that is part of
the stratigraphic section rather than a younger plug or sill.
Volcaniclastic diamictite of Daugherty Gulch
Diamictite in the Daugherty Gulch drill hole is of two types charac-
terized by lower greenschist facies and prominent cleavage. The
top 2.4 m (8 ft) of this unit in the drill core is light greenish, chlori-
tic, unsorted siltite to metasandstone matrix with grit to pebble-
sized clasts that are dominantly quartzite and metasandstone.
Below this, the drill hole cut 72.5 m (181 ft) of cleaved, hetero-
geneous-clast volcaniclastic diamictite to the bottom of the hole.
The matrix of this lower predominant diamictite comprises
unsorted material in the clay to sand size fractions. The matrix con-
tains 20–30% fractured quartz grains, some strained. Another,
about 25%, is lithic fragments, including quartz siltites, volcanic
chips and carbonate grains. The rest of the matrix is a mixture of
fine-grained sericite, quartz, opaque grains and indistinguishable
sericitic and carbonate altered material that originated as a tuffac-
eous fraction. Towards the bottom of the core, the matrix becomes
more tuffaceous, although there are few relict feldspar or mafic
minerals. Clasts make up c. 50% of the rock and are as large as
25 cm in diameter. These are dominantly rhyolite, lesser quartzite
and metasandstone (but these form the large clasts), minor carbon-
aceous phyllite and dolostone, and a few plutonic clasts. The pro-
portion of volcanic to sedimentary clasts increases downward in the
core. Volcanic clasts are light yellow, pink, green and grey. Relict
volcanic textures, such as fiamme, are common in the clasts. Meta-
sandstone clasts are dominantly pink (although some are buff ), and
well rounded. Quartzite pebbles are similar to the known Mesopro-
terozoic Gunsight and Swauger formations exposed in the vicinity
of Daugherty Gulch. The less abundant phyllite and dolostone
pebbles did not originate from the Mesoproterozoic formations of
the region but, rather, are from other Neoproterozoic units. Volca-
nic clasts are markedly elongated and aligned parallel to cleavage.
They are commonly plastically deformed where in contact with
rounded but not stretched quartzitic clasts.
Moores Station Fm.
The Moores Station Fm. overlies the Edwardsburg Fm. across
west-central Idaho (Fig. 39.1). It is composed of two dominant
442 K. LUND ET AL.
rock types with a total thickness of c. 1000 m. The major rock type
is dark grey to black (weathering to orange), fine-grained, graphi-
tic, pyrite- and apatite-rich siliceous and calcareous phyllite. The
phyllite is interlayered with grey, calc-silicate bearing calcitic
and dolomitic marble lenses that, although probably both
thickened and thinned by deformation, are as much as 30 m
thick. Although fine-scale layering is readily observed, bedding
character is not preserved because of the metamorphic mineral
growth and deformation.
In the Daugherty Gulch drill core of east-central Idaho, there is
a 1082-m-thick section of lower greenschist facies, interlayered
with carbonate and carbonaceous phyllite that is also considered
part of the Moores Station Fm. (Lund et al. 2010). The lowest
part of this section includes 59.7 m of white, muscovite-bearing,
medium-grained, dolomitic marble. Above this marble, 873.7 m
of cleaved, dark grey, pyrite-bearing, carbonaceous phyllite is
gradationally interlayered on the centimetre scale with lighter
grey, calcareous and dolomitic phyllitic marble. Soft-sediment
bedding features are preserved, including load structures that
indicate the section is upright. The phyllite is folded and some
folds are preserved in the core. Spaced cleavage is defined by
sericite foliation and is generally about 508 steeper than
bedding. An upper 148.3 m of sandy, phyllitic dolostone was
not examined for this study (because this part of the core was
not archived).
Moores Lake Fm.
The Moores Lake Fm. is an c. 1000-m-thick unit that includes an
upper quartzite and a lower diamictite. Lenses of quartzite-cobble
diamictite with cobbles up to 15 cm across occur near the base of
the formation. The diamictite matrix is muscovite-feldspar-quartz
metasandstone and biotite-muscovite-felspar-quartz schist. The
diamictite is massive and there is also no apparent compositional
layering. Upper parts of the Moores Lake Fm. are more than
95% recrystallized quartz; minor feldspar and mica make up the
rest of the rock. The upper part of the unit is massive and
bedding is difficult to identify.
Boundary relations with overlying and underlying
non-glacial units
The Edwardsburg Fm. is underlain by three formations of undeter-
mined age within the late Mesoproterozoic to early Neoproterozoic
period (Fig. 39.1). These units are staurolite-garnet schist, felds-
pathic quartzite, and calc-silicate gneiss to marble, from oldest
to youngest. The clasts in the diamictite of the Wind River
Meadows Member of the Edwarsdsburg Fm. probably originated
from these underlying older metamorphic rocks. This suggests
that the subjacent older units were exposed to erosion prior to
deposition of the Edwardsburg Fm. Local, low-angle discordance
between the Edwardsburg Fm. and underlying rocks also suggest
that the base of the Edwardsburg Fm. is an unconformity. At the
top of the Edwardsburg Fm., the Placer Creek Fm. and Hogback
Rhyolite are variably preserved, suggesting an erosional top to
the formation or lateral facies variation.
The base of the diamictite of Daugherty Gulch was not reached
by drilling so there is no information about the complete thickness
of the unit or its relations to underlying rocks. The heterogeneous
clasts of the diamictite, which includes volcanic, metasandstone,
and carbonaceous and carbonate rocks, indicate that a varied rock-
type source terrain of several ages was exposed and eroded to form
this unit. The upper contact with the overlying white dolostone was
described as fractured but conformable (Jacob 1990). However,
without exposures that can be mapped in three dimensions,
interpretations of an erosional unconformity or a thrust fault for
the upper contact cannot be ruled out.
At the top of the Edwardsburg Fm., the coarse unsorted debris of
the diamictite abruptly changes to the fine grained carbonaceous
phyllite and carbonate rocks of the Moores Station Fm. A
similar thick section of pyritic, carbonaceous phyllite and carbon-
ate rocks above the Daugherty Gulch diamictite and is believed to
be correlative to the Moores Station Fm.
Local preservation of the underlying fine-grained metasand-
stone of the Goldman Cut Fm. suggests that the Moores Lake
Fm. overlies an angular unconformity. The upper contact of the
Moores Lake Fm. with the overlying Missouri Ridge Fm. is a
narrow gradational zone of calc-silicate-dominant metasiltites.
The 450-m-thick Missouri Ridge Fm. is a ribbon-laminated calc-
silicate gneiss and marble section. This succession from a diamic-
tite unit to a carbonate-silicate is similar to the cycle of Edwards-
burg Fm. diamictites to the Moores Station carbonate rocks except
that the Missouri Ridge carbonate rocks are not as carbonaceous
and pyritic as the lower-cycle Moores Station carbonate rocks.
Chemostratigraphy
The metamorphic overprint and amount of regional magmatic and
hydrothermal activity probably make the central Idaho sections
inappropriate for chemostratigraphic analyses. However, C-
isotopic constraints available for the correlative Pocatello Fm.,
southeastern Idaho (Smith et al. 1994; Lorentz et al. 2004), bear
on interpretations for similar aged central Idaho units (see discus-
sion). A 667–686 Ma thin carbonate layer, lying directly above
diamictite in the upper Scout Mountain Member of the Pocatello
Fm., yields negative d13C values and is interpreted as a cap carbon-
ate (Corsetti et al. 2007). Carbonate rocks associated with a
667 Ma tuff and a sandstone –shale succession at the top of the
Scout Mountain Member also have negative d13C values similar
to values from cap-carbonate rocks overlying Neoproterozoic
glacial deposits (Lorenz et al. 2004). However, this upper carbon-
ate does not directly overlie rocks interpreted as glaciogenic and
the negative d13C values have been interpreted as the possible
result of overturn of an anoxic ocean rather than the result of post-
glacial processes (Lorenz et al. 2004).
Other characteristics
Limited whole-rock geochemical data are available for volcanic
rocks interlayered with diamictite in the Edwardsburg Fm. (Lund
et al. 2003, 2010). Although interpretation of this whole-rock
data is not ideal because of the metamorphic and hydrothermal
overprints, trace-element data for alternative analyses in these
rocks are incomplete (Lund et al. 2010). The metatuffs of the
Edwardsburg Fm. are marginally alkalic to alkali-calcic and
range in composition from dacite to trachydacite (as plotted on
IUGS system, Le Bas & Strecheisen 1991). SiO2 concentrations
average about 65%. Two of the samples are Ti-rich (average
TiO2 of 1.3%), which is in agreement with the alkalic to alkali-
calcic classification. Samples are mildly peraluminous, sodic to
very sodic (average 4.6%), have average to rich Fe concentrations
(average 6.8%). The mafic volcanic rocks of the Edwardsburg
Fm. are basaltic (by IUGS system of Le Bas & Strecheisen
1991; or picritic based on De la Roche et al. 1980), metaluminous,
average to potassic (average 0.35%), and have average to rich Mg
concentrations (average 12.0%). High TiO2 concentrations
(average of 2.4%) support an alkalic parentage from a basaltic to
ultramafic source. The samples range from strongly to marginally
undersaturated in silica. The Cryogenian mafic volcanic rocks in
the correlative Pocatello Fm. of southeastern Idaho (Harper &
Link 1986) and the broadly correlative Huckleberry Fm. of north-
eastern Washington and southern British Columbia (Devlin et al.
1988) are chemically similar to these metabasalts of the Edwards-
burg Fm.
 THE EDWARDSBURG FORMATION 443

Palaeolatitude and palaeogeography (a) 800

206Pb/238U age (Ma)

0.080
700
The metamorphism and multiple deformation events that have
1100
overprinted the central Idaho Windermere Supergroup rocks 0.076 600

have likely reset any palaeomagnetic signatures. As a result,

207Pb/206Pb
500
1000
direct evidence of palaeolatitude cannot be determined from 0.072 Age = 687 ± 10 Ma
(MSWD = 2.7)
central Idaho rocks. A palaeopole for Laurentia was developed
0.068 900
using data from 723 Ma sills thought to be correlative with
Rapitan Group diamictite in the Mackenzie Mountains, northern Concordia Age = 685 ± 7 Ma
800
0.064 prob. of concordance = 0.93
Canadian Cordillera (Park 1997). From these data, a low palaeola- n=11 (of 17)
titude was determined for Laurentia at 723 Ma, the presumed age 00KL040
0.060
of the Rapitan Group diamictites and of the traditionally correlated rhyodacite, Wind River
500
Cryogenian diamictite units and associated mafic volcanic rocks of Meadows Member
0.056
the Cordillera. However, the ages of, and possible correlations for, 4 6 8 10 12
the central Idaho diamictites (Fig. 39.4) put this interpretation in 238U/206Pb
question (Lund et al. 2003).
(b)

Pb/238U age (Ma)

0.080 700
Geochronological constraints
1100 660
0.076
U –Pb Sensitive High Resolution Ion Microprobe (SHRIMP)

207Pb/206Pb
620
zircon dates were obtained from volcanic members of the 1000

206
0.072 Age = 682 ± 6 Ma
Edwardsburg Fm. An additional zircon date was acquired from tuf- (MSWD = 0.97)
faceous matrix near the base of the diamictite at Daugherty Gulch. 0.068 900
A sample from the base of the rhyodacite flows in the Wind Concordia Age = 684 ± 4 Ma
800
River Meadows Member was dated by Lund et al. (2003) from a 0.064 prob. of concordance = 0.37
n=15 (of 15)
locality about 2 km north of Big Creek (location 00KL040,
97KE074
Fig. 39.3). Zircons from this rhyodacite contain low concentrations 0.060
Hogback Rhyolite 600
of U (,49 –606 ppm, mostly ,200 ppm). The weighted average 500
of the 206Pb– 238U ages, using 14 of 17 analyses, is 687 + 10 Ma 0.056
4 6 8 10 12
(Fig. 39.4a). The data are somewhat scattered, as indicated by 238U/206Pb
the mean square of weighted deviates (MSWD) of 2.7. One analy-
sis provides a significantly younger age (595 + 24 Ma), and two
analyses yield older ages (742 + 20 and 1181 + 28 Ma). A (c) 0.085

Pb/238U age (Ma)

720

Concordia Age calculated from 11 (of 17) analyses is 680

0.075
685 + 7 Ma (probability of concordance ¼ 0.93).
640
A sample from near the middle of the Hogback Rhyolite was
207Pb/206Pb

dated by Lund et al. (2003) from a locality about 0.5 km east of 0.065
600
Age = 664 ± 6 Ma
206
Big Creek (location 97KE074, Fig. 39.3). Zircons from the 750 (MSWD = 0.75)
Hogback Rhyolite contain relatively high U concentrations 550
0.055
(,327–1394 ppm) and have isotopic ratios that form a single 450
population on a concordia plot (Fig. 39.4b). The weighted
average of the 206Pb – 238U ages from 15 analyses is 682 + 6 Ma 0.045 DG1 Concordia Age = 667 ± 6 Ma
(MSWD ¼ 0.97). Alternatively, the Concordia Age calculated tuffaceous matrix prob. of concordance = 0.78
from 15 (of 15) analyses is 684 + 4 Ma (probability of n=15 (of 16)
daimictite of Daugherty Gulch
concordance ¼ 0.37). Thus, the two volcanic samples from the 0.035
7 9 11 13 15
Edwardsburg Fm. are about the same age (within uncertainty):
(i) Wind River Meadows Member, 685 + 7 Ma and (ii) 238U/206Pb
Hogback Rhyolite Member, 684 + 4 Ma (Lund et al. 2003).
Fig. 39.4. Terra-Wasserburg plots of SHRIMP U–Pb data for Cryogenian
Tuffaceous matrix from the 1140.3 m (3741 ft) depth in the
volcanic rocks in central Idaho. (a) Rhyodacite, Wind River Meadows Member,
Daugherty Gulch core (c. 55.2 m below the top of the diamictite
sample 00KL040. (b) Rhyolite, Placer Creek Member, Edwardsburg Fm.,
and 1.5 m above the bottom of the hole) was dated by the sample 97KE074. (c) Tuffaceous matrix, diamictite of Daugherty Gulch,
SHRIMP U –Pb method (Lund et al. 2010). Using weighted sample DG1. Locations for (a) and (b) in Figure 39.3. Location for (c) in
averages of selected individual 206Pb– 238U ages, SHRIMP U –Pb Figure 39.2.
isotopic data from 15 grains form a coherent grouping with an
age of 664 + 6 Ma (Fig. 39.4c). A Concordia Age calculated
from 15 (of 16) analyses is 667 + 6 Ma (probability of
Clast compositions in the diamictite units provide some evi-
concordance ¼ 0.78). This age is approximately the same as ages
dence of the provenance of the rocks. Many clasts are generally
for syenite-diorite plutonic suites that crop out across central
recognizable as locally derived from underlying sedimentary
Idaho, suggesting a regional magmatic event (Lund et al. 2010).
rocks. In the Placer Lake Member and the diamictite of Daugherty
Gulch, clasts of sedimentary rock are mixed with clasts of contem-
Discussion poraneous or slightly older volcanic rocks and closely associated
rip-up sedimentary clasts. These relations suggest that Neoproter-
Depositional setting ozoic and Mesoproterozoic rocks were locally exposed to erosion
and then included as clasts.
Because metamorphism and deformation overprinted original Direct indicators of glacial sedimentation, such as glacial facet-
sedimentary structures, palaeoenvironmental and palaeoclimatic ing of clasts, striated clasts or bedding compaction below drop-
interpretations for the central Idaho Windermere rocks are not stones, are not recognized in the diamictites of the Edwardsburg
straightforward. Fm., Daugherty Gulch drill core, or Moores Lake Fm. Possible
444 K. LUND ET AL.
origins for such poorly sorted deposits include local debris flows,
fault-scarp debris or glaciogenic conditions (see discussion in
Miller 1985). Using the available data from Cryogenian rocks of
central Idaho, a glaciogenic origin is a permissible, yet uncertain
interpretation. The palaeoenvironmental significance of these
deposits is primarily based on the stratigraphic patterns of diamic-
tite succeeded by thick, fine-grained carbonaceous siliciclastic and
carbonate units in the central Idaho sections. This stratigraphic
pattern is similar to stratigraphic successions described from
better preserved, Cordilleran glaciogenic rocks to the north and
south (Figs 39.1 & 39.5).
Textures (at microscopic to outcrop scales) in the volcanic rocks
of the Edwardsburg Fm. provide evidence that these rocks are
extrusive and interbedded with diamictites and metasandstones.
In the case of the diamictite of Daugherty Gulch, where meta-
morphic grade is low, evidence of tuffaceous matrix is preserved.
These data indicate that the diamictite was deposited synchro-
nously with local volcanism and that dates on the volcanic rocks
also provide ages for formation of associated diamictite units
(Fig. 39.4).
Geochronological constraints
The ages for the two Edwardsburg Fm. diamictites are 685+7 to
684 + 4 Ma and, given the over lapping age constraints, may rep-
resent two closely spaced events or just a single event. The age for
volcanism synchronous with diamictite at Daugherty Gulch is
664 + 4 Ma, but because of the lack of exposure, the exact relation
between this diamictite and the older Edwardsburg Fm. diamictites
are undetermined.
Correlations along the North American Cordillera
Regional corrrelations based on lithostratigraphic similarity and
geochronological data are shown in Figure 39.5. Diamictites in
the upper part of the Scout Mountain Member, Pocatello Fm.,
southeastern Idaho, are bracketed between 701+4 (Bannock
Volcanic Member clast in diamictite) and 667 + 5 Ma (tuff over-
lying diamictite) in one section and determined to be younger than
686 + 4 Ma in another section (Fanning & Link 2004, 2008).
Thus, overlapping SHRIMP U –Pb ages of about 685 Ma now
provide chronostratigraphic correlations between Edwardsburg
Fm. diamictite units in central Idaho and upper Scout Mountain
Member diamictite units in southeastern Idaho (Lund et al.
2003; Fanning & Link 2008).
Diamictites and underlying mafic volcanic rocks in the Scout
Mountain Member are correlated traditionally with other diamic-
tites associated with mafic volcanic rocks along the entire Cordil-
lera (Fig. 39.5). However, only the mafic volcanics in central Idaho
are bracketed closely by dated units (685+7 to 684 + 4 Ma). The
age of the Bannock Volcanic Member in southeastern Idaho is
interpreted based on dates of 701+4 and 717 + 4 Ma from volca-
nic clasts in an overlying diamictite, rather than based on direct
radiometric dating of the volcanic unit itself (Fanning & Link
2004, 2008). The generally accepted correlations between diamic-
tites from the Rapitan Group in the Mackenzie Mountains and
other diamictite-bearing units along the Cordillera down to the
Great Basin (Figs 39.1 & 39.5) are permissive within the current
geochronological database. However, as the ages of the other Cor-
dilleran mafic volcanic and diamictite units are poorly constrained,
the diamictites in the Canadian Cordillera (Hoffman & Halverson
2011; Smith et al. 2011) and SW USA (Mrofka & Kennedy 2011;
Petterson et al. 2011) could be either older or younger than those in
the Edwardsburg Fm.
The diamictite of Daugherty Gulch (664 + 4 Ma, Figs 39.4 &
39.5) is about the same age as a thin tuff in post-glacial sandstone-
shale rocks near the top of the Scout Mountain Member
(667 + 5 Ma, Fanning & Link 2008). The Scout Mountain tuff
is interpreted to provide a minimum age for underlying Scout
Mountain glaciogenic rocks that are thought to be correlative
with the Edwardsburg diamictites (Fanning & Link 2004, 2008).
Because this Scout Mountain tuff is not associated with recognized
glaciogenic rocks (it is stratigraphically separated from the two
underlying glaciogenic diamictites), a directly overlying carbonate
unit with negative d13C values was interpreted in the context of
oceanic processes rather than as a ‘cap’ carbonate related to a Neo-
proterozoic glaciation (Lorentz et al. 2004). The correlative ages
of the Daugherty Gulch diamictite and this Scout Mountain tuff
presents the possibility that the negative d13C values were
related to a 665 Ma glacial event or alternatively that the Daugh-
erty Gulch diamictite is non-glacial. If the former is correct, the
diamictite of Daugherty Gulch presents a potentially different,
younger (c. 665 Ma), correlation point for undated glaciogenic
diamictites that underlie thick carbonaceous shale and carbonate
successions along the Cordillera (Fig. 39.5).
The thick carbonaceous shale and carbonate rocks of the Moores
Station Fm. which lie above the Edwardsburg Fm. and Daugherty
Gulch diamictites are important for reconstructing the Windermere
Supergroup stratigraphy of central Idaho because their special
layering and composition characteristics are not seen in older or
younger rocks in the region. For the same reasons, these rocks
serve as units that can be correlated along the Cordillera
(Fig. 39.5). Elsewhere, carbonate rocks at the base of these sections
are interpreted as cap carbonates and the section is interpreted
as the result of deepening water levels in post-glacial times (e.g.
Fanning & Link 2004).
Correlation between the Moores Lake Fm. in central Idaho
and the Caddy Canyon Quartzite in southeastern Idaho is based
on lithostratigraphic similarity and context (Fig. 39.5). The ages
of those deposits are only loosely constrained by dates between
685 and 665 Ma on the underlying volcanic rocks (Lund et al.
2003; Fanning & Link 2004, 2008) and 570 and 580 Ma on over-
lying volcanic rocks (Levy & Christie-Blick 1991; Link et al.
1993; Levy et al. 1994; Christie-Blick 1997). The Caddy
Canyon Fm. is thought to represent late Cryogenian glaciogenic
rocks on the basis of these geochronological data and the preserved
incised and refilled valleys that are interpreted as a product of
sea-level fall during a glacial event (Link et al. 1993; Levy et al.
1994). Because of this interpretation, Moores Lake and Caddy
Canyon rocks can in turn be correlated with the late Cryogenian
glacial interval in the northern Canadian Cordillera (Icebrook
Fm., Fig. 39.5; Link et al. 1993; Lund et al. 2003; Fanning &
Link 2008). However, without geochronological ties, these corre-
lations (especially for the Moores Lake Fm.) are possible, but
uncertain.
Broader implications of Central Idaho geological record
The dates of c. 685 Ma for interlayered volcanic and diamictite
rocks in the Edwardsburg Fm. provide the first direct dates for Cor-
dilleran diamictite-bearing units of Cryogenian age. Previously,
the older Cryogenian glaciogenic rocks of the Windermere Super-
group were thought to be c. 760– 700 Ma on the basis of the few
available dates from underlying basement, intrusions that cut
older rocks, and inconclusively correlated rocks (see discussion
in Lund et al. 2003). Although the Cordilleran Cryogenian glacio-
genic deposits were thought to have formed during an older Cryo-
genian (‘Sturtian’) glaciation, the recent dates from the
Edwardsburg and Pocatello formations indicate that (much or all
of) the glaciation in Idaho occurred later than the commonly
cited ages for a global, older Cryogenian glacial event. In the
absence of direct dating in many of the Cordilleran sections and
the limited sedimentary evidence in central Idaho, it cannot be
determined how widespread this event was. However, it is possible
that the geochronological data for the Edwardsburg Fm. constrains
 THE EDWARDSBURG FORMATION
Fig. 39.5. Correlation chart for Windermere Supergroup strata in central Idaho and the nearby Cordillera based on lithostratigraphic and geochronological data. Locations of strata are shown in Figure 39.1. Ages are
U– Pb zircon, except where otherwise noted. Diagram is constructed to tie rock units to geological time, but lack of dates throughout Cordillera results in much interpretation for duration of units and correlations. Cgl,
Conglomerate; Cyn, Canyon; Cr, Creek; Fm, Formation; Grp, Group; Lk, Lake; M, Member; Mdws, Meadows; Mtn, Mountain; Qtzt, Quartzite; Rhy, Rhyolite; Rv, River; Volc, Volcanic. 1Aitken (1991), Rainbird et al. (1996)
Ross et al. (1995), Kendall et al. (2004); 2 Devlin & Bond (1988); Hein & McMechan (1994); 3Devlin & Bond (1988), Devlin (1989), Hein & McMechan (1994), Colpron et al. (2002); 4Evans (1987), Miller & Whipple

445
(1989), Lindsey et al. (1990), Miller (1994); 5Lund et al. (2003); 6McCandless (1982), Evans & Green (2003), Lund et al. (2010); 7Crittenden et al. (1971, 1983), Link et al. (1993), Christie-Blick (1997); 8Crittenden et al. (1971,
1983), Link et al. (1993).
446 K. LUND ET AL.
not only the age of diamictite-bearing rocks in Idaho but also those
of other diamictite units along the Cordillera.
Revising the age of the diamictite-bearing deposits in the north-
ern US Cordillera from 750– 700 Ma to c. 685– 665 Ma may affect
palaeolatitude reconstructions and global climate change models.
Park (1997) used 723 Ma sills thought to be correlative with the
glaciogenic Rapitan Group in the northern Canadian Cordillera
to develop the palaeopole for Laurentia at 723 Ma. From these
data, he interpreted a low palaeolatitude for this older Cryogenian
glaciation (see discussion in Lund et al. 2003). However, direct age
constraints for the Cryogenian deposits in Idaho (Lund et al. 2003;
Fanning & Link 2004, 2008) indicate that the 723 Ma palaeopole
for Laurentia would be too old for time of formation of the
Idaho diamictites and thus would not provide an accurate palaeo-
latitude for that glaciation. Additionally, the palaeogeography for
the glaciogenic deposits needs to be reinterpreted due to the differ-
ence in supercontinent configurations at 750 Ma compared to
685 Ma. Ultimately, there is a broad range of possible ages for dia-
mictites in the Cordillera. Constraints are best for diamictites of
685 to 665 Ma, but some less direct constraints may extend the
maximum age to c. 720 Ma. As such, models for Neoproterozoic
palaeogeography, isotopic composition of seawater curves, and
global glaciations (Snowball Earth events) that are partly based
on Cordilleran data must be used with care.
An early version of this manuscript was improved by comments from D. I.
Bleiwas. Comments and suggestions by reviewers P. K. Link and M. Colpron
and suggestions and guidance from volume editor E. Arnaud are greatly appreci-
ated. This represents a contribution of the IUGS- and UNESCO-funded IGCP
(International Geoscience Programme) project #512.
References
Aitken, J. D. 1991. The Ice Brook Formation and post-Rapitan, Late Pro-
terozoic glaciation, Mackenzie Mountains, Northwest Territories.
Geological Survey of Canada Bulletin 404, 43.
Christie-Blick, N. 1997. Neoproterozoic sedimentation and tectonics in
west-central Utah. In: Link, P. K. & Kowallis, B. J. (eds) Brigham
Young University Geology Studies. Salt Lake City, Utah, 1 –30.
Christie-Blick, N. & Levy, M. 1989. Stratigraphic and tectonic frame-
work of Upper Proterozoic and Cambrian rocks in the western
United States. In: Christie-Blick, N. & Levy, M. (eds) Late Proter-
ozoic and Cambrian Tectonics, Sedimentation, and Record of
Metazoan Radiation in the Western United States. American
Geophysical Union, Washington, DC, 28th International Geological
Congress, Field Trip Guidebook T331, 7 –21.
Colpron, M., Logan, J. M. & Mortensen, J. K. 2002. U –Pb zircon age
constraint for late Neoproterozoic rifting and initiation of the lower
Paleozoic passive margin of western Laurentia. Canadian Journal
of Earth Sciences, 39, 133– 143.
Corsetti, F. A., Link, P. K. & Lorentz, N. J. 2007. d13C chemostratigra-
phy of the Neoproterozoic succession near Pocatello, Idaho, USA:
implications for glacial chronology and regional corellations. In:
Link, P. K. & Lewis, R. S. (eds) Proterozoic Geology of Western
North America and Siberia, Belt Symposium IV, Society of Econ-
omic Paleontologists and Mineralogists, Special Publications, 86,
193– 205.
Crittenden, M. D. Jr, Schaefer, F. E., Trimble, D. E. & Woodward,
L. A. 1971. Nomenclature and correlation of some Upper Precam-
brian and basal Cambrian sequences in western Utah and southeastern
Idaho. Geological Society of America Bulletin, 82, 581– 602.
Crittenden, M. D. Jr, Stewart, J. H. & Wallace, C. A. 1972. Regional
correlation of Upper Precambrian strata in western North America.
24th International Geological Congress, 1972, 1, 334–341.
Crittenden, M. D. Jr & Wallace, C. A. 1973. Possible equivalents of
the Belt Supergroup in Utah. Belt Symposium 1973 Volume 1,
Idaho Bureau of Mines and Geology, Moscow, Idaho, 116– 138.
Crittenden, M. D. Jr, Christie-Blick, N. & Link, P. K. 1983. Evidence
for two pulses of glaciation during the late Proterozoic in northern
Utah and southeastern Idaho. Geological Society of America Bulletin,
94, 437– 450.
Devlin, W. J. 1989. Stratigraphy and sedimentology of the Hamill Group
in the northern Selkirk Mountains, British Columbia: evidence for
latest Proterozoic – Early Cambrian extensional tectonism. Canadian
Journal of Earth Sciences, 26, 515–533.
Devlin, W. J. & Bond, G. C. 1988. The initiation of the early Paleozoic
Cordilleran miogeocline: evidence from the uppermost Proterozoic –
Lower Cambrian Hamill Group of southeastern British Columbia.
Canadian Journal of Earth Sciences, 25, 1 –19.
Evans, J. G. 1987. Geology of the Stensgar Mountain quadrangle, Stevens
County, Washington. US Geological Survey Bulletin, 1679, 23.
Evans, K. V. & Green, G. N. 2003. Geologic Map of the Salmon National
Forest and vicinity, east-central Idaho. US Geological Survey Geolo-
gic Investigations, I-2765.
Fanning, C. M. & Link, P. K. 2004. U–Pb SHRIMP ages of Neopro-
terozoic (Sturtian) glaciogenic Pocatello Formation, southeastern
Idaho. Geology, 32, 881– 884.
Fanning, C. M. & Link, P. K. 2008. Age constraints for the Sturtian gla-
ciation; data from the Adelaide geosyncline, South Australia and
Pocatello Formation, Idaho, USA. Selwyn Symposium 2008, Neo-
proterozoic climates and origin of early life. Geological Society of
Australia, Abstracts, 91, 57 – 62.
Fisher, F. S., McIntyre, D. H. & Johnson, K. M. 1992. Geologic Map of
the Challis 1oX2o quadrangle, Idaho. US Geological Survey Miscel-
laneous Investigation Series Map I-1819.
Gabrielse, H. 1972. Younger Precambrian of the Canadian Cordillera.
American Journal of Science, 272, 521–536.
Harper, G. D. & Link, P. K. 1986. Geochemistry of Upper Proterozoic
rift-related volcanics, northern Utah and southeastern Idaho.
Geology, 14, 864–867.
Hein, F. J. & McMechan, M. E. 1994. Proterozoic– Lower Cambrian
strata of the western Canada sedimentary basin. In: Mossop, G. D.
& Shetsen, I. (eds) Geological Atlas of the Western Canada Sedi-
mentary Basin. Canadian Society of Petroleum Geologists and
Alberta Research Council, 47 –57.
Hobbs, S. W. & Hays, W. H. 1990. Ordovician and older rocks of the
Bayhorse area, Custer County, Idaho. US Geological Survey Bulletin,
1891, 40.
Hobbs, S. W., Hays, W. H. & McIntyre, D. H. 1991. Geologic Map of the
Bayhorse area, central Custer County, Idaho. US Geological Survey
Miscellaneous Geologic Investigations Map I-1882.
Hoffman, P. F. & Halverson, G. P. 2011. Neoproterozoic glacial record
in the Mackenzie Mountains, northern Canadian Cordillera. In:
Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geo-
logical Record of Neoproterozoic Glaciations. Geological Society,
London, Memoirs, 36, 397–411.
Jacob, T. 1990. Late Proterozoic tuff near Challis, Idaho. In: Moye, F. J.
(ed.) Geology and Ore Deposits of the Trans-Challis Fault System/
Great Falls Tectonic Zone. Guidebook of the Fifteenth Annual
Tobacco Root Geological Society field Conference, 97 – 106.
Kendall, B. S., Creaser, R. A., Ross, G. M. & Selby, D. 2004. Con-
straints on the timing of Marinoan ‘Snowball Earth’ glaciation by
187
Re – 187Os dating of a Neoproterozoic, post-glacial black shale in
western Canada. Earth and Planetary Science Letters, 222, 729– 740.
Leonard, B. F. 1962. Old metavolcanic rocks of the Big Creek area,
central Idaho, in Short papers in geology, hydrology and topography.
US Geological Survey Professional Paper 450B, B11 –B15.
Levy, M. & Christie-Blick, N. 1991. Late Proterozoic paleogeography
of the eastern Great Basin. In: Cooper, J. D. & Stevens, C. H.
(eds) Paleozoic Paleogeography of the Western United States – II.
Pacific Section, Society of Economic Paleontologists and Mineralo-
gists, Los Angeles, California, 371– 386.
Levy, M., Christie-Blick, N. & Link, P. K. 1994. Neoproterozoic
incised valleys of the eastern Great Basin, Utah and Idaho: fluvial
response to changes in depositional base level. In: Dalrymple,
R. W., Boyd, R. & Zaitlin, B. A. (eds) Incised-Valley Systems:
Origin and Sedimentary Sequences. Society for Sedimentary
Geology, Tulsa, OK, 369–382.
Lindsey, K. A., Gaylord, D. R. & Groffman, L. R. 1990. Geology of the
Upper Proterozoic to Lower Cambrian Three Sisters Formation,
Gypsy Quartzite, and Addy Quartzite, Stevens and Pend Oreille
 THE EDWARDSBURG FORMATION
counties, Washington, Washington Division of Geology and Earth
Resources Report of Investigations, 30, 37.
Link, P. K. & Christie-Blick, N. 2011. Neoproterozoic strata of South-
eastern Idaho and Utah: record of Cryogenian rifting and glaciation.
In: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The
Geological Record of Neoproterozoic Glaciations. Geological
Society, London, Memoirs, 36, 425–436.
Link, P. K., Christie-Blick, N. et al. 1993. Middle and Late Proterozoic
stratified rocks of the western US Cordillera, Colorado Plateau, and
Basin and Range province. In: Reed, J. C. Jr, Bickford, M. E.,
Houston, R. S., Link, P. K., Rankin, D. W., Sims, P. K. & Van
Schmus, W. R. (eds) Precambrian: Conterminous US. Geological
Society of America, Boulder, Colorado, The Geology of North
America, 463– 595.
Lund, K. 2004. Geology of the Payette National Forest and vicinity,
west-central Idaho. US Geological Survey Professional Paper,
1666, 89.
Lund, K. 2008. Geometry of the Neoproterozoic and Paleozoic rift
Margin of western Laurentia: implications for mineral deposit
settings. Geosphere, 4, 429– 444.
Lund, K., Aleinikoff, J. N., Evans, K. V. & Fanning, C. M. 2003.
SHRIMP U–Pb geochronology of Neoproterozoic Windermere
Supergroup, central Idaho: implications for regional synchroneity
of Sturtian glaciation and associated rifting. Geological Society of
America Bulletin, 115, 349–372.
Lund, K., Aleinikoff, J. N., Evans, K. V., duBray, E. A., Dewitt, E. H.
& Unruh, D. M. 2010. SHRIMP U–Pb dating of recurrent Cryogen-
ian and Late Cambrian – Early Ordovician alkalic Magmatism in
central Idaho: implications for Rodinian rift tectonics. Geological
Society of America Bulletin, 122, 430– 453.
McCandless, D. O. 1982. A reevaluation of Cambrian through Middle
Ordovician stratigraphy of the southern Lemhi Range. Master’s thesis,
The Pennsylvania State University, University Park, Pennsylvania, 157.
Miller, J. M. G. 1985. Glacial and syntectonic sedimentation: the upper
Proterozoic Kingston Peak Formation, southern Panamint Range,
eastern California. Geological Society of America Bulletin, 96,
1537– 1553.
Miller, F. K. 1994. The Windermere Group and Late Proterozoic tec-
tonics in northeastern Washington and northern Idaho. Washington
Division of Geology and Earth Resources, 1 –19.
Miller, F. K. & Whipple, J. W. 1989. The Deer Trail Group – is it part of
the Belt Supergroup. In: Joseph, N. L. (ed.) Geologic Guidebook for
447
Washington and Adjacent Areas. Washington Division of Geology
and Earth Resources, 1– 21.
Mrofka, D. & Kennedy, M. 2011. The Kingston Peak Formation in
the eastern Death Valley Region. In: Arnaud, E., Halverson,
G. P. & Shields-Zhou, G. (eds) The Geological Record of Neo-
proterozoic Glaciations. Geological Society, London, Memoirs, 36,
449– 458.
Oliver, D. H. & Blackwell, D. D. 2002. Neoproterozoic diamictites in
central Idaho: evidence of felsic volcanism from embayed quartz
grains. Geological Society of America Abstracts with Programs, 34,
89– 90.
Park, J. K. 1997. Paleomagnetic evidence for low-latitude glaciation
during deposition of the Neoproterozoic Rapitan Group, Mackenzie
Mountains, N.W.T., Canada. Canadian Journal of Earth Science,
34, 34– 49.
Petterson, R., Prave, A. R. & Wernicke, B. P. 2011. Glaciogenic and
related strata of the Neoproterozoic Kingston Peak Formation in the
Panamint Range, Death Valley region, California. In: Arnaud, E.,
Halverson, G. P. & Shields-Zhou, G. (eds) The Geological
Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 459–465.
Rainbird, R. H., Jefferson, C. W. & Young, G. M. 1996. The early
Neoproterozoic sedimentary succession B of northwestern Laurentia:
correlations and paleogeographic significance. Geological Society
of America Bulletin, 108, 454–470.
Ross, G. M., Bloch, J. D. & Krouse, H. R. 1995. Neoproterozoic strata
of the southern Canadian Cordillera and the isotopic evolution of
seawater sulfate. Precambrian Research, 73, 71 –99.
Smith, L. H., Kaufman, A. J., Knoll, A. H. & Link, P. K. 1994.
Chemostratigrapy of predominantly siliciclastic Neoproterozoic
successions: a case study of the Pocatello Formation and
lower Brigham Group, Idaho, USA. Geological Magazine, 131,
301– 314.
Smith, M. D., Arnaud, E., Arnott, R. W. C. & Ross, G. M. 2011. The
record of Neoproterozoic glaciation in the Windermere Supergroup,
southern Canadian Cordillera. In: Arnaud, E., Halverson, G. P.
& Shields-Zhou, G. A. (eds) The Geological Record of Neopro-
terozoic Glaciations. Geological Society, London, Memoirs, 36,
413– 424.
Stewart, J. H. 1972. Initial deposits in the Cordilleran geosyncline;
evidence of a Late Precambrian (,850 m.y.) continental separation.
Geological Society of America Bulletin, 83, 1345– 1360.
 Chapter 40

The Kingston Peak Formation in the eastern Death Valley region

DAVID MROFKA* & MARTIN KENNEDY

1
Department of Earth Sciences, University of California Riverside, Riverside, CA 92521, USA
*Corresponding author (e-mail: mrofkd01@ucr.edu)

Abstract: The late Neoproterozoic Kingston Peak Formation (Fm.) is a several-kilometre-thick sedimentary succession primarily influ-
enced by syndepositional tectonism and located in the region around Death Valley, California (Fig. 40.1). Its distribution is divisible into
an eastern facies assemblage, the subject of this paper, and a western facies assemblage covered in a separate chapter. The diamictite-
bearing Kingston Peak Fm. is bounded by the underlying shallow platform carbonates of the Beck Spring Fm. and overlain by the
Noonday Dolomite. There is an absence of direct palaeolatitude or radiometric age constraints and any correlation is based on broad
similarities with other coarse-grained strata (diamictite) located in a northward trending belt along the Cordilleran miogeocline. The
overlying Noonday Dolomite has been interpreted to be a late Cryogenian ‘cap carbonate’ and shares a set of unique facies associations
and isotopic and lithological characteristics with other late Neoproterozoic post-glacial carbonate intervals in Namibia, Canada, Australia
and Brazil. Research to date has focused on understanding local basin evolution, glacial sedimentology, correlation between the eastern
and western facies assemblages and initiation and development of the North American Cordillera. The intimate association of tectonic
and glacial facies with rapid local thickness and facies changes corresponding with syn-sedimentary faulting is the most distinctive strati-
graphic characteristic of the Kingston Peak Fm. The complex local stratigraphy complicates correlation both within the Death Valley
region as well as globally, and pending absolute age dates, does not fit easily with conventional Cryogenian Period glacial models
identifying two or more discrete ice ages.

The Kingston Peak Fm. (KPF) is the uppermost of the three for-
mations in the Pahrump Group (Hewett 1940) (Fig. 40.1), which
comprises the oldest sedimentary rocks preserved in the region.
It crops out throughout the Death Valley region of southeastern
California, where active extensional tectonism associated with
the Basin and Range province along with an arid climate allow
clear lateral stratigraphic and facies relations to be observed. The
mixed carbonate–siliciclastic sediments of the Crystal Spring
Fm. and overlying shallow marine carbonate of the Beck Spring
Dolomite comprise the basal and middle formations, respectively,
of the Pahrump Group. The KPF is overlain by the Noonday dolo-
mite (Fig. 40.1)
Description of the KPF is complicated by its division into a
distinctive western facies assemblage (Petterson et al. 2011;
western-KPF) along the western boundary of Death Valley in the
Panamint Range and a distinctive eastern facies assemblage (this
chapter; eastern-KPF) that itself is subdivided into northern and
southern facies (Fig. 40.1). The KPF also crops out in the
Funeral Mountains further north (Miller 1983) but has undergone
amphibolite-grade metamorphism (Mattinson et al. 2007) and is
not well-studied. Eastern and western facies assemblages of the
KPF are dominated by similar coarse-grained siliciclastic rocks,
but lithostratigraphic correlation is complicated by lateral facies
changes within each facies assemblage (Miller 1983), an overall
difference in the appearance of specific facies between both
regions and the presence in each region of distinctive carbonate
intervals bounded by dissimilar facies.
The eastern facies assemblage (also referred to as southeastern
KPF in the literature) crops out in the southern Black Mountains
and in a number of hills and ranges SE of the Black Mountains,
primarily in the Kingston Range (Fig. 40.1). More specifically, it
crops out extensively in a readily accessible 30-km-long belt
along the northern and eastern flanks of the Kingston Range
(Hewett 1940) and is superbly exposed in a panel extending
from the Silver Rule Mine (358480 1700 N, 1158560 3900 W, #6 in
Fig. 40.1) to Beck Canyon Divide (358480 1900 N, 1158550 3100 W,
#7 in Fig. 40.1). The northern Kingston Range is the location of
three characteristic KPF sections published in Hewett (1956;
Fig. 40.1). It is the best location to examine the eastern-KPF
because of a relative lack of metamorphism, laterally persistent
outcrops and visibility of rapid lateral facies changes.
Much of the interval of coarse-grained sediments comprising the
KPF was first described in detail from the Panamint Range and
informally named in the Telescope group by Murphy (1930,
1932) as the Surprise Fm., Sourdough Limestone, Middle Park
Fm. and Wildrose Fm., from bottom to top respectively. These
names have been retained but subsequently assigned to formal
member status (Johnson 1957; Carlisle et al. 1980; Labotka
et al. 1980) within Hewett’s (1940) Kingston Peak Fm.
(Fig. 40.1). Noble (1934, 1941) described the eastern-KPF and
likened it to the ‘Algonkian’ series in the Grand Canyon,
suggesting a correlation to Murphy’s (1932) Telescope group
(later the KPF) in the Panamint Range. Hazzard (1939) interpreted
the KPF to be glaciogenic based on the presence of striated clasts at
the Gunsight Mine in the uppermost KPF.
From 1950 to the mid-1980s, relevant publications focused on
the stratigraphy, sedimentology, palaeogeography and source
regions for the different facies of the KPF (Wright 1952, 1968;
Wright & Troxel 1966; Troxel 1967, 1982b; Wright et al. 1976,
1978, 1984). From the early 1980s onwards, publications primarily
addressed the glacial and tectonic features in the eastern and
western facies assemblages (Miller 1982, 1983, 1985, 1987;
Christie-Blick & Levy 1989; Link et al. 1993), providing interpret-
ations for the role of glaciation and extension in the deposition of
the KPF as well as using the succession to help interpret the evol-
ution of the developing passive margin of the western Cordillera
(Levy & Christie-Blick 1989, 1991; Fedo & Cooper 2001).
Prave (1999) proposed a chemo- and tectono-stratigraphic corre-
lation for the eastern and western facies assemblages of the KPF,
suggesting the entire formation fit into a ‘Snowball Earth’ model
(Hoffman et al. 1998) with a stratigraphic record of two discrete
ice ages. Most recently, Mrofka (2010) discussed the usefulness
of carbonate intervals as timelines and the important relationship
between syndepositional tectonism and preservation of the
climate record in both the eastern- and western-KPF.
Structural framework
The KPF is concentrated in extensional basins, products of syn-
depositional tectonism (Mrofka 2010) likely associated with

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 449– 458. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.40
450 D. MROFKA & M. KENNEDY
rifting along the western margin of North America between 850
and 600 Ma (Stewart 1972; Christie-Blick & Levy 1989). During
extension and uplift along normal faults, resulting basins were
filled with sediments derived from successively older stratigraphic
levels of the underlying Pahrump Group (Wright et al. 1976, 1992;
Burchfiel et al. 1992). Evidence for extension includes a prepon-
derance of coarse-grained sediments including kilometre-scale
olistoliths (Miller 1985), syndepositional normal faults in the
Kingston Range (Mrofka 2010), the southern Nopah Range
(Wright et al. 1976, 1978) and the Panamint Range (Prave 1999)
and tholeiitic (Hammond 1983) pillow lavas in the Panamint
Range (Miller 1985). Vertical offsets on the faults are .400 m
in the Kingston Range at Jupiter Hills (Fig. 40.1), as demonstrated
by removal of the entire Beck Spring Dolomite on the footwall side
of faults, and possibly the removel of as much as 3 km of rock
elsewhere (Burchfiel et al. 1992).
Sedimentation in the Death Valley region was relatively con-
tinuous throughout the Palaeozoic (Abolins et al. 2000) but under-
went compressional shortening in the Permian (Snow 1992) and
again during the Mesozoic (Levy & Christie-Blick 1989), with
accompanying metamorphism in the Panamint Range
(western-KPF) (Labotka et al. 1980). In the Kingston Range,
north-northeastwards tilting strata of the eastern-KPF are generally
offset along north –south trending normal faults and, in the
southern Kingston Range, a detachment system occurs related to
Fig. 40.1. (a) Map showing regional
distribution of the western (white; see
Petterson et al. 2011 for more detail) and
eastern (northern, light-shaded; southern,
dark-shaded) facies assemblages of the
Kingston Peak Fm. Numbers represent
measured sections used for this study: 1,
Alexander Hills; 2, Gunsight Mine; 3, War
Eagle Mine; 4, Beck Canyon West; 5,
Crystal Spring; 6, Silver Rule Mine; 7, Beck
Canyon Divide; 8, Horsethief Spring; 9,
Kingston North; 10, Jupiter Mine; 11, Snow
White Mine; 12, Horsethief Mine; 13,
Mesquite North; 14, Mesquite South; 15,
Mesquite Small Block; 16, Winters Pass;
17, Southern Valjean Hills; 18, Sperry
Hills; 19, Saddle Peak Hills; 20, Saratoga
Hills; 21, Ibex Hills; 22, Eclipse Mine;
23, Virgin Spring Wash south; 24, Virgin
Spring Wash north; 25, Silurian Hills; 26,
southern Saddle Peak Hills; 27, Galena
Canyon; 28, Goler Wash; 29, Wood
Canyon; 30, Sourdough Canyon. Type
sections from Hewett (1956) are #7, #8 and
3 km south of #8. (b) Generalized
stratigraphy of the Pahrump Group showing
predominance of coarse-grained facies in
the Kingston Peak Fm. relative to the
bounding carbonate strata. (c) Schematic
stratigraphic logs comparing the different
facies assemblages of the Kingston Peak
Fm. across the Death Valley region.
Cenozoic extension of the Basin and Range province (Davis
et al. 1993). Levy & Christie-Blick (1989) estimated c. 150%
extension to the original pre-Mesozoic basin after Mesozoic com-
pression and later Cenozoic extension. Estimates for Cenozoic
extension range from c. 50–500% (Wright & Troxel 1967;
Wernicke et al. 1988; Miller 1991), with higher estimates recently
called into question by the findings of Renik et al. (2008). At its
western boundary, in the southern Black Mountains, the eastern-
KPF has undergone significant structural complication (Noble
1941; Troxel & Wright 1987; Miller 1991).
Stratigraphy
Underlying Pahrump Group
The lower and middle formations in the Pahrump Group
(Fig. 40.1b) record a transition from mixed carbonate –siliciclastic
marine and fluvial facies of the Crystal Spring Fm. (Roberts 1974)
to the Beck Spring Dolomite, comprised of shallow-water carbon-
ates in the north and mixed carbonate –siliciclastic fluvial –tidal
deposits south of the central Saddle Peak Hills (Marian &
Osborne 1992). In the Kingston Range, Marion & Osborne
(1992) divided the Beck Spring Dolomite into (i) a lower, lami-
nated, cherty member, (ii) a laminated member with angular
 KINGSTON PEAK FORMATION, EASTERN DEATH VALLEY
intraclasts and columnar stromatolites, (iii) a relatively thinner
oolitic –pisolitic member and (iv) a partially silicified upper
member with abundant chert, shale lenses and stromatolites.
Kingston Peak Fm.: Eastern Facies Assemblage
The coarse-grained siliciclastic eastern-KPF is separated into a
northern and southern facies based primarily on the distinct lithol-
ogy of clasts in diamictite from each area (Troxel 1967). The north-
ern facies of the eastern-KPF is the focus of this contribution and is
informally separated into the KP1 through KP4 members (Wright
1974) (Fig. 40.1a, c). The southern facies is limited in outcrop rela-
tive to the northern facies. Other prominent units within the KPF
include (i) a black laminated karstic limestone unit (Virgin
Spring Limestone) that locally separates the KP1 and KP2
members, (ii) a discontinuous dolostone bed in the lower half of
the KP3 Member in the Kingston Range (Corsetti & Kaufman
2003) and (iii) lenses of diamictite that often overlie the KP4
Member (referred to informally as the Gunsight Member).
The basal KP1 and Virgin Spring Limestone of the northern
facies are separated from the glaciogenic units of KP2 through
KP4 by a regional unconformity (Fig. 40.1c). Based on strati-
graphic and sedimentary characteristics, these two basal units
should constitute separate formations, but this has only come to
light recently with the identification of the regional unconformity
(Mrofka 2010). The current stratigraphic scheme (Fig. 40.1c) and
member names are retained here until formal stratigraphic nomen-
clature is published elsewhere.
Noonday Dolomite
The Noonday Dolomite overlying the eastern-KPF is divided into a
lower, cream-coloured, laminated, microbial, dolomite member
and an upper, laminated, silty, dolomite member (Wright et al.
1978). The laminations at the base several metres of the lower
member are parallel and horizontal, transitioning upwards to
arching laminations that define larger-scale microbial mounds
with synoptic relief of up to 200 m (Williams et al. 1974). Alterna-
tively, Summa (1993) suggested apparent mound ‘topography’
was due to an intra-formational erosion surface. The lower
member contains distinctive vertical tubes possibly related to ver-
tical transport of fluids (Cloud et al. 1974; Kennedy et al. 2001b) or
microbial processes (Corsetti & Grotzinger 2005) and centimetre-
to decimetre-scale pockets of sparry cement (Cloud et al. 1974;
Williams et al. 1974). In the east, the Noonday Dolomite tran-
sitions southwards to a siliciclastic-rich facies (Ibex Fm.), which
includes an arkosic siltstone member, a shaley-limestone
member and a quartz-dolomite member (Williams et al. 1974).
Glaciogenic deposits and associated strata
Glaciogenic strata occur in the KP2 –KP4 members of the
Kingston Peak Fm. as well as in the southern facies assemblage
(Fig. 40.1c). Deposition of the basal KP1 Member and overlying
Virgin Spring limestone was not influenced by glacial processes
and both are therefore described subsequently within this section
as associated strata.
KP2 Member, northern facies: regional unconformity and
diamictite deposition
The KP2 Member comprises a regionally extensive blanket of
massive- to diffusely-bedded cobble-boulder diamictite, which
varies from 10 to 250 m in thickness. It sharply overlies a region-
ally extensive unconformity (Mrofka 2010), which defines the top
451
of the underlying KP1 Member or Virgin Spring limestone and
contains striated and faceted clasts. Clasts within the diamictite
are derived from the underlying Pahrump Group or basement
and the matrix is composed of coarse angular quartz sand and
illite (mica) or chlorite. Basal diamictite commonly contains
black limestone clasts and a carbonate-rich matrix likely derived
from the underlying Virgin Spring limestone. In the Saratoga
Hills, southern Saddle Peak Hills and Alexander Hills, the diamic-
tite facies is interrupted by a 5– 20 m interval of finer-grained
facies, variably including siltstone and sandstone with parallel
laminations, trough cross-bedding, steep bimodal cross-lamination
and normally graded pebble conglomerates with sandy tops.
KP3 Member, northern facies: sandstone, breccia and
conglomerate
The KP3 Member is 15–2000 m thick, consists of interbedded silt-
stone and sandstone, diamictite with striated clasts, normally
graded conglomerate beds, kilometre-scale olistoliths and channel-
filling sedimentary breccia. The lower KP3 Member is comprised
primarily of siltstone and sandstone interbedded with minor dia-
mictite and conglomerate. Pebble- to cobble-sized outsized clasts
commonly float in sandstone beds that grade laterally to conglom-
erate or diamictite. Sedimentary structures include normally
graded beds, convolute laminations and siltstone with intraclasts
and flame structures. A 2– 3-m-thick oncolitic dolostone in the
Kingston Range (358460 2700 N, 1158520 5900 W) is found near the
top of the finer-grained lower KP3 Member and marks a
coarsening-upwards transition to interbedded sandstone, normally
graded conglomerate and diamictite. As with the KP2 Member,
diamictite intervals commonly contain black limestone clasts in
a black calcitic matrix. The middle to upper KP3 Member is
characterized by metre- to kilometre-scale mega-clasts and olisto-
liths of the underlying Pahrump Group that form prominent ridges
in the Kingston Range around 358440 4300 N, 1158510 500 W and
358440 3500 N, 1158500 2200 W (Wright et al. 1976). Fe-rich (45%
iron by weight) units at Sperry Wash (Abolins et al. 2000) are
found near Tertiary-age faulting and volcanic intrusions.
KP4 Member, northern facies: sedimentary breccia and
conglomerate
The KP4 Member gradationally overlies the KP3 Member and
comprises 200– 1300 m of conglomerate, sedimentary breccia
and monomictic mega-breccia (Figs 40.1 & 40.2). Conglomerate
and breccia beds are commonly normally graded and fill chan-
nels 1–2 m deep and 10–50 m wide. Mega-breccia is massive,
dominated by up to metre-scale angular blocks of Beck Spring
Dolomite and filling steeper and narrower channels than the
graded breccia and conglomerate beds. In the Kingston Range,
sedimentary breccia composed entirely of Beck Spring Dolomite
was deposited adjacent to a syndepositional fault (358470 2200 N,
1158500 100 W, #10 in Fig. 40.1). Laterally, sedimentary breccia
and conglomerate is interbedded with normally graded to
massive sandstone beds with sharp planar lower bed contacts. In
footwall sections, the KPF is composed entirely of a diamictite
interval informally named the Gunsight Member (Troxel pers.
comm.). Near the Jupiter Mine (358470 2800 N, 1158500 100 W,
#10 in Fig. 40.1), the Gunsight Member includes channelized
sandstone with mudcracks and ripples overlain by diamictite
with striated clasts.
Glaciogenic deposits of the southern facies
Outcrops of the southern facies of the eastern-KPF (dark grey
shading in Fig. 40.1) are limited to a c. 8-km-wide belt of quartzite,
452 D. MROFKA & M. KENNEDY
conglomerate and diamictite that crops out in the southern Saddle
Peak Hills, the southern Salt Spring Hills and in the Silurian Hills.
In the southern Salt Spring Hills, the southern facies is 1000 m
thick and dominantly composed of quartzite with c. 75 m of base-
ment clast-bearing diamictite capping the lower third of the section
(Troxel 1967). In the Silurian Hills, the eastern-KPF is .2000 m
thick (Kupfer 1960) and is floored by a dark parallel-laminated
karsted limestone. Kupfer (1960) correlated this limestone to the
Beck Spring Dolomite, but Prave (1999) correlated it to a discon-
tinuous limestone below the Wildrose sub-Member in the western-
KPF. Prave’s (1999) correlation was based on similarities in d13C
values and the appearance of a quartzite cobble conglomerate
below both limestone intervals. The section above the limestone
coarsens upwards and contains normally graded sandstone and
conglomerate, diamictite and megaclasts (tens of metres scale).
Similarities between the northern and southern facies include
karsted and laminated limestone facies overlain by sandstone
beds, a coarsening-upwards trend above the karsted limestone
intervals, and the presence of megaclasts. These similarities
Fig. 40.2. (a) Cross-section of four
measured sections showing footwall to
hanging wall transition and relationship
between northern and southern facies.
Numbers next to section names represent
approximate section thicknesses. Note:
different vertical scales are used. Northern
facies sections rest on the Beck Spring
Dolomite, except at the War Eagle mine,
where the Gunsight Member rests on the
Crystal Spring Fm. Uppermost unit shown
is the Noonday Dolomite. (b) Map showing
location of measured sections: 4, Beck
Canyon West; 7, Beck Canyon Divide;
12, Horsethief Mine; 25, Silurian Hills.
(c) Schematic diagram showing typical
extensional geometry of graben deposits
thought to apply to the eastern-KPF (after
Faerseth et al. 1997).
suggest the limestone in the Silurian Hills may alternatively be cor-
related to the Virgin Spring Limestone in the northern facies.
Unlike northern facies diamictite, the diamictite in the southern
facies is dominated by clasts of granite and gneiss (Troxel
1967). Troxel (1982a) suggested this diamictite is interbedded
with the northern facies in the southern Saddle Peak Hills.
KP1 Member, northern facies: sandstone and siltstone
The non-glaciogenic KP1 Member is 1 to 180 m thick, composed
of centimetre-scale beds of parallel laminated sandstone and silt-
stone and underlies a regional erosional unconformity. Sedimen-
tary structures include low-angle cross-lamination, beds with
scoured bases, rare massive sandstone beds with mudstone chips
and a general coarsening and increase in carbonate cement upsec-
tion. In the southern Black Mountain (358540 4500 N, 1168380 5000 W,
#23 and 24 in Fig. 40.1), the KP1 Member varies in thickness by
c. 30 m over a lateral distance of 100 m due to erosional truncation.
 KINGSTON PEAK FORMATION, EASTERN DEATH VALLEY 453

Virgin Spring limestone, northern facies Field studies by the authors (see also Mrofka 2010) provide evi-
dence for uninterrupted deposition beginning at the base of the
The Virgin Spring limestone (Tucker 1986) sharply overlies the KP2 Member and continuing through the Noonday Dolomite,
KP1 Member, is erosionally truncated and karsted, dark, parallel- demonstrated by the following four sedimentary relationships.
laminae, and is preserved in only three localities. The limestone is First, KP4 Member sedimentary breccias in the Alexander Hills
17 m thick and best exposed in the Ibex Hills (358450 1800 N, are interbedded with the basal Noonday Dolomite (358450 5600 N,
1168260 1200 W, #21 in Fig. 40.1), but also crops out at Virgin 116860 5500 W). Second, in footwall sections the contact between
Spring Wash and in the Saratoga Hills where it is ,4 m thick the Noonday Dolomite and underlying strata is commonly inter-
and gradually truncated to the south (Fig. 40.1). Tucker (1986) rupted by a 1– 10 m layer of Gunsight Member diamictite; a
described the Virgin Spring limestone as comprising of similar diamictite interval appears conformable with the under-
centimetre- to decimetre-scale beds of parallel laminated lime- lying KP4 Member in hanging wall sections. Third, in the southern
stone interbedded with ,1-mm-thick sandstone laminae with Valjean Hills (358390 4000 N, 116870 2200 W, #17 in Fig. 40.1) and in
scoured bases and occasional normal grading. Petrographic and the Ibex Hills (DeYoung pers. comm.), Noonday Dolomite clasts
sedimentary features include ooids and convoluted or overturned are included in diamictite of the KP4 member or are in diamictite
beds attributed to mass sediment movement down the palaeoslope interbedded with KP4 Member sedimentary breccia. Alternatively,
(Tucker 1986). Corsetti & Kaufman (2005) interpreted this Ibex Hills interbedded
diamictite to post-date KPF deposition. Fourth, the base of the
Noonday Dolomite commonly contains clasts from the Beck
Boundary relations with overlying and underlying Spring Dolomite and Crystal Spring Fm. (i.e. 358450 4600 N,
non-glacial units 116860 5000 W). This relationship is consistently observed when
the Noonday Dolomite overlies the KP4 Member. Furthermore,
Contact with underlying units the Beck Spring Dolomite or Crystal Spring Fm. has not been
found in direct contact with the Noonday Dolomite. This likely
The contact underlying the glaciogenic interval of the KPF is indicates that during incipient Noonday Dolomite deposition,
complex and does not represent a single timeline, as it is located loose clasts from the surface of an unlithified KP4 Member were
above both an older and younger unconformity. The older uncon- reworked along with carbonate material from the flanks of
formity is preserved in hanging wall sections and separates the Noonday Dolomite mounds. Alternatively, the Pahrump Group
KP2 Member from a beveled contact with either the Virgin may have been exposed during Noonday Dolomite deposition,
Spring limestone or the underlying KP1 Member. The younger serving as a source of clasts in the basal Noonday Dolomite, and
unconformity occurs in footwall sections and variably separates contacts between the two are simply not exposed.
the KP4 Member from a beveled contact with either of the two
underlying formations of the Pahrump Group (Beck Springs
Dolomite or Crystal Springs Fm.) or the granitic basement. This Chemostratigraphy
younger unconformity developed due to erosion from uplift of
footwall blocks, which removed the older unconformity as well The most prominent carbonate units within the KPF succession are
as variable intervals of the Pahrump Group, down to the underlying the Virgin Spring limestone and the KP3 Member oncolitic dolo-
basement. stone bed in the northern facies, the Silurian Hills limestone in
In hanging wall sections where the contact between the non- the southern facies and the Sourdough Limestone in the
glacial units of the eastern-KPF and the underlying Beck Spring western-KPF (Fig. 40.1). Statistics for published isotopic data
Dolomite is preserved, it has been described as conformable, inter- for prominent carbonate units within and bounding the KPF are
fingering or unconformable (Christie-Blick & Levy 1989) and listed in Table 40.1 (Tucker 1983, 1986; Bergfeld et al. 1996;
shows no evidence of erosional truncation, although Kenny & Kennedy et al. 2001a; Corsetti & Kaufman 2003). Eastern-KPF
Knauth (2001) describe karstification of the upper Beck Spring carbonates have relatively enriched d13C and depleted d18O
Dolomite in some localities. In the Alexander Hills and Saratoga values, and data from published stratigraphic profiles (Tucker
Hills, the KP1 Member is described as transitional with the top
of the Beck Spring Dolomite over 10 m (Wright et al. 1992).
This relationship can be seen in the Alexander Hills (358460 200 N,
Table 40.1. Range of C- and O-isotopic values for carbonate associated with
116870 1000 W, #1 in Fig. 40.1) and in the southern Black Mounta-
the KPF
ins (358540 4500 N, 1168380 5000 W, #23 in Fig. 40.1) where there
is a sharp contact between the Beck Spring Dolomite and the
Unit Delta Min. Max. Average Median SD
KP1 Member sandstone, followed by interbedding between
centimetre-scale dolomite and sandstone beds over the next Beck Spring d13Ccarb – 4.4 5.8 2.7 3.0 1.8
several metres. Dolomite d18O –18.0 1.3 –6.0 – 5.5 3.8
n ¼ 259 d13Corg (n ¼ 45) –25.9 0.7 – 18.4 –18.6 4.3
KPF Oncolite d13Ccarb – 4.0 1.1 –1.6 – 2.0 1.5
Contact with the overlying Noonday Dolomite
Marker Bed d18O –11.2 – 2.1 –6.1 – 5.8 2.7
n ¼ 13 d13Corg
The contact between the eastern-KPF and the overlying Noonday Virgin Spring d13Ccarb 1.0 2.4 2.4 2.1 1.3
Dolomite is contentious and has been reported as regionally uncon- Limestone d18O –16.2 –12.5 – 15.4 –15.6 1.0
formable (Noble 1934; Wright et al. 1978), locally unconformable n ¼ 11 d13Corg
(Christie-Blick & Levy 1989) and locally conformable (Miller Sourdough d13Ccarb –7 3.9 –0.2 – 0.8 2.7
1987). An unconformable relationship has been suggested Limestone d18O –16.9 – 8.3 – 13.3 –13.6 2.1
because the Noonday Dolomite seems to cap successively older, n ¼ 45 d13Corg (n ¼ 9) –15.0 – 4.4 –7.9 – 6.7 3.7
seemingly tilted, strata (Cloud et al. 1974; Wright et al. 1976) Noonday d13Ccarb – 4.2 – 0.9 –2.8 – 2.7 1.1
between the Alexander Hills and the southern Nopah Range, ulti- Dolomite d18O –11.3 – 4.0 –6.9 – 6.5 3.1
mately straddling the contact between the Crystal Spring Fm. and n ¼ 22 d13Corg (n ¼ 9) –25.6 –17.6 – 21.3 –21.3 2.5
the basement at the War Eagle Mine. Alternatively, Prave (1999)
suggested the Gunsight Member infills erosional topography and Data from Tucker (1983, 1986); Bergfeld et al. (1996); Kennedy et al. (2001a)
is conformable with the overlying Noonday Dolomite. and Corsetti & Kaufman (2003).
454 D. MROFKA & M. KENNEDY
1986; Prave 1999; Corsetti & Kaufman 2003) for d13C and d18O
identify no clear stratigraphic trends within the KPF succession.
On the other hand, distinctive isotopic values and spatial continu-
ity of the Beck Spring Dolomite and the Noonday Dolomite bracket
the KPF and provide a clear stratigraphic and geochemical frame-
work for the KPF. The Beck Spring Dolomite shows a several per
mille enrichment in d13C above the base of the Formation (Corsetti
& Kaufman 2003) and has enriched d13C and d18O values overall
(Table 40.1). Noonday Dolomite d13C and d18O values for the over-
lying Noonday Dolomite show a moderate positive co-variation
(r 2 ¼ 0.3) and Kennedy et al. (1998, 2001a) pointed out the simi-
larity between the negative-to-positive d13C trend in the base of
the Noonday Dolomite and similar trends in other late Cryogenian
cap carbonates. Hurtgen et al. (2004) published 34Ssulphate values
for the Beck Spring Dolomite and the Noonday Dolomite in the
range 11.0–27.4‰ and 15–35‰, respectively, and suggested the
values were evidence of ocean sulphate concentrations at 10% of
modern values during the mid-Proterozoic with a transition to
higher values in the late Proterozoic, possibly due to glaciation.
Other characteristics
Corsetti et al. (2003) documented complex microfossils preserved
in chert and carbonate from the oncolitic dolostone bed within the
KP3 Member, similar to microfossils identified in chert nodules
from the Beck Spring Dolomite (Pierce & Cloud 1979; Horodyski
& Knauth 1994). Microfossil evidence was used to support the
existence of active shallow-water microbial biota during a Cryo-
genian glaciation, in which biological activity was thought to
have been non-existent (Hoffman et al. 1998).
Palaeolatitude and palaeogeography
There is no published palaeolatitude data for the KPF itself and
palaeomagnetism test holes in the KP3 Member in the Alexander
Hills yielded data that showed later remagnetization (Wright
2002, pers. comm.). Evans (2000, p. 365, fig. 3) estimated a c. 98
palaeolatitude for the KPF at 723 + 3 Ma based on correlation
with other Neoproterozoic strata along the North American
Cordillera and proposed a nearly equivalent line of latitude with
the Toby Fm. (Christie & Fahrig 1983; Heaman et al. 1992). How-
ever, given Cryogenian Period glacial episodes occur between
c. 750– 634 Ma (Condon et al. 2005; Kendall et al. 2006) and
the lack of geochronological constraints on KPF deposition, the
98 palaeolatitude is highly equivocal.
Geochronological constraints
The KPF is poorly constrained by a date of 1.08 Ga from diabase
within the middle member of the Crystal Spring Fm. (Heaman &
Grotzinger 1992), 500–1000 m below the base of the KPF, and
the Precambrian–Cambrian Boundary in the lower member of the
Wood Canyon Fm. (Corsetti & Hagadorn 2000), 2000 m above
the base of the Noonday Dolomite (Fig. 40.1). Deposition of
broadly similar (but not necessarily correlative) coarse-grained
strata in Idaho is bounded by a lower limit of 717 + 4 Ma and
701 + 4 Ma (Fanning & Link 2004, 2008; U–Pb SHRIMP on a
volcanic clast in diamictite) and upper limit of 685 + 7 Ma (Lund
et al. 2003, U–Pb SHRIMP) and 667 + 5 Ma (Fanning & Link
2004; U–Pb SHRIMP based on tuff above overlying carbonate).
Discussion
Tectonic and glacial deposits of the eastern-KPF are associated
with the underlying Beck Spring Dolomite and overlying
Noonday Dolomite carbonate platforms, overlie a regional
unconformity and record an abrupt change to siliciclastic sedi-
mentation resulting from tectonic uplift. The wedge-shaped
packaging of the KPF strata next to tilted and erosionally trun-
cated segments of the underlying Pahrump Group and basement
is consistent with successions that accumulate during rotation of
hanging wall and footwall sections in extensional systems
(Jackson & White 1988; Faerseth et al. 1990, p. 1291, fig. 7;
Jackson et al. 2005). Coarse-grained facies, initiated with depo-
sition of the KP2 Member, are primarily the product of local tec-
tonic activity as indicated by (i) syndepositional normal faults
and erosional bevelling of footwall blocks, (ii) a systematic
and consistent pattern of coarsening upwards from sand and
angular cobble debrites to sedimentary breccia and kilometre-
scale olistoliths, (iii) a transition from marine debris-flow
facies to terrestrial fanglomerates and (iv) a systematic pattern
of unroofing of the underlying Pahrump group as seen in the
sequence of dominant clasts. A coarsening-upwards trend begin-
ning in the upper KP3 Member and the transition to terrestrial
fanglomerates in the KP4 Member, especially near basin
margins, indicate shallowing as deposition outpaces subsidence
and creation of accommodation space. The occurrence of
striated clasts in the KP2 and KP3 members indicates a glacio-
genic influence that is sporadic and limited to specific
stratigraphic intervals.
The KP1 Member parallel-laminated sandstone was interpreted
by Tucker (1986) as having been deposited on a shelf by storm cur-
rents and the overlying laminated Virgin Spring limestone repre-
senting a relatively deeper water carbonate environment with
periodic input of sand and ooids by storms. The unconformity
that truncates the Virgin Spring limestone might be a result of sea-
level fall from glaciation or tectonic uplift. Erosional truncation in
hanging wall sections indicates that while clasts from the under-
lying Pahrump Group in the KP2 Member provide a clear signal
for initiation of tectonism, initial exposure and erosion across the
region may have been the result of ice growth and sea-level
change prior to local tectonism, as indicated by the presence of
striated clasts.
The KP2 Member was likely deposited in a glaciomarine
setting and underwent downslope reworking (Boulton &
Deynoux 1981) on the same broad shallow shelf on which the
KP1 Member and Virgin Spring limestone were deposited.
There is no evidence of the clinoform geometry or rapid lateral
facies changes expected with ice-proximal debris aprons and
no evidence of laminations within the diamictite facies hosting
outsized clasts expected from rainout of ice-rafted debris. The
KP3 Member records the greatest lateral thickness changes of
all the KPF members and represents the most active phase of tec-
tonism in the KPF, perhaps equivalent to a rift climax (Prosser
1993). Diamictite intervals within the KP3 Member with striated
clasts are interbedded with coarse-grained deposits interpreted to
be a result of gravity-driven debris flows of glaciogenic debris
down a rapidly tectonically steepening margin. The KP4
Member is dominated by commonly monomictic channelized
sedimentary breccia and likely represents terrestrial fanglomerate
(Hewett 1956). Lateral inter-fingering relations evident in the
Kingston Range between fanglomerate facies and turbiditic sand-
stone suggest the KP4 Member is a terrestrial equivalent of the
more distal, marine KP3 Member, indicating fan-building pro-
gressed sub-aqueously and mixed with finer-grained distal
density deposits.
The Gunsight Member records a final pulse of glaciation and
subsequent deglacial flooding of KP4 Member terrestrial deposits.
The abrupt and conformable transition to the overlying, regionally
continuous, platformal Noonday Dolomite, as well as the thickness
of the dolomite’s microbial mounds (.200 m) suggest flooding
was a result of deglacial sea-level rise, and that continued trans-
gression and cessation of tectonism completely cut off any
source of siliciclastic sediments.
 KINGSTON PEAK FORMATION, EASTERN DEATH VALLEY
Association with Neoproterozoic glaciation
The only direct evidence for glaciation in the KPF is striated clast-
bearing diamictite (Hazzard 1939; Miller 1985) found throughout
the KP2 Member, interbedded in the lower half of the KP3
Member and commonly comprising the Gunsight Member.
Striated clasts are polished, comprised of siltstone, chert and quart-
zite, and rarely faceted. There is no systematic relationship
between striated clast-bearing diamictite and bounding units; in
the KP2 Member diamictite is overlain sharply by sandstone of
the basal KP3 Member, in the KP3 Member diamictite is inter-
bedded with sandstone and sedimentary breccia and in the Gun-
sight Member it sharply overlies sedimentary breccia. Indirect
evidence for a glacial influence (Crowell 1999) in the KPF was
documented by Miller (1985) in both the western and eastern-KPF
and includes evidence for rapid deposition, presence of diamictite
and rapid lateral changes in facies and thickness.
Sediments with striated clasts are often attributed to ice-rafted
debris in glaciomarine settings (Crowell 1999). At Sperry Wash
(358420 1300 N, 1168140 3400 W, #18 in Fig. 40.1) outsized clasts
within turbidite facies (Troxel 1982b) of the KP3 Member have
been interpreted as dropstones (Abolins et al. 2000; Corsetti &
Kaufman 2003) or as lone clasts rolling down tectonically produced
slopes (Troxel 1982b). Several clasts appear to deform underlying
sediments and at least one pierces underlying sediments and may
have splash-marks (Corsetti pers. comm. 2008). However, outsized
clasts at Sperry Wash are grouped along common bedding
planes and laterally associated with diamictites and normally
graded conglomerate beds interpreted to be debris flows (Troxel
1982b). In other sections, similar outsized clast-bearing facies
are interbedded with conglomerate and host megaclasts (tens of
metres scale) and kilometre-scale olistoliths. The association
with tectonically emplaced olistoliths, the bedding plane parallel
orientation of many clasts and close association of outsize clasts
to debrites beds with lonestones, suggests outsized clasts may
alternatively represent the distal edges of debris flows or loose
clasts that tumbled down clast-laden slopes (Postma et al. 1988).
The KP3 Member at Sperry Wash contains up to 45% Fe by
weight (Abolins et al. 2000) and due to association between
Fe-rich sediments and Neoproterozoic glacial intervals (Young
1976), has been interpreted to represent evidence for glaciation
(Stewart 1972; Abolins et al. 2000; Awramik et al. 2000).
However, there is no direct evidence for an oceanographic origin
for the Fe. Alternatively, the Fe might be associated with nearby
Tertiary-aged volcanic intrusions.
Neoproterozoic glacial sediments are commonly intimately
associated with overlying carbonate facies (Hoffman et al.
1998). These cap carbonates typically overlie an older and
younger glacial interval in many sections and host a variety of dis-
tinctive sedimentary and geochemical features (Kennedy et al.
1998; Hoffman et al. 2002). The Sourdough Limestone Member
in the western-KPF is characterized by graphite-rich parallel
laminations and depleted d13C values (Table 40.1), characteristics
shared among older Cryogenian cap carbonates (Kennedy et al.
1998), and has been interpreted to represent an older cap carbonate
in the western-KPF (Prave 1999). In the eastern-KPF, the Virgin
Spring and Silurian Hills limestone units are also characterized
by graphite-rich parallel laminations but have enriched d13C
values (Table 40.1). The Noonday Dolomite, interpreted to rep-
resent a younger cap carbonate (Prave 1999), overlies the KPF
and shares a number of characteristics with other younger cap car-
bonates, including its cream colour, tubestones, abundant marine
cements and depleted d13C values (Kennedy et al. 1998; Nogueira
et al. 2003).
Carbonate intervals in both eastern and western KPF intervals
are associated with coarse-grained facies but litho-stratigraphic
correlation is complicated by dissimilar bounding lithofacies.
Deposition and a sporadic glacial influence are continuous from
the KP2 Member through the Noonday Dolomite, so the interval
455
does not conform to a glacial – cap carbonate –interglacial–
glacial –cap carbonate cycle typical of other Neoproterozoic
successions. Although associated with coarse-grained facies, the
palaeoclimatic significance of the carbonate intervals in the
eastern KPF is unclear.
Conclusion
The Kingston Peak Fm. in the Death Valley region is one of the few
areas in the southern Cordillera where the association between
extension and resulting Neoproterozoic sedimentation is clear
(Burchfiel et al. 1992). This allows three competing hypotheses
for the origin of Neoproterozoic glacial deposits to be tested: (1)
glacial sediments record a globally synchronous climate event
(Snowball Earth); (2) glacial sediments record regional glaciation
attributed to tectonism and terminated with the cessation of exten-
sion (i.e. zipper-rift model of Eyles & Januszczak (2004)); and (3)
glacial sediments comprise an incomplete record of either regional
glaciation or a long-term (50 –100 Ma) Cryogenian glacial era
(Allen & Etienne 2008) with sedimentary evidence of glaciation
only preserved during periods of tectonism and generation of
accommodation space. In hypothesis two and three, the preser-
vation of glaciogenic sediments is linked with regional tectonism
and therefore results in a diachronous record of glaciation.
Each of the three hypotheses results in specific predictions for
the sedimentary record. In the first case, if the transgressive depos-
its of the Noonday Dolomite are related to global deglaciation and
not cessation of extension, syn-depositional faulting should con-
tinue through the Noonday Dolomite. This has not been observed.
In the second case, carbonate deposition is related to transgression
during thermal subsidence when tectonism ends; rising sea level
confines glacial evidence up-dip or may be entirely lacking
because of slowdown in adiabatic cooling provided by uplift. In
the last case, extension and uplift provide accommodation space
in shallow marine to terrestrial environments that rapidly preserve
glacial sediments and provide a record of glaciation and climate
change whose continuity and completeness is controlled by
local tectonism.
The intimate association between localized tectonically created
uplift, accommodation space, and conglomeratic facies bring into
question the suggestion that KPF strata record a regional com-
ponent of synchronous glacial sedimentation associated with
global Neoproterozoic ice ages (Prave 1999). The complete lack
of geochronological data makes any global inference problematic.
Evidence in the Kingston Peak Fm. provides more support for
hypotheses two and three and consequently a strong argument
for diachronous glacial deposition and cap carbonate deposition.
These data support the argument that the Kingston Peak Fm.,
along with an increasing number of other Neoproterozoic deposits
worldwide (Allen & Etienne 2008), record part of a continual and
diachronous climate record (Mrofka 2010) spanning the Cryogen-
ian Period.
The National Science Foundation (EAR 0345207), NASA Exobiology
(NWG04GjJ42G), Geological Society of America Student Research Grant and
the American Association for Petroleum Geologists Grant-in-Aid provided
support for D.M.’s field studies on the KPF. The authors appreciate the advice
and direction provided by B. Troxel and L. Wright. The authors also wish to
thank the following for their assistance in the field: K. Thompson, D. deYoung,
T. Bristow, Ganqing Jiang and C. Partin. This represents a contribution of
the IUGS- and UNESCO-funded IGCP (International Geoscience Programme)
project #512.
References
Abolins, M., Oskin, R., Prave, A. R., Summa, C. & Corsetti, F. A.
2000. Neoproterozoic glacial record in the Death Valley region,
456 D. MROFKA & M. KENNEDY
California and Nevada. In: Lageson, D. R., Peters, S. G. & Lahren,
M. M. (eds) GSA Field Guide 2: Great Basin and Sierra Nevada,
319– 335.
Allen, P. A. & Etienne, J. L. 2008. Sedimentary challenge to Snowball
Earth. Nature Geoscience, 1, 817–825.
Awramik, S. M., Corsetti, F. A. & Shapiro, R. 2000. Stromatolites and
the pre-Phanerozoic to Cambrian history of the area south east of
Death Valley. Bulletin of the San Bernardino County Museum, 47,
65 –74.
Bergfeld, D., Nabelek, P. I. & Labotka, T. C. 1996. Carbon isotope
exchange during polymetamorphism in the Panamint Mountains,
California, USA. Journal of Metamorphic Geology, 14, 199– 212.
Boulton, G. S. & Deynoux, M. 1981. Sedimentation in glacial environ-
ments and the identification of tills and tillites in ancient sedimentary
sequences. Precambrian Research, 15, 397– 422.
Burchfiel, B. C., Cowan, D. S. & Davis, G. A. 1992. Tectonic overview
of the Cordilleran Orogen in the Western United States. In: Burch-
fiel, B. C., Lipman, P. W. & Zoback, M. L. (eds) The Cordilleran
Orogen; Conterminous U.S.: The Geology of North America.
Geological Society of America, Boulder, G-3, 407– 480.
Carlisle, D., Kettler, R. M. & Swanson, S. C. 1980. Geological study
of uranium potential of the Kingston Peak Fm., Death Valley region,
California. US Department of Energy Open File Report.
Christie, K. W. & Fahrig, W. F. 1983. Paleomagnetism of the Borden
dykes of Baffin Island and its bearing on the Grenville Loop. Cana-
dian Journal of Earth Sciences, 20, 275–289.
Christie-Blick, N. & Levy, M. 1989. Stratigraphic and tectonic frame-
work of the upper Proterozoic and Cambrian rocks in the western
United States. In: Christie-Blick, N., Levy, M., Mount, J. F.,
Signor, P. W. & Link, P. K. (eds) Late Proterozoic and Cambrian
Tectonics, Sedimentation, and Record of Metazoan Radiation in the
Western United States, Field Trip Guidebook T331, American
Geophysical Union, Washington, DC, 7 –21.
Cloud, P., Wright, L. A., Williams, E. G., Diehl, P. E. & Walter, M. R.
1974. Giant stromatolites and associated vertical tubes from the
upper Proterozoic Noonday Dolomite, Death Valley region, eastern
California. Geological Society of America Bulletin, 85, 1869– 1882.
Condon, D., Zhu, M., Bowring, S., Wang, W., Yang, A. & Jin, Y. 2005.
U –Pb ages from the Neoproterozoic Doushantuo Formation, China.
Science, 308, 95 –98.
Corsetti, F. A. & Hagadorn, J. W. 2000. Precambrian – Cambrian
transition: Death Valley, United States. Geology, 28, 299– 302.
Corsetti, F. A. & Kaufman, A. J. 2003. Stratigraphic investigations of
carbon isotope anomalies and Neoproterozoic ice ages in Death
Valley, California. Geological Society of America Bulletin, 115,
916– 932.
Corsetti, F. A. & Grotzinger, J. P. 2005. Origin and significance of tube
structures in Neoproterozoic post-glacial cap carbonates: example
from Noonday Dolomite, Death Valley, United States. Palaios, 20,
348– 362.
Corsetti, F. A. & Kaufman, A. J. 2005. The relationship between the
Neoproterozoic Noonday Dolomite and the Ibex Formation: New
observations and their bearing on ‘snowball Earth’. Earth-Science
Reviews, 73, 63– 78.
Corsetti, F. A., Awramik, S. M. & Pierce, D. 2003. A complex micro-
biota from snowball Earth times: microfossils from the Neoprotero-
zoic Kingston Peak Formation, Death Valley, USA. Proceedings of
the National Academy of Sciences of the United States of America,
100, 4399– 4404.
Crowell, J. C. 1999. Pre-Mesozoic Ice Ages; Their Bearing on Under-
standing the Climate System. Geological Society of America,
Memoirs, 192, 1– 112.
Davis, G. A., Fowler, T. K. et al. 1993. Pluton pinning of an active
Miocene detachment fault system, eastern Mojave Desert, California.
Geology, 21, 627–630.
Evans, D. A. 2000. Stratigraphic, geochronological, and paleomagnetic
constraints upon the Neoproterozoic climatic paradox. American
Journal of Science, 300, 347– 433.
Eyles, N. 1993. Earth’s glacial record and its tectonic setting. Earth-
Science Reviews, 35, 1– 248.
Faerseth, R. B., Knudsen, B. E., Liljedahl, T., Midboe, P. S. &
Soderstrom, B. 1997. Oblique rifting and sequential faulting in
the Jurassic development of the northern North Sea. Journal of Struc-
tural Geology, 19, 1285–1302.
Fanning, C. M. & Link, P. K. 2004. U –Pb SHRIMP ages of Neoproter-
ozoic (Sturtian) glaciogenic Pocatello Formation, southeastern Idaho.
Geology, 32, 881–884.
Fanning, C. M. & Link, P. K. 2008. Age constraints for the Sturtian
Glaciation; data from the Adelaide Geosyncline, South Australia
and Pocatello Formation, Idaho, USA. In: Gallagher, S. J. &
Wallace, M. W. (eds) Neoproterozoic extreme climates, the origin
of early metazoan life. Geological Society of Australia Extended
Abstracts, 91, 57 – 62.
Fedo, C. M. & Cooper, J. D. 2001. Sedimentology and sequence stratigra-
phy of Neoproterozoic and Cambrian units across a craton-margin
hinge zone, southeastern California, and implications for the early
evolution of the Cordilleran margin. Sedimentary Geology, 141,
501– 522.
Hammond, J. C. 1983. Late Precambrian diabase intrusions in the
southern Death Valley region California: their petrology, geochemis-
try, and tectonic significance. PhD dissertation, University of
Southern California.
Hazzard, J. C. 1939. Possibility of pre-Cambric glaciation in southeast-
ern California. Pan-American Geologist, 71, 47– 48.
Heaman, L. M. & Grotzinger, J. P. 1992. 1.08 Ga diabase sills in the
Pahrump Group, California; implications for development of the
Cordilleran Miogeocline. Geology, 20, 637–640.
Heaman, L. M., Lecheminanta, A. N. & Rainbird, R. H. 1992. Nature
and timing of Franklin igneous events, Canada: implications for a
Late Proterozoic mantle plume and the break-up of Laurentia.
Earth and Planetary Science, 109, 117–131.
Hewett, D. F. 1940. New formation names to be used in the Kingston
Range, Ivanpah Quadrangle, California. Journal of the Washington
Academy of Sciences, 30, 239–240.
Hewett, D. F. 1956. Geology and mineral resources of the Ivanpah Quad-
rangle, California and Nevada. United States Geological Survey
Professional Paper, 275, 23– 99.
Hoffman, P. F. & Schrag, D. P. 2002. The snowball earth hypothesis:
testing the limits of global change. Terra Nova, 14, 129–155.
Hoffman, P. F., Kaufman, A. J., Halverson, G. P. & Schrag, D. P.
1998. A Neoproterozoic snowball earth. Science, 281, 1342–1346.
Horodyski, R. J. & Knauth, L. P. 1994. Life on land in the Precambrian.
Science, 263, 494– 498.
Hurtgen, M. T., Arthur, M. A. & Prave, A. R. 2004. The sulfur isotope
composition of carbonate-associated sulfate in Mesoproterozoic
to Neoproterozoic carbonates from Death Valley, California. In:
Amend, J. P., Edwards, K. J. & Lyons, T. W. (eds) Sulfur Biogeo-
chemistry – Past and Present. Geological Society of America
Special Paper, 379, 177–194.
Jackson, J. A. & White, N. J. 1988. Normal faulting in the upper conti-
nental crust: observations from regions of active extension. Journal
of Structural Geology, 11, 15– 36.
Jackson, C. A. L., Gawthorpe, R. L., Carr, I. D. & Sharp, I. R. 2005.
Normal faulting as a control on the stratigraphic development of
shallow marine syn-rift sequences: the Nukhul and Lower Rudeis
Formations, Hammam Faraun fault block, Suez Rift, Egypt. Sedimen-
tology, 52, 313– 338.
Johnson, B. K. 1957. Geology of a part of the Manly Peak Quadrangle,
southern Panamint Range, California. University of California Publi-
cations in Geological Sciences, 30, 353–423.
Kendall, B., Creaser, R. A. & Selby, D. 2006. Re– Os geochronology
of postglacial black shales in Australia; constraints on the timing of
‘Sturtian’ glaciation. Geology, 34, 729–732.
Kennedy, M. J., Runnegar, B., Prave, A. R., Hoffmann, K. H. &
Arthur, M. A. 1998. Two or four Neoproterozoic glaciations.
Geology, 26, 1059–1063.
Kennedy, M. J., Christie-Blick, N. & Prave, A. R. 2001a. Carbon iso-
topic composition of Neoproterozoic glacial carbonates as a test
of paleoceanographic models for snowball Earth phenomena.
Geology, 29, 1135–1138.
Kennedy, M. J., Christie-Blick, N. & Sohl, L. E. 2001b. Are Protero-
zoic cap carbonates and isotopic excursions a record of gas hydrate
destabilization following Earth’s coldest intervals? Geology, 29,
443– 446.
 KINGSTON PEAK FORMATION, EASTERN DEATH VALLEY
Kenny, R. & Knauth, L. P. 2001. Stable isotope variations in the Neo-
proterozoic Beck Spring Dolomite and Mesoproterozoic Mescal
Limestone paleokarst: Implications for life on land in the Precam-
brian. Geological Society of America Bulletin, 113, 650–658.
Kupfer, D. H. 1960. Thrust faulting and chaos structure, Silurian Hills,
San Bernardino County, California. Geological Society of America
Bulletin, 71, 181–214.
Labotka, T. C., Albee, A. L., Lanphere, M. A. & McDowell, S. D.
1980. Stratigraphy, structure, and metamorphism in the central
Panamint Mountains (Telescope Peak Quadrangle), Death Valley
area, California. Geological Society of America Bulletin, 91 Part I,
125– 129.
Levy, M. & Christie-Blick, N. 1989. Pre-Mesozoic palinspastic recon-
struction of the eastern Great Basin (Western United States). Science,
245, 1454–1462.
Levy, M. & Christie-Blick, N. 1991. Tectonic subsidence of the early
Paleozoic passive continental margin in eastern California and
southern Nevada. Geological Society of America Bulletin, 103,
1590– 1606.
Link, P. K., Christie-Blick, N. et al. 1993. Middle and late Proterozoic
stratified rocks of the western U.S. Cordillera, Colorado Plateau,
and Basin and Range Province. In: Reed, J. C. Jr, Bickford, M. E.
et al. (eds) Precambrian-Conterminous U.S.: The Geology of
North America, United States Geological Survey, Denver, CO,
463– 595.
Lund, K., Aleinikoff, J. N., Evans, K. V. & Fanning, C. M. 2003.
SHRIMP U–Pb geochronology of Neoproterozoic Windermere
Supergroup, central Idaho: implications for rifting of western Lauren-
tia and synchroneity of Sturtian glacial deposits. Geological Society
of America Bulletin, 115, 349–372.
Mattinson, C. G., Colgan, J. P., Metcalf, J. R., Miller, E. L. &
Wooden, J. L. 2007. Late Cretaceous to Paleocene metamorphism
and magmatism in the Funeral Mountains metamorphic core
complex, Death Valley, California. In: Cloos, M., Carlson,
W. D., Gilbert, M. C., Liou, J. G. & Sorensen, S. S. Special
Paper – Geological Society of America, 419, 205– 223.
Marian, M. L. & Osborne, R. H. 1992. Petrology, petrochemistry, and
stromatolites of the middle to late Proterozoic Beck Spring Dolomite,
eastern Mojave Desert, California. Canadian Journal of Earth
Sciences, 29, 2595–2609.
Miller, J. M. G. 1982. Kingston Peak Formation in the southern Panamint
Range; a glacial interpretation. In: Cooper, J. D., Troxel, B. W. &
Wright, L. A. (eds) Geology of Selected Areas in the San Bernardino
Mountains, Western Mojave Desert, and Southern Great Basin,
California: Geological Society of America Cordilleran Section Field
Trip Guidebook and Volume. Death Valley Publishing Company,
Shohone, California, 155–164.
Miller, J. M. G. 1983. Stratigraphy and sedimentology of the upper
Proterozoic Kingston Peak Formation, southern Panamint Range,
eastern California. PhD dissertation, University of California,
Santa Barbara.
Miller, J. M. G. 1985. Glacial and syntectonic sedimentation; the upper
Proterozoic Kingston Peak Formation, southern Panamint Range,
eastern California. Geological Society of America Bulletin, 96,
1537– 1553.
Miller, J. M. G. 1987. Paleotectonic and stratigraphic implications of the
Kingston Peak –Noonday contact in the Panamint Range, eastern
California. Journal of Geology, 95, 75 – 85.
Miller, M. G. 1991. High-angle origin of the currently low-angle
Badwater turtleback fault, Death Valley, California. Geology, 19,
372– 375.
Mrofka, D. D. 2010. Competing models for the timing of Cryogenian Gla-
ciation: evidence from the Kingston Peak Formation, southeastern
California. PhD dissertation, University of California, Riverside.
Murphy, F. M. 1930. Geology of the Panamint silver district, California.
Economic Geology and the Bulletin of the Society of Economic Geol-
ogists, 25, 305– 325.
Murphy, F. M. 1932. Geology of a part of the Panamint Range, report 27
of the state mineralogist. California Journal of Mines and Geology,
28, 329– 356.
Noble, L. F. 1934. Rock formations of Death Valley, California. Science,
80, 173– 178.
457
Noble, L. F. 1941. Structural features of the Virgin Spring area, Death
Valley, California. Geological Society of America Bulletin, 52,
941– 999.
Nogueira, A. C. R., Riccomini, C., Sial, A. N., Moura, C. A. V. & Fair-
child, T. R. 2003. Soft-sediment deformation at the base of the
Neoproterozoic Puga cap carbonate (southwestern Amazon craton,
Brazil): confirmation of rapid icehouse to greenhouse transition in
snowball Earth. Geology, 31, 613– 616.
Petterson, R., Prave, A. R. & Wernicke, B. P. 2011. Glaciogenic and
related strata of the Neoproterozoic Kingston Peak Formation in the
Panamint Range, Death Valley Region, California. In: Arnaud, E.,
Halverson, G. P. & Shields-Zhou, G. (eds) The Geological
Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 459–465.
Pierce, D. & Cloud, P. 1979. New microbial fossils from approximately
1.3 billion-year-old rocks of eastern California. Geomicrobiology
Journal, 1, 295– 309.
Postma, G., Nemec, W. & Kleinspehn, K. L. 1988. Large floating clasts
in turbidites: a mechanism for their emplacement. Sedimentary
Geology, 58, 47 –61.
Prosser, S. 1993. Rift-related linked depositional systems and their seis-
mic expression. Geological Society Special Publications, 71, 35– 66.
Prave, A. R. 1999. Two diamictites, two cap carbonates, two d13C excur-
sions, two rifts; the Neoproterozoic Kingston Peak Formation, Death
Valley, California. Geology, 27, 339–342.
Renik, B., Christie-Blick, N., Troxel, B., Wright, L. & Niemi, N.
2008. Re-evaluation of the middle Miocene Eagle Mountain For-
mation and its significance as a piercing point for the interpretation
of extreme extension across the Death Valley Region, California,
U.S.A. Journal of Sedimentary Research, 78, 199–219.
Roberts, M. T. 1974. Stratigraphy and depositional environments of the
Crystal Spring Formation, Southern Death Valley Region, California.
Guidebook: Death Valley region, California and Nevada, Death
Valley Publishing Company, Shoshone, 49 –57.
Snow, J. K. 1992. Large-magnitude Permian shortening and continental-
margin tectonics in the southern Cordillera. Geological Society of
America Bulletin, 104, 80 –105.
Stewart, J. H. 1972. Initial deposits in the Cordilleran geosyncline:
evidence of a late Precambrian (,850 m.y.) continental separation.
Geological Society of America Bulletin, 83, 1345– 1360.
Summa, C. L. 1993. Sedimentologic, stratigraphic, and tectonic controls
of a mixed carbonate-siliciclastic succession; Neoproterozoic
Johnnie Formation, Southeast California. PhD dissertation, Massa-
chusetts Institute of Technology, Cambridge, MA.
Troxel, B. W. 1967. Sedimentary rocks of late Precambrian and Cam-
brian age in the southern Salt Spring Hills, southeastern Death
Valley, California. Short Contributions to California Geology:
Special Report, California Division of Mines and Geology, 2, 33– 41.
Troxel, B. W. 1982a. Basin facies (Ibex Formation) of the Noonday
Dolomite, southern Saddle Peak Hills, southern Death Valley,
California. In: Cooper, J. D., Troxel, B. W. & Wright, L. A.
(eds) Geology of Selected Areas in the San Bernardino Mountains,
Western Mojave Desert, and Southern Great Basin, California: Geo-
logical Society of America Cordilleran Section Field Trip Guidebook
and Volume. Death Valley Publishing Company, Shoshone, 43– 48.
Troxel, B. W. 1982b. Description of the uppermost part of the Kingston
Peak Formation, Amargosa Rim Canyon, Death Valley region,
California. In: Cooper, J. D., Troxel, B. W. & Wright, L. A. (eds)
Geology of Selected Areas in the San Bernardino Mountains, Western
Mojave Desert, and Southern Great Basin, California: Geological
Society of America Cordilleran Section Field Trip Guidebook and
Volume. Death Valley Publishing Company, Shoshone, CA, 61– 70.
Troxel, B. W. & Wright, L. A. 1987. Tertiary extensional features,
Death Valley region, eastern California. In: Hill, M. L. (ed.)
Centennial Field Guide-Cordilleran Section, Geological Society of
America, Boulder, CO, 1, 121–132.
Tucker, M. E. 1983. Diagenesis, geochemistry, and origin of a Precam-
brian dolomite: The Beck Spring Dolomite of eastern California.
Journal of Sedimentary Petrology, 53, 1097– 1119.
Tucker, M. E. 1986. Formerly aragonitic limestones associated with til-
lites in the Late Proterozoic of Death Valley, California. Journal of
Sedimentary Petrology, 56, 818–830.
458 D. MROFKA & M. KENNEDY
Wernicke, B., Axen, G. J. & Snow, J. K. 1988. Basin and range exten-
sional tectonics at the latitude of Las Vegas, Nevada. Geological
Society of America Bulletin, 100, 1738– 1757.
Williams, E. G., Wright, L. A. & Troxel, B. W. 1974. The Noonday
Dolomite and equivalent stratigraphic units, southern Death Valley
region, California. Guidebook: Death Valley Region, California
and Nevada. Death Valley Publishing Company, Shoshone, 73 –77.
Wright, L. A. 1952. Geology of the Superior talc area, Death Valley,
California. Special Report – California Division of Mines and
Geology, 20, 1– 22.
Wright, L. A. 1968. Talc deposits of the southern Death Valley – King-
ston Range region, California. Special Report – California Division
of Mines and Geology, 95, 1 –79.
Wright, L. A. 1974. Geology of the S.E. 1/4 Tecopa 15-minute quadran-
gle, San Bernardino and Inyo Counties, California. California
Division of Mines and Geology Map Sheet 20, 1:24,000.
Wright, L. A. & Troxel, B. W. 1966. Strata of late Precambrian –
Cambrian age, Death Valley region, California –Nevada. Bulletin of
the American Association of Petroleum Geologists, 50, 846– 857.
Wright, L. A. & Troxel, B. W. 1967. Limitations on right-lateral, strike–
slip displacement, Death Valley and Furnace Creek Fault Zones,
California. Geological Society of America Bulletin, 78, 933–950.
Wright, L. A., Troxel, B. W., Williams, E. G., Roberts, M. T. &
Diehl, P. E. 1976. Precambrian sedimentary environments of the
Death Valley region, eastern California. California Division of
Mines and Geology Special Report, 106, 7 –15.
Wright, L. A., Williams, E. G. & Cloud, P. 1978. Algal and cryptalgal
structures and platform environments of the late pre-Phanerozoic
Noonday Dolomite, eastern California. Geological Society of
America Bulletin, 89, 321–333.
Wright, L. A., Williams, E. G. & Troxel, B. W. 1984. Type section of
the newly-named Proterozoic Ibex Formation, the basinal equivalent
of the Noonday Dolomite (Appendix II). Geology of the northern half
of the Confidence Hills 15-minute Quadrangle, Death Valley region,
eastern California, California Division of Mines and Geology, Map
Sheet 34, scale 1:24,000, 25 –31.
Wright, L. A., Troxel, B. W. & Prave, A. R. 1992. Field traverse of
Proterozoic rock units, Alexander Hills and southern Nopah Range,
Death Valley region, California. Geological Society of America
Penrose Conference: Late Precambrian Plate Tectonics and the
Dawn of the Phanerozoic, Geological Society of America, 1 –11.
Young, G. M. 1976. Iron-formation and glaciogenic rocks of the Rapitan
Group, Northwest Territories, Canada. Precambrian Research, 3,
137– 158.
 Chapter 41

Glaciogenic and related strata of the Neoproterozoic Kingston Peak Formation

in the Panamint Range, Death Valley region, California

RYAN PETTERSON1*, ANTHONY R. PRAVE2 & BRIAN P. WERNICKE1

1
Division of Geological and Planetary Sciences, California Institute of Technology, Mail Stop 100-23, Pasadena, CA, 91125, USA
2
Department of Earth Sciences, University of St Andrews, St Andrews, KY16 9AL, UK
*Corresponding author (e-mail: ryanpetterson@gmail.com)

Abstract: Glaciogenic deposits in the Death Valley region occur within the Neoproterozoic Kingston Peak Formation (Fm.). In the
Panamint Range, immediately west of Death Valley, the formation is as much as 1000 m thick and is continuously exposed for
nearly 100 km along the strike of the range. Although the strata are variably metamorphosed and locally exhibit pronounced ductile
strain, original sedimentary textures are well preserved in many places. Diamictite occurs in two distinct intervals, a lower one compris-
ing the Limekiln Spring and Surprise members, and an upper one, the Wildrose Sub-member of the South Park Member. Lonestones,
bullet-shaped and striated clasts, and rare dropstones within these members, along with the impressive lateral continuity of diamictic
units, support a glacial origin. Both diamictic intervals are succeeded by well-defined carbonates, the oldest is the Sourdough
Member of the Kingston Peak Fm. and the younger one is the Sentinel Peak Member of the overlying Noonday Dolomite. The strati-
graphic succession between the Sourdough Member and the Wildrose Sub-member (i.e. the Middle Park, Mountain Girl and Thorndike
sub-members of the South Park Member) is c. 300 m thick and includes lithologies recording deposition in braided fluvial to platform
carbonate settings. Lithostratigraphic and chemostratigraphic profiles of d13C for the Sourdough (– 3‰ to þ2‰, increasing upward) and
Sentinel Peak (– 3‰ +1‰) members suggest correlation with, respectively, the older Cryogenian (commonly referred to as ‘Sturtian’ in
previous literature) and younger Cryogenian (commonly referred to as ‘Marinoan’ in previous literature) cap-carbonate sequences recog-
nized worldwide. Potentially economic uranium deposits (secondary brannerite) occur in graphitic schist of the Limekiln Spring Member
and sub-economic uranium and thorium (hosted by detrital monazite) occur within quartz-pebble conglomerate in the South Park
Member. The Kingston Peak Fm. strata in the Panamint Range contain no fossils, radiometric age control or primary magnetizations.

Murphy (1932) was the first person to infer a glacial origin for a
thick succession (1000 m) of Precambrian conglomeratic strata
exposed in the Telescope Peak area in the Panamint Range, a
major Basin and Range fault block on the western flank of Death
Valley (Figs 41.1 & 41.2). Subsequently, Hewett (1940) documen-
ted equivalent rocks in the Kingston Range, about 50 km east of
Death Valley, and named them the Kingston Peak Fm., the young-
est of three formations he assigned to the newly defined Pahrump
Group (the other two being the Crystal Spring Fm. and Beck
Spring Dolomite). This nomenclature has since been adopted
and used throughout the Death Valley region.
Kingston Peak Fm. strata are highly variable in thickness and
sedimentary facies, owing to syn-depositional tectonism during
part (but not all) of their sedimentation history. Therefore, regional
correlation of units at the sub-formational level has proved chal-
lenging (Miller et al. 1988; Prave 1999). Further, the recognition
and interpretation of the Sourdough Member (commonly refereed
to as Sourdough Limestone in the literature) of the Kingston Peak
Fm. and the Noonday Dolomite as cap-carbonate sequences related
to an older and younger Cryogenian glacial period, respectively,
has remained a subject of debate (Prave 1999; Corsetti &
Kaufman 2003). This controversy, however, is becoming increas-
ingly resolved by Petterson’s (2009) work, to be published in detail
elsewhere, which shows that the Noonday Dolomite is the likely
equivalent to the younger Cryogenian cap sequence (commonly
referred to as ‘Marinoan’), recognized worldwide and known to
be c. 635 Ma in age (Hoffmann et al. 2004).
The strata east of Death Valley are described in detail in Mrofka
& Kennedy (2011). In that area, units are well exposed and have
experienced only minor metamorphism. However, lateral continu-
ity of exposures in eastern Death Valley is limited owing to strati-
graphic variability, structural deformation, and the presence of
wide areas of alluvium separating relatively small fault blocks.
In contrast, Kingston Peak Fm. strata in the Panamint Range
are well exposed for nearly 100 km along strike (Fig. 41.2). This
outcrop belt is the largest single exposure of Cryogenian strata
in the southwestern part of Laurentia, providing an unusual oppor-
tunity to examine lateral transitions in facies and thickness in
detail.
Structural framework
The Panamint Range is a north –south-trending, east-tilted range
block within the Basin and Range extensional province. The
general structure of the range defines a NNW-trending anticline.
The anticline core is 1.7 Ga gneiss, which is locally intruded by
a 1.4 Ga porphyritic quartz monzonite (Albee et al. 1981).
Various units of the Pahrump Group and younger strata overlie
these basement rocks (collectively known as the World Beater
Complex; Labotka et al. 1980). Syn-depositional faulting and
mafic magmatism have been identified within the Kingston Peak
Fm. and are interpreted to be a manifestation of continental
rifting (e.g. Stewart 1972; Wright et al. 1974; Hammond 1983),
Contractile deformation, granitic magmatism and associated meta-
morphism, and development of an east-directed thrust fault system
in the Tucki Mountain area (Fig. 41.2; Wernicke et al. 1993)
affected the Panamint Range in Jurassic and Cretaceous times
(Labotka et al. 1985). The final major deformation event was east-
ward tilting of fault blocks along west-side-down normal faults
during Late Tertiary time, accompanied by intrusion of granitic
plutons and associated dykes (Labotka et al. 1980; Hodges et al.
1989). Although the Neoproterozoic strata experienced green-
schist to amphibolite facies metamorphism and locally exhibit
pronounced ductile strain, original sedimentary textures are well
preserved in many places.
The palaeogeographic setting of the Kingston Peak Fm. is that
of an episodically active cratonic rift basin (Wright et al. 1974;
Labotka et al. 1980). The stratigraphic position of the Kingston
Peak Fm. below more widespread deposits of the Cordilleran

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 459– 465. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.41
460 R. PETTERSON ET AL.
miogeocline has suggested to many workers that this phase of
extensional deformation represents the early stages of continental
break-up leading to the development of the early Palaeozoic
passive margin (Burchfiel & Davis 1972; Stewart 1972; Wright
et al. 1974; Levy & Christie-Blick 1991).
Stratigraphy
In the Panamint Range, the Kingston Peak Fm. has been sub-
divided into the Limekiln Spring, Surprise, Sourdough and South
Park members (Labotka et al. 1980; Miller 1985). The latter, orig-
inally named by Murphy (1932), has been further sub-divided into,
from oldest to youngest, the Middle Park, Mountain Girl, Thorn-
dike and Wildrose sub-members (Fig. 41.1). Diamictic strata
occur throughout the Limekiln Spring Member, the Surprise
Member and the Wildrose Sub-member. With the exception of
the Limekiln Spring and Surprise members, which are lithologi-
cally variable, the units of the Kingston Peak Fm. are laterally
Fig. 41.1. Composite stratigraphic column
of Neoproterozoic units in the Death Valley
region (Stewart 1970; Wright et al. 1974;
Labotka et al. 1980; Heaman & Grotzinger
1992), showing the age constraints and key
markers discussed in the text. The left-hand
column shows a compilation of published
C-isotopic data discussed in text.
persistent and can be traced along most of the length of the Pana-
mint Range.
The Limekiln Spring Member
The Limekiln Spring Member is the most stratigraphically
complex member of the Kingston Peak Fm. in the Panamint
Range. Despite continuity of exposure of the Kingston Peak Fm.
as a whole, detailed mapping of this member is limited to a
small area in the southern part of the range and a somewhat
larger area in the vicinity of Telescope Peak (Fig. 41.2). Conse-
quently, the stratigraphic architecture of the Limekiln Spring
Member is poorly understood. In southern exposures, which are
not strongly deformed or metamorphosed, it consists of laterally
and vertically variable, interbedded diamictite, sandstone and
siltstone (Miller et al. 1988). In northern exposures, which are
strongly deformed and metamorphosed, it is mostly immature
sandstone, pelitic schist, amphibolite, minor dolomitic marble
 KINGSTON PEAK FORMATION, PANAMINT RANGE
and discontinuous lenses of metaconglomerate and breccia
(Labotka 1978; Labotka et al. 1980; Miller 1985). Even though
structural thicknesses are highly variable, ranging from 0 to
greater than 1000 m, at the scale of the Panamint Range, the Lime-
kiln Spring Member thins eastward beneath the overlying Surprise
Member. At least part of the thickness variability can be attributed
to the filling of pre-Limekiln-Spring palaeotopography formed on
the underlying crystalline basement (Labotka 1978; Labotka et al.
1980; Miller 1985).
461
Fig. 41.2. (a) Map showing distribution of
the Kingston Peak Fm. in the Death Valley
region. (b) Map showing distribution of the
four members of the Kingston Peak Fm. and
the Noonday Dolomite in the Panamint
Range.
The Surprise Member
The Surprise Member is poorly bedded to massive diamictite in the
southern Panamint Range and appears to pass northward into finer-
grained facies consisting of argillite and immature sandstone
(Labotka et al. 1980; Miller 1985). Locally, it contains a few
tens of metres of tholeiitic pillow basalt (Hammond 1983). The
Surprise Member ranges from 35 m in thickness near Goler
Wash in the southern Panamint Range (Miller 1985) to as much
462 R. PETTERSON ET AL.
as 1300 m west of Telescope Peak, although this may be partly the
result of structural thickening (Labotka et al. 1980). In the central
and southern part of the Panamint Range, the Surprise Member is
mostly massive diamictite dominated by quartzite and carbonate
clasts supported by an argillaceous, sandy matrix; thin grey to
dark grey argillite beds are present locally, but these are minor.
Clasts range from pebbles to boulders, with the quartzite clasts
generally rounded to sub-rounded whereas the carbonate clasts
are sub-angular (Miller 1985). North of the latitude of Telescope
Peak, diamictite is uncommon and the dominant lithology is
dark grey to black argillaceous siltstone and fine sandstone. In
addition, there is a significant amount of interbedded laminated
limestone and argillaceous limestone (Labotka et al. 1980).
The transition between the diamictite-dominated facies in the
southern and central parts of the Panamint Range and the argillite-
dominated, diamictite-poor facies in the northern part of the range
is structurally complex (Labotka 1978; Labotka et al. 1980). This
leads to uncertainty as to whether the two facies are coeval.
The Sourdough Member
The Sourdough Member is a 0.5 –45-m-thick, grey to dark grey
laminated limestone. It commonly exhibits strong internal defor-
mation, inhibiting confident identification of primary structures
(Miller 1985). Laminations are defined by changes in calcite
grain size and abundance of muscovite and quartz, and, in areas
of higher metamorphic grade, finely disseminated graphite. Syn-
sedimentary deformation, isolated clasts, and lenticular beds of
clastic material have been reported (Miller 1985). Typically, car-
bonate beds become interbedded with thin argillite beds upwards
from the base of the member.
South Park Member (exclusive of the Wildrose Sub-member)
Below the Wildrose Sub-member, the South Park Member
averages c. 300 m in thickness and contains three lithologically
distinct units that are mappable over the entire outcrop belt in
the Panamint Range. They include 150–200 m of sandstone and
pelite of the Middle Park Sub-member, 50–100 m of quartzite
and conglomerate of the Mountain Girl Sub-member, and 50–
200 m of limestone and dolostone of the Thorndike Sub-member
(formerly referred to as the ‘un-named’ Limestone; Prave 1999;
Corsetti & Kaufman 2003). Due to the details of stratigraphic
architecture both within and above these units, thickness variations
and stratigraphic omissions of one or more units are common. For
example, in the Wildrose Canyon area, the base of the Mountain
Girl Sub-member defines an angular unconformity (c. 158) and
variably sits on the Middle Park Sub-member, then on the Sour-
dough and Surprise members (as discussed by Petterson 2009).
In the Telescope Peak area, an erosion surface at the base of the
Wildrose Sub-member locally omits the entire lower portion of
the South Park Member (Miller 1987), although as discussed
below, this surface does not appear to be an angular unconformity.
Wildrose Sub-member of the South Park Member
The Wildrose Sub-member is variably preserved at the top of the
South Park Member and below the Noonday Dolomite. Where
present, it is a massive, matrix-supported diamictite. Its thickness
varies from 0 to 190 m (Miller 1985), but it is typically only a few
tens of metres thick. The diamictite in the Wildrose Sub-member is
distinct from that in the Limekiln Spring-Surprise members in that
it typically contains abundant gneiss clasts. However, there is a sig-
nificant internal variation in both clast and matrix compositions
such that, locally, Wildrose diamictite can be dominated by clasts
and matrix derived from the immediately underlying substrate.
Noonday Dolomite
In contrast to the siliciclastic-dominated and diamictic-bearing
Kingston Peak Fm., the overlying Noonday Dolomite is predomi-
nately carbonate. In the Panamint Range, the Noonday Dolomite
has been subdivided into three members, from oldest to youngest,
the Sentinel Peak, Radcliff, and Redlands (Murphy 1932; Johnson
1957; Labotka et al. 1980; Albee et al. 1981). Work by the authors
has modified this nomenclature to include a fourth member, the
Mahogany Flats Member, which lies between the Radcliff
Member and strata equivalent to the type Redlands Member
(Petterson 2009); only the first three members will be discussed
here.
The Sentinel Peak Member is an easily identifiable carbonate
marker unit exposed throughout the Panamint Range. It ranges
from 0.5 m to as much as 70 m in thickness and, in its most com-
plete development, can be sub-divided into three parts. Both the
lower and upper parts consist of laminated dolostone and each is
typically 1– 5 m thick. The middle part makes up the remaining
thickness and is generally fine-grained dolostone with irregular
spar-filled vugs, locally containing mound-like features containing
tube structures (Hunt & Mabey 1966; Cloud et al. 1974; Corsetti &
Grotzinger 2005). In many places, the middle part is absent and the
upper and lower parts are amalgamated to form a single carbonate
interval. In several localities where this occurs, a thin (as much as a
few metres thick) intraformational breccia occurs interbedded
within the laminated dolostone. We want to highlight that in
earlier reports, in areas where the Wildrose Sub-member is
absent, the Thorndike Sub-member and the Sentinel Peak
Member were inadvertently mapped together, and their combined
thickness was incorrectly assigned to the latter (e.g. the mapping of
Albee et al. 1981 in the Telescope Peak area includes both units as
Sentinel Peak Member).
The Radcliff Member is highly variable in both thickness and
lithology. It ranges from 100 to 250 m in thickness, and includes
limestone, dolostone, siltstone, sandstone and carbonate breccia.
All lithologies are thin-bedded, a characteristic feature of the Rad-
cliff Member. Its lower part is typically siliciclastic and consists of
laminated argillite and/or argillaceous arkose. These lithologies
pass gradationally upward to more abundant limestone with a dis-
tinctive pale yellowish orange colour; metre-thick layers and
lenses of intraformational breccia occur locally. The top of the
member includes quartz sandstone overlain by siltstone containing
coarse carbonate breccia.
The Mahogany Flats Member is primarily stromatolitic dolo-
stone. It is about 200 m thick and exhibits a gradually increasing
component of quartz sand in the upper half of the unit. Microbial
structures are ubiquitous, and include (i) metre-scale mound struc-
tures; (ii) laterally linked heads; and (iii) branching columnar
stromatolites. Intermound fill includes both carbonate and quartz
sandstone.
Glaciogenic deposits and associated strata
Diamictic units of the Surprise Member and the Wildrose Sub-
member are usually a few tens to a few hundred metres thick.
One of the remarkable aspects of these units is their lateral persist-
ence, which permits them to be traceable along the strike of the
Panamint Range for at least 50–100 km. Clasts in both of the dia-
mictic intervals range from pebbles to cobbles in size, although
larger clasts up to 0.5 m are common and can be as much as 3 m
in maximum dimension (Miller 1987). Clast shapes vary from
rounded to angular for all lithologies (although basement and
quartzite clasts tend to be mostly the former whereas carbonate
clasts are typically the latter), and rare bullet-shaped quartzite
and/or basement clasts can be found. In general, the diamictic
units of the Surprise Member tend to be dominated by clasts
derived from the underlying Pahrump Group sedimentary rocks
 KINGSTON PEAK FORMATION, PANAMINT RANGE
with the basement rocks being only minor. The opposite pro-
portion is generally the case for the clast composition in the Wild-
rose Sub-member, although clast composition can be nearly 100%
of the lithology of the immediately underlying substrate. The
matrix for the Surprise Member is a greenish grey argillite to argil-
laceous sandstone. This is in contrast to the Wildrose Sub-member
in which the matrix varies from dark grey to black argillite and
arkosic fine sandstone where basement clasts predominate, to
carbonate-cemented medium to coarse greywacke sandstone
where carbonate clasts (generally derived from the Thorndike Sub-
member) predominate. The rarity of any type of layering is striking
in both the Surprise and Wildrose diamictic units; the diamictite is
largely massive, grading is poorly developed at best, and finer-
grained, laterally discontinuous interbeds are rare. It is in the
latter intervals that rare lonestones and clasts that have pierced
laminae can be observed.
Boundary relations with overlying and underlying
non-glacial units
Boundary relations of the Limekiln Spring/Surprise Members
In the Panamint Range, the Kingston Peak Fm. variably overlies
the Crystal Spring and Beck Spring formations and crystalline
basement. In the southern part of the range, the Limekiln Spring
Member sits directly on the crystalline basement (Miller 1985).
Throughout much of the rest of the Panamint Range, the base of
the Kingston Peak Fm. is not exposed and/or is modified by intru-
sion of Mesozoic and Cenozoic granitoids. In the Telescope Peak
area, the base of the Kingston Peak Fm. is obscured by metamorph-
ism and structural complexity, which hinders assessment of the
nature of the contact. The unit underlying the Kingston Peak Fm.
in this area is a dolomitic marble correlated with the Beck
Spring Dolomite (Albee et al. 1981). The contact between them
purportedly displays interfingering (Labotka et al. 1980, fig. 2a).
However, in areas of lesser structural and metamorphic complex-
ity, the base of the Kingston Peak Fm. is observed to be uncon-
formable, including an angular discordance with underlying
units of the Pahrump Group, and in other places, resting noncom-
formably on the crystalline basement (Labotka et al. 1980).
The contact between the Surprise and the Sourdough members is
sharp. However, in a number of places, the contact is reportedly
marked by interbedded diamictite and limestone (Miller 1985);
these occurrences, however, may be due to structural interleaving.
Boundary relations of the Wildrose Sub-member
The base of the Wildrose Sub-member in the Panamint Range is
sharp and erosive. Its substrate ranges from the Thorndike Sub-
member to crystalline basement (Miller 1987, fig. 2). In the Tele-
scope Peak area, the base of the Wildrose Sub-member cuts down
section (from west to east) from the Thorndike Sub-member to the
Surprise Member (Miller 1987). In the southern part of the range, it
cuts down section from the Thorndike Sub-member to basement
(Miller 1987; Prave 1999). Despite the diversity of substrates
along this contact, there does not appear to be a significant
angular discordance between pre-Wildrose units and the lower
part of the Noonday Dolomite. Rather, as documented by Miller
(1987, figs 3 & 5), erosion of the pre-Wildrose units is
accompanied by an increase in thickness of the lower parts of
the Wildrose Sub-member, which reflects infilling of pre-Wildrose
topographic lows, some with as much as 300 m of local relief.
The top of the Kingston Peak Fm. in the Panamint Range is
defined by the occurrence of the Sentinel Peak Member of the
Noonday Dolomite. The contact is sharp, and the depositional
substrate is either the Wildrose or Thorndike sub-members,
463
although the Sentinel Peak Member locally sits on older units (Pet-
terson 2009).
Chemostratigraphy
The Kingston Peak Fm. in the Panamint Range contains two main
intervals of carbonate rocks, the Sourdough Member and Thorn-
dike Sub-member; minor occurrences of carbonate-bearing strata
also occur in parts of the Limekiln Spring, Surprise and South
Park Members. In this discussion, we will also include the over-
lying Sentinel Peak Member of the Noonday Dolomite (for discus-
sion of the chemostratigraphy of the underlying Beck Spring Fm.
in the eastern Death Valley region, see Mrofka & Kennedy 2011).
Although all these units have undergone variable metamorph-
ism, the C-isotopic data are consistent from section to section
and the trends are reproducible, independent of sedimentary
facies and the degree of metamorphism (Petterson 2009).
Further, the values obtained from the rocks in the Panamint
Range are nearly identical to those obtained by other workers in
the correlative, non-metamorphosed sections in eastern Death
Valley (Prave 1999; Corsetti & Kaufman 2003; Petterson 2009).
Thus, the carbonate C-isotopic data can be confidently interpreted
as recording depositional values, and is therefore useful for both
intrabasinal and global correlations.
The Sourdough Member in the Redlands Canyon area of the
southern Panamint Range yields d13C values between –2.6‰
and –1.1‰, although a single þ2.2‰ value is reported from the
upper part of the unit (Prave 1999). Data from the Goler Wash
and Pleasant Canyon areas provide d13C values ranging from
–2.4‰ to –1.6‰ and –3.1‰ to –2.9‰, respectively (Corsetti
& Kaufman 2003).
Compared to the other carbonate units in the Kingston Peak
Fm., the d13C values of the Thorndike Sub-member are strongly
positive. Values range from þ5.7‰ to þ6.3‰ in the Redlands
Canyon area (southern Panamint Range, Prave 1999) and
þ4.7‰ to þ5.3‰ in the Tucki Mountain area (northern Panamint
Range, Corsetti & Kaufman 2003).
d13C values for the Sentinel Peak Member in the Panamint
Range are between þ1.0‰ and – 3.0‰ (Petterson 2009). These
values match well those for the lower unit of the Noonday Dolo-
mite (i.e. the tube-bearing dolostone) in the eastern Death Valley
region, which has values near – 3.0‰ (Prave 1999; Corsetti &
Kaufman 2003), and recent work by Petterson (2009), to be pub-
lished elsewhere, has shown that this unit is correlative with the
Sentinel Peak Member.
Other characteristics
A number of Neoproterozoic microfossil assemblages have been
described from the Beck Spring and Kingston Peak formations
in the eastern Death Valley region (Pierce & Cloud 1978;
Horodyski & Mankiewicz 1990; Awramik et al. 2000; Corsetti
et al. 2003). However, no fossils have been reported from
Pahrump Group strata in the Panamint Range. This lack of
fossils is not surprising given both the degree of metamorphism
and the general scarcity of fossils elsewhere.
Marginally economic levels of uranium (as secondary branner-
ite, UTi2O6) occur in a graphitic schist unit of the Limekiln Spring
Member and sub-economic levels of uranium and thorium (as det-
rital monazite) are found in quartz pebble conglomerates in the
lower part of the Mountain Girl Sub-member (Carlisle et al.
1980). Iron ore has been mined from the Kingston Peak Fm. in
eastern Death Valley, but no such deposits occur in the Panamint
Range. Although many intrusion-related precious metal deposits
are present in the Panamint Range, no economically viable depos-
its directly related to depositional processes or diagenesis
are known from the Kingston Peak Fm. in the Panamint Range.
464 R. PETTERSON ET AL.
Palaeolatitude and palaeogeography
No palaeolatitude estimates have been determined from any of the
Kingston Peak Fm. rocks, either in the Panamint Range or in cor-
relative strata from eastern Death Valley. Numerous attempts have
been made to isolate primary magnetizations, but have proved
unsuccessful, mostly displaying Mesozoic overprinting (J. Kirsch-
vink 2007, pers. comm.).
Geochronological constraints
The geochronological constraints that exist for the Precambrian
sedimentary succession in Death Valley are robust, but very
sparse. The maximum age of the section comes from diabase
sills in the Crystal Spring Fm., which also occur as clasts in the
Limekiln Spring Member. Samples of the sills from two locations
in eastern Death Valley yield U –Pb baddeleyite ages of
1087 + 3 Ma and 1069 + 3 Ma (Heaman & Grotzinger 1992).
The Precambrian –Cambrian boundary occurs nearly 5 km above
the Crystal Spring Fm., within the lower member of the Wood
Canyon Fm. (Corsetti & Hagadorn 2000). Chemostratigraphic
and lithological correlations to well-dated sections in Namibia
(Hoffmann et al. 2004) may indicate an age of 635 Ma for the
base of the Noonday Dolomite (Prave 1999; Petterson 2009).
Discussion
As noted previously, the palaeogeographic setting of the Kingston
Peak Fm. is inferred to be one in which episodic extensional tec-
tonism influenced basinal development, replacing the relatively
stable carbonate platform setting of the underlying Beck Spring
Dolomite with the predominantly clastic deposition of the King-
ston Peak Fm. (Wright et al. 1974; Labotka et al. 1980). The King-
ston Peak Fm. in the Panamint Range does contain evidence for
deposition occurring along a surface that had pre-existing palaeo-
topography, which was infilled during deposition of the Limekiln
Spring and Surprise members. For example, in the Telescope Peak
area, a subaerially exposed basement high (the ‘World Beater
Island’ of Labotka et al. 1980) was postulated to have existed
during lower Kingston Peak time. However, the uniform character
and widespread development of the overlying Sourdough Member
and Middle Park Sub-member implies a more uniform substrate
and period of relative tectonic quiescence. Subsequently, a
renewed phase of tectonism during Mountain Girl Sub-member
time resulted in tilting and development of an angular unconfor-
mity with associated locally developed coarse-grained facies (Pet-
terson 2009). This gave way to a setting where widespread
development of carbonate strata occurred, as evident from the
Thorndike Sub-member. It was during these phases of relative tec-
tonic quiescence that the Wildrose Sub-member was deposited and
overlain by the Noonday Dolomite.
Evidence for a direct influence of glaciation in the Kingston
Peak Fm. in the Panamint Range is sparse. A glaciogenic origin
has been suggested for parts of the Limekiln Spring Member, the
Surprise Member and the Wildrose Sub-member based on the pres-
ence of laterally extensive and uniformly massive diamictic units,
the occurrence of rare lonestones and bullet-shaped and striated
clasts, and the even rarer presence of clasts that can be observed
to pierce laminae. In the eastern Death Valley sections, striated
and faceted lonestones and dropstones are more abundant
(Troxel 1982, p. 62; Miller 1985; Miller et al. 1988, fig. 6;
Mrofka & Kennedy 2011). However, in both regions, no direct
influence of ice on deposition has yet been documented, such as
striated pavements at the base of diamictite units or glaciotectonic
deformation features. Nevertheless, due to the difficulty of gener-
ating sediment gravity flows of the dimensions shown by the dia-
mictic units by non-glacial mechanisms, deposition is presumably
glacially influenced, either as glacimarine deposition, or in the case
of the Wildrose, perhaps as lodgement till (Miller 1985).
A critical line of evidence that enables linking the Kingston
Peak Fm. to Cryogenian strata elsewhere is the presence of two
carbonate units that overlie each of the diamictite-bearing units,
the Sourdough Member of the Kingston Peak Fm. and the Sentinel
Peak Member of the Noonday Dolomite. Although direct age
control is lacking, the lithological and chemostratigraphical
characteristics of these two units permit correlations to older and
younger Crygenian cap carbonates known elsewhere (Prave
1999; Petterson 2009). If correct, then the strongly positive d13C
values recorded in the Thorndike Sub-member suggest temporal
correlation with the Trezona anomaly (Halverson et al. 2005).
The timing of the older Cryogenian glacial– cap carbonate
couplet (previously commonly referred to as ‘Sturtian’) remains
unresolved (ages range from c. 750 to c. 643 Ma, and may
consist of multiple glaciations, Kendall et al. 2009) but the
younger one (previously commonly referred to as ‘Marinoan’) is
now known to be c. 635 Ma in age (Hoffmann et al. 2004).
This global correlation suggests that the Kingston Peak Fm.
spans a significant interval of time without deposition of diamictite
or other glaciogenic rocks, as evident from the c. 300-m-thick
South Park Member (exclusive of the Wildrose Sub-member).
This permits defining at least two distinct intervals of glaciogenic
sedimentation, between which deposition of braided fluvial to
shallow marine carbonate units took place (Petterson 2009). In
the eastern Death Valley sections, the South Park strata are not
recognized, thereby raising the inevitable questions as to how
the sections in eastern Death Valley correlate to those in the
Panamint Range.
Alternative global correlations have also been proposed for the
Kingston Peak Fm. In one interpretation, the Beck Spring Fm. is
considered to be an older cap carbonate (Halverson et al. 2005),
implying that the entire Kingston Peak is part of the younger Cryo-
genian glacial. In another, the Kingston Peak Fm.– Sentinel Peak
Member is inferred to be the older Cryogenian glacial–cap carbon-
ate couplet (Corsetti et al. 2007). All these ideas are currently
being tested.
We acknowledge support for this research from NSF grants EAR-01-07123 and
EAR-03-10413. This represents a contribution of the IUGS- and UNESCO-
funded IGCP (International Geoscience Programme) project #512.
References
Albee, A. L., Labotka, T. C., Lanphere, M. A. & McDowell, S. D.
1981. Stratigraphy, structure, and metamorphism in the central Pana-
mint Mountains (Telescope Peak quadrangle), Death Valley area,
California: Summary. Geological Society of America Bulletin, 91,
125– 129.
Awramik, S. M., Corsetti, F. A. & Shapiro, R. 2000. Stromatolites and
the pre-Phanerozoic to Cambrian history of the area south east of
Death Valley. Bulletin of the San Bernardino County Museum, 47,
65 – 74.
Burchfiel, B. C. & Davis, G. A. 1972. Structural framework and evol-
ution of the southern part of the Cordilleran orogen, western United
States. American Journal of Science, 272, 97 –118.
Carlisle, D., Kettler, R. M. & Swanson, S. C. 1980. Uranium- and
thorium-bearing facies of the late Proterozoic Kingston Peak For-
mation, Death Valley region, California. In: Fife, D. L. & Brown,
A. R. (eds) Geology and Mineral Wealth of the California Desert.
South Coast Geological Society, Santa Ana, CA, 31 –51.
Cloud, P., Wright, L. A., Williams, E. G., Diehl, P. E. & Walter, M. R.
1974. Giant stromatolites and associated vertical tubes from the Upper
Proterozoic Noonday Dolomite, Death Valley region, Eastern Califor-
nia. Geological Society of America Bulletin, 85, 1869–1882.
Corsetti, F. A. & Grotzinger, J. P. 2005. Origin and significance of tube
structures in Neoproterozoic post-glacial cap carbonates: Example
from Noonday Dolomite, Death Valley, United States. Palaois, 20,
348– 362.
 KINGSTON PEAK FORMATION, PANAMINT RANGE
Corsetti, F. A. & Hagadorn, J. W. 2000. Precambrian – Cambrian tran-
sition: Death Valley, United States. Geology, 28, 299– 302.
Corsetti, F. A. & Kaufman, A. J. 2003. Stratigraphic investigations
of carbon isotope anomalies and Neoproterozoic ice ages in Death
Valley, California. Geological Society of America Bulletin, 115,
916– 932.
Corsetti, F. A., Awramik, S. M. & Pierce, D. 2003. A complex micro-
biota from snowball Earth times: microfossils from the Neoprotero-
zoic Kingston Peak Formation, Death Valley, USA. Proceedings of
the National Academy of Sciences of the United States of America,
100, 4399–4404.
Corsetti, F. A., Link, P. K. & Lorentz, N. J. 2007. d13C chemostratigra-
phy of the Neoproterozoic succession near Pocatello, Idaho, U.S.A.:
implications for glacial chronology and regional correlations. In:
Link, P. K. & Lewis, R. S. (eds.) Proterozoic Geology of Western
North America and Siberia. Society for Sedimentary Geology
Special Publication, 86, 193–205.
Halverson, G. P., Hoffman, P. F., Schrag, D. P., Maloof, A. C. & Rice,
A. H. N. 2005. Toward a Neoproterozoic composite carbon-isotope
record. Geological Society of America Bulletin, 117, 1181–1207.
Hammond, J. L. G. 1983. Late Precambrian diabase intrusions in the
southern Death Valley region, California; their petrology, geochem-
istry, and tectonic significance. PhD thesis, University of Southern
California.
Heaman, L. M. & Grotzinger, J. P. 1992. 1.08 Ga Diabase sills in
the Pahrump Group, California: implications for development of
the Cordilleran miogeocline. Geology, 20, 637– 640.
Hewett, D. F. 1940. New formation names to be used in the Kingston
Range, Ivanpah Quadrangle, California. Journal of the Washington
Academy of Sciences, 30, 239–240.
Hodges, K. V., McKenna, L. W. et al. 1989. Evolution of extensional
basins and Basin and Range topography west of Death Valley,
California. Tectonics, 8, 453–467.
Hoffmann, K.-H., Condon, D. J., Bowring, S. A. & Crowley, J. L.
2004. U– Pb zircon date from the Neoproterozoic Ghaub Formation,
Namibia: constraints on Marinoan glaciation. Geology, 32, 817– 820.
Horodyski, R. J. & Mankiewicz, C. 1990. Possible Late Proterozoic
skeletal algae from the Pahrump-Group, Kingston Range, southeast-
ern California. American Journal of Science, 290A, 149– 169.
Hunt, C. B. & Mabey, D. R. 1966. General geology of Death Valley,
California; stratigraphy and structure. US Geological Survey Pro-
fessional Paper, 494-A, 162.
Johnson, B. K. 1957. Geology of a part of the Manly Peak Quadrangle,
southern Panamint Range, California. University of California Publi-
cations in Geological Sciences, 30, 353– 423.
Kendall, B., Creaser, R. A., Calver, C. R., Raub, T. D. & Evans,
D. A. D. 2009. Correlation of Sturtian diamictite successions in
southern Australia and northwestern Tasmania by Re– Os black
shale geochronology and the ambiguity of ‘sturtian’-type diamic-
tite – cap carbonate pairs as chronostratigraphic marker horizons.
Precambrian Research, 172, 301– 310.
Labotka, T. C. 1978. Geology of the Telescope Peak Quadrangle, Cali-
fornia and late Mesozoic regional metamorphism, Death Valley
area, California. PhD thesis, California Institute of Technology.
Labotka, T. C., Albee, A. L., Lanphere, M. A. & Mcdowell, S. D.
1980. Stratigraphy, structure, and metamorphism in the Central
Panamint Mountains (Telescope-Peak Quadrangle), Death-Valley
Area, California – summary. Geological Society of America Bulletin,
91, 125– 129.
Labotka, T. C., Warasila, R. L. & Spangler, R. R. 1985. Polymeta-
morphism in the Panamint Mountains, California: a Ar39 –Ar40
study. Journal of Geophysical Research – Solid Earth and Planets,
90, 359– 371.
465
Levy, M. & Christie-Blick, N. 1991. Tectonic subsidence of the early
Palaeozoic passive continental margin in eastern California and
southern Nevada. Geological Society of America Bulletin, 103,
1590–1606.
Miller, J. M. G. 1985. Glacial and syntectonic sedimentation; the upper
Proterozoic Kingston Peak Formation, southern Panamint Range,
eastern California. Geological Society of America Bulletin, 96,
1537–1553.
Miller, J. M. G. 1987. Palaeotectonic and stratigraphic implications
of the Kingston Peak-Noonday contact, Panamint Range, eastern
California. Journal of Geology, 95, 75– 85.
Miller, J. M. G., Troxel, B. W. & Wright, L. A. 1988. Stratigraphy and
palaeogeography of the Proterozoic Kingston Peak Formation, Death
Valley region, eastern California. In: Gregory, J. L. & Baldwin,
E. J. (eds) Geology of the Death Valley Region. South Coast Geologi-
cal Society, Santa Ana, CA, 118–142.
Mrofka, D. & Kennedy, M. 2011. The Kingston Peak Formation in
the eastern Death Valley Region. In: Arnaud, E., Halverson,
G. P. & Shields-Zhou, G. (eds) The Geological Record of Neopro-
terozoic Glaciations. Geological Society, London, Memoirs, 36,
449– 458.
Murphy, F. M. 1932. Geology of a part of the Panamint Range,
California, Report XXVII of the State Mineralogist. California
Department of Natural Resources, Division of Mines and Geology,
28, 329– 356.
Petterson, R. 2009. I. Glaciogenic and related strata of the Neopro-
terozoic Kingston Peak Formation in the Panamint Range,
Death Valley region, California. II. The basal Ediacaran Noonday
Formation, and implications for Laurentian equivalents. III. Rifting
of southwest Laurentia during the Sturtian-Marinoan intergla-
cial: The Argenta orogeny. PhD thesis, California Institute of
Technology.
Pierce, D. & Cloud, P. 1978. New microbial fossils from 1.3 billion-
year-old rocks of eastern California. Geomicrobiology Journal, 1,
295– 309.
Prave, A. R. 1999. Two diamictites, two cap carbonates, two d13C excur-
sions, two rifts: the Neoproterozoic Kingston Peak Formation, Death
Valley, California. Geology, 27, 339–342.
Stewart, J. H. 1970. Upper Precambrian and Lower Cambrian strata in
the southern Great Basin, California and Nevada. US Geological
Survey Professional Paper, 620, 206.
Stewart, J. H. 1972. Initial deposits in Cordilleran geosyncline: evidence
of a Late Precambrian (less than 850 My) continental separation.
Geological Society of America Bulletin, 83, 1345– 1360.
Troxel, B. W. 1982. Description of the uppermost part of the
Kingston Peak Formation, Amargosa Rim, Canyon, Death
Valley region, California. In: Cooper, J. D., Troxel, B. W. &
Wright, L. A. (eds) Geology of Selected Areas in the San Bernardino
Mountains, Western Mojave Desert, and Southern Great Basin,
California. The Death Valley Publishing Company, Shoshone, CA,
61– 70.
Wernicke, B., Snow, J. K., Hodges, K. V. & Walker, J. D. 1993.
Structural constraints on Neogene tectonism in the southern
Great Basin. In: Lahren, M. M., Trexler, J. H. Jr. & Spinosa, C.
(eds) Crustal Evolution of the Great Basin and Sierra Nevada.
Geological Society of America Cordilleran/Rocky Mountain
Sections Meeting Fieldtrip Guidebook, University of Nevada,
Reno, 453– 479.
Wright, L. A., Troxel, B. W., Williams, E. G., Roberts, M. T. &
Diehl, P. E. 1974. Precambrian sedimentary environments of the
Death Valley region, eastern California. Guidebook, Death Valley
Region, California and Nevada. Death Valley Publishing Company,
Shoshone, CA, 27 –36.
 Chapter 42

The deep-marine glaciogenic Gaskiers Formation, Newfoundland, Canada

SHANNON L. CARTO* & NICK EYLES

Department of Geology, University of Toronto, Toronto, Ontario, M1C 1A4 Canada
*Corresponding author (e-mail: shannon.carto@gmail.com)

Abstract: In eastern Canada, the Neoproterozoic Gaskiers Formation (Fm.) consists of a thick diamictite– turbidite succession (250–
300 m thick) that occurs within the deep-marine predominantly volcaniclastic turbidite units of the Conception Group (4– 5 km thick).
These rocks are well exposed on the coast of the Avalon Peninsula in eastern Newfoundland. The thick succession is considered by some
to represent the final Neoproterozoic glacial event, known as the Gaskiers Glaciation c. 582– 585 Ma. The Gaskiers Fm. appears to have
accumulated in a volcanically-influenced arc-related basin within the confines of the Proterozoic peri-Gondwanan Avalon terrane, before
rifting away from the Gondwanan margin in the early Palaeozoic, but it is not clear whether it was adjacent to the West African Craton or
the Amazon Craton. It has also been argued that Avalon had already rifted from Gondwana by the late Proterozoic.
Glacially striated clasts, dropstones, and chatter-marked garnets identify a glaciated source area, and clasts and matrix are of volcanic
origin; pyroclastic flows and volcanic bombs are also present. As a result, there is a growing consensus that the Gaskiers Fm. records local
glaciation of a high relief volcanic topography, owing its origin to the episodic downslope reworking of volcanic and glacial debris into a
deep, rapidly subsiding basin. The volcanic nature of these deposits implies that local volcanic activity was coeval with deposition. A
minimum age of the Gaskiers Fm. is constrained by a diverse assemblage of Ediacaran-type fossils in the upper Conception Group
dated at c. 565 + 3 Ma and in the lower St. John’s Group, which overlies the Conception Group. A maximum age for the Gaskiers is
provided by a U –Pb date of 606 þ3.7/ –2.9) Ma in the Harbour Main volcanic rocks underlying the Gaskiers Fm. A thin limestone
bed has also been identified at two localities in Conception Bay directly overlying the diamictite yielding strongly negative d13C values.

The Neoproterozoic Gaskiers Fm. is a 250 –300-m-thick succes-
sion of glaciogenic diamictite with interbedded turbidite units,
which crop out on the coast of the southern Avalon Peninsula of
eastern Newfoundland, Canada (Fig. 42.1). The Gaskiers Fm.
occurs towards the base of the 4– 5-km-thick succession of deep-
marine volcaniclastic sedimentary rocks belonging to the Con-
ception Group (Fig. 42.2a; Williams & King 1975). There are
excellent cliff exposures south of Cape St. Francis, along the
southern margin of Conception Bay, in the south on the eastern
shoreline of St. Mary’s Bay, and on the Colinet Islands. Outcrops
in the St. Mary’s Bay area in the vicinity of Double Road Point are
widely and informally regarded as the type section (Figs 42.1 &
42.2b). There, the diamictite comprises more than 80% of the
total thickness of the exposed section, and is interbedded with tur-
bidite units containing dropstones as well as thin volcanic agglom-
erate units; volcanic bombs are also present (Fig. 42.2b; Eyles &
Eyles 1989; Myrow 1995). Within the Avalon terrane of New-
foundland, a number of other diamictite deposits have been con-
sidered to be potential time-equivalents to the Gaskiers Fm.,
such as within the Musgravetown Group, Rock Harbour Group
and Connecting Point Group but no evidence has yet been ident-
ified that these are coeval or glacially influenced (Myrow 1995).
The Gaskiers Fm. is commonly considered to represent the last
of the major Neoproterozoic glaciations known as the Gaskiers
Glaciation c. 582 –585 Ma (Bowring et al. 2003; Knoll et al.
2006). The Gaskiers diamictite units were labelled as ‘tilloids’
by McCartney (1967) in the sense of ‘non-glacial conglomeratic
mudstone resulting from extensive slides, slumps or mudflows’.
Bruckner & Anderson (1971) identified striated and faceted
pebbles and Williams & King (1979) discovered outsized drop-
stones within the interbedded siltstone and sandstone laminations,
demonstrating a glaciogenic source of debris. Other evidence of a
glacial origin rests on the presence of ‘chatter-marked’ garnets
(Gravenor 1980) and ‘flat-iron’, bullet-shaped clasts produced by
subglacial abrasion in the source area (Bruckner & Anderson
1971). Williams & King (1975) used the term ‘mixtite’, introduced
by Schermerhorn (1966) for a ‘coarse, poorly sorted clastic rock
without regard for composition or origin’ and interpreted the dia-
mictite units as largely the product of ice rafting. Gravenor
(1980) and King (1988) argued that the Gaskiers deposits were
debris flows composed of glaciogenic debris transported down-
slope from a nearby grounded ice margin. A detailed lithofacies
analysis by Eyles & Eyles (1989) similarly concluded that the
Gaskiers Fm. was deposited as debris flows and turbidites in a
deep-marine, volcanically influenced, mid-slope setting. This
interpretation drew on the depositional context provided by turbi-
dite strata present conformably within, above and below the
Gaskiers Fm. indicating a deep-marine slope setting (Gardner &
Hiscott 1988). Because of the reworked deep-marine nature of the
Gaskiers Fm., it is difficult to infer much of the parent glacial
environment onland other than to positively identify the presence
of wet-based glaciers on the peaks of an active volcanic arc.
The Gaskiers Fm. is well known for the associated diverse
assemblage of Ediacara-type fossils present through the upper
Drook and Mistaken Point formations dated at c. 565 + 3 Ma
(Benus 1988; Clapham & Narbonne 2002; Narbonne 2004) and
in the lower St. John’s Group with U –Pb dates from 565 to
543 Ma (Myrow 1995). Moreover, recent analysis of the Fe
content of the deep-marine sedimentary rocks of the Conception
Group and overlying St. John’s Group by Canfield et al. (2007)
reveals that anoxic conditions persisted in the deep ocean before
and during the deposition of the Gaskiers Fm., subsequently
becoming oxic afterward. Canfield and colleagues argue that this
is suggestive of a causal link between the evolution of Edicaran
biota and this oxygenation event.
Structural framework
The Avalon terrane of eastern Newfoundland, Canada, is com-
posed of c. 15-km-thick relatively unmetamorphosed Proterozoic
sedimentary, volcanic and plutonic rocks (c. 750–570 Ma)
locally overlain by Cambrian – Ordovician (excluding the Carbon-
iferous sedimentary rocks of the Burin Peninsula) sedimentary
rocks (King 1988). Murphy & Nance (1989) cited the close associ-
ation of calc-alkaline and rift-related continental volcanic rocks as
evidence that the Avalon terrane developed in a volcanic-arc
setting along the active continental margin of West Gondwana
(O’Brien et al. 1983; Nance et al.1991; McMenamin 1987;
Burrett et al. 1990; Forty & Cocks 2003). The abundance of

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 467– 473. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.42
468 S. L. CARTO & N. EYLES

(a)

Cape St. Francis

ay
nB
p tio
ce
n
Co
St. John’s (a)
Kitchuses

Avalon Harbour
Peninsula Main

Little
Colinet (b)
Island

Great
Colinet St. Mary’s
Island 46°55'

St. Mary’s Gaskiers

(c) Double Road Point
Bay Outcrop
D
0 km 20

M 40

St. Mary’s
47° 10' 90

(b)
M Mall Bay

Little
Colinet Gaskiers Point La Haye
Island
ne

Fig. 42.1. (a) General map of the Avalon

ticli
ne

Great
D Drook
ticli

Colinet
An

Island Peninsula in eastern Newfoundland, Canada

An

ce

V showing location of principal outcrops of

Pla
e Pt.

ing

the stratigraphic units of the Conception

Hay

ss

South
Cro

Group. (b) Map showing location of both

La

Point (c) Gaskiers

N
St. Mary’s
hidden and exposed outcrops of the
St. Mary’s
Bay Gaskiers Fm Gaskiers Fm. in the St. Mary’s Bay area,
Gaskiers Hidden
Holyrood Outcrop 90 Holyrood eastern Newfoundland, Canada. (c)
Pond Pond Enlargement of St. Mary’s Bay area
False 0 km 10 0 km 1 showing location of outcrops of Mall Bay,
Cape
St. Vincents Drook and Gaskiers formations on the east
46° 46' 46°52' side of St. Mary’s Bay (modified from Eyles
53°45' 53°20' 53°37' 53°34' & Eyles 1989).

bimodal Avalonian volcanic suites underlying the Conception inconsistencies observed in the Drook and Mistaken Point for-
Group, the high level of plutonic rocks and the abundance of ash mations exposed in the Bay Roberts and Spaniard’s Bay area, in
layers within the succession suggests the Gaskiers Fm. accumu- the north-central part of the Avalon Peninsula (palaeoflow direc-
lated in an arc-related basin (Myrow 1995). Possible basin types tion towards the east and SE) and exposures of these units at Mis-
include intra-arc (Dec et al. 1992), back-arc (Myrow et al. 1999) taken Point in southern Avalon (palaeoflow toward the NE). This
and fore-arc (Narbonne et al. 2001) basins. Based on petrological, topographical high is believed to be responsible for steering the
structural and geochronological data collected from the Con- turbidity currents in the NE direction along the axis of the basin
ception Group, it has been argued that the Gaskiers Fm. most throughout deposition of most of the Drook, Briscal and Mistaken
likely accumulated during the transition from a fore-arc (c. 630– Point formations in the southern Avalon Peninsula.
560 Ma) to strike –slip pull-apart basin (c. 560 –540 Ma) The rocks of the Avalon Peninsula were deformed during the
(Murphy et al. 1999; Nance et al. 2002; Wood et al. 2003). The late Proterozoic Avalon orogeny and subsequently deformed and
Fermeuse Fm. is thought to mark this change in tectonic regime metamorphosed during the mid-Palaeozoic Acadian orogeny
with the remainder of the St. John’s and Signal Hill groups (Hatcher & Goldberg 1991). The Acadian orogeny produced
accumulating in the strike –slip pull-apart basin. Support for this most of the folding, faulting and metamorphism recognized on
theory stems from the scarcity of volcanic ash in the Fermeuse the eastern Avalon Peninsula. The rocks of the Avalon Peninsula,
Fm. and higher stratigraphic units, indicating the cessation of including the diamictite units, have been metamorphosed under
volcanic activity expected if subduction no longer occurred. prehnite-pumpellyite facies, but sedimentary structures are very
Recent work by Ichaso et al. (2007) suggests that the present- well preserved (Papezik 1974). Outcrops of the Conception
day outcrops of the Harbour Main volcanic rocks may have been Group in the southern Avalon Peninsula on the east side of
a topographical high that segmented the Avalonian fore-arc St. Mary’s Bay have been folded into anticlines and synclines
region, creating a relatively confined sub-basin (termed the ‘west that have NE-trending axes (Eyles & Eyles 1989). The rocks of
Conception sub-basin’). This theory is based on palaeocurrent the Avalon Peninsula, including the Gaskiers Fm., are cleaved
 THE GASKIERS FORMATION
and cut by numerous faults, although the extent to which the
Gaskiers Fm. has been deformed varies throughout the region
(King 1980). Deformation from the Acadian Orogeny is over-
printed by compression from the Alleghanian Orogeny (290 –
250 Ma) in the Appalachian region (Osberg et al.1989; Rast 1989).
Stratigraphy
The Harbour Main Group (.1500 m thick) rests below the
Conception Group and consists of mafic to acidic lavas, pyroclastic
rocks with interbedded continental sedimentary rocks, intruded by
469
Fig. 42.2. Schematic representation of
stratigraphy of the eastern Avalon Peninsula
and a detailed log through the Gaskiers Fm.
and associated Mall Bay and Drook
formations at Double Road Point, east side
of St. Mary’s Bay (see Fig. 40.1c for
location of detailed log; modified from
Eyles & Eyles 1989).
the Holyrood plutonic complex of granite and gabbro (Anderson &
King 1981).
The Conception Group (4– 5 km thick) is divided into the Mall
Bay, Gaskiers, Drook, Briscal and Mistaken Point formations
(Fig. 42.2), but there is little lithofacies difference between these
siliceous and volcanogenic deep-water turbidite units, which
contain occasional tuff, rhyolite flows, agglomerates, mafic
pillow lavas and mafic dykes (King 1988). The deep-marine silic-
eous turbidite units of the Mall Bay Fm. (.800 m) are overlain by
the diamictite-bearing Gaskiers Fm. (250 –300 m), which in turn
is overlain by the siliceous turbidite sequence of the Drook Fm.
(1500 m) (King 1988). The Briscal Fm. (1200 m) is a grey to
470 S. L. CARTO & N. EYLES
green sandy thick-bedded unit that lies between the Drook and
Mistaken Point formations in the southeastern Avalon Peninsula
(Ichaso et al. 2007). Pebbly mudstone diamictites similar to
those of the Gaskiers Fm. occur in the overlying Drook and
Briscal formations, but these do not include striated clasts and
may be non-glaciogenic. The Mistaken Point Fm. (400 m) consists
of a deep-marine siliclastic succession of graded beds composed
of grey to pink sandstone units and green, red and purple shale
units (King 1988). The upper part of the formation is composed
of finely laminated light yellow to greenish grey tuffaceous
layers (King 1988). Fossiliferous horizons are present throughout
the formation.
The Conception Group is overlain conformably by the
8-km-thick molasses facies Hodgewater Group in the central
Avalon Peninsula and by the 8 –9-km-thick St. John’s and Signal
Hill groups to the east (Fig. 42.2). The St. John’s Group is domi-
nated by black shale, slate and silty sandstone. The Signal Hill
Group consists of grey sandstone and siltstone units at the base,
passing up into red sandstone and siltstone units, which are over-
lain by red conglomerate and sandstone units (O’Brien et al.
1983). These record shoaling of the basin prior to deposition of
extensive Cambrian –Ordovician shallow marine lithofacies
(Smith & Hiscott 1984; Myrow 1995).
Glaciogenic deposits and associated strata
Mall Bay Fm.
The Mall Bay Fm. (.800 m thick) consists of mostly thin-bedded
(10– 150 mm), fine-to-medium-grained, grey, brown and green
graded sandstone units (Eyles & Eyles 1989). These units are
well exposed at Point La Haye and can be traced to the NE as a con-
tinuous exposure up to the base of the overlying Gaskiers Fm. at
Double Road Point. Polygonal systems of clastic dykes, suggesting
rapid de-watering, occur on upper bedding surfaces of the Mall
Bay Fm. (Eyles & Eyles 1989). Palaeocurrent directions are
southerly, and slump folds and other soft-sediment deformation
structures are common (Eyles & Eyles 1989).
The Gaskiers Fm.
The Gaskiers Fm. is 250 –300 m thick and occurs conformably
between the Mall Bay and Drook formations (Fig. 42.2).
Detailed sedimentological analysis is provided by Eyles &
Eyles (1989). The diamictite is mostly massive and unsorted,
with a distinct planar tabular geometry typical of debris flows.
Some beds exhibit weak grading and are interbedded with thin
intraformational turbidite and tuff horizons (Anderson & King
1981). In some places, the diamictite shows subtle stratification,
suggesting the amalgamation of thinner beds. Beds of grey to
green thinly stratified and graded sandstone, siltstone and mud-
stone units separate diamictite beds, the latter of which are as
much as 40 m thick (Williams & King 1979; Eyles & Eyles
1989). These thinly stratified beds are continuous and can be
traced for several tens of metres along strike, although some
beds are extensively folded and show evidence of slumping
(Eyles & Eyles 1989). Clasts are commonly 30–100 mm in
diameter, but rare larger clasts up to 0.8 m in diameter have
also been observed (Myrow & Kaufman 1999). Clasts are com-
posed of a wide variety of igneous lithologies such as basalt,
granite, diorite and granophyre, as well as sedimentary litholo-
gies such as quartzite, siltstone and conglomerate fragments
(Eyles & Eyles 1989). Intraclasts of mudstone, laminated silt-
stone and carbonate are also present. Clast shape varies from
well-rounded to highly angular; striated and faceted clasts are
common. The diamictite is also characterized by large-scale
slump structures.
Drook Fm.
The Drook Fm. (1500 m) consists of a succession of medium- and
thick-bedded, fine-to coarse-grained, graded and massive sand-
stone units interbedded with graded argillite units, similar to
those of the Mall Bay Fm. (Eyles & Eyles 1989). Outcrops of
the Drook Fm. are widely exposed in eastern St. Mary’s Bay, Gask-
iers, St. Vincent’s and Cape English, but the outcrop at Double
Road Point is poor (Eyles & Eyles 1989; King 1988). Soft-
sediment deformation structures and slump folds are common
(Eyles & Eyles 1989). Beds vary in thickness from a few milli-
metres to 10 cm. Load-flame structures are moderately abundant,
and ash beds are present but in small amounts (Ichaso et al. 2007).
Associated volcanic rocks
Evidence of contemporaneous volcanic activity is present at
Double Road Point outcrop in the form of air-fall bombs, and
agglomerate and tuff layers. Beds of weathered volcanic agglom-
erates (,30 mm) occur within the thin-bedded, fine-grained sand-
stone units that are interbedded with the lower part of the
diamictite facies at this locality. Intercalations of tuff, as well as
a thin (,3 cm) bed of orange-weathering volcanic agglomerate
also occurs at Great Colinet Island (King 1980). A volcanic
air-fall bomb, volcanic blocks and lapilli also occur within this
section (Anderson & King 1981; Eyles & Eyles 1989). Volcanic
ash beds, ranging from 0.2 to 30 cm in thickness, have been ident-
ified in low abundance in the Drook Fm. and in high abundance
in the Mistaken Point Fm., exposed in the coastal sections in the
Spaniard’s Bay –Bay Roberts area (Williams & King 1979;
Ichaso et al. 2007). Crystal tuffs are also present in the Trepassey
Fm., but are thinner and less common (Wood et al. 2003).
Associated carbonate
Myrow & Kaufman (1999) reported a thin (,50 cm) limestone
bed and concretions overlying a thick diamictite succession
exposed discontinuously at Harbour Main and Kitchuses sites
in south Conception Bay. No limestone beds occur at the
St. Mary’s Bay outcrops of the Gaskiers Fm. (or at other outcrops
on Little Colinet Island). The limestone at both localities have a
mottled nodular fabric with no preserved primary sedimentary or
biogenic structures, and are locally overprinted with tectonic
fabrics (Myrow & Kaufman 1999). These outcrops were tenta-
tively assigned to the Gaskiers Fm. on the basis of striated clasts,
dropstones and evidence for deep-water conditions.
Boundary relations with overlying and underlying
non-glacial units
Throughout the Avalon terrane, the Gaskiers Fm. is conformable
with the underlying and overlying Mall Bay and Drook formations,
respectively. Boundaries between the interbedded laminations and
the underlying diamictite facies at the Double Road Point outcrop
are generally sharp, planar contacts, but some units are extensively
folded (Eyles & Eyles 1989). Basal contacts of massive and
crudely stratified diamictite facies are mostly sharp, planar or
gently undulating, and are locally erosive. The upper contacts of
diamictite units are sharp and planar (Eyles & Eyles 1989). The
Drook Fm. is conformably overlain by the Briscal Fm. at the
outcrop at Portugal Cove of the southern Avalon Peninsula
(King 1988). In turn, the Briscal Fm. in the southern Avalon
Peninsula is conformably overlain by the Mistaken Point Fm.
(Williams & King 1979). The Conception Group is overlain con-
formably by the St. John’s Group and Signal Hill Group in the
eastern Avalon Peninsula (e.g. Smith & Hiscott 1984).
 THE GASKIERS FORMATION
Chemostratigraphy
The oxygenation state of the late Neoproterozoic global ocean was
analysed by Canfield et al. (2007) by examining the Fe content
(reactive Fe:total Fe ratio, FeHR/FeT) of the deep-marine rocks
in Newfoundland. Using Fe extraction techniques, Canfield et al.
discovered that both the upper Mall Bay Fm. and Gaskiers diamic-
tite samples contained highly reactive FeHR/FeT ratios exceeding
0.38, indicative of anoxic deposition. Marine sediments deposited
in oxygen-containing water columns are consistently below 0.38
(Poulton & Raiswell 2002). The sedimentary rocks overlying the
Gaskiers diamictite yielded FeHR/FeT ratios lower than 0.38,
indicative of a long period of oxic marine conditions.
The isotopic composition of sulphide in pre-Gaskiers Fm. sedi-
ments was found to be greater than zero, indicating relatively small
fractionations from seawater sulphate caused by low rates of
sulpate reduction under sulphate-limiting concentrations. Higher
fractionations were found in the Gaskiers Fm., and particularly
in the overlying Drook Fm., indicative of an increase in sulphate
concentration. To explain these results, Canfield et al. (2007)
suggest that increased oxygen enhanced the oxidative weathering
of sulphide to sulphate on the continents, thus increasing the flux
of sulphate to the ocean.
Isotopic analyses of a thin limestone bed overlying Gaskiers Fm.
deposits discovered at the Harbour Main and Kitchuses sites have
yielded negative values of d13C (– 7.5 to – 1.5‰), and has been
interpreted as slope-deposited carbonate rocks that accumulated
in response to sea-level rise (Myrow & Kaufman 1999).
Other characteristics
A diverse assemblage, at least 30 species (Anderson & Conway
Morris 1982) of Ediacaran-type fossils, exists within the upper
Conception Group (Drook and Mistaken Point formations) and
lower St. John’s Group (Trespassey and Fermeuse formations)
(Fig. 42.2a; Myrow 1995). The fossil assemblages are also
known from Charnwood Forest in central England (Jenkins
1992). As a result, a full taxonomic description has never been
presented. At present, the following Ediacaran taxa have been
reported from the Mistaken Point Fm. (Bradgatia, Clapham
et al. 2004; Charnia masoni, Anderson 1978; Ivesia, Clapham
et al. 2004; Thectardis avalonensis, Clapham et al. 2004). The
Mistaken Point Fm. has yielded a date of 565 + 3 Ma for a vol-
canic ash that covers the most fossilerous surface near the top
of the formation (Benus 1988). Two species of the Ediacaran
frond Charnia – C. masoni and C. wardi – represent the most
complex of the four species in the Drook Fm. (Narbonne &
Gehling 2003).
Palaeolatitude and palaeogeography
The common view is that along with other exotic terranes, Avalo-
nia originated along the margin of Gondwana, rifted from the
Gondwanan margin in the early Palaeozoic and tectonically
accreted to eastern Laurentia in the mid-Palaeozoic (Nance &
Murphy 1996). There is debate as to whether the Avalon terrane
was adjacent to the West African or Amazonian margins of Gond-
wana (Rast & Skehan 1983; Dalziel et al. 1994). In contrast,
Landing (1996) has argued that there is no lithological, biotic or
geological evidence that supports the contention that Avalon was
contiguous with the margins of West Gondwana in the late Proter-
ozoic –Cambrian. Instead, Landing (1996) argued that Avalon was
an insular continent by the Early Palaeozoic, having already rifted
from Gondwana by the late Proterozoic. Palaeomagnetic investi-
gation of the Neoproterozoic Marystown Group (Burin Peninsula)
of southeastern Newfoundland suggests western Avalonia had a
471
palaeolatitude of c. 348 þ88/– 78 (mean tilt-corrected inclination
of 538) at c. 580–570 Ma, similar to the palaeolatitude of the
northern margin of the West African Craton (McNamara et al.
2001). These results were based on palaeomagnetic core samples
collected from 21 sites, all of which yielded a positive conglomer-
ate test, confirming a primary age of magnetization for the Marys-
town Group as a whole (McNamara et al. 2001). Some samples
consisted of multiple specimens; magnetization directions from
the samples within a site are averaged to produce the site mean
of that site.
In contrast, Nance & Murphy (1996) have argued, based on Nd
isotopic data from crustally derived felsic igneous rocks of West
Avalonia, that this terrane was close to the northern margin of
the Amazon Craton. Initial 1Nd values of West Avalonia (used to
identify mantle source from an igneous rock) are strongly positive
( –0.4 to þ5.0), and the Tdm model age (age at which crust was
extracted from mantle) for West Avalonia is c. 0.8– 1.1 Ga.
Similar positive initial 1Nd values (þ0.2 to þ6.9) and similar
Tdm model ages of c. 0.9–1.2 Ga were identified in the Tocantins
province of central Brazil (Pimental & Fuck 1992). As a result, a
palaeoposition adjacent to South America is favoured for West
Avalonia during the Neoproterozoic. Samson et al. (2005) has
also shown that the main age groups of detrital zirons from Avalo-
nia are predominantly Mesoproterozoic in age (c. 1.25– 1.15 Ga
and 1.65– 1.50 Ga), similar to the age pattern of detrital zircons
found in the Amazonian Craton. Mesoproterozoic crust has not
been identified in the West African Craton.
Other palaeogeographical reconstructions place Avalonia near
Laurentia at either a low latitude very near to the equator, or at a
high-latitude position in the southern hemisphere at c. 580 Ma
(Symons & Chaisson 1991; Trindade & Macouin 2007), but
these reconstructions are incompatible with strongly different
Cambrian –Ordovician biotas in the two regions (e.g. Rehmer
1981; Conway Morris & Rushton 1988).
Geochronological constraints
A maximum age for the Gaskiers Fm. is constrained by U – Pb
zircon ages of 606.8 þ3.7/–2.9 Ma, 622.6 þ2.3/–2.0 Ma and
631 + 2 Ma from the underlying basement rocks of the underlying
Harbour Main Group (Krogh et al. 1988). Zircons from ash beds in
the Ediacaran fossil-bearing Mistaken Point Fm. of the upper
Conception Group, on the eastern Avalon Peninsula, have yielded
a U –Pb date of 565 + 3 Ma (Benus 1988; Clapham et al. 2004;
Narbonne 2004). The main age groups of detrital zirons from the
basement rocks of Avalonia are predominantly Mesoproterozic
in age c. 1.25–1.15 Ga and 1.65 –1.50 Ga (Samson et al. 2005).
Discussion
Traditionally, the Gaskiers Fm., which is conformable with the
overlying and underlying marine sedimentary sequences of the
Mall Bay and Drook formations, has been interpreted as the sedi-
mentary product of floating ice in a marine shelf environment
(King 1988). A glacial origin for the Gaskiers Fm. rests heavily
on the presence of dropstones, chatter-marked garnets, and striated
and faceted stones in the diamictite (Williams & King 1979; King
1980). In addition to glaciogenic sediments, the Gaskiers Fm. also
contains pyroclastic and tuffaceous debris, as well as volcanic air
fall bombs and interbeds of agglomerate flows indicating volcanic
eruptions coeval with glaciation (Eyles & Eyles 1989; Ichaso et al.
2007). The Mall Bay and Drook formations have also been inter-
preted as base-of-slope or basin plain turbidite deposits (Gardiner
& Hiscott 1988); the same turbidite facies are interbedded with
the diamictite of the Gaskiers Fm. As a result, a clear picture
emerges of a deep basin receiving large volumes of volcaniclastic
472 S. L. CARTO & N. EYLES
sedimentary rocks that during glaciation was subjected to episodic
downslope resedimentation of muddy glaciogenic sediment. It is
highly likely that repeated episodes of submarine mass wasting
and downslope flow of glacially sourced debris may have been
induced by slope failure, triggered by volcanic eruptions and
melting of ice cover, producing mudflows akin to modern-day
lahars (Fisher 1984).
The Conception Group is one of several Neoproterozoic succes-
sions deposited between 620 and c. 570 Ma in arc-related basins
marginal to Gondwana, reflecting oblique subduction below the
Gondwanan margin (Murphy et al. 2004). They are all character-
ized by great thicknesses of volcanogenic turbidite, diamictite
and volcanic units. Debate has long surrounded the glacial status
of some age-equivalent diamictite units within several basins
such as in Armorica (Eyles 1990) and eastern North America
(Squantum Tillite; Carto & Eyles 2011). The case for a glaciated
hinterland can be convincingly made for the Gaskiers Fm. with
its numerous dropstones and striated clasts. The glacial origin of
the age-equivalent deposits are arguable indicating that the
extent of glaciation around the outer volcanic Gondwanan
margin (during the so-called Gaskiers Glaciation) was probably
limited in extent, topographically constrained to volcanic peaks
and unrelated to any global cooling event. The Quaternary and
modern Earth affords numerous examples of glaciated volcanic
peaks (Ehlers & Gibbard 2004). These ice masses do not extend
to sea level, but poorly sorted glaciogenic debris may be swept
into marine basins by lahars to be preserved as deep-marine
debris flows intercalated with thick turbidite successions. The pres-
ence of non-glaciogenic debrites (lahars?) intercalated with turbi-
dite units of the Mall Bay and Drook formations, above and below
the Gaskiers Fm. respectively, indicates continuity of marine
depositional processes through time (Eyles & Eyles 1989). Sub-
marine debris flow and turbidity current activity continued
throughout, but only within the Gaskiers Fm. was poorly sorted
glacial debris with striated clasts supplied to the marine margin.
The Ediacaran biota found in the Drook and Mistaken Point for-
mations and lower St. John’s Group (Trespassey and Fermeuse for-
mations) (Myrow 1995) are thought to represent the oldest known
complex, soft-bodied organisms, including the oldest definitive
animals on Earth (Narbonne 1998). The Edicaran fossils are
thought to mark a critical stage in the evolution of life on Earth,
and the presence of these fossils above the Gaskiers Fm. has led
to the theory that Neoproterozoic glaciations influenced eukaryote
evolution and/or acted as an impetus for the evolution of animals
(Hoffman et al. 1998; Runnegar 2000).
N. Eyles thanks the Natural Sciences and Engineering Research Council of
Canada for long-term financial support and a postgraduate scholarship to
S. Carto, R. Slatt and B. Rogerson (formerly of Memorial University of New-
foundland). R. Hiscott and C. Eyles are thanked for their assistance and discus-
sions in the field. This represents a contribution of the IUGS- and UNESCO-
funded IGCP (International Geoscience Programme) project #512.
References
Anderson, M. M. 1978. Ediacara fauna. In: Lapedes, D. N. (eds)
McGraw-Hill Yearbook of Science and Technology. McGraw-Hill,
New York, 146– 149.
Anderson, M. M. & King, A. F. 1981. Precambrian tillites of the Con-
ception Group on the Avalon Peninsula, southeastern Newfoundland.
In: Hambrey, M. J. & Harland, W. B. (eds) Pre-Pleistocene Glacial
Record. Cambridge University Press, Cambridge, 760–763.
Anderson, M. M. & Conway Morris, S. 1982. A review, with descrip-
tions of four unusual forms, of the soft-bodied fauna of the
Conception and St. John’s Groups (Late Precambrian), Avalon
Peninsula, Newfoundland. Proceedings of the Third North American
Paleontology Convention, 1, 1 –8.
Benus, A. P. 1988. Sedimentological context of a deep-water Ediacaran
fauna: Mistaken Point, Avalon Zone, Eastern Newfoundland.
In: Landing, E., Narbonne, G. M. & Myrow, P. (eds) Trace
Fossils, Small Shelly Fossils and the Precambrian– Cambrian
Boundary. New York State Museum and Geological Survey Bulletin
463, 8– 9.
Bowring, S. A., Myrow, P. M., Landing, E. & Ramezani, J. 2003.
Geochronological constraints on terminal Neoproterozoic events
and the rise of metazoans. NASA Astrobiology Institute General
Meeting Abstracts, 113– 114.
Bruckner, M. D. & Anderson, M. M. 1971. Late Precambrian glacial
deposits in southeastern Newfoundland, a preliminary note.
Geology Association of Canada, Proceedings, 24, 95– 102.
Burrett, J., Long, J. A. & Stait, B. 1990. Early– Middle Palaeozoic bio-
geography of Asian terranes derived from Gondwana. In: McKer-
row, W. S. & Scotese, C. R. (eds) Palaeozoic Palaeogeography
and Biogeography. Geological Society, London, Memoirs, 12,
163– 174.
Canfield, D. E., Poulton, S. W. & Narbonne, G. M. 2007. Late
Neoproterozoic deep ocean oxygenation and the rise of animal life.
Science, 315, 92 – 94.
Carto, S. L. & Eyles, N. 2011. The Squantum Member of the Boston
Basin, Massachusetts, USA. In: Arnaud, E., Halverson, G. P. &
Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic
Glaciations. Geological Society, London, Memoirs, 36, 475–480.
Clapham, M. E. & Narbonne, G. M. 2002. Ediacaran epifaunal tiering.
Geology, 30, 627–630.
Clapham, M. E., Narbonne, G. M., Gehling, J. G., Greentree, C. &
Anderson, M. M. 2004. Thectardis avalonensis: a new Ediacaran
fossil from the Mistaken Point biota, Newfoundland. Journal of
Paleontology, 78, 1031– 1036.
Conway Morris, S. & Rushton, A. W. A. 1988. Precambrian to Trema-
doc biotas in the Caledonides In: Harris, A. L. & Fettes, D. J. (eds)
The Caledonian– Appalachian orogen. Geological Society, London,
Special Publications, 38, 93– 109.
Dalziel, I. W. D., Dalla Salda, L. H. & Gahagan, L. M. 1994.
Paleozoic Laurentia – Gondwana interaction and the origin of the
Appalachian – Andean mountain system. Geological Society
American Bulletin, 106, 243–252.
Dec, T., O’Brian, S. J. & Knight, I. 1992. Late Precambrian volcani-
clastic deposits of the Avalonian Eastport Basin (Newfoundland
Appalachians): petrofacies, detrital clinopyroxene geochemistry
and plate tectonic implications. Precambrian Research, 59,
243– 262.
Ehlers, J. & Gibbard, P. L. 2004. Quaternary Glaciations: Extent and
Chronology, Part II: North America. Elsevier Science, Amsterdam,
450.
Eyles, N. 1990. Marine debris flows: Late Precambrian ‘tillites’ of the
Avalonian – Cadomian orogenic belt. Palaeogeography, Palaeocli-
matology, Palaeoecology, 79, 73 –98.
Eyles, N. & Eyles, C. H. 1989. Glacially-influenced deep-marine
sedimentation of the Late Precambrian Gaskiers Formation, New-
foundland, Canada. Sedimentology, 36, 601–620.
Fisher, R. V. 1984. Submarine volcaniclastic rocks. In: Kokelaar, B. P.
& Howells, M. F. (eds) Marginal Basin Geology: Volcanic and
Associated Sedimentary and Tectonic Processes in Modern and
Ancient Marginal Basins. Geological Society of London, Special
Publications, 16, 5 –27.
Forty, R. A. & Cocks, L. R. M. 2003. Palaeontological evidence bearing
on global Ordovician– Silurian continental reconstructions. Earth
Science Reviews, 67, 247–307.
Gardiner, S. & Hiscott, R. N. 1988. Deep-water facies and depositional
setting of the lower Conception Group (Hadrynian), southern Avalon
Peninsula, Newfoundland. Canadian Journal of Earth Sciences, 25,
1579– 1594.
Gravenor, C. P. 1980. Heavy minerals and sedimentological studies on
the glaciogenic late Precambrian Gaskiers Formation of Newfound-
land. Canadian Journal of Earth Sciences, 17, 1331–1341.
Hatcher, R. D. & Goldberg, S. A. 1991. The Blue Ridge province.
In: Horton, J. W. & Zullo, V. M. (eds) Geology of the
Carolinas. Carolina Geological Society, University of Tennessee
Press, 11– 35.
Hoffman, P. F., Kaufman, A. J., Halverson, G. P. & Schrag, D. P.
1998. A Neoproterozoic Snowball Earth. Science, 281, 1342– 1346.
 THE GASKIERS FORMATION
Ichaso, A. A., Dalrymple, R. W. & Narbonne, G. M. 2007. Paleoenvir-
onmental and basin analysis of the late Neoproterozoic (Edicaran)
upper Conception and St. John’s groups, west Conception Bay,
Newfoundland. Canadian Journal of Earth Sciences, 44, 25– 41.
Jenkins, R. J. F. 1992. Functional and ecological aspects of Ediacarian
assemblages. In: Lipps, J. H. & Signor, P. (eds) Origin and Early
Evolution of the Metazoa. Plenum, New York, 131–176.
King, A. F. 1980. The birth of the Caledonides: Late Precambrian rocks
of the Avalon Peninsula, Newfoundland, and their correlatives in
the Appalachian Orogen. In: Wones, D. R. (ed.) Proceedings of the
Caledonides in the USA Memoirs of the Department of Geological
Science, Virginia Polytechnic Institute, Blacksburg, Virginia, 2, 3– 8.
King, A. F. 1988. Late Precambrian sedimentation and related orogenesis
of the Avalon Peninsula, Eastern Avalon Zone. Fieldtrip Guidebook
A4, Geological Association of Canada, 84.
Knoll, A. H., Walter, M. R., Narbonne, G. M. & Christie-Blink, N.
2006. The Ediacaran Period: a new addition to the geologic time
scale, Lethaia, 39, 13 – 30.
Krogh, T. E., Strong, D. F., O’Brien, S. J. & Papezik, V. S. 1988.
Precise U– Pb zircon dates from the Avalon terrane in Newfoundland.
Canadian Journal of Earth Sciences, 25, 442– 453.
Landing, E. 1996. Avalon – insular continent by the latest Precambrian.
In: Lance, R. D. & Thompson, M. (eds) Avalonian and Related
Peri-Gondwanan Terranes of the Circum-North Atlantic. Geological
Society of America, Special Paper, 304, 27 – 64.
McCartney, W. D. 1967. Whitbourne Map Area, Newfoundland.
Geological Survey of Canada, Memoir, 341, 1– 133.
McMenamin, M. A. S. 1987. On the emergence of animals. Scientific
American, 256, 84 – 92.
McNamara, A. K., Mac Niocaill, C., Van der Pluijm, B. A. & Van der
Voo, R. 2001. West African proximity of Avalon in the latest Precam-
brian. Geological Society of America Bulletin, 113, 1161– 1170.
Murphy, J. B. & Nance, R. D. 1989. Model for the evolution of the
Avalonian – Cadomian belt. Geology, 17, 735– 738.
Murphy, J. B., Keppie, J. D., Dostal, J. & Nance, R. D. 1999. Neopro-
terozoic –early Paleozoic evolution of Avalonia. In: Ramos, V. A. &
Keppie, J. D. (eds) Laurentia –Gondwana Connections Before
Pangea. Geological Society of America, Special Papers, 336,
253– 266.
Murphy, J. B., Pisarevsky, S. A., Nance, R. D. & Keppie, J. D. 2004.
Neoproterozoic – Early Paleozoic evolution of peri-Gondwanan ter-
ranes: implications for Laurentia – Gondwana connections. Inter-
national Journal of Earth Sciences, 93, 659–682.
Myrow, P. M. 1995. Neoproterozoic rocks of the Newfoundland Avalon
zone. Precambrian Research, 73, 123– 136.
Myrow, P. M. & Kaufman, A. J. 1999. A newly discovered cap carbonate
above Varanger-age glacial deposits in Newfoundland, Canada.
Journal of Sedimentary Research, 69, 784–793.
Myrow, P. M., Taylor, J. F., Miller, J. F., Ethington, R. L., Ripper-
dan, R. L. & Branchle, C. M. 1999. Stratigraphy, sedimentology,
and paleontology of the Cambrian – Ordovician of Colorado and
adjacent areas. Geological Society of America Field Guide, 1,
157– 176.
Narbonne, G. M. 1998. The Ediacaran biota: a terminal Neoproterozoic
experiment in the evolution of life. Geological Survey of America
Today, 8, 1– 6.
Narbonne, G. M. 2004. Modular construction of early Ediacaran
complex life forms. Science, 305, 1141– 1144.
Narbonne, G. M. & Gehling, J. G. 2003. Life after Snowball: the oldest
complex Ediacaran fossils. Geology, 31, 27– 30.
Narbonne, G. M., Dalrymple, R. W., Gehling, J. G., Wood, D. A.,
Clapham, M. E. & Sala, R. A. 2001. Neoproterozoic fossils and
environments of the Avalon Peninsula, Newfoundland. Field Trips
B5, Geological Association of Canada– Mineralogical Association
of Canada Joint annual Meeting, St. John’s, NL.
473
Nance, R. D., Murphy, J. B., Strachan, R. A., Lemos, R. S. & Taylor,
G. K. 1991. Late Proterozoic tectonostratigraphic evolution of the
Avalonian and Cadomian terranes. Precambrian Research, 53,
41– 78.
Nance, R. D. & Murphy, J. B. 1996. Basement isotopic signatures and
Neoproterozoic paleogeography of Avalonian– Cadomian and
related terranes in the circum-North Atlantic. In: Nance, R. D. &
Thompson, M. D. (eds) Avalonian and Related Peri-Gondwanan
Terranes of the Circum-North Atlantic. Geological Society of
America Special Papers, 304, 333–346.
Nance, R. D., Murphy, J. B. & Keppie, J. D. 2002. Cordilleran model for
the evolution of Avalonia. Tectonophysics, 352, 11 –32.
O’Brien, S. J., Wardle, R. J. & King, A. F. 1983. The Avalon zone: a
Pan-African terrane in the Applachian orogen of Canada. Geological
Journal, 18, 195–222.
Osberg, P. H., Tull, J. F., Robinson, P. H. R & Butler, J. R. 1989. The
Acadian orogen. In: Hatcher, R. D., Jr, Thomas, W. A. & Viele,
G. W. (eds) The Appalachian– Ouachita Orogen in the United
States: The Geology of North America, Geological Society of
America, F-2, 179–232.
Papezik, V. S. 1974. Prehnite-pumpellyite facies metamorphism of the
Late Precambrian rocks of the Avalon Peninsula, Newfoundland.
Canadian Minerology, 12, 463–468.
Pimentel, M. M. & Fuck, R. A. 1992. Neoproterozoic crustal accretion
in central Brazil. Geology, 20, 375– 379.
Poulton, S. W. & Raiswell, R. 2002. The low-temperature geochemical
cycle of iron: from continental fluxes to marine sediment deposition,
American Journal of Science, 302, 774–805.
Rast, N. 1989. The evolution of the Appalachian chain. In: Bally, A. W.
& Palmer, A. R. (eds) The Geology of North America – An Over-
view. Geological Society of America, 4, 323–347.
Rast, N. & Skehan, J. W. 1983. The evolution of the Avalon plate.
Tectonophysics, 100, 257– 286.
Rehmer, J. 1981. The Squantum tilloid Member of the Roxbury Conglom-
erate of Boston, Massachusetts. In: Hambrey, M. J. & Harland,
W. B. (eds) Earth’s Pre-Pleistocene Glacial Record. Cambridge
University Press, Cambridge, 756–759.
Runnegar, B. N. 2000. Loophole for Snowball Earth. Nature, 405,
403– 404.
Samson, S. D., D’lemos, R. S., Miller, B. V. & Hamilton, M. A. 2005.
Neoproterozic palaeogeography of the Cadomia and Avalon terranes:
constraints from detrital zircons U– Pb ages. Journal of the Geologi-
cal Society, 162, 65– 71.
Schermerhorn, L. J. G. 1966. Terminology of mixed coarse-fine sedi-
ments. Journal of Sedimentary Petrology, 36, 831–835.
Smith, S. A. & Hiscott, R. N. 1984. Latest Precambrian to Early
Cambrian basin evolution, Fortune Bay, Newfoundland: fault-
bounded basin to platform. Canadian Journal of Earth Sciences,
21, 1379–1392.
Symons, D. T. A. & Chaisson, A. D. 1991. Palaeomagnetism of the
Callander complex and the Cambrian apparent polar wander path
for North America. Canadian Journal of Earth Sciences, 28,
355– 363.
Trindade, R. I. F. & Macouin, M. 2007. Palaeolatitude of glacial depos-
its and palaeogeography of Neoproterozoic ice ages. Comptes Rendus
Geoscience, 339, 200–211.
Williams, H. & King, A. F. 1975. Southern Avalon, Newfoundland,
Trepassey map area. Geological Survey of Canada Paper, 75, 11– 15.
Williams, H. & King, A. F. 1979. Trepassey Map Area, Newfoundland.
Geological Survey of Canada, Memoir, 389, 1 –24.
Wood, D. A., Dalrymple, R. W., Narbonne, G. M., Gehling, J. G. &
Clapman, M. E. 2003. Paleoenvironmental analysis of the Late
Neoproterozoic Mistaken Point and Trepassey formations, south-
eastern Newfoundland. Canadian Journal of Earth Sciences, 40,
1375–1391.
 Chapter 43

The Squantum Member of the Boston Basin, Massachusetts, USA

SHANNON L. CARTO* & NICK EYLES

Department of Geology, University of Toronto, Toronto, Ontario, M1C 1A4, Canada
*Corresponding author (e-mail: shannon.carto@gmail.com)

Abstract: The Neoproterozoic diamictite-bearing Squantum Member is located in the Boston Basin in eastern Massachusetts, USA. The
Boston Basin forms part of the Avalonia island arc terrane (c. 650 Ma), and appears to have originated as a rift-type basin in an exten-
sional setting along the northern margin of Gondwana, although its exact position is debated. Inferred palaeoenvironmental reconstruc-
tions of the Boston Basin have alternated between a fluvial basin where ice played a major role in transporting much of the coarse material
and an evolving marine basin dominated by non-glacial subaqueous mass flow, submarine fans and turbidity-current deposition.
The age of the Squantum is bracketed between c. 595 and 570 Ma, and is correlated by some to the glaciogenic diamictite succession of
the Gaskiers Formation (eastern Newfoundland) as part of the putative global Gaskiers Glaciation c. 582–585 Ma. However, the Squan-
tum Member consists of diamictite, graded sandstone and siltstone units, and fine-grained laminated argillite/mudstone units typical of
debris flow and turbidite facies that accumulate in a submarine setting. A glacial influence is not readily identified and revolves around
early interpretations of the diamictite as being ‘till-like’, the presence of laminated horizons that resemble glaciolacustrine ‘varvites’ and
the disputed recognition of ice-rafted dropstones. There are no associated carbonates and, consequently, no geochemical data are
available in connection with the Squantum Member.

The famous Squantum diamictite occurs within the sedimentary
rocks of the Boston Bay Group, preserved within the Neoprotero-
zoic Boston Basin in eastern Massachusetts, USA (Figs 43.1 &
43.2). It is formally known as the Squantum Member of the
Roxbury Conglomerate, but has also been referred to as the Squan-
tum ‘Tillite’, Squantum Tillite, Squantum Tilloid, the Roxbury/
Brighton complex and Massive diamictite association (Billings
1976; Smith & Socci 1990; Thompson 1993). The Squantum
Member has been very influential in the evolution of ideas con-
cerning pre-Pleistocene glacial deposits (Dodge 1875; Sayles
1914; Coleman 1926), continental drift and the reconstruction of
Pangea (Wegner 1912), the recognition of submarine mass flow
processes (Crowell 1957; Dott 1961) and the terminology used
for poorly sorted rocks (Pettijohn 1957).
Dodge (1875) was first to suggest a glacial origin for the Squan-
tum Member. Shaler (1869) suggested an exclusively marine
origin in view of the fine laminations, current structures and
graded bedding observed in the overlying Cambridge Formation
(Fm.), and parts of the Roxbury assemblage. He later suggested
a glacial origin for the conglomerate units within the Roxbury
assemblage (Shaler et al. 1899). Sayles & LaForge (1910)
regarded the diamictite and laminated facies of the Squantum as
a Carboniferous –Permian glacial and varved glaciolacustrine
deposit. Wegner (1912) considered the Squantum to be non-glacial
on the basis that poorly sorted facies are not exclusively of glacial
origin and also because the deposit did not fit with any simple lati-
tudinally constrained ice sheet on a reconstructed Pangaea for
which he received much undue criticism. Its age was later deter-
mined to be Neoproterozoic (for a review of historical ideas see
Eyles 2004).
Some interpretations of the Boston Bay Group have character-
ized all or part of the succession as a subaerial alluvial fan or
braided river system close to an ice margin (Rehmer & Roy
1976; Kaye 1984). Others attribute the Squantum Member to sub-
aqueous debris flow processes entirely unrelated to ice (Pettijohn
1957; Crowell 1957; Caldwell 1964; Dott 1961; Lindsay et al.
1970; Billings 1976). Crowell (1957) proposed that the Squantum
diamictite had been generated by submarine slumping and mixing
of gravel and mud downslope. His work, and especially that of Dott
(1961), highlighted the limitations of ‘tillite’ as a descriptive term
for such deposits because of its implication of direct glacial depo-
sition. He recommended the use of purely descriptive terms for
poorly sorted and laminated facies (e.g. diamictite, mixtite, rhyth-
mite, symmictite; see Schermerhorn 1966) that avoided specific
genetic references. Some authors suggest the Squantum Member
was deposited by gravity-debris flows that reworked glacial sedi-
ment into deep water (Stuart et al. 1975).
Conventionally, the Boston Bay Group is divided into a lower
Roxbury Conglomerate assemblage containing the diamictite-
dominated Squantum Member and an upper Cambridge Argillite
Formation (Fig. 43.2; Billings et al. 1939; Skehan 1964; Rehmer
& Roy 1976). The type area and best exposures of the Squantum
Member are in the low coastal cliffs at Squantum Head in
Quincy, Massachusetts, in the southern part of the Basin
(718000 W, 428180 N; Rehmer 1981). Outcrops are of limited
extent and many locations cited in earlier literature have been
lost to urbanization. Outcrops at Squantum Head expose two dia-
mictite facies interbedded with thinly laminated diamictite and
mudstone units.
Structural framework
The Squantum Member is one of several diamictite successions
preserved within the Late Proterozoic Avalonia– Cadomian oro-
genic belt. This belt consists of a series of exotic terranes that occu-
pied positions adjacent to the margin of Gondwana during the
Neoproterozoic (the so-called peri-Gondwana terranes) (e.g.
Murphy & Nance 1989). The remnants of these terranes can be
widely traced in North America, Europe and Africa (McNamara
et al. 2001). The Squantum Member and its associated sedimentary
and volcanic rocks, which occur within the 750 km2 fault-bounded
Boston Basin (Rehmer 1981), are part of the Western Avalonia
terrane. The Boston Basin appears to be a rift-related basin, repre-
sentative of an intra-arc (Nance 1990; Socci & Smith 1990) or a
back-arc (Cardoza et al. 1990) basin that developed in an exten-
sional or transtensional setting following the closing of the
Cadomian Ocean (Avalonian orogeny) (Rast & Skehan 1983). It
subsequently rifted from the Gondwanan margin in the early
Palaeozoic and drifted northward, to be later accreted to eastern
Laurentia c. 440 Ma (Murphy & Nance 1989; Murphy et al.
2004). The arc basin interpretation stems from the calc-alkaline
character and trace element geochemistry of the granitoids

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 475– 480. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.43
476 S. L. CARTO & N. EYLES

45’N
Boston Basin

.
N.H
a ss.
Boston
M

Squantum Head
N Boston Basin,
Nantasket Massachusetts
50 km

R.I.
ss.
Ma
n.
Con

36’N Fig. 43.1. General schematic map of the

45’W 30’W location of the Boston Basin in
Massachusetts, USA, indicating the
location of principal outcrops mentioned in
the text (after Socci & Smith 1987).

(Dedham Granite) and felsic volcanic rocks (Mattapan-Lynn Vol-
canic Complex) (Hermes et al. 1981) that underlie the strata of the
Boston Bay Group.
The Boston Bay Group was affected by subsequent Alleghenian
deformation in the form of broad east –NE striking thrust faults and
NE-trending folds (Billings 1979; Skehan & Murray 1980). The
metamorphic grade of the Boston Bay Group is subgreenschist
facies, and a slaty, well-developed, spaced cleavage, oriented
approximately perpendicular to bedding, is present (Rehmer
1981). Deformation induced by the Acadian orogeny is recorded
Cambridge Argillite
(2300 - 5500m)
Squantum Member
Dorchester Member
(100 - 500m)
Roxbury
Conglomerate
Brookline Member
(150 - 1300m)
Fig. 43.2. Original ‘layer-cake’ stratigraphic framework of the
lithostratigraphic units of the Boston Bay Group (modified from Socci & Smith
1987). Each stratigraphic unit was defined and recognized on the basis of the
percentage of conglomerate, diamictite, argillite, sandstone, siltstone and
matrix present. Recent sedimentological studies indicate that the sedimentary
units of the Boston Bay Group are complexly intercalated, with repetition of
facies within different members suggesting that the sedimentation of these rocks
was diachronous and genetically related.
in rocks outside the Boston Basin to the NW (e.g. Clinton-
Newbury and Bloody Bluff faults) (Skehan & Murray 1980).
Stratigraphy
The Boston Bay Group (c. 5 km thick; Bailey 1987) has tradition-
ally been divided into two distinctive units: a lower Roxbury
Conglomerate assemblage and the upper Cambridge Argillite For-
mation (Rehmer & Roy 1976). The Roxbury is subdivided into
three sub-members: the basal Brookline Member (clast-supported
conglomerate with minor sand), the medial Dorchester Member
(mostly sandstone with minor conglomerate) and the upper Squan-
tum Member (diamictite-dominated) (Emerson 1917; LaForge
1932). There is agreement that the basin fill cannot simply be sub-
divided in layer-cake fashion on the basis of correlative facies
types and the complex interfingering of the various clastic litholo-
gies within the Roxbury assemblage (Fig. 43.2; Dott 1961; Billings
1976, 1979; Socci & Smith 1990).
The granitic basement of the Boston Basin, represented by the
c. 610 Ma Dedham Granite (Hepburn et al. 1993) and the
c. 599 Ma Westwood Granite (Thompson et al. 1996) is thought
to have been intruded into the older crust (Middlesex Fells Volca-
nic Complex). The basement rocks are also represented by the
bimodal Mattapan-Lynn Volcanic Complex (c. 596 Ma), which
unconformably underlies and pre-dates the Boston Bay Group in
most areas (Hon & Hepburn 1986). This complex is composed
of more than 300 m of rhyolitic, andesitic, basaltic, local kerato-
phyre flows, coarse granitic and syenitic dykes, volcanic breccia,
scoria, and volcanic mudflows and tuffs (La Forge 1932; Thomp-
son 1993), and records the transition from a compressional to
extensional tectonic regime (Skehan & Murray 1980; Hon &
Hepburn 1986). These basement units are overlain by the
calc-alkaline Brighton Volcanic rocks/Melaphyre (c. 580–
650 Ma), which consist of the altered basalt and andesite flows,
pyroclastic rocks, breccia, tuff, and intrusive rocks that sporadi-
cally interfinger the Brookline and Dorchester members through-
out the basin (Dott 1961; Billing 1976; Zartman & Naylor
1984). The Brighton Volcanic rocks record a later phase of
volcanic activity.
 THE SQUANTUM MEMBER OF THE BOSTON BASIN
Glaciogenic deposits and associated strata
Brookline Member
The Brookline Member (150 –1300 m thick), at the base of the
Roxbury Conglomerate, has been described as a ‘puddingstone’
consisting of a massive clast-supported pebble and cobble con-
glomerate with interbedded argillite, sandstone and melaphyre
units (Billing 1976; Reymer & Roy 1976). The matrix is composed
of grey feldspathic sandstone, supporting well-rounded pebbles
and cobbles of quartzite granite, felsite, and quartz monzonite
with clast size ranging from 1 to 15 cm (Billing 1976). Smith &
Socci (1990) describe this unit as a diamictite/sandstone/argillite
association, in reference to complex interbedding of laminated and
graded argillite and sandstone units and massive, matrix- and clast-
supported diamictite units.
Dorchester Member
In the Boston Basin, purplish, greenish and grey siltstone and sand-
stone units, as well as shale units above the Brookline Member, are
traditionally assigned to the Dorchester Member (180– 500 m
thick) (Billings 1976). Smith & Socci (1990) also describe this
unit as commonly containing full or partial Bouma sequences
and dominated by medium-to fine-grained argillite, containing
lesser amounts of sandstone and conglomerate relative to the
Brookline Member. Evidence of penecontemporaneous fragmen-
tation and contortion due to downslope slumping are common in
these units. The same facies occur interbedded within the Squan-
tum diamictite units (Bailey 1987).
Squantum Member
The Squantum diamictite has been described as a polymictic, het-
erogeneous and poorly sorted admixture of rare boulders up to
1.2 m in diameter with pebbles, cobbles and sand in a silty-clay
matrix (Dott 1961; Socci & Smith 1990). It has also been described
as a matrix-supported conglomerate (Billings 1976; Rehmer &
Roy 1976) and as a pebbly mudstone (Crowell 1957). At the
type area at Squantum Head, these facies are interbedded with
laminated mudstone units (bed thickness ranging from 2 to
10 cm) and lamina-thick diamictite horizons (bed thickness
ranging from a few millimetres to 1 cm; Dott 1961; Socci &
Smith 1990), some of which contain a few outsized clasts that
have depressed the underlying laminae, having the appearance of
dropped pebbles (Lindsay et al. 1970).
Most outcrops of the diamictite show chaotic bedding in the
form of contorted and folded patches of sand and local clusters
of clasts, as well as coherent slump blocks of mudstone (Dott
1961; Lindsay et al. 1970). These chaotically mixed facies are
readily distinguished from more homogeneous diamictite facies.
Diamictite units reportedly range in thickness from 18 to 215 m
(Rehmer 1981). These units are mostly massive and lenticular in
form; some are crude to moderately well sorted and some
exhibit normal grading. There are also thinner diamictite and intra-
clastic rich pebbly horizons exposed in other parts of the Boston
Basin area (Bailey, pers. comm. 2008).
Clasts consist of multicoloured, locally derived felsic and mafic
volcanic rocks (Mattapan and/or Brighton type), granodiorite
(Dedham-type), quartzite and intrabasinal clasts of massive,
graded and laminated sandstone and siltstone (Rehmer 1981).
Much of the matrix has been derived from the latter facies.
Clasts range from sub-rounded to angular clasts, 5– 60 cm in diam-
eter, to well-rounded clasts 3 –8 cm in diameter (Billings 1976).
Striated clasts (Sayles & LaForge 1910; Sayles 1914), chatter-
marked quartz grains (Rehmer & Hepburn 1974) and dropstones
(Cameron & Jeanne 1976; Wolfe 1976; Cameron 1979) have
477
been reported by early workers but have not been confirmed by
later observers. Moreover, several previously identified dropstones
have been re-interpreted as having been emplaced by lateral
sediment-gravity or current processes (Dott 1961; Bailey &
Bland 2001). The sand- and gravel-sized detritus is composed of
volcanic, granitic and metasedimentary lithic fragments and is
essentially the same composition as that of the other Roxbury
units (Dott 1961). The presence of lapilli tuff beds has been ident-
ified in thin sections of strata genetically related to, and in close
association with, the Squantum diamictite units at Squantum
Head, showing that further analysis of the volcanic character of
this deposit is warranted.
Cambridge Fm.
Overlying the Squantum Member, the Cambridge Fm. (known
formally as the Cambridge Argillite or the Cambridge Slate) con-
sists of up to 5 km of laminated, dark to olive grey, graded siltstone
and sandstone beds (Billings 1976). Graded beds, starved ripples,
scour marks, load casts and micro-faults are numerous. Soft-
sediment deformation structures, such as mega slump folds
many metres in amplitude, and pinch and swell bedding, are also
common (Bailey 1987). Petrographically, quartz, feldspar,
melaphyre fragments, chlorite and epidote dominate these facies,
indicating a volcanic/igneous source area. Discrete ash
beds measuring a few centimetres to tens of centimetres in
thickness have also been documented in this unit (LaForge 1932;
Thompson & Bowring 2000).
Boundary relations with overlying and underlying
non-glacial units
The Boston Bay Group strata are fault-bounded to the north and
west, and to the south rest unconformably on the bimodal Mattapan
Volcanic Complex, which, in turn, overlies the Dedham Granite
(Rehmer 1981). The upper and lower contacts of the Squantum
Member are gradational in outcrop (LaForge 1932). Significantly,
many authors note intricate facies variations and complex
interfingering between the clastic lithofacies of the Roxbury Con-
glomerate and volcanic lithologies within the Roxbury assem-
blage. Sayles (1914), Dott (1961), Rahm (1962) and Socci &
Smith (1987) have noted that Squantum-type rocks (less well-
sorted facies of the Brookline and Dorchester lithologies as
defined by Dott 1961) appear to inter-finger with the Roxbury
and overlying Cambridge Argillite, and become more common
towards the NE (Dott 1961). The Brighton Volcanic rocks are
also interbedded throughout the Brookline and Dorchester
Members in the southern portion of the basin (Zartman & Naylor
1984; Thompson 1993).
Chemostratigraphy
Recently, Passchier & Erukanure (2010) determined the chemical
index of alteration (CIA) to assess if any evidence of climatic
changes is present in the Squantum Member. The geochemical
work targeted the fine-grained facies, including the matrix of the
diamictite units at the type locality. The researchers found CIA
to be relatively high for the Squantum Member (CIA of 61 –75).
This value is higher than those of Pleistocene glacial diamictite,
which has a typical CIA of 50– 55 (Nesbitt & Young 1982; Passch-
ier & Krissek 2008). The CIA values from the Squantum diamictite
were interpreted to indicate that, at the time of deposition, the con-
tinental areas could not have been entirely ice covered and that
significant weathering occurred before the muds were supplied
to the basin (Passchier & Erukanure 2010).
478 S. L. CARTO & N. EYLES
Other characteristics
Ediacaran (Vendian as per Lenk et al. 1982) microfossils (Bavli-
nella cf. faveolata) occur in the Cambridge Argillite Formation
(Lenk et al.1982). These fossils are preserved as petrifactions in
pyrite interspersed within organic laminae (Lenk et al. 1982).
The microfossils were not fragmented and do not show any evi-
dence of having been reworked from sediment. Raised ring struc-
tures (5 –35 mm) have been described from outcrops of Boston
Bay Group sediments at Hewitts Cove and Slate Island (Bailey
& Bland 2001). Although similar in form and Neoproterozoic in
age, these are thought to be distinct from Ediacaran Aspidella ter-
ranovica described in Newfoundland and elsewhere (Bailey &
Bland 2001). More recently, Passchier & Erukanure (2010) have
reported two dislocated stromatolite hemispheroids in outcrop in
the laminae exposed at the type section.
Palaeolatitude and palaeogeography
Palaeogeographical reconstructions of Avalonia based on palaeo-
magnetic pole data are inconclusive, as they place Avalonia
either near Laurentia at low latitudes or at a high palaeolatitude
(Wu et al. 1986; McNamara et al. 2001; Trindade & Macouin
2007). Most palaeomagnetic reconstructions have long placed
Avalonia along the northern margin of Gondwana (Van der Voo
1993; Torsvik et al. 1996), but whether it was adjacent to the
West African craton or Amazonian craton remains uncertain.
Recent palaeomagnetic results from the c. 580 to 570 Ma volcanic
and sedimentary rocks of the Marystown Group of the Burin
Peninsula, Newfoundland (western Avalonia), has revealed that
Avalonia had a palaeolatitude of 348 þ88/–78 (mean tilt-corrected
inclination of 538), coinciding with that of the northern margin of
West Africa (McNamara et al. 2001). This palaeolatitude was
based on palaeomagnetic core samples collected from 21 sites,
all of which yielded a positive conglomerate test, confirming a
primary age of magnetization for the Marytown Group as a
whole. Demagnetization results revealed that two or more com-
ponents of magnetization were present in almost all samples, yield-
ing curved trajectories in orthogonal plots. Some samples consist
of multiple specimens; magnetization directions from the samples
within a site are averaged to produce the site mean for that site.
A similar palaeoposition of Avalonia is supported by Thompson
et al. (2007), who suggested Avalonia occupies a midlatitude pos-
ition of 38 + 88 between c. 600 Ma and 575 Ma. This palaeolati-
tude is based on samples collected at six locations (142 cores)
from Lynn and Mattapan volcanic complexes and three sites (33
cores) from the Squantum Member. Results from the Squantum
Member were corrected for bedding tilt and plunge. The samples
from each site passed reversal and fold tests, and are considered
to record primary Neoproterozoic magnetization at c. 595 Ma.
Thompson et al. (2007) noted that two permissible palaeomagnetic
positions exist for Avalonia. If placed at a palaeolatitude of 388S
it would border West Africa, whereas if it were placed 388N it
would border Amazonia.
Conversely, Samson et al. (2005) have shown that the main age
groups of detrital zircons from Avalonia cluster within the Meso-
proterozic, similar to the age pattern of detrital zircons also found
in the Amazonian craton, and therefore place Avalonia at 388N
along side the Amazonian craton at c. 580 Ma.
Geochronological constraints
U –Pb zircon data derived from a sandstone bed from the Squan-
tum Member on the western end of Squantum Head (12 detrital
zircon samples) yielded a weighted mean 207Pb/206Pb date
ranging from c. 593 + 3 Ma (weighted mean 207Pb/206Pb date
for the two youngest grains) to 618 Ma (date for the oldest grain
from the remaining nine grains; an older grain with a
207
Pb/206Pb date of 818 Ma was not included) (Thompson &
Bowring 2000; Thompson et al. 2007). In this study, similar
ages were yielded from a welded tuff clast, crystal-poor tuff
sample, and a sample of granophyre removed from the Squantum
diamictite, c. 595 + 2 Ma, 600 + 1 Ma and 610 + 2 Ma, respect-
ively (Thompson & Bowring 2000). The minimum age limit of
the Squantum Member is thought to be c. 570 Ma (207Pb/206Pb
date) or younger based on the youngest detrital zircon component
from an ash bed in the overlying Cambridge Argillite, which
yielded a 207Pb/206Pb date of c. 570 Ma (Thompson & Bowring
2000). Based on these results, the Squantum deposition is
bracketed between c. 595 and 570 Ma.
Discussion
There is an agreement that the Squantum Member is the product of
the downslope slumping and intermixing of fine and coarse facies in
a marine environment (Crowell 1957; Pettijohn 1957; Dott 1961;
Caldwell 1964; Lindsay et al. 1970; Billings 1976; Smith & Socci
1990). The close similarity of the Roxbury conglomerate and Squan-
tum clast compositions has been cited by Dott (1961) as evidence
that Squantum Member diamictite units were derived by mixing of
Roxbury gravels with mud as they slumped downslope and were
redeposited (see also Crowell 1957), known to be a common submar-
ine process (e.g. Nemec et al. 1984). These diamictites and the
associated thick turbidite succession of the Cambridge Argillite
are readily placed into a submarine fan or slope environment. The
volcanic and coarsely clastic character of the Boston Bay Group
points to deposition associated with volcanic activity, whereas the
extent of any glacial influence is arguable and was supported by
indirect evidence at best. Past identification of a glaciogenic
source rested entirely on the identification of ‘dropstones’ (Wolfe
1976; Cameron & Jeanne 1976; Cameron 1979) and striated
pebbles (Sayles & LaForge 1910; Sayles 1914), which have not
been substantiated by later observers (Dott 1961; Bailey 1987). At
present, no carbonates have been identified in this succession.
Stratigraphic relationships among the units of the Boston Bay
Group have, as stated above, traditionally been described in terms
of a simple, vertical layer-cake succession (e.g. Mattapan-Lynn
Complex/Dedham-Brookline-Dorchester-Squantum-Cambridge).
This scheme was made on the basis of correlating like-for-like
facies and in ignorance of any understanding of the origin of the
facies and the depositional system in which these facies occur.
The Boston Basin is one of several Neoproterozoic successions
deposited in arc-related basins that were found within peri-
Gondwana terranes (e.g. Armorica). They are characterized by volu-
minous accumulations of volcanogenic turbidites, debrites and vol-
canic rocks reflecting oblique subduction below the Gondwanan
margin sometime between 620 and c. 590 Ma (Murphy et al.
2004). The degree of any glacial influence on debrites elsewhere
within Armorica has long been debated, much like the controversy
surrounding the Squantum Member deposits. Given its resemblance
to till, past practice has simply been to label as glacial any deposit
that is poorly sorted (see Eyles 1990, 1993, for review). It is signifi-
cant that Crowell’s (1999) authoritative review of Earth’s Neoproter-
ozoic glacial record does not include the Squantum Member.
There are nonetheless close sedimentological similarities with
the glacially and volcanically influenced debris flow/turbidite
succession of the Gaskiers Fm. in Newfoundland, Canada (Carto
& Eyles 2011), but the case for a glacial setting for the Boston
Basin strata is tenuous, unlike for the Gaskiers Fm., where
undoubted glacially striated and faceted clasts, as well as drop-
stones, are found. The extent of glaciation around the outer Gond-
wanan margin (the so-called Gaskiers Glaciation c. 580 Ma) was
probably limited to topographically controlled small glacier com-
plexes on higher standing, arc-related volcanoes.
 THE SQUANTUM MEMBER OF THE BOSTON BASIN
This represents a contribution of the IUGS- and UNESCO-funded IGCP (Inter-
national Geoscience Programme) project #512.
References
Bailey, R. H. 1987. Stratigraphy of the Boston Bay Group, Boston Area,
Massachusetts. In: Roy, D. C. (ed.) Northeastern Section of the Geo-
logical Society of America: Boulder, Colorado. Geological Society of
America, Centennial Field Guide, 5, 209–212.
Bailey, R. H. & Bland, B. H. 2001. Recent developments in the study of
the Boston Bay Group. In: West, D. P. Jr. & Bailey, R. H. (eds)
Guidebook for Geological Field Trips in New England, 2001
Annual Meeting of the Geological Society of America, Boston,
Massachusetts, U-1 –U-23.
Billings, M. P. 1976. Geology of the Boston Basin. In: Skehan, J. W. &
Murray, D. P. (eds) Studies in New England Geology. Geological
Society of America, Memoir, 46, 5– 30.
Billings, M. P. 1979. Bedrock geology of the Boston Basin. In:
Cameron, B. (ed.) Carboniferous Basins in Southeastern New
England. Field guidebook for Trip No. 5, 9th International Congress
of Carboniferous Stratigraphy and Geology, 46 – 64.
Billings, M. P., Loomis, F. B. & Stewart, G. W. 1939. Carboniferous
topography in the vicinity of Boston, Massachusetts. Geological
Society of American Bulletin, 50, 1867– 1884.
Caldwell, D. W. 1964. The Squantum Formation: Paleozoic tillite or
tilliod? In: Skehan, S. J. (ed.) Guidebook to Field Trips in the
Boston Area and Vicinity. New England Intercollegiate Geological
Conference, 56th Annual Meeting, 53 –60.
Cameron, B. 1979. General geology and the Carboniferous Basins of
eastern Massachusetts and Rhode Island. In: Cameron, B. (ed.)
Carboniferous Basins of Southeastern New England, Field Guide-
book for Trip No. 5, 9th International Congress of Carboniferous
Stratigraphy and Geology, 1– 6.
Cameron, B. & Jeanne, R. A. 1976. New evidence for glaciation during
deposition of the Boston Bay. In: Cameron, B. (ed.) Geology of
Southern New England, 68th Annual Meeting New England Inter-
collegiate Geological Conference, Science Press, Princeton, NJ,
117– 134.
Cardoza, K. D., Hepburn, J. C. & Hon, R. 1990. Geochemical con-
straints on the paleotectonic settings of two late Proterozoic mafic
volcanic suites, Boston-Avalon Zone, Eastern Massachusetts. In:
Socci, A. D., Skehan, J. W. & Smith, G. W. (eds) Geology of the
Composite Avalon Terrane of Southern New England. Geological
Society of America Special Paper, 245, 113–131.
Carto, S. L. & Eyles, N. 2011. The deep marine glaciogenic Gaskiers
Formation: Newfoundland, Canada. In: Arnaud, E., Halverson,
G. P. & Shields-Zhou, G. (eds) The Geological Record of Neopro-
terozoic Glaciations. Geological Society, London, Memoirs, 36,
467– 473.
Coleman, A. P. 1926. Ice Ages: Recent and Ancient. Macmillan, London.
Crowell, J. C. 1957. Origin of pebbly mudstones. Geological Society of
America Bulletin, 68, 993–1010.
Crowell, J. C. 1999. Pre-Mesozoic ice ages: their bearing on under-
standing the climate system. Geological Society of America
Memoir, 192, 106.
Dodge, W. W. 1875. Notes on the geology of eastern Massachusetts.
Boston Society of Natural History, 17, 388–419.
Dott, R. H. 1961. Squantum ‘Tillite’, Massachusetts, evidence of glacia-
tion or subaqueous mass movement? Geological Society of America
Bulletin, 72, 1289–1306.
Emerson, B. K. 1917. Geology of Massachusetts and Rhode Island.
United States. Geological Survey Bulletin, 597, 56– 57.
Eyles, N. 1990. Late Precambrian ‘tillites’ of the Avalonian –Cadomian
belt; marine debris flows in an active tectonic setting. Palaeogeo-
graphy, Palaeoclimatology, Palaeoecology, 79, 73 –98.
Eyles, N. 1993. Earth’s glacial record and its tectonic setting. Earth
Science Reviews, 35, 1 –248.
Eyles, N. 2004. Frozen in time: concepts of ‘global glaciation’ from 1837
(die Eiszeit) to 1998 (the Snowball Earth). Geoscience Canada, 31,
157– 166.
479
Hermes, O. D., Barosh, P. J. & Smith, P. V. 1981. Contact relationships
of the late Paleozoic Narragansett Pier Granite and country rock,
In: Boothroyd, J. C. & Hermes, O. D. (eds) Guidebook for Field
Studies in Rhode Island and Adjacent Areas: New England Intercol-
legiate Geological Conference, 73rd Annual Meeting, University of
Rhode Island, Kingston, Rhode Island, 47– 67.
Hepburn, J. C., Hon, R., Dunning, G. R., Bailey, R. H. & Galli, K.
1993. The Avalon and Nashoba terranes (eastern margin of the Appa-
lachian orogen in southeastern New England). In: Cheney, J. T. &
Hepburn, J. C. (eds) Field Trip Guidebook for the Northeastern
United States. Boston GSA, Volume 2. Geology Department, Univer-
sity of Massachusetts, Contribution 67, X.1– X.31.
Hon, R. & Hepburn, J. C. 1986. Igneous geochemistry and its impli-
cations for terrane analysis of the Avalonian event in southeastern
New England, U.S.A. Atlantic Geoscience Society Abstract, 1986
colloquium. Maritime Sediments and Atlantic Geology, 22, 331.
Kaye, C. A. 1984. Boston Basin restudied. In: Hanson, L. S. (ed.)
Geology of the Coastal Lowlands Boston, Massachusetts to Kenne-
bunk, Maine. New England Intercollegiate Geological Conference
76th Annual Meeting, 124–140.
LaForge, L. 1932. Geology of the Boston area, Massachusetts. US
Geological Survey Bulletin, 839, 105.
Lenk, C., Strother, P. K., Kaye, C. A. & Barghoorn, E. S. 1982. Pre-
cambrian age of the Boston Basin: new evidence from microfossils.
Science, 216, 619– 620.
Lindsay, J. F., Summerson, C. H. & Barrett, P. J. 1970. A long-axis
clast fabric comparison of the Squantum ‘Tillite’ and the Gowganda
Formation, Ontario. Journal of Sedimentary Petrology, 40, 475– 479.
McNamara, A. K., Niocaill, C. M., Van der Pluijm, B. A. & Van der
Voo, R. 2001. West African proximity of the Avalon terrane in the
latest Precambrian. Geological Society of America Bulletin, 113,
1161–1170.
Murphy, J. B. & Nance, R. D. 1989. Model for the evolution of the
Avalonian– Cadomian Belt. Geology, 17, 735– 738.
Murphy, J. B., Pisarevsky, S. A., Nance, R. D. & Keppie, J. D. 2004.
Neoproterozoic– Early Paleozoic evolution of peri-Gondwanan
terranes: implications for Laurentia– Gondwana connections.
International Journal of Earth Sciences, 93, 659–682.
Nance, R. D. 1990. Late Precambrian – Early Paleozoic arc –platform
transitions in the Northern Appalachians; review and implications,
In: Socci, A. A., Skehan, J. W. & Smith, G. W. (eds) Geology
of the Composite Avalon Terrane of Southern New England:
Boulder, Colarado. Geological Society of America Special Paper,
245, 1– 11.
Nemec, W., Steel, R. J., Porebski, S. J. & Spinnangr, A. 1984. Domba
Conglomerate, Devonian, Norway: process and lateral variability in
a mass flow-dominated lacustrine fan-delta. In: Koster, E. H. &
Steel, R. J. (eds) Sedimentology of Gravels and Conglomerates.
Canadian Society of Petroleum Geologists, Calgary, 295– 330.
Nesbitt, H. W. & Young, G. M. 1982. Early Proterozoic climates and
plate motions inferred from major element chemistry of lutites.
Nature, 299, 715– 717.
Passchier, S. & Krissek, L. A. 2008. Oligocene –Miocene Antarctic
continental weathering record and paleoclimatic implications, Cape
Roberts drilling Project, Ross Sea, Antarctica. Palaeogeography,
Palaeoclimatology, Palaeoecology, 260, 30– 40.
Passchier, S. & Erukanure, E. 2010. Palaeoenvironments and weather-
ing regime of the Neoproterozoic Squantum ‘Tillite’, Boston Basin:
no evidence of a snowball Earth. Sedimentology, 57, 1526–1544.
Pettijohn, F. J. 1957. Sedimentary Rocks. Harper & Brothers, New York.
Rahm, D. A. 1962. Geology of the main drainage tunnel, Boston, Massa-
chusetts. Journal of the Boston Society of Civil Engineering, 49,
319– 368.
Rast, N. & Skehan, J. W. 1983. The evolution of the Avalonian Plate.
Tectonophysics, 100, 257– 286.
Rehmer, J. A. 1981. The Squantum tilloid member of the Roxbury
Conglomerate of Boston, Massachusetts. In: Hambrey, M. J. &
Harland, W. B. (eds) Earth’s Pre-Pleistocene Glacial Record.
Cambridge University Press, London, 756– 759.
Rehmer, J. A. & Hepburn, J. C. 1974. Quartz sand surface textural
evidence for a glacial origin of the Squantum ‘Tillite’, Boston
Basin, Massachusetts. Geology, 2, 413– 415.
480 S. L. CARTO & N. EYLES
Rehmer, J. A. & Roy, D. C. 1976. The Boston Bay Group: the boulder
bed problem. In: Cameron, B. (ed.) Geology of Southeastern New
England. New England Intercollegiate Geology Conference
Science Press, Princeton, New Jersey, 71 –91.
Samson, S. D., D’lemos, R. S., Miller, B. V. & Hamilton, M. A. 2005.
Neoproterozic palaeogeography of the Cadomia and Avalon terranes:
constraints from detrital zircons U –Pb ages. Journal of the Geologi-
cal Society, 162, 65 –71.
Sayles, R. W. 1914. The Squantum Tillite. Bulletin Harvard Museum
Comparative Zoology, 66, 141–175.
Sayles, R. W. & LaForge, L. 1910. The glacial origin of the Roxbury
Conglomerate. Science, 32, 723–724.
Schermerhorn, L. J. G. 1966. Terminology of mixed coarse-fine
sediments. Journal of Sedimentary Petrology, 36, 831– 836.
Shaler, N. S. 1869. Note on the geological section at Chestnut Hill Reser-
voir, Massachusetts. Boston Society of Natural History Proceedings,
13, 172– 177.
Shaler, N. S., Woodworth, J. B. & Foerste, A. F. 1899. Geology of
the Narragansett Basin. United States Geological Survey, Mono-
graph, 33.
Skehan, S. J. 1964. Guidebook to Field Trips in the Boston Area and
Vicinity. New England Intercollegiate Geological Conference
(NEIGC), 56th Annual Meeting, Boston, Massachusetts.
Skehan, S. J. & Murphy, D. P. 1980. Geological profile across southeast-
ern New England. Tectonophysics, 69, 285– 319.
Smith, G. W. & Socci, A. D. 1990. Late Precambrian sedimentary
geology of the Boston Basin. In: Socci, A. D., Skehan, J. W. &
Smith, G. W. (eds) Geology of the Composite Avalon Terrane of
Southern New England. Geological Society of America, Special
Papers, 245, 75 – 84.
Socci, A. D. & Smith, G. W. 1987. Recent sedimentological interpret-
ations in the Avalon terrane of the Boston Basin, Massachusetts.
Maritime Sediments and Atlantic Geology, 23, 75– 84.
Socci, A. D. & Smith, G. W. 1990. Stratigraphic implications of
facies within the Boston Basin. In: Socci, A. D., Skehan, J. W. &
Smith, G. W. (eds) Geology of the Composite Avalon Terrane of
Southern New England. Geological Society of America Special
Paper, 245, 55 –74.
Stuart, J., Baker, H. W., Dott, H. R., Rehmer, J. A. & Hepburn, J. C.
1975. Quartz sand surface textural evidence for a glacial origin of
the Squantum ‘Tillite’, Boston Basin, Massachusetts: comments
and reply. Geology, 3, 153– 155.
Thompson, M. D. 1993. Late Proterozoic stratigraphy and structure in the
Avalonian magmatic arc southwest of Boston, Massachusetts.
American Journal of Science, 293, 725–743.
Thompson, M. D. & Bowring, S. A. 2000. Age of the Squantum ‘tillite’,
Boston Basin, Massachusetts: U– Pb zircon constraints on terminal
Neoproterozoic glaciation. American Journal of Science, 300,
630– 655.
Thompson, M. D., Hermes, O. D., Bowring, S. A., Isachsen, C. E.,
Besancon, J. R. & Kelly, K. L. 1996. Tectonostratigraphic impli-
cations of Late Proterozoic U–Pb zircon ages in the Avalon Zone
of southeastern New England. In: Nance, R. D. & Thompson, M.
D. (eds) Avalonian and Related Peri-Gondwanan Terranes of the
Circum-North Atlantic. Geological Society of America Special
Papers, 304, 179– 191.
Thompson, M. D., Grunow, A. M. & Ramezani, J. 2007. Late
Neoproterozoic paleogeography of the Southeastern New England
Avalon Zone: insights from U– Pb geochronology and paleomagnet-
ism. Geological Society of America Bulletin, 119, 681– 696.
Torsvik, T. H., Smethurst, M. A. & Meert, J. G. 1996. Continental
breakup and collision in the Neoproterozoic and Palaeozoic: a tale
of Baltica and Laurentia. Earth Science Reviews, 40, 229– 258.
Trindade, R. I. F. & Macouin, M. 2007. Paleolatitude of glacial deposits
and paleogeography of Neoproterozoic ice ages. Geoscience, 339,
200– 211.
Van der Voo, R. 1993. Paleomagnetism of the Atlantic, Tethys and
Iapetus Oceans. Cambridge University Press, Cambridge.
Wegner, A. 1912. The origin of the continents. Geologische Rundschau,
3, 276– 292.
Wolfe, W. 1976. Geology of Squaw Head, Squantum, Massachusetts In:
Cameron, B. (ed.) Geology of Southeastern New England: Prince-
ton, New Jersey, New England Intercollegiate Geologic Conference,
68th Annual Meeting, Guidebook, 107– 116.
Wu, F., Van der Voo, R. & Johnson, R. J. E. 1986. Eocambrian
palaeomagnetism of the Boston basin: evidence for a displaced
terrane. Geophysical Research Letters, 13, 1450.
Zartman, R. E. & Naylor, R. S. 1984. Structural implications of some
radiometric ages of igneous rocks in southeastern New England.
Geological Society of America Bulletin, 95, 522–539.
 Chapter 44

The Chiquerı́o Formation, southern Peru

DAVID CHEW1* & CHRISTOPHER KIRKLAND2

1
Department of Geology, Trinity College Dublin, Dublin 2, Ireland
2
Laboratory for Isotope Geology, Swedish Museum of Natural History, S-104 05 Stockholm, Sweden
*Corresponding author (e-mail: chewd@tcd.ie)

Abstract: The Chiquerı́o Formation (Fm.) is a thick glaciogenic succession deposited unconformably on gneisses of the Arequipa massif
in southern Peru. It has undergone greenschist facies metamorphism during Early Palaeozoic orogenesis. The Chiquerı́o Fm. consists of
nearly 400 m of diamictite, sandstone, mudstone and carbonate, with a thin (11 m) cap dolostone at the top of the formation. It is overlain
by the San Juan Fm., a 2-km-thick carbonate succession. The thick glacially influenced succession was deposited in deep marine conditions
and consists mainly of massive diamictites (representing either ice-rafted debris or submarine debris flows) interbedded with turbiditic sand-
stones. Where internal lamination is present (e.g. bedding in the turbiditic packages), abundant dropstones can be recognized. There is no
evidence of shallow marine reworking of the succession. No absolute age constraints on the depositional timing of the Chiquerı́o Fm. exist,
because no volcanic tuffs have yet been identified. U–Pb dating of detrital zircons (U–Th–Pb SIMS) from the Chiquerı́o Fm. and the over-
lying San Juan Fm. suggest it is autochthonous with respect to Amazonia, as the detrital zircon age spectra suggest derivation from the
Amazonian craton. Detrital grains as young as c. 700 Ma have been documented in the post-glacial San Juan Fm. The sparse (chemo)strati-
graphic data available for the Chiquerı́o Fm. exhibit patterns similar to those observed generally in Neoproterozoic post-glacial carbonate
sequences. Palaeogeographic models for the deposition of the Chiquerı́o Fm. are critically dependent on the timing of the docking of the
basement of the Arequipa massif with the South American craton (Amazonia). Presently there are no palaeomagnetic constraints. More
research on the chronological and palaeogeographical constraints of this succession is required.

Supplementary material: Data are available at http://www.geolsoc.org.uk/SUP18479.

The Chiquerı́o Fm. crops out locally on the western coast of
southern Peru (Fig. 44.1a). It is best exposed at its type locality
5 km SE of the town of San Juan (Fig. 44.1b, 75880 W, 158240 S,
UTM 18L 482500 8301000). It rests unconformably on basement
gneiss (Fig. 44.1b), termed the Arequipa massif (Cobbing &
Pitcher 1972; Ramos 2008), and is cut by Early Palaeozoic intru-
sions (Loewy et al. 2004). The Chiquerı́o Fm. is overlain uncon-
formably by Jurassic sedimentary rocks, although the oldest
cover rocks at its type locality (Fig. 44.1b) are Neogene in age.
There are very few data from other sections of the Chiquerı́o
Fm. Caldas (1978) documents the presence of a well-exposed
section of the Chiquerı́o Fm. in the Quebadra Jahuay (Fig. 44.1a,
748510 W, 158280 S), while sporadic outcrops are encountered over-
lying the Arequipa massif basement in the vicinity of Marcona
Mine (Fig. 44.1a, 75870 W, 158120 S).
The first detailed study of these rocks was undertaken by the
Marcona Mining Company (1968). They defined the Marcona
Fm. to include all the low-grade metasedimentary rock overlying
the basement gneisses, and they considered the Marcona Fm. to
be Carboniferous in age. The basal member of the Marcona Fm.
was termed the Justa Member, and described as conglomerate
with pebbles of gneissic basement. Wilson (1975) obtained four
K –Ar ages for the San Nicolas batholith, which cross-cuts the low-
grade metasedimentary rocks in the region. The ages obtained (K –
Ar hornblende ages of 442 + 10 Ma and 438 + 9 Ma and K –Ar
biotite ages of 428 + 12 Ma and 421 + 11 Ma) demonstrated
that the Marcona Fm. must be pre-Late Ordovician in age.
Caldas (1978), in the course of a regional mapping programme,
reinterpreted the stratigraphy of the Marcona Mining Company
(1968). The basal Justa Member of the Marcona Fm. was termed
the Chiquerı́o Fm., and the glaciogenic nature of this part of the
sequence was recognized for the first time (Caldas 1978, 1979).
The overlying dolomitic rocks were also assigned to a new for-
mation, the San Juan Fm. The Marcona Fm. was redefined to rep-
resent the phyllitic, siliciclastic unit lying above the San Juan Fm.
Caldas (1978, 1979) inferred a Late Precambrian age for the Chi-
querı́o Fm. and the overlying dolomitic San Juan Fm. based on the
stratigraphic position of the Chiquerı́o Fm. above the Precambrian
basement gneiss and the Late Ordovician age for the cross-cutting
San Nicolas batholith (Wilson 1975). Caldas (1978, 1979) con-
sidered the Marcona Fm. to be Late Precambrian –Early Palaeo-
zoic in age.
Subsequent workers (Shackleton et al. 1979; Cobbing 1981) res-
urrected the terminology of the Marcona Mining Company (1968),
with the Marcona Fm. representing all the low-grade metasedi-
mentary rock overlying the basement gneiss including the basal
glaciogenic strata. Shackleton et al. (1979) regarded the Marcona
Fm. to be entirely Early Palaeozoic based on structural consider-
ations. He traced deformation events from the Marcona Fm. into
the underlying basement gneiss and established that the crystalline
basement rocks had experienced an older deformation history. The
glaciogenic rocks were described by Cobbing (1981) in the IGCP
38 volume on the Earth’s Pre-Pleistocene Glacial Record
(Hambrey & Harland 1981). Cobbing (1981) considered them to
be Early Palaeozoic in age. Subsequently, Injoque & Romero
(1986) described possible Precambrian stromatolites in the San
Juan Fm.
Loewy et al. (2003, 2004) undertook whole-rock Pb and U – Pb
zircon geochronological analyses on the Chiquerı́o Fm. and the
underlying gneisses of the Arequipa massif. They used, with modi-
fications, the stratigraphic terminology of Caldas (1978, 1979).
The Chiquerı́o Fm. was defined as the basal, glaciogenic portion
of the sequence and was considered to be Neoproterozoic in age.
Loewy et al. (2003, 2004) were unable to find any significant
differences between the San Juan and Marcona formations of
Caldas (1978, 1979), and considered the San Juan Fm. to include
all the low-grade metasedimentary rock overlying the Chiquerı́o
Fm. at this locality. This stratigraphic nomenclature was adopted
by Chew et al. (2007a) and is used in this chapter. Chew et al.
(2007a) presented sedimentary observations, chemostratigraphic
data and U –Pb detrital zircon analyses from the Chiquerı́o and
San Juan formations. They considered the Chiquerı́o and San
Juan formations Late Neoproterozoic in age, and to be autochtho-
nous with respect to the Amazonian craton. These considerations

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 481– 486. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.44
482
(a) (b)
76° 75° 480
Early Palaeozoic plutons
San Juan and Chiquerío N
Formations
Arequipa massif
50
basement gneisses
Marcona 15° Punta
0 20 km
Mine San Juan
8300
Ecuador
San Juan
Fig. 44.1b
Brazil
Quebadra
Peru Jahuay
N
16°
8296
also constrained the docking history of the underlying Arequipa
massif to be Late Neoproterozoic or older in age, most likely jux-
taposing during the 1.3–1.0 Ga Grenville –Sunsas orogeny (Chew
et al. 2007a), as first postulated by Loewy et al. (2004).
Structural framework
Very limited work has been published on the nature of the basin in
which the sediments of the Chiquerı́o and San Juan formations
accumulated. The present-day outcrop distribution of the Chi-
querı́o and San Juan formations is restricted to a thin zone,
20 km wide, along a strike length of 200 km on the western
margin of the Arequipa massif in southern Peru (Fig. 44.1a).
However, the initial geometry and tectonic setting of this Late
Neoproterozoic –?Early Palaeozoic basin is uncertain. Juvenile
extensional magmatism (dacite dykes) has been dated at 635 +
5 Ma in the basement of the Antofalla terrane in Northern Chile
(Loewy et al. 2004). Late Neoproterozoic extension-related
volcanism related to Rodinia break-up has been identified in the
Puncoviscana fold belt of northwestern Argentina (Omarini et al.
1999), although the Puncoviscana Basin has a complex history
of extension, compression and magmatism (Ramos 2008) that
may not be directly comparable to the tectonic evolution of the
basin where the sediments of the Chiquerı́o and San Juan for-
mations accumulated. Considering a Late Neoproterozoic age for
the Chiquerı́o and San Juan formations (e.g. Caldas 1978, 1979;
Chew et al. 2007a; Loewy et al. 2003, 2004), an extensional
basin setting is most likely.
Shackleton et al. (1979) proposed a structural evolution of the
Chiquerı́o and San Juan formations. Two deformation events are
recognized within these rocks. Two earlier deformation events
(D1 and D2) are restricted to the underlying Arequipa massif base-
ment, so the deformation events affecting the Chiquerı́o and San
Juan formations are attributed to the regional D3 and D4 defor-
mation events (Shackleton et al. 1979). The earlier event (D3) pro-
duced a bedding-parallel schistosity (S3) formed by fine-grained
muscovite and biotite. Where suitable strain markers are present
(such as in the conglomeratic portions of the Chiquerı́o Fm.), a
strong lineation is aligned along this foliation surface, plunging
moderately to the south. The S3 schistosity is crenulated and
folded by a second deformation (D4), which controls the large-
scale distribution of the units. Peak metamorphic conditions
were attained during the D3 event, and the muscovite and biotite
assemblages present are indicative of the greenschist facies
(Shackleton et al. 1979). Undeformed fine-grained granite dykes
cut the F3 and F4 fold axial planes, and one such granite dyke
has yielded a U –Pb zircon lower intercept age of between 468
D. CHEW & C. KIRKLAND
484 488
45
SAN
5 JUAN
30
85
54 48
Punta El Cenicero 50
.3
Fig. 44
Fault
Neogene
Fig. 44.1. (a) Location map and geological
San Juan Fm. 80 70
Chiquerío Fm. map of the western coast of southern Peru.
Punta
Arequipa massif . 44.2 (b) Geological map of the region around
Chiquerío Fig
basement gneisses San Juan, modified from Caldas (1978,
0 1km
1979).
and 440 Ma (Loewy et al. 2004). The regional D3 and D4 events
are considered to be Early Palaeozoic in age (Loewy et al.
2004), and are probably coeval with Famatinian (Early Ordovi-
cian) metamorphism and subduction-related magmatism on the
western Gondwanan margin (Chew et al. 2007b).
Stratigraphy
The Chiquerı́o Fm. rests unconformably on gneisses of the Are-
quipa massif and consists of nearly 400 m of diamictite, sandstone,
mudstone and carbonate, with a thin (11 m) finely laminated dolos-
tone and dolomicrite unit at the top of the formation (Fig. 44.2,
Chew et al. 2007a). It is overlain by the San Juan Fm., a
2-km-thick carbonate succession (Fig. 44.3). The San Juan Fm.
consists of several hundred metres of massive beige dolomite,
with subordinate thinly bedded limestone, black shale, graded
pebbly dolostone and phyllite (Fig. 44.3, Chew et al. 2007a). No
unconformities have been observed in either formation. At many
localities, the San Juan Fm. rests directly on the basement gneiss
(Caldas 1978).
Glaciogenic deposits and associated strata
The following description of the glaciogenic deposits of the Chi-
querı́o and San Juan formations is derived chiefly from Chew
et al. (2007a), based on the well-exposed coastal section SE of
the town of San Juan (Fig. 44.1b). The basal siliciclastic section
of the Chiquerı́o Fm. is 348 m thick (Fig. 44.2), and consists pri-
marily of massive diamictite with poorly developed internal strati-
fication. The matrix of the diamictite is a dark meta-siltstone,
whereas the majority of the clasts are granitic gneiss that superfi-
cially resemble the underlying Arequipa massif basement
(Loewy et al. 2004; Chew et al. 2007a). The clasts display no evi-
dence of faceting or striation. Clast types include weakly foliated,
K-feldspar rich granites, foliated grey-pink gneiss (sometimes
megacrystic), fine-medium grained sandstone blocks and clasts
of amphibolite. Some of the gneissic clasts are greater than
50 cm across. There is only very occasional evidence of stratifica-
tion in the massive diamictite portion of the sequence. This strati-
fication is present in the form of thin, discontinuous mudstone
layers and occasional lenses, up to 3 m thick, of boulder conglom-
erate. These conglomeratic lenses are poorly sorted, and have a
high concentration of clasts, approaching being clast supported.
Between 76 and 152 m above the basement contact there is a
sequence of stratified diamictite interbedded with thin siltstone
and graded sandstone beds. The stratification in the diamictite is
 THE CHIQUERÍO FORMATION
Fig. 44.2. (a) Stratigraphic section of the Chiquerı́o Fm. and the basal portion
of the San Juan Fm. The line of section is illustrated in Figure 44.1b. (b) Log of
the carbonate portion of the upper Chiquerı́o Fm. This section (359–410 m) is
indicated by a box in (a).
defined by either thin siltstone beds or mudstone lenses. This
internal stratification is much more pronounced and more continu-
ous than that seen in the underlying massive diamictite. The stra-
tified diamictite contains abundant outsized clasts of granitic
gneiss that deflect underlying lamina. A large angular megaclast
of bedded sandstone, greater than 6 m across (Fig. 44.2) is also
present within diamictite in this part of the sequence. It appears
to pierce the crude lamination in the underlying diamictite.
The upper part of the Chiquerı́o Fm. and the overlying San Juan
Fm. are predominantly carbonate, with a relatively abrupt switch
483
Fig. 44.3. Stratigraphic section and C-isotopic trends through the upper part of
the Chiquerı́o Fm. and the San Juan Fm. Stratigraphic heights are in metres
above the Chiquerı́o Fm.– Arequipa massif basement contact. The line of the
section is illustrated in Figure 44.1b.
(transitional over 1 m) to carbonate-dominated sedimentation
occurring at 348 m (Fig. 44.2). The dominant lithology is a calcar-
eous diamictite with white dolostone and limestone clasts, and
only minor amounts of granitic gneiss. The carbonate clasts are
484 D. CHEW & C. KIRKLAND
strongly flattened and stretched along the main bedding-parallel
tectonic fabric (S3). Overlying the carbonate diamictite are 11 m
of finely laminated (0.2 –5 cm), fine-grained pink dolostone and
dark dolomicrite in the upper Chiquerı́o Fm. (Chew et al. 2007a).
This dolomite unit shows no internal structure apart from prominent
lamination; it has no evidence of outsized clasts. The San Juan Fm.
overlies this fine-grained laminated dolostone –dolomicrite unit.
The basal portions of the formation consist of several hundred
metres of predominantly massive beige dolomite (Fig. 44.3,
Chew et al. 2007a). This is overlain by a lithologically varied
unit of black shale, massive dolomite, and thinly bedded graded
pebbly dolostone (950 –1093 m; Fig. 44.3). The overlying unit is
170 m thick, and consists of thinly bedded limestone and dark
micrite (Fig. 44.3). Above this unit, there is a thick (nearly 1 km)
sequence of massive dolomite that is only briefly interrupted by
the deposition of a thin package of graded pebbly dolostone and
mudstone (1395 – 1487 m; Fig. 44.3).
Boundary conditions with overlying and underlying
non-glacial units
The basal contact of the Chiquerı́o Fm. is an unconformity with
gneissic rock of the Arequipa massif basement, best seen at low
tide, along the coast, 5 km SE of San Juan. At the regional scale,
it is unknown whether the unconformity surface is planar or has
topography. At many other localities (e.g. Punta San Juan, 2 km
west of the town of San Juan, Fig. 44.1b), the contact between
the Chiquerı́o Fm. and the basement gneiss is a fault. In addition,
the entire Chiquerı́o Fm. is frequently excised, with the overlying
San Juan Fm. resting directly on the basement gneiss (Caldas
1978).
The upper contact of the Chiquerı́o Fm. has been placed at either
the base (e.g. Caldas 1978) or at the top of the finely laminated,
11-m-thick unit of pink dolostones and dark dolomicrites (e.g.
Cobbing 1981; Chew et al. 2007a). The Chiquerı́o and San Juan
formations are overlain unconformably by unmetamorphosed
Mesozoic and Cenozoic sediments (Caldas 1978).
Chemostratigraphy
Chew et al. (2007a) presented C- and O-isotope data for the
upper part of the Chiquerı́o Fm. and the overlying San Juan
Fm. (Fig. 44.3). These are the only chemostratigraphic data pre-
sently available for the Chiquerı́o and San Juan formations.
Diagenetic overprinting of the original seawater isotopic signa-
tures is difficult to assess. Detailed textural evidence to evaluate
diagenesis within the carbonate rocks is lacking as the rocks
have undergone some recrystallization during greenschist-facies
metamorphism, and there is no alternative complete section
with which to compare lateral variations in the stable
isotope profile. The beginning of carbonate-dominated sedimen-
tation in the Chiquerı́o Fm. occurs at 348 m (Fig. 44.2), where
the dominant lithology is a calcareous diamictite with white
dolostone and limestone clasts. Both the clasts within the dia-
mictite and the interbedded limestone beds yield d13C values
between 0‰ and þ2‰ (VPDB) (Fig. 44.3). Finely laminated
pink dolostone and dark dolomicrite overlie the carbonate dia-
mictite. This dolostone unit yields consistent negative d13C
values of –2‰ (Fig. 44.3, Chew et al. 2007a).
The overlying San Juan Fm. exhibits a recovery in d13C values
to between þ1‰ and þ2‰ (Fig. 44.3). The basal portions of the
formation consist of several hundred metres of predominantly
massive beige dolomite. Between 1075 m and 1250 m, a thinly
bedded limestone and dark micrite unit exhibits strongly negative
d13C values from –5‰ to –8‰ (five data points, Fig. 44.3, Chew
et al. 2007a). Above this unit, there is a return to deposition of
massive dolomite and the d13C values range between þ1‰ and
þ2.5‰. So far, no unequivocal glaciogenic strata nor significant
sequence boundary associated with this younger, strongly nega-
tive ( –5‰ to – 8‰) d13C excursion have been identified (Chew
et al. 2007a).
Other characteristics (e.g. economic deposits, biomarkers)
The Marcona deposit (20 km north of San Juan, Fig. 44.1a) and the
associated Pampa de Pongo deposit (35 km east of San Juan) are
the largest Fe accumulations, with associated copper and gold,
along the western South America margin. The deposit substan-
tially post-dates the deposition of the glaciogenic strata, and is con-
sidered to have formed during a phase of Mesozoic arc magmatism
(Hawkes et al. 2002). Approximate resources include more than
1400 Mt of iron ore at Marcona and 1000 Mt of magnetite miner-
alization at Pampa de Pongo (Hawkes et al. 2002). The two depos-
its form part of a cluster of similar occurrences that together define
the ‘Marcona Fe –Cu District’. The larger Fe bodies are located
within the Chiquerı́o, San Juan and Marcona formations, and
also by basaltic andesite, andesite and volcaniclastic rock of the
Middle to Upper Jurassic Rio Grande Fm. (Hawkes et al. 2002).
There are no biostratigraphic data available for the Chiquerı́o
Fm. ‘Stromatolite-like’ structures have been recorded in the over-
lying San Juan Fm. (Injoque & Romero 1986) and are corre-
lated by the authors with late Neoproterozoic– Early Cambrian
stromatolites.
Palaeolatitude and palaeogeography
Palaeolatitudinal and palaeogeographic constraints for the Chi-
querı́o Fm. are sparse. There have been no palaeomagnetic
studies on the Chiquerı́o Fm. Given that the Chiquerı́o and San
Juan formations have experienced greenschist-facies metamorph-
ism (Shackleton et al. 1979) and were subsequently intruded by
Early Palaeozoic plutons (Loewy et al. 2004), remagnetization
by Early Palaeozoic-age metamorphism is likely.
The detrital zircon data of Chew et al. (2007a) from the Chi-
querı́o and San Juan formations are consistent with derivation
from the Proto-Andean margin (Chew et al. 2007b). This would
imply that both the glaciogenic strata and its underlying gneissic
basement were proximal to the South American craton (Amazonia)
during Late Neoproterozoic times (Chew et al. 2007a). This juxta-
position indicates that the Arequipa massif basement must have
accreted earlier, probably during the 1.3– 1.0 Ga Grenville –
Sunsas Orogeny (Chew et al. 2007a) as first postulated by
Loewy et al. (2004). Ramos (2008) and Loewy et al. (2004)
provide a detailed synthesis on the tectonic evolution and
docking history of the Arequipa massif and the Antofalla terrane.
An autochthonous origin for the Chiquerı́o Fm., with respect to
cratonic South America, places crude palaeolatitudinal constraints
on these rocks. There is presently only one palaeomagnetic pole
for the Amazon craton in the Late Neoproterozoic (Tohver et al.
2006), derived from the palaeomagnetic study of the Neoprotero-
zoic Puga cap carbonate (Trindade et al. 2003). The dolomite and
limestone of the Puga Fm. from the SE Amazon craton preserve a
dual-polarity component that is interpreted as a primary magneti-
zation. This implies a low palaeolatitude of 22 þ 6/ –58 for the
Amazonian block just after deposition of the Puga diamictites.
Although direct ages for the Puga Fm. are not yet available,
87
Sr/86Sr ratios and d13C results presented by de Alvarenga et al.
(2004) suggest correlation with the c. 635 Ma (Hoffmann et al.
2004; Condon et al. 2005) post-glacial units of the Congo craton.
Geochronological constraints
To date, no tuffs have been recorded from the Chiquerı́o Fm.
Existing age constraints include a minimum age of 468– 440 Ma
 THE CHIQUERÍO FORMATION 485

c Proterozoic
subsidiary peak at c. 1000 Ma (Chew et al. 2007a, Fig. 44.4).
SJ-16 is a sample of diamictite matrix from the Chiquerı́o Fm. It
SJ-57, n=55/66 Pebbly limestone,
is also characterized by a restricted age distribution from 950 to
5 6 1300 Ma, with a prominent peak at c. 1200 Ma and a subsidiary
697 ± 11 Ma San Juan Formation peak at c. 1000 Ma (Chew et al. 2007a, Fig. 44.4). The detrital
4 5 zircon data from these Chiquerı́o Fm. samples yield very
minimal detritus, which could potentially be derived from the
4 underlying basement (the Palaeoproterozoic Arequipa massif,
3 1790 –2020 Ma; Loewy et al. 2004). Sample SJ-57 (55 grains) is
3 from a coarse pebbly limestone bed from the San Juan Fm.,
2 2 1412 m above the Chiquerı́o Fm. – Arequipa massif basement
contact and 178 m above the second negative C-isotope excursion
1 1 (Fig. 44.4). The majority of grains from this sample also lie in the
950– 1300 Ma range, with peaks at c. 1000 Ma and c. 1200 Ma.
There are also minor peaks within the c. 1600–2000 Ma and
9 SJ-16, n=62/71 Siltstone (tillite matrix) 14 c. 700 –830 Ma intervals (Chew et al. 2007a, Fig. 44.4). A
8 Chiquerío Formation c. 700 Ma grain provides a maximum age constraint for the depo-
932 ± 28 Ma 12 sition of this portion of the San Juan Fm.
7
6 10
5 8 Discussion
4
6 The depositional environment of the Chiquerı́o Fm. is most likely
3
2 4 deep marine, based on the lithofacies and the lack of high-energy
sedimentary structures. The stratified portion of the siliciclastic
1 2 section (between 76 and 152 m above the basement contact,
Fig. 44.2) contains abundant dropstones of granitic gneiss,
9
Turbiditic sandstone 10
8 SJ-11, n=35/46
which disrupt the lamination in graded turbiditic sandstone beds,
9 thus suggesting a glacially influenced marine environment
Chiquerío Formation 8
7 955 ± 18 Ma (Caldas 1978; Cobbing 1981; Loewy et al. 2004; Chew et al.
7 2007a). The depositional environment of the massive diamictite
6
6 portions straddling this stratified interval (Fig. 44.2) may represent
Probability x 10-3

5 ice-rafted debris and suspension settling of fine grained sediment,

5
4 or alternatively may have been produced by submarine debris
4
Frequency

3 flows. There is no sedimentary evidence in the siliciclastic

3 portion of the sequence (e.g. wave ripples, cross-bedding) of
2 2 shallow-water conditions (Chew et al. 2007a). The San Juan Fm.
1 1 consists predominantly of massive beige dolomite with little
0 0 internal structure. The possible Precambrian stromatolites
described by Injoque & Romero (1986) would suggest a shallow
400
500
600
700
800
900
1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
2000

marine or intertidal environment for portions of the San Juan Fm.

Age (Ma)
Fig. 44.4. Zircon probability density distribution diagrams from the Chiquerı́o
Fm. (SJ-11, SJ-16) and the San Juan Fm. (SJ-57) (Chew et al. 2007a). Light
grey curves represent all ages from each sample, and the dark curves represent
ages that are .90% concordant. The youngest detrital zircon age in each
sample is shown within a black box.
based on a loosely defined U –Pb TIMS zircon lower intercept
from the cross-cutting, post-tectonic San Juan granite of the San
Nicolas batholith (Loewy et al. 2004), and maximum ages of
932 + 28 Ma and 955 + 18 Ma (the youngest detrital U –Pb
SIMS zircon ages from the study of Chew et al. 2007a).
Loewy et al. (2004) also dated three clasts from the Chiquerı́o
Fm. SE of San Juan by U –Pb TIMS zircon. Two clasts of pink,
weakly foliated, K-feldspar rich megacrystic granite yielded ages
of 1168 þ 9/–6 Ma, 1162 + 6 Ma, while a third clast of similar
composition but with a gneissic foliation yielded a poorly con-
strained upper intercept of c. 1165 Ma. Chew et al. (2007a) under-
took U –Th –Pb SIMS analyses of detrital zircons from three
samples from the Chiquerı́o and San Juan formations. Combined
age (probability-density distribution) plots and histograms for
the three samples from that study are illustrated in Figure 44.4.
Sample SJ-11 is from a thin graded turbiditic sandstone bed
from the Chiquerı́o Fm. It yields a restricted age distribution of
950– 1300 Ma, with a prominent peak at c. 1200 Ma and a
The laminated dolostone facies at the top of the Chiquerı́o Fm.
and its associated large negative d13C excursion are characteristic
of cap dolostone associated with Late Neoproterozoic glacials
(Kennedy et al. 1998; Hoffman & Schrag 2002; Halverson et al.
2005; Shields 2005), although negative anomalies in the late Neo-
proterozoic are not exclusively linked to ice ages (Le Guerroué
et al. 2006). Chew et al. (2007a) considered the Chiquerı́o Fm.
and the pronounced negative C-isotope excursion in the San Juan
Fm. to represent two distinct glacial events correlated to a ‘Stur-
tian–Marinoan’ couplet elsewhere in the world. Although no
unequivocal glaciogenic strata nor a significant sequence boundary
indicative of a glacial event have been identified with the second
negative C-isotope excursion, it may correlate with the negative
Trezona anomaly, which immediately preceded the Marinoan gla-
ciation (Halverson et al. 2005). Alternatively, if Chiquerı́o Fm.
and the C-isotope excursion represent a Marinoan-age glacial
event and the Shuram/Wonoka isotopic anomaly (Halverson et al.
2005; Le Guerroué et al. 2006), then a depositional age of c.
635 Ma for the Chiquerio Fm. (Chew et al. 2007a) is inferred. In
either case, a Late Neoproterozoic (,700 Ma) age is supported by
the youngest detrital zircon population in the San Juan Fm. The
Proto-Andean margin of Amazonia is characterized by abundant
zircon detritus between 1300–900 Ma and 650–550 Ma (Chew
2007b). The absence of Late Neoproterozoic (700 Ma and
younger) zircon in the Chiquerı́o Fm. may simply reflect that
these glaciogenic strata were too old to accumulate such detritus,
or alternatively had a restricted sediment source, mainly derived
from local basement.
486 D. CHEW & C. KIRKLAND
Further global correlation of the Chiquerı́o Fm. and the pro-
nounced negative C-isotope excursion in the San Juan Fm. are
hampered by the lack of consensus on the temporal range of the
‘Sturtian’ glacial episode (see Hoffman & Li 2009 for a review),
the lack of absolute age constraints for the Chiquerı́o and San
Juan formations, and the relatively low-resolution sampling
employed for the stable isotope study. Further areas of research
that might prove beneficial in the future include higher-resolution
sampling for stable isotope analysis in the San Juan Fm. and a
comprehensive search for tuffs in the glaciogenic strata.
This study was funded by the Swiss National Science Foundation under a grant
held by U. Schaltegger. We are extremely grateful to Urs Schaltegger at the Uni-
versity of Geneva, the Geological Survey of Peru (INGEMMET) and C. Moreno
of S. Marcos University in Lima for scientific advice and logistical support during
the field seasons in Peru. The careful and insightful reviews of V. A. Ramos, E. Le
Guerroué and editor E. Arnaud are gratefully acknowledged. This work represents
a contribution of the IUGS- and UNESCO-funded IGCP (International
Geoscience Programme) project #512.
References
Caldas, J. 1978. Geologı́a de los Cuadrángulos de San Juan, Acarı́ y
Yauca, Hojas: 31-m, 31-n, 32-n. Instituto de Geologı́a y Minerı́a,
Lima.
Caldas, J. 1979. Evidencias de una glaciación Precambriana en la costa
sur del Perú. Segundo Congreso Geológico Chileno J, Arica, 29 – 37.
Chew, D. M., Kirkland, C. L., Schaltegger, U. & Goodhue, R.
2007a. Neoproterozoic glaciation in the Proto-Andes: tectonic impli-
cations and global correlation. Geology, 35, 1095– 1099.
Chew, D. M., Schaltegger, U., Košler, J., Whitehouse, M. J.,
Gutjahr, M., Spikings, R. A. & Miškovic, A. 2007b. U– Pb geo-
chronologic evidence for the evolution of the Gondwanan margin
of the north-central Andes. Geological Society of America Bulletin,
119, 697– 711.
Cobbing, E. J. 1981. Tillites at the base of the possible Early Palaeozoic
Marcona Formation, southwest coastal Peru. In: Hambrey, M. J. &
Harland, W. B. (eds) Earth’s Pre-Pleistocene Glacial Record.
Cambridge University Press, Cambridge, 899– 901.
Cobbing, E. J. & Pitcher, W. S. 1972. Plate tectonics and the Peruvian
Andes. Nature, 246, 51– 53.
Condon, D., Zhu, M. Y., Bowring, S., Wang, W., Yang, A. H. & Jin,
Y. G. 2005. U–Pb ages from the Meoproterozoic Doushantuo For-
mation, China. Science, 308, 95 –98.
de Alvarenga, C. J. S., Santos, R. V. & Dantas, E. L. 2004. C –O– Sr
isotopic stratigraphy of cap carbonates overlying Marinoan-age
glacial diamictites in the Paraguay Belt, Brazil. Precambrian
Research, 131, 1 –21.
Halverson, G. P., Hoffman, P. F., Schrag, D. P., Maloof, A. C. &
Rice, A. H. N. 2005. Toward a Neoproterozoic composite carbon-
isotope record. Geological Society of America Bulletin, 117,
1181– 1207.
Hambrey, M. J. & Harland, W. B. 1981. Earth’s Pre-Pleistocene
Glacial Record. Cambridge University Press, London & New York.
Hawkes, N., Clark, A. H. & Moody, T. C. 2002. Marcona and Pampa de
Pongo: Giant Mesozoic Fe-(Cu, Au) deposits in the Peruvian Coastal
Belt. In: Porter, T. M. (ed.) Hydrothermal Iron Oxide Copper –Gold
& Related Deposits: A Global Perspective. PGC Publishing, Ade-
laide, Australia, 2, 115–130.
Hoffman, P. F. & Schrag, D. P. 2002. The snowball Earth hypothesis:
testing the limits of global change. Terra Nova, 14, 129–155.
Hoffman, P. F. & Li, Z.-X. 2009. A palaeogeographic context for Neopro-
terozoic glaciation. Palaeogeography, Palaeoclimatology, Palaeo-
ecology, 277, 158– 172.
Hoffmann, K. H., Condon, D. J., Bowring, S. A. & Crowley, J. L.
2004. U– Pb zircon date from the Neoproterozoic Ghaub
Formation, Namibia: constraints on Marinoan glaciation. Geology,
32, 817– 820.
Injoque, J. & Romero, L. 1986. Estromatolitos (?) en la formación San
Juan, San Juan de Marcona. Evidencia de estructuras fósiles precám-
bricas en el Perú. De re metallic.De la minerı́a y los metales: Revista
del Instituto Geológico Metalúrgico, 11, 4 –5.
Kennedy, M. J., Runnegar, B., Prave, A. R., Hoffmann, K. H. &
Arthur, M. A. 1998. Two or four Neoproterozoic glaciations?
Geology, 26, 1059–1063.
Le Guerroué, E., Allen, P. A., Cozzi, A., Etienne, J. L. & Fanning,
C. M. 2006. 50 Myr recovery from the largest negative d13C excur-
sion in the Ediacaran ocean. Terra Nova, 18, 147– 153.
Loewy, S. L., Connelly, J. N., Dalziel, I. W. D. & Gower, C. F. 2003.
Eastern Laurentia in Rodinia: constraints from whole-rock Pb and
U/Pb geochronology. Tectonophysics, 375, 169– 197.
Loewy, S. L., Connelly, J. N. & Dalziel, I. W. D. 2004. An orphaned
basement block: The Arequipa –Antofalla basement of the central
Andean margin of South America. Geological Society of America
Bulletin, 116, 171–187.
MARCONA MINING COMPANY. 1968. Geologic Map of the Marcona
iron deposits.
Omarini, R. H., Sureda, R. J., Götze, H. J., Seilacher, A. & Pflüger,
F. 1999. Puncoviscana folded belt in northwestern Argentina: testi-
mony of Late Proterozoic Rodinia fragmentation and pre-Gondwana
collisional episodes. International Journal of Earth Sciences, 88,
76 – 97.
Ramos, V. A. 2008. The basement of the Central Andes: the Arequipa and
related terranes. Annual Review of Earth and Planetary Sciences, 36,
289– 324.
Shackleton, R. M., Ries, A. C., Coward, M. P. & Cobbold, P. R. 1979.
Structure, metamorphism and geochronology of the Arequipa
Massif of coastal Peru. Journal of the Geological Society, 136,
195– 214.
Shields, G. A. 2005. Neoproterozoic cap carbonates: a critical appraisal
of existing models and the plumeworld hypothesis. Terra Nova, 17,
299– 310.
Tohver, E., D’Agrella-Filho, M. S. & Trindade, R. I. F. 2006. Paleo-
magnetic record of Africa and South America for the 1200–500 Ma
interval, and evaluation of Rodinia and Gondwana assemblies.
Precambrian Research, 147, 193– 222.
Trindade, R. I. F., Font, E., D’Agrella-Filho, M. S., Nogueira,
A. C. R. & Riccomini, C. 2003. Low-latitude and multiple geomag-
netic reversals in the Neoproterozoic Puga cap carbonate, Amazon
craton. Terra Nova, 15, 441–446.
Wilson, P. A. 1975. Potassium– argon age studies in Peru with particular
reference to the chronology of emplacement of the coastal batholith.
PhD thesis, University of Liverpool.
 Chapter 45

Glacially influenced sedimentation of the Puga Formation, Cuiabá Group

and Jacadigo Group, and associated carbonates of the Araras
and Corumbá groups, Paraguay Belt, Brazil

CARLOS J. S. ALVARENGA1 *, PAULO C. BOGGIANI2, MARLY BABINSKI2, MARCEL A. DARDENNE1,

MILENE F. FIGUEIREDO2, ELTON L. DANTAS1, ALEXANDRE UHLEIN3, ROBERTO V. SANTOS1,
ALCIDES N. SIAL4 & ROLAND TROMPETTE5
1
Instituto de Geociências, Universidade de Brası́lia, Campus Darcy Ribeiro, CEP 70910-900, Brası́lia, Brazil
2
Instituto de Geociências, Universidade de São Paulo, Rua do Lago, 562, CEP 05508-080, São Paulo, Brazil
3
Instituto de Geociências, Universidade Federal de Minas Gerais, Av. Antônio Carlos, 6627, Campus da Pampulha,
CEP 31270-901, Belo Horizonte, Brazil
4
Departmento de Geologia, Universidade Federal de Pernambuco, CEP 50732-970, Recife, PE, Brazil
5
CEREGE, University d’Aix-Marseille III, CNRS-FU017, BP 13545, Aix-en-Provence, Cedex 04, France
*Corresponding author (e-mail: alva1@unb.br)

Abstract: Discontinuous exposures of diamictite (.1000 km) termed the Puga Formation (Fm.) and interpreted as being related to a late
Cryogenian glacial event are known in the Paraguay Belt (Brazilian– Pan-African Orogeny) and part of the Amazon craton and Rio Apa
Block. These diamictite units and a mixed assemblage of sandstone, conglomerate and claystone were first named the Jangada Group.
Recent interpretations have shown that the diamictite of the Puga Formation passes laterally into metasediments of the Cuiabá Group,
interpreted as glacially influenced turbidites. The correlative Jacadigo Group in the southern Paraguay Belt includes a thick succession of
banded iron formation (BIF) bearing large boulders that have been interpreted as recording glacially influenced sedimentation. The dia-
mictite of the Puga Fm. is overlain by two different carbonate-bearing successions, the Corumbá Group (Cadieus, Cerradinho, Bocaina,
Tamengo and Guaicurus formations) in the south and the Araras Group (Mirassol d’Oeste, Guia, Nobres formations) in the north. Evi-
dence of glaciation in the Puga Fm. consists of striated and faceted pebbles and blocks, and dropstones in the turbidites. Sedimentary and
geochemical data from the associated carbonate reinforce the interpretation of a glacial origin. C, O and Sr isotope data from the northern
Paraguay Belt are consistent with the proposed late Cryogenian age for the Puga Fm. sedimentation.

Discontinuous outcrops of the Puga Fm. extend for more than
1000 km along the Paraguay Belt and over the Amazonian
Craton and Rio Apa Block (Fig. 45.1). The glacial diamictite
succession was first described from the isolated Puga Hill (198
370 20.0300 S, 578 310 40.0100 W), SE of Corumbá (Fig. 45.1; Maciel
1959), but its best exposures are located in the northern Paraguay
Belt, where the diamictite of the Puga Fm. passes laterally into tur-
bidites with dropstones of the Cuiabá Group (Fig. 45.2) (Alvarenga
& Trompette 1992). A glaciogenic origin was first proposed by
Maciel (1959), but Alvarenga & Trompette (1992) were the first
to suggest a glaciomarine setting; they interpreted the fine-grained
sediments of the Cuiabá Group as glacially influenced turbidites.
The diamictite exposures in the northern Paraguay Belt were
first described by Almeida (1964a, b) as part of Acorizal,
Engenho, Bauxi and Marzagão formations within the Jangada
Group. The type section was described between Jangada and
Bauxi (Fig. 45.1). Quartzite and sandstone units below the diamic-
tite (Puga Fm.) and above the slate (Cuiabá Group) were described
as the Bauxi Fm. (Almeida 1964a; Vieira 1965), whereas diamic-
tite and associated conglomerate, quartzite and slate were named
the Jangada Group (Almeida 1964a, b; Rocha-Campos & Hasuı́
1981). The name ‘Jangada Group’ is not used in the current
nomenclature.
In the southern Paraguay Belt, glacial deposits include the Jaca-
digo Group and the Puga Fm. These two units occur in isolated
hills making it difficult to confirm the stratigraphic relationships
between them. The Jacadigo Group (Almeida 1945; Dorr II
1945; Almeida 1946) is exposed in some hills south of the town
of Corumbá extending into Bolivia. The Urucum Fm. (lower
unit) is characterized by conglomerate and arkose and is overlain
by the economic manganesiferous bed and banded iron formation
(BIF) of the Santa Cruz Fm. (type section, 198 110 24.3600 S, 578
360 31.4500 W). Fe formations bearing large granite boulders have
been interpreted as recording a glacial influence (Barbosa 1949;
Walde et al. 1981; Hoppe et al. 1987; Urban et al. 1992; Klein
& Ladeira 2004; Walde & Hagemann 2007). The Jacadigo
Group, in contrast to the huge exposure area of the Puga Fm.
and Cuiabá Group in the northern Paraguay Belt, is restricted to
a few hills on the border of Bolivia and Brazil near Corumbá
(Fig. 45.1).
There are two Fe formation units in the southern Paraguay Belt.
The most significant is the economic Fe formation and Mn ore
deposits of the Jacadigo Group in Urucum Hill (Almeida 1945l
Dorr II 1945); the second is an Fe formation in diamictite of the
Puga Formation (Boggiani et al. 2006; Piacentini et al. 2007).
Carbonate rocks of the Araras Group and mixed clastic-
carbonate rocks of the Corumbá Group overlie diamictite of the
Puga Fm. and in some areas are interpreted as cap carbonate
(Nogueira et al. 2003; Boggiani et al. 2003; Alvarenga et al.
2004, 2008). Historically, ‘Corumbá and Arara Limestones’
were the first names given to these carbonate rocks in the Paraguay
Belt (Evans 1894). Most studies about carbonate rocks that overlie
the diamictite of the Puga Fm. focus on isotope geochemistry,
chemostratigraphy (Boggiani et al. 2003; Alvarenga et al. 2004,
2008; Font et al. 2006; Nogueira et al. 2007; Riccomni et al.
2007) and palaeomagnetic data (Trindade et al. 2003).
Outcrops of the Araras Group extend for more than 600 km
along the Northern Paraguay Belt. This group is subdivided into
three carbonate formations (Fig. 45.1, Table 45.1): (i) the Mirassol
d’Oeste Fm., described as a cap dolomite (Nogueira et al. 2003;

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 487– 497. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.45
488 C. J. S. ALVARENGA ET AL.

Fig. 45.1. Geological map of the Paraguay

Belt (modified from Alvarenga &
Trompette 1993; Boggiani 1998).

Alvarenga et al. 2004, 2008), (ii) the Guia Fm. (Almeida 1964a;
Hennies 1966), and (iii) the Nobres Fm. (Hennies 1966; Luz &
Abreu 1978). Good exposures of the Mirassol d’Oeste Fm. over-
lying the Puga Fm. in the Northern Paraguay Belt can be found
at the Terconi (158 400 41.8700 S, 588 040 18.2200 W) and Tangará
(148 440 13.700 S, 578 490 52.4500 W) quarries and in a drill core
from the Bauxi region (158 090 06.1400 S, 568 410 00.7400 W). The
Guia Fm. can be found in the Terconi and Tangará quarries
as well as at the Cimento Tocantins Mine (148 380 12.0200 S,
568 150 57.5000 W) in the Nobres region and the Nossa Senhora
 PUGA FORMATION, PARAGUAY BELT 489

Platformal glaciomarine depositional system Glacial diamictite-turbidite

Slope depositional system Outer slope depositional system

Araras Group

Puga Formation
Bauxi Formation
Amazonian Craton
(3.0 - 1.0 Ga.)
Cuiabá Group

Diamictite Conglomerate and Fine-grained sediments

graded pebbly sandstone with occasional clasts

Fig. 45.2. Schematic cross-section showing the glacial facies relationships for the late Cryogenian glacial event (c. 635 Ma) along the southeastern edge of the
Amazonian Craton (modified from Alvarenga & Trompette 1992). Diamictites of the Puga Fm. are in the platformal glaciomarine system, while the diamictites of the
Cuiabá Group are in the slope and outer slope depositional system.

da Guia Quarry (158 210 15.0700 S, 568 110 24.4700 W). The Nobres
Fm. is exposed in the Provı́ncia Serrana.
The Corumbá Group, in the southern Paraguay Belt, is subdi-
vided into five formations (Tables 45.2 & 45.3). The basal
Cadieus and Cerradinho formations occur only in the Serra da Bod-
oquena (Fig. 45.1, Table 45.2; Boggiani 1998). Above this unit lies
the grey dolomite of the Bocaina Fm. (Almeida 1945), which is
exposed in the Serra da Bodoquena and Corumbá regions and
which outlines the geographical extent of exposures of the
Corumbá Group in the southern Paraguay Belt. The pink carbonate
of the Bocaina Fm. overlies the Puga Fm. at Puga Hill (Table 45.3).
The uppermost two units of the Corumbá Group are the Tamengo
and Guaicurus formations (Almeida 1965) exposed in the
Corumbá region and in the eastern part of Serra da Bodoquena.
In the northern Paraguay belt, two Neoproterozoic diamictite-
bearing glacial intervals are described (Table 45.1): the units
described in this chapter and attributed to a late Cryogenian glacia-
tion (c. 630 Ma), as well as the Serra Azul Fm. attributed to the
Ediacaran glaciation (580 Ma; Alvarenga et al. 2007; Figueiredo
et al. 2008, 2011). Most rocks in the region have experienced
low-grade metamorphism. In order to facilitate subsequent rock
descriptions, the prefix ‘meta’ has been omitted.
Structural framework
Outcrops of the Neoproterozoic rocks extend along the southeast-
ern border of the Amazonian Craton as well as along the eastern
margin of the Rio Apa Block, and contain a thick sequence of gla-
ciomarine, turbidite, carbonate and siliciclastic sedimentary rocks
formed on an extensional continental margin. In the northern Para-
guay belt, the Puga Fm. is mostly subhorizontal and c. 100 m thick
on the craton, whereas in the central part of the belt, in the deepest
part of the basin, the glacially influenced sediments (Cuiabá
Group) are more than 3000 m thick (Alvarenga & Saes 1992;
Alvarenga & Trompette 1992, 1993; Dantas et al. 2009). In the
southern Paraguay Belt, the NE – SW-trending extensional struc-
tures, with sub-vertical faults of a half-graben system, were filled
by conglomerate, arkose sandstone and jaspilite of the Jacadigo
Group. The present landscape in the south is due to a later
tectonic event and erosion during the Neogene subsidence of the
Pantanal Basin, with blocks defined by vertical faults (Trompette
et al. 1998).
The overlying Corumbá and Araras groups were deposited in a
passive margin environment. The total thickness of the sedimen-
tary pile increases from c. 200 m in the western cratonic area

Table 45.1. Neoproterozoic lithostratigraphy nomenclature in the northern Paraguay Belt

Period Group Formation Lithology Ages

Diamantino Siltstone and arkose sandstone

Alto Paraguay Sepotuba Siltstone Rb–Sr: 569 ± 20 Ma

Raizama Sandstone
Ediacaran
Serra Azul Diamictite and mudstone

Nobres Dolostone and sandy dolostone

Araras Guia Limestone and mudstone
Mirassol d'Oeste Pink dolostone Pb/Pb: 633 ± 25 Ma

Puga Formation Diamictite

Cryogenian Diamictite,
conglomerate,
Cuiabá Group sandstone and
fine-grained rocks
490 C. J. S. ALVARENGA ET AL.

Table 45.2. Neoproterozoic lithostratigraphy nomenclature in the Serra da Bodoquena region, southern Paraguay Belt

Period Group Formation Lithology U–Pb Ages

Guaicurus Shale
Tamengo Limestone, mudstone and shale 543 ± 2 Ma

Corumbá Bocaina Dolostone

Ediacaran
Cerradinho Limestone, shale, sandstone
Cadieus Conglomerate, arkose and shale

Puga Diamictite
Cryogenian

(inner-shelf domain in Mirassol d’Oeste region) to more than Stratigraphy

1000 m eastwards (middle-outer shelf domain) along the Provı́ncia
Serrana (Alvarenga & Saes 1992; Alvarenga & Trompette 1993;
Trompette 1994; Alvarenga et al. 2008, 2009).
The Neoproterozoic rocks along the Paraguay Belt were folded
and metamorphosed to very low-grade metamorphism during the
Brasiliano/Pan-African Orogeny, in the Ediacaran – Cambrian
Period (Alvarenga 1990; Trompette 1994; Pimentel et al. 1996;
Trindade et al. 2003; Tohver et al. 2006). This event was followed
by post-orogenic sub-alkaline granitic magmatism at c. 500 Ma
(Almeida & Mantovani 1975).
Deformation and metamorphism increase from imperceptible
effects on flat-lying rocks overlying the stable Amazonian craton
border to low-grade metamorphism within the Paraguay Belt,
determined by folding and crystallization of the illite parallel to
cleavage planes (Alvarenga et al. 1990; Alvarenga & Trompette
1993). Despite low metamorphism, diamictite, quartzite, slate
and phyllite of the Cuiabá Group and Puga Fm., as well as the
limestone, marl, dolostone and shale of the Araras and Corumbá
groups, have well-preserved sedimentary structures in most
outcrops.
The Puga Fm. and the Cuiabá and Jacadigo Groups include diamic-
tite and fine-grained rocks with outsized clasts. In the north Para-
guay Belt, the Puga Fm. and Cuiabá Group are overlain by the
carbonate rocks of the Mirassol d’Oeste Fm., Araras Group
(Table 45.1; Nogueira et al. 2003; Alvarenga et al. 2008, 2009).
In the south Paraguay Belt glaciogenic deposits are named the
Puga Fm. and the Jacadigo Group; these are in turn overlain by
mixed clastic and carbonate rocks of the Corumbá Group
(Tables 45.2 & 45.3).
Puga Fm. and Cuiabá Group
The diamictite lying concordantly below the dolomite of the
Bocaina Fm. originally described in the southern Paraguay Belt
at Puga Hill (Maciel 1959) has been found in other areas of the
Paraguay Belt, and is also named Puga Fm.
In the northern Paraguay Belt, the Puga Fm. consists of
diamictite, sandstone and claystone units that are part of a

Table 45.3. Neoproterozoic lithostratigraphy nomenclature in the Corumbá and Puga hill region, southern Paraguay Belt

Period Group Formation Lithology U–Pb Ages

Guaicurus Shale

Corumbá Tamengo Limestone, mudstone and shale 543 ± 2 Ma

Ediacaran

Bocaina Dolostone

Puga* Diamictite

Cryogenian Santa Cruz BIF, Mn beds and arkose

Jacadigo
Urucum
Conglomerate and arkose

*Puga Fm. is thought to be coeval with the sedimentation of the Jacadigo Group, although isolated outcrops make it difficult to
confirm their stratigraphic relationship.
 PUGA FORMATION, PARAGUAY BELT 491

marine succession (c. 100 m thick) close to the western border of

the Neoproterozoic basin. To the east, in the deeper parts of the
basin, layers of diamictite intercalated with conglomerate, sand-
stone and fine-grained rocks of the Cuiabá Group (c. 3000 m

Santa Cruz Formation

thick) were deposited as sediment gravity flows (Table 45.1;
Alvarenga & Trompette 1992).
In the Bauxi region (Fig. 45.1), the Puga Fm. contains diamictite
associated with conglomerate, sandstone and siltstone in its lower Mn
part followed by a massive diamictite. At this site, the massive dia-
mictite, interbedded with conglomerate and sandstone, passes to
the east into a thick fine-grained facies with a few isolated clasts
of the Cuiabá Group.

Jacadigo Group
The exposed section of the Puga Fm., at its type section on Puga Mn
Hill (Fig. 45.1), consists of a 95-m-thick massive diamictite with a
sandy matrix, which in the lower 50 m, contains a sandstone layer Mn
intercalated with a clay matrix massive diamictite (Maciel 1959).
A pink limestone of the Bocaina Fm. rests on the diamictite
(Table 45.3; Maciel 1959) and was interpreted as a cap carbonate
typical of other Neoproterozoic successions (Boggiani & Coimbra

Urucum Formation
1996; Boggiani et al. 2003).
Diamictite of the Puga Fm. at the Serra da Bodoquena (Fig. 45.1,
Table 45.2) are found along the axis of folded rocks, overlain by 200
carbonate rocks of the Corumbá Group (Almeida 1965). To the
east of the Serra da Bodoquena a massive foliated diamictite
with a ferruginous matrix is intercalated with centimetre-thick
hematite and magnetite layers, and sub-millimetre layers of 100
quartz and chert (Piacentini et al. 2007).

Jacadigo Group 0 (m)

Between Corumbá and Puga Hill (Fig. 45.1), the Neoproterozoic

exposures of the Jacadigo Group occur in isolated hills (up to
900 m high), surrounded by lowlands of the wetland Pantanal
plain (Neogene Basin). The Jacadigo Group (Fig. 45.3) differs
from the Puga Fm. in the presence of significant ferruginous
and manganesiferous chemical deposits and the absence of
diamictite. The lower unit of the Jacadigo Group (Urucum
Fm.) comprises conglomerate and arkose sandstone, with
discontinuous conglomerate interbeds, c. 200 m thick. The Santa
Cruz Fm. is c. 400 m thick, and forms the upper part of the Jaca-
digo Group. The lower part of this formation starts with a continu-
ous layer of red-violet ferruginous-manganese arkose with a
jaspilite/hematite cement along Urucum Hill (Almeida 1945;
Walde & Hagemann 2007); this unit was named the Banda
Alta Fm. by Dorr II (1945). Towards the top of the Santa
Cruz Fm., there are bedded hematite-rich rocks and jasper
(BIF) with arkose and manganese ore intercalations bearing
large granite boulders. In Bolivia, the lower part of the Boqui
Group has been interpreted as correlative to the Jacadigo Group
(Litherland et al. 1986).
Araras Group, northern Paraguay Belt
The thickness of the Araras Group varies from 100 to 150 m in the
western area, overlying the cratonic region (inner shelf), to more
than 1300 m eastwards and in the middle-outer shelf domain.
Deeper-water rocks are chiefly marl limestone, rhythmite and silt-
stone that mark the slope deposits of the basin (Alvarenga et al.
2008). This unit was subdivided into three formations
(Table 45.1). The lower Mirassol d’Oeste Fm. consists mostly of
dolostone. The 18- to 32-m-thick type section at the Terconi
Quarry is characterized by laminated dolostone associated with
microbialites, giant wave ripples, tube-like structures and
fan-like crystals (Nogueira et al. 2003; Nogueira & Riccomini
2006; Alvarenga et al. 2008).
The overlying Guia Fm. consists of dark grey laminated micrite
limestone with thin interbedded shale or marl laminae. This
Basement
(granite)
1 2 3 4 5
Fig. 45.3. Lithostratigraphy of the Jacadigo Group (modified from Almeida
1945). 1, Arkose sandstone; 2, conglomerate and arkose sandstone; 3, siltstones;
4, BIF with five Mn beds; 5, ferruginous arkose sandstone, sometimes with
boulders and clasts.
formation reaches c. 250 m in thickness in the Nobres region. In
the Terconi and Tangará quarries (inner shelf carbonate basin),
this unit overlies the dolostone of the Mirassol d’Oeste Fm. and
presents at its base fan-like crystals interpreted as aragonite pseu-
domorphs (Alvarenga et al. 2008). Interbedded graded grainstone
and cross-bedded grainstone occur in the middle part of this
formation. Locally, in the fore slope basin, an interbedded
12-m-thick layer of dolostone (grainstone and breccia) was
described at the Nossa Senhora da Guia Quarry (Alvarenga et al.
2004, 2008).
The uppermost Nobres Fm. consists of a uniform, light grey
dolostone (c. 1100 m) that forms the main karstic relief
of the region. This formation is exposed in the Provı́ncia
Serrana but is absent in the eastern part of the Paraguay Belt
(Fig. 45.1). The base of this unit is characterized by 2– 4 m of
dolomite brecciated layers intercalated with fine-grained and
laminated dolomite. Most of the sequence consists of a thick
succession of grainstone and packstone dolostone (Alvarenga
et al. 2000, 2004).
492 C. J. S. ALVARENGA ET AL.
The Corumbá Group, southern Paraguay belt
The Corumbá Group is exposed both in unfolded and folded
domains. In the unfolded domains, this unit overlies the Rio Apa
Craton, located on the western margin of the Paraguay
River, whereas in the folded domains, it crops out in the eastern
part of the Serra da Bodoquena (Fig. 45.1). The Corumbá
Group is the uppermost Neoproterozoic succession exposed in
the southern Paraguay Belt and was subdivided into five for-
mations: Cadieus, Cerradinho, Bocaina, Tamengo and Guaicurus
(Tables 45.2 & 45.3).
The Cadieus and Cerradinho formations occur only in the Serra
da Bodoquena (Fig. 45.1) and comprise a mixed assemblage of
clastic and carbonate rocks that unconformably overlies the Rio
Apa Craton (Boggiani 1998). The Cadieus Fm. consists of con-
glomerate and arkosic sandstone that was deposited in an alluvial
fan atop the igneous-metamorphic basement. The proximal facies
of this fan grades upwards and laterally into the arkosic sandstone,
shale and grainstone that comprises the more distal alluvial facies
of the Cerradinho Fm. (Boggiani et al. 1993; Boggiani 1998). Cross-
bedding and hummocky cross-stratification found within grainstone
facies in the upper portion of the Cerradinho Fm. suggest a rise in sea
level at the end of the rift phase (Boggiani 1998).
The Bocaina Fm. is present in the Serra da Bodoquena, Corumbá
region and in Puga Hill, where it may reach thicknesses up to
300 m (Maciel 1959). This unit is placed directly above the lime-
stone of the Cerradinho Fm. at the Serra da Bodoquena and the
diamictite of the Puga Fm. at Puga Hill.
The Tamengo Fm. is up to 100 m thick and consists of dark,
organic-rich limestone and shale rhythmically interbedded with
uncommon limestone grainstone. In the Corumbá area, the
organic-rich limestone and shale of this formation contain a rich
Ediacaran microfossil assemblage that will be described below.
The dominance of shale in the Bodoquena region suggests
deeper waters when compared to the Corumbá region (Boggiani
1998), thus explaining the absence of Cloudina lucianoi in the
Serra da Bodoquena limestone-shale.
The Guaicurus Fm. was first described by Almeida (1965) as a
thick shale succession on the top of the Corumbá Group that
extends over the eastern part of the Serra da Bodoquena and
along the Miranda River valley. It consists of fine-grained silici-
clastic rocks, mainly shale. Diamictite and pelite with clasts that
occur as lenses in the pelite of the Guaicurus Formation at
Laginha Mine, Corumbá, Mato Grosso do Sul, have been inter-
preted as sediment gravity-flow deposits in a slope setting (Bog-
giani et al. 2004). There are no equivalents of both the Tamengo
and Guaicurus formations in the northern part of the Paraguay
belt (Fig. 45.1; Tables 45.1, 45.2 & 45.3).
Glaciogenic deposits and associated strata
Puga Fm.
In the northern Paraguay Belt, the Puga Fm. consists of a succes-
sion of diamictite units intercalated with sandstone, conglomerate
and mudstone. The matrix of the diamictite varies from sandy to
clayey. Diamictite is massive or stratified and contains clasts
from a few centimetres up to 1 m in diameter. Clasts mainly of
basement rocks (granite, gneiss, quartzite, quartz, schists, etc.),
some of them deflecting the underlying laminae, faceted and
striated, and abundant detrital mica, are found in outcrop, close
to the Palaeoproterozoic basement rocks. Massive diamictite
shows great variations in the relative proportions of matrix
(muddy sand) and clasts. Occasionally, a crude stratification can
be observed. Some stratified diamictite consists of few-
centimetres-thick, massive diamictite beds associated with mud-
stone with graded bedding and outsized clasts clearly deflecting
the underlying laminae.
The mudstone comprises an alternation of silt-shale, siltstone
and very fine-grained sandstone, with clasts cutting or disturbing
the bedding (Alvarenga & Trompette 1992). Some thin beds and
lenses, up to 0.3 m thick, made up of clast-rich diamictite, have
erosive basal contacts. Lenses of fine-grained sandstone com-
monly show load structures. Some massive or laminated sandstone
intercalations, up to 15 m thick, include isolated clasts.
Cuiabá Group
The Cuiabá Group includes a thick sequence of fine-grained rocks,
sandstone, conglomerate and diamictite (Fig. 45.2; Alvarenga &
Trompette 1992). Diamictite in this group is massive, with an
abundant clay-silt matrix. Most clasts are millimetre to centimetre
scale, although a few reach 1 m in diameter. Some clasts of carbon-
ate rocks are found in clast-poor diamictite.
Sandstone and coarse-grained rocks are widely distributed in
this group. Conglomerate with beds 0.3–15 m thick is commonly
interbedded with sandstone and diamictite. The clast size ranges
from less than 1 cm to c. 10 cm. The matrix represents less than
10%, and consists of a clay-silt-sand mixture (Alvarenga &
Trompette 1992). Pebbly sandstone exhibits normal grading, start-
ing with a pebbly conglomerate and passing upward to a fine-
grained conglomerate followed by sandstone. The individual
graded bed is normally 0.5–3.0 m thick, with a sharp erosive
basal contact. Clasts include quartz, feldspar, quartzite, mudstone,
limestone and some granite. Flame, load casts and ball and pillow
structures are observed at the contact between two graded
sequences, with an injection of sandy material into the overlying
pebbly sandstone layer. Fine-grained sandstone layers grade
upwards into mudstone. Occasionally, isolated clasts can be
found cutting the underlying layers of laminated claystone.
In the Puga Fm., east of the Serra da Bodoquena, a BIF occurs
as a bed, c. 2 m thick, confined within a massive diamictite with
a ferruginous matrix. The BIF is formed by centimetre-thick
layers of hematite and magnetite, alternating with millimetre-thick
layers of quartz and chert (Piacentini et al. 2007).
Jacadigo Group
A BIF at the top of the Jacadigo Group (Santa Cruz or Banda Alta
formations) consists of a thick succession (up to 450 m thick) of
interbedded (i) ferruginous arkose sandstone, (ii) hematite-rich
jaspilite (BIF), (iii) arkose and feldspathic sandstone, sometimes
containing isolated granite blocks and boulders, (iv) siltstone and
ferruginous mudstone, (v) mudstone with banded jaspilite, and
(vi) four manganese ore beds between 0.3 and 4 m thick
(Fig. 45.3). Beds vary from centimetres to a few metres in thick-
ness. Arkose and feldspathic sandstone ranges from massive to
bedded, and is rarely graded. Some arkose beds exhibit isolated
granite-gneiss boulders, some of them bigger than 1.5 m in diam-
eter. Deformation around the boulders is generally more pro-
nounced above them (Trompette et al. 1998).
Mirassol d’Oeste Fm.
The Mirassol d’Oeste Fm. (20 –32 m thick) is the basal pink
dolostone of the Araras Group, resting on the diamictite of the
Puga Fm. (Nogueira et al. 2003). The first 8 m of the formation
consists of laminated pinkish dolostone that grades upward
through a diffuse and transitional contact to a grey laminated
dolostone. The basal dolostone has stratiform and wave microbia-
lite with dispersed tube-like structures (Nogueira et al. 2003; Font
et al. 2006; Elie et al. 2007). The upper part of the dolostone has
an enigmatic wave bed form that was interpreted as a giant wave
ripple by Allen & Hoffman (2005) but described as a tepee-like
structure by Nogueira et al. (2003). Fan-like crystals interpreted
 PUGA FORMATION, PARAGUAY BELT 493

as aragonite pseudomorphs are commonly found in the upper part -5 0 +5 +10

of the dolostone and in the overlying limestone of the Guia For-
mation (Nogueira et al. 2003; Alvarenga et al. 2008). Carbonate
has also been described from a drill core in the Bauxi area 1500
(Fig. 45.1) in the middle-outer shelf domain where the Araras
Group is c. 1300 m thick. In this area, 18-m-thick white laminated
dolostone overlies the diamictite of the Puga Formation in sharp
contact (Alvarenga et al. 2008). Other common sedimentary struc-
tures found in this dolostone include stromatolite, and breccia.

Basal member of the Bocaina Fm.

Maciel (1959) was the first to describe the carbonate of the Bocaina
Fm. that overlies the diamictite of the Puga Fm. at Puga Hill. The A
succession starts with a succession of 4-m-thick massive pink 1000
limestone overlaid by c. 10 m of a purple limestone/dolostone, R
mudstone and sandstone that grade upwards to grey dolostone A Nobres
and light grey dolostone. This upper dolostone is the common R FM.
rock of the Bocaina Fm., which also presents grainstone, pack- A
stone, breccia and stromatolites with poorly developed stratifica- S
tion (Almeida 1945; Boggiani et al. 1993).

Boundary relations with overlying and underlying G

non-glacial units R
The basal diamictite of the Puga Fm. lies unconformably above the 500 O
Palaeoproterozoic metamorphic rocks, in the western part of the U
basin, around Mirassol d’Oeste (Fig. 45.1). This contact is P 87
Sr/ 86
Sr
usually poorly exposed, but a road cut section exhibits 12 m of
0.7077
massive diamictite above the unconformity, followed upwards
by mudstone with isolated cobbles and pebbles, including some
0.7078
intercalations of massive diamictite, up to 0.80 m thick (Alvarenga Mirassol 0.7076
1990; Alvarenga & Trompette 1992). Guia d' Oeste Fm. 0.7075
Good exposures of the upper contact of the Puga Fm. with the Fm
Araras Group can be found at the Terconi and Guia quarries and 0.7075
0.7078
in a drill core in the Bauxi region (Alvarenga & Trompette 0.7075
1992; Nogueira et al. 2003; Alvarenga et al. 2004, 2008). The 0
best exposure of this contact (Terconi Quarry) exhibits dolostone (m) Puga Fm.
of the Mirassol d’Oeste Fm. overlying diamictite along a conform-
able sharp contact. This contact exhibits soft-sediment defor- -10 -5 0 +5 +10
mation (Nogueira et al. 2003). In the southern Paraguay Belt, the
13
pink limestone from the Bocaina Fm. overlies the diamictite at d C (PDB)
Puga Hill (Maciel 1959; Boggiani et al. 2003).
Fig. 45.4. Stratigraphic section and variations of d13Cpdb and 87Sr/86Sr for the
Araras Group (data from Nogueira et al. 2003, 2007; Alvarenga et al. 2004,
Chemostratigraphy 2008; Font et al. 2006; Figueiredo 2006).

The chemostratigraphic data discussed here take into account all

available C, O and Sr isotope data from the Araras and Corumbá
carbonate successions (Gaucher et al. 2003; Boggiani et al.
2003; Nogueira et al. 2003, 2007; Pinho et al. 2003; Alvarenga
et al. 2004, 2008; Figueiredo 2006; Font et al. 2006; Misi et al.
2007; Riccomini et al. 2007). The discussion will focus on C iso-
topes, because they are more resistant to post-depositional altera-
tion. The Sr-isotope ratios were also considered for those
samples in which the Sr concentration was higher than 400 ppm.
Northern Paraguay Belt
The d13C profiles of the Mirassol d’Oeste Fm. were constructed
from three sections in the northern Paraguay Belt: Terconi and
Tangará quarries, both located on the western cratonic domain,
and the João Santos borehole, which is located in the fold belt
near Bauxi (Figs 45.1 & 45.4). While the d13C values of the first
20– 30 m in the cratonic region range between – 10.5‰ and
–3.0‰, the d13C values along the profile located in the folded
domain range from –4.8‰ to –1.7‰. Corresponding d18O ratios
range from –8.2 to –1.3‰, whereas 87Sr/86Sr ratios are high
and variable with low Sr content (44 – 96 ppm) and high Mn –Sr
ratios (.18).
Dark grey laminated micritic limestone interbedded with the
thin shale or marl layers of the Guia Fm. lie above the Mirassol
d’Oeste Fm. This grey limestone presents increasing d13C values
across the first 20 m from the base, with a narrow range of C-
isotopic values (between –1.6 and þ0.1‰; Fig. 45.4). The
87
Sr/86Sr ratios for limestone samples with high Sr content
(750 –4351 ppm) and low Mn –Sr (,0.2) range between 0.70763
and 0.70780 (Fig. 45.4).
The dolostone of the overlying Nobres Fm. presents a wide
range of C-isotopic values (Alvarenga et al. 2004; Nogueira
et al. 2007). In the Bauxi-Nobres area, for instance (Fig. 45.1),
the dolostone at the base of the sequence (900 m thick) is charac-
terized by homogeneous and positive d13C values (þ1.9 to
þ2.7‰), while the upper dolostone, which consists of thin
layers of sandy dolostone, exhibit increasing d13C values up to
þ9.6‰ (Fig. 45.4; Alvarenga et al. 2004; Figueiredo 2006). On
the other hand, in the Cáceres area, the lowermost 300 m of the
494
dolostone placed on top of the Guia Fm. have d13C values ranging
from – 2.2 to þ0.3‰ (Fig. 45.4; Nogueira et al. 2007).
Southern Paraguay Belt
Boggiani et al. (2003) presented d13C values close to –5‰ for the
first 12 m of laminated carbonate rocks overlying the Puga Fm. in
the Puga Hill (Fig. 45.5). In the Jacadigo Group, thin limestone
intercalations in the BIF show d13C ratios between –5.2 and
–7.0‰ (Klein & Ladeira 2004).
The Tamengo Fm. occurs in the Corumbá area and along the
eastern part of the Serra da Bodoquena. The isotopic data of lime-
stone from this unit were obtained at the Laginha, Saladeiro and
Corcal quarries, located in the Corumbá area (Figs 45.1 & 45.5;
Boggiani 1998; Boggiani et al. 2003; Misi et al. 2007). Limestone
of the lower part of the Tamengo Formation records a negative
d13C excursion (–3.3 to –2.5‰), which is followed by limestone
with positive d13C values (up to þ5.8‰). In the Serra da Bodo-
quena, located about 200 km SE of Corumbá, limestone correlated
to the Tamengo Fm. shows homogeneous d13C values around
þ3‰ (Boggiani 1998). The 87Sr/86Sr ratios of the Tamengo car-
bonate rocks from both areas are clustered between 0.7084 and
0.7085 (Boggiani 1998; Babinski et al. 2008).
Other characteristics
Mn and Fe ore bodies with 36 billion metric tonnes of hematite,
and estimated Mn ore reserves of 608 million metric tonnes
(Urban et al. 1992) are known around south Corumbá.
This mineralization occurs throughout the Santa Cruz
Fm., as banded, hematite-rich, jaspilite with arkose sandstone
-5 0 +5
87Sr/ 86Sr
60
40
Tamengo Fm
30
20
0 & Trompette 1992; Dantas et al. 2009).
20
Bocaina Fm
0 1982; Walde et al. 1982), Babvinella faveolata, Vandalosphaeri-
10
0 sistent with the U –Pb age obtained in the Tamengo Fm.
Puga Fm. (m)
-10 -5 0 +5
13
d C (PDB)
Fig. 45.5. Composite of d13Cpdb and 87Sr/86Sr record for the Corumbá Group
from Corumbá area, Puga Hill and Serra da Bodoquena (data from Boggiani
1998; Boggiani et al. 2003; Misi et al. 2007 and unpublished data).
C. J. S. ALVARENGA ET AL.
intercalations (Walde et al. 1981; Hoppe et al. 1987; Trompette
et al. 1998; Walde & Hagemann 2007). The manganese deposits
consist of four individual layers ranging from 0.3 to 4.0 m in thick-
ness. Beds of Mn ore, up to 5 m thick, are mined in the under-
ground Urucum Mine. Isolated granite clasts up to 1.5 m in
diameter have been identified in the arkose intercalations
(Trompette et al. 1998; Walde & Hagemann 2007).
The BIFs in the eastern part of the Serra da Bodoquena are thin
beds (2 –3 m thick) in a massive diamictite with a ferruginous
(magnetite) matrix, suggesting the relationship of these sediments
with Fe precipitation (Piacentini et al. 2007). The magnetite matrix
has an Fe content ranging from 15 to 72%, with an average of 27%
(Piacentini 2008).
Palaeolatitude and palaeogeography
Palaeomagnetic data are only available for the Neoproterozoic
Mirassol d’Oeste Fm. and are restricted to 19 samples from the
Terconi Quarry, in the northern part of the Paraguay belt (Trindade
et al. 2003). These palaeomagnetic data indicated five polarity
reversals within the first 20 m of the dolomite, suggesting a
primary magnetization and a low palaeolatitudes (22 þ 6/– 58)
for these rocks.
Geochronological constraints
A U– Pb zircon age of 543 + 2 Ma (Ediacaran) has been obtained
from an ash bed intercalated with the Cloudina bearing limestone
of the Tamengo Fm. near Corumbá city (Babinski et al. 2008), thus
providing a precise Upper Ediacaran age for this unit. In the north-
ern Paraguay belt (Terconi Quarry), a Pb/Pb isochron age of
633 + 25 Ma was obtained for the dolostone of the Mirassol
d’Oeste Fm. and the limestone of the Guia Fm. (Babinski et al.
+10 unpublished). This age is interpreted as the time of deposition
for these carbonate rocks. There are otherwise no radiometric
data available to directly constrain the age of the Puga Fm.
Other radiometric ages include the Nd isotope signature
obtained from clasts in diamictite, quartzite and phyllite in the
0.7086 Cuiabá Group and Puga Fm. by Dantas et al. (2009). The Tdm
model ages display an irregular distribution. Diamictite present
variable Tdm values (1.4 –2.15 Ga), but homogeneous 1nd (t)
around – 8. The sandstone show Tdm model ages (c. 2.0 –2.1 Ga)
older than the ones determined for fine-grained rocks (Tdm
values of 1.7 and 1.8 Ga). This difference of Nd isotopes may be
0.7085 related to variations in the source region or may be the result of
mixing of material from different sources. Nevertheless, all of
these data indicate sources within the Amazonian Craton, in agree-
ment with the palaeogeographic interpretation, which points to the
northwestern basement as the main source of detritus (Alvarenga
The limestone and shale of the Tamengo and Guaicurus for-
mations near the city of Corumbá contain a record of Ediacaran
fossils (Hahn et al. 1982; Walde et al. 1982; Zaine & Fairchild
1985, 1987; Gaucher et al. 2003). The microfossil assemblage is
dominated by scyphozoan Corumbella werneri (Hahn et al.
dium sp., Cloudina lucianoi (Zaine & Fairchild 1985, 1987), Sol-
dadophycus bossii, Titanotheca and the vendotaenid Eoholynia
corumbensis sp. (Gaucher et al. 2003). The presence of these
fossils suggests a deposition during the 570–545 Ma period, con-
+10
Discussion
Northern Paraguay Belt depositional settings
The glacially influenced succession of the Puga Fm. and Cuiabá
Group in the northern Paraguay Belt displays three main
 PUGA FORMATION, PARAGUAY BELT
depositional settings: platform, slope and outer slope (Fig. 45.2).
The platformal deposits are reworked by sedimentary gravity
flows. The deposits on the inner shelf show dominant massive dia-
mictite, alternating with sandstone and fine-grained sedimentary
rocks that contain striated and faceted clasts (Alvarenga & Tromp-
ette 1992). On the outer shelf, there is an association of massive
diamictite, stratified diamictite and fine-grained sedimentary
rocks with some clasts disrupting the underlying beds. The outer
shelf succession has been interpreted as a succession of resedimen-
ted glacial deposits with alternations of debris-flows and shallow
water turbidite events (Alvarenga & Trompette 1992).
Glaciomarine deposits reworked by sediment gravity flows and
related to submarine fans characterize the slope depositional
system. Progressive sorting in the deeper portions of the fans are
demonstrated by the transition from diamictite to massive con-
glomerate and sandstone with local inverse and/or normal
grading and normally graded, fine-grained turbidites. Sandstone
and siltstone intercalations are consistent with inter-channel
deposits. Deposition on the outer slope system was dominated by
fine-grained deposits related to low-density turbidity currents, in
which a direct glacial influence is only indicated by isolated
clasts (dropstones).
This type of basin-filling suggests that the sedimentary source
was located on the Amazonian Craton (Alvarenga & Trompette
1992) and this was confirmed by Nd isotopic analysis (Dantas
et al. 2009). Shelf reworking during the glacial event helped
develop submarine channels and turbidite deposits on the slope
and outer slope (Alvarenga & Trompette 1992). Diamictite at the
top of this succession is interpreted as the final stage of glaciation,
during which all deposits were transported into the basin by sedi-
ment gravity flows and iceberg melting.
The Araras Group records carbonate deposition and is divided
into three lithostratigraphic units. The basal Mirassol d’Oeste
Fm. was deposited directly on top of glacial diamictites of the
Puga Fm., and is interpreted as a transgression over previously gla-
ciated landscapes. The laminated limestone and shale of the middle
Guia Fm. are related to a deep-water system. The upper carbonate
unit, Nobres Fm., consists of more than 1100 m of shallow-shelf
dolostone (breccia, grainstone and packstone) indicating high-
energy environments (Alvarenga et al. 2004).
Southern Paraguay Belt depositional settings
In the southern Paraguay Belt (Fig. 45.1), the rocks interpreted as
deposited during the late Cryogenian glacial event are the Jacadigo
Group and the Puga Fm. The Jacadigo Group around Corumbá
is interpreted as deposited in a Neoproterozoic tectonic graben
system, coeval with the sedimentation of the Puga Formation
(Trompette et al. 1998). The Santa Cruz Formation shows alterna-
tions of autochthonous chemical sediments and siliciclastic sedi-
ments (fine-grained and arkose sandstone), which have been
interpreted as glacially influenced based on the presence of large
granite boulders thought to be dropstones (Barbosa 1949; Walde
et al. 1981; Urban et al. 1992; Klein & Ladeira 2004). Trompette
et al. (1998), however, remarked that these granite pebbles and
boulders generally do not cross-cut the underlying bedding, and
the deformation around the boulders is associated with compaction
from overlying strata, being more pronounced above the boulders
rather than beneath them. These data were used to deny a glacial
origin of these deposits, suggesting reworked gravitational fluxes
(Trompette et al. 1998). The diamictite of the Puga Fm. in the
southern Paraguay Belt does not show evidence of striated clasts
or dropstones, but their widespread areal distribution underlying
a carbonate succession (Araras and Corumbá groups) is strong evi-
dence for their stratigraphic correlations to other glacial deposits
of the Puga Fm. in the Paraguay Belt.
The Corumbá Group records mixed clastic and carbonate sedi-
mentation in a passive margin setting. The Cadieus and Cerradinho
495
formations record proximal and distal alluvial deposition. The pink
limestone of the Bocaina Fm. has been described as a ‘cap’ carbon-
ate (Boggiani et al. 2003; see discussion below). The Tamengo
and Guaicurus formations comprise a trangressive– highstand
sequence, which drowned the carbonate shelf of the Tamengo
Fm. (Boggiani & Coimbra 1996). Diamictite and pelite with clasts
that occur as lenses in the pelites of the Guaicurus Fm. at Laginha
Mine, Corumbá, Mato Grosso do Sul, have been interpreted as sedi-
ment gravity flow deposits in a slope setting (Boggiani et al. 2004).
Correlations and geotectonic evolution of the Paraguay Belt
Despite the widespread areal distribution of diamictite underlying
both carbonate successions (Araras and Corumbá groups), there is
no consensus about the lithostratigraphic correlation between the
northern and southern parts of the Paraguay Belt. In addition, the
evolution of these two segments could have been diachronic.
Some differences and similarities observed between the rocks
from the north and south will be mentioned here in order to
provide a picture of the current knowledge and controversies
surrounding the geotectonic evolution of the Paraguay Belt.
Glacial episodes have been proposed for both sectors of the
Paraguay Belt, but some differences are observed. In the northern
Paraguay Belt, two glacial events can be found: an older one rep-
resented by the Puga Fm. and the Cuiabá Group rocks (Alvarenga
& Trompette 1992), and a younger one represented by the Serra
Azul Fm. and thought to be related to the Ediacaran glaciation
(Alvarenga et al. 2007, 2009; Figueiredo et al. 2008, 2011). In con-
trast, the glacial origin of the Puga Fm. in the southern Paraguay
Belt is debatable and no younger diamictite-bearing unit has
been found there.
Whereas carbonate lithofacies and sedimentary structures
typical of other Neoproterozoic cap-carbonate rocks have been
described in the North (Nogueira et al. 2003), none have been
found in the southern belt.
Although there is some evidence of diagenetic alteration of Sr
values in the lower Mirassol d’Oeste Fm., the carbonate units
found overlying diamictite of the Puga Fm. in both regions
exhibit d13C and 87Sr/86Sr values interpreted as representing
original seawater geochemistry; these data are consistent with a
north – south correlation for the Paraguay Belt and are typical of
other Neoproterozoic cap-carbonate rocks.
In contrast, the upper part of the Corumbá and Araras Groups
differ in their lithostratigraphy. The limestone and shale from the
Tamengo and Guaicurus formations, both with microfossils
(Gaucher et al. 2003), represent a transgressive phase at the top
of Corumbá Group. This is contrasted with the 1300 m of
carbonate-dominated deposition and no microfossils in the top of
the Araras Group, in the northern part of the belt. A depositional
age of 543 + 2 Ma (zircon U –Pb SHRIMP dating) was obtained
for the Tamengo Fm. (Babinski et al. 2008). If we consider that
the Bocaina Fm. was deposited after the late Cryogenian glacia-
tion, which is assumed to be 630 Ma, an unconformity would
have to exist between the Bocaina and Tamengo formations,
because the deposition of the Bocaina Fm. (200 m thick, at
most) could not have taken such a long time (c. 100 Ma). In con-
trast, the 1300-m-thick carbonate succession between the two
glacial units in the northern Paraguay Belt is thought to have
lasted 50 Ma, assuming the two glacial units represent a late Cryo-
genian (c. 630 Ma) and an Ediacaran age glaciation (580 Ma),
respectively (Alvarenga et al. 2007; Figueiredo et al. 2008).
Further evidence that undermines a north –south correlation for
at least the upper part of the succession comes from the age of the
Alto Paraguay Group, which overlies the Araras Group. Although
it is poorly constrained, a Rb –Sr whole-rock isochron age of c.
569 Ma, was determined from the shale of the Sepotuba Formation
(Table 45.1; Cordani et al. 1985), suggesting that the units of the
uppermost formations of the northern part of the belt are older
496 C. J. S. ALVARENGA ET AL.
than the ones from the southern part (e.g. the Tamengo Fm.,
Table 45.2).
In addition, the geotectonic evolution of the two regions appears
to differ during deposition of the upper part of the succession.
40
Ar– 39Ar ages ranging from 541 and 531 Ma were determined
from the biotite of metavolcanics (Araés gold deposit in the
eastern end of the belt) and interpreted as the cooling ages follow-
ing regional metamorphism in the northern Paraguay Belt
(Geraldes et al. 2008). In contrast, in the southern part of the
belt, the deposition of carbonate rocks and shale from the
Tamengo (c. 543 Ma) and Guaicurus formations was taking
place in a passive margin environment. These drastic differences
in the two segments of the belt are still not understood, and a
more detailed and profound study is needed in order to better
constrain the evolution of the Paraguay belt.
Radiometric constraints on the Puga Fm. are limited to Edia-
caran ages (543 Ma) obtained from the Tamengo Fm. in the
southern Paraguay Belt, although the significance of this data
depends on stronger evidence for north –south stratigraphic corre-
lations. In the end, the Puga Fm. and the correlated Cuiabá Group
in the northern Paraguay Belt are interpreted as related to the late
Cryogenian glaciation because of the overlying dolomite lithofa-
cies, palaeomagnetic data and negative d13C values of the Mirassol
d’Oeste Fm. (Nogueira et al. 2003; Alvarenga et al. 2004, 2008;
Allen & Hoffman 2005). Limestone and mud-limestone deposited
above the Mirassol d’Oeste Fm. have 87Sr/86Sr ratios between
0.70763 and 0.70780, consistent with a seawater composition
from c. 630 Ma (Alvarenga et al. 2004, 2008). The age and
glacial origin of the Puga Fm. in the southern Paraguay Belt
awaits further study to resolve existing controversies.
This chapter is a contribution to the International Geological Programme (IGCP)
Project 512 ‘Neoproterozoic Ice Ages’ and Project 478 ‘Neoproterozoic– Early
Palaeozoic Events in SW-Gondwana’. This research was supported by the Con-
selho Nacional de Desenvolvimento Cientı́fico e Tecnológico (CNPq) and Funda-
ção de Amparo à Pesquisa do Estado de São Paulo (FAPESP). We thank
Mineração Corumbaense Reunidas S.A. and Urucum Mineração S.A. for their
support during fieldwork. A. C. Rocha-Campos, N. M. Chumakov, E. Tohver,
C. Riccomini, G. Shields and E. Arnaud are thanked for their careful and
helpful reviews. This represents a contribution of the IUGS- and UNESCO-
funded IGCP (International Geoscience Programme) project #512.
References
Allen, P. A. & Hoffman, P. F. 2005. Extreme winds and waves in the
aftermath of a Neoproterozoic glaciation. Nature, 433, 123– 127.
Almeida, F. F. M. de 1945. Geologia do sudoeste matogrossense. Boletim
da Divisão de Geologia e Mineralogia, DNPM, 116, 1 –118.
Almeida, F. F. M. de 1946. Origem dos minérios de ferro e manganês de
Urucum (Corumbá Estado de Mato Grosso). Boletim Divisão de
Geologia e Mineralogia, DNPM, 119, 1– 57.
Almeida, F. F. M. de 1964a. Geologia do centro-oeste matogrossense.
Boletim da Divisão de Geologia e Mineralogia, DNPM, 215, 1 – 137.
Almeida, F. F. M. de 1964b. Glaciação Eocambriana em Mato Mato
Grosso. Notas Preliminares e Estudos, DNPM, 117, 10.
Almeida, F. F. M. de 1965. Geologia da Serra da Bodoquena (Mato
Grosso), Brasil. Boletim da Divisão de Geologia e Mineralogia,
DNPM, 219, 1 –96.
Almeida, F. F. M. de & Mantovani, M. S. M. 1975. Geologia e geocro-
nologia do granito São Vicente, Mato Grosso. Anais da Academia
Brasileira de Ciências, 47, 451–458.
Alvarenga, C. J. S. de 1990. Phénomènes sédimentaires, structuraux et
circulation de fluides développés à la transiction chaı̂ne-craton.
Exemple de la chaı̂ne Paraguai d 0 âge proterozoı̈que supérieur,
Mato Grosso, Brésil. PhD thesis, Université d0 Aix-Marseille III,
France.
Alvarenga, C. J. S. de & Saes, G. S. 1992. Stratigraphy and sedimentol-
ogy of the middle and Late Proterozoic in the southeast of the
Amazonian Craton. Revista Brasileira de Geociências, 22, 493– 499.
Alvarenga, C. J. S. & Trompette, R. 1992. Glacially influenced
sedimentation in the Later Proterozoic of Paraguay belt (Mato Grosso,
Brazil). Palaeogeography, Palaeoclimatology, Palaeoecology, 92,
85–105.
Alvarenga, C. J. S. & Trompette, R. 1993. Brasiliano tectonic of the
Paraguay Belt: the structural development of the Cuiabá Region.
Revista Brasilieira de Geociências, 23, 18 –30.
Alvarenga, C. J. S. de, Cathelineau, M. & Dubessy, J. 1990. Chron-
ology and reorientation of N2 – CH4, CO2 – H2O, and H2O-rich
fluid-inclusions trails in intra-metamorphic quartz veins from the
Cuiabá gold district, Brazil. Mineralogical Magazine, 54, 245–255.
Alvarenga, C. J. S. de, Moura, C. A. V., Gorayeb, P. S. S. & Abreu,
F. A. M. 2000. Paraguay and Araguaia Belts. In: Cordani, U. G.,
Milani, E. J., Thomaz Filho, A. & Campos, D. A. (eds) Tectonic
Evolution of South America, Rio de Janeiro, 31st International
Geological Congress, 183– 193.
Alvarenga, C. J. S., Santos, R. V. & Dantas, E. L. 2004. C-O-Sr iso-
topic stratigraphy of cap carbonates overlying Marinoan-age glacial
diamictites in the Paraguay Belt, Brazil. Precambrian Research,
131, 1– 21.
Alvarenga, C. J. S. de, Figueiredo, M. F., Babinski, M. & Pinho,
F. E. C. 2007. Glacial diamictites of Serra Azul Formation (Edia-
caran, Paraguay Belt): evidence of the Gaskiers glacial event in
Brazil. Journal of South American Earth Science, 23, 236–241.
Alvarenga, C. J. S. de, Dardenne, M. A. et al. 2008. Isotope stratigra-
phy of Neoproterozoic cap carbonate in the Araras Group, Brazil.
Gondwana Research, 13, 469–479.
Alvarenga, C. J. S. de, Boggiani, P. C., Babinski, M., Dardenne,
M. A., Figueiredo, M. F., Santos, R. V. & Dantas, E. L. 2009.
The Amazonian Paleocontinent. In: Gaucher, C., Sial, A. N., Hal-
verson, G. P. & Frimmel, H. E. (eds) Neoproterozoic–Cambrian
Tectonics, Global Change and Evolution: A Focus on Southwestern
Gondwana. Developments in Precambrian Geology, 16, Elsevier,
15 – 28.
Babinski, M., Boggiani, P. C., Fanning, M., Simon, C. M. & Sial, A. N.
2008. U– Pb SHRIMP geochronology and isotope chemostratigraphy
(C, O, Sr) of the Tamengo Formation, southern Paraguay belt, Brazil.
In: South American Symposium on Isotope Geology, 6, San Carlos de
Bariloche, Argentina. Proceedings, CD-ROM.
Barbosa, O. 1949. Contribuição à geologia da região Brasil-Bolı́via.
Mineração e Metalurgia, 13, 271– 278.
Boggiani, P. C. 1998. Análise estratigráfica da bacia Corumbá
(Neoproterozóico)-Mato Grosso do Sul. PhD thesis. Universidade
de São Paulo, Instituto de Geociências, Brazil.
Boggiani, P. C. & Coimbra, A. M. 1996. The Corumbá Group
(Central South America) in the context of Late Neoproterozoic
global changes. Anais da Academia Brasileira de Ciências, 68,
595– 596.
Boggiani, P. C., Fairchild, T. R. & Coimbra, A. M. 1993. O Grupo
Corumbá (Neoproterozóico–Cambriano) na região central da Serra
da Bodoquena (Faixa Paraguai) Mato Grosso do Sul. Revista Brasi-
lieira de Geociências, 23, 301– 305.
Boggiani, P. C., Ferreira, V. P. et al. 2003. The cap carbonate of the
Puga hill (central South America) in the context of the Post-Varanger
glaciation. In: South American Symposium on Isotope Geology, 4,
Salvador, Brazil, Short Papers, 324–327.
Boggiani, P. C., Fairchild, T. R. & Riccomini, C. 2004. New level of
diamictites in the Corumbá Group (Ediacaran), Paraguay Belt,
South America. In: 1st Symposium on Neoproterozoic-Early Paleo-
zoic Events in SW-Gondwana. Extended Abstract, IGCP, Project
478, Second Meeting, Brazil, 10 –13.
Cordani, U. G., Filho, T. A., Brito Neves, B. B. & Kawashita, K.
1985. On the applicability of the Rb– Sr method to argillaceous
sedimentary rocks: some examples from Precambrian sequences of
Brazil. Giornale di Geologia, 47, 253–280.
Dantas, E. L., Alvarenga, C. J. S., Santos, R. V. & Pimentel, M. M. 2009.
Using Nd isotopes to understand the provenance of sedimentary rocks
from a continental margin to a foreland basin in the Neoproterozoic
Paraguay Belt, Central Brazil. Precambrian Research, 170, 1–12.
Dorr II, J. V. N. 1945. Manganese and iron deposits of Morro do Urucum,
Mato Grosso, Brazil. United States Geological Survey Bulletin,
946A, 1 – 47.
 PUGA FORMATION, PARAGUAY BELT
Elie, M., Nogueira, A. C. R., Nédélec, A., Trindade, R. I. F. & Kenig,
F. 2007. A red algal bloom in the aftermath of the Marinoan Snowball
Earth. Terra Nova, 19, 303– 308.
Evans, J. W. 1894. The geology of Matto Grosso (particulary the region
drained by Upper Paraguay). Quaternary Journal of the Geological
Society, London, 50, 85 –104.
Figueiredo, M. F. 2006. Quimioestratigrafia das rochas ediacaranas no
extremo norte da Faixa Paraguai, Mato Grosso. MSc thesis, Instituto
de Geociências, Universidade de São Paulo, Brazil.
Figueiredo, M. F., Babinski, M., Alvarenga, C. J. S. & Pinho, F. E. C.
2008. Nova unidade litoestratigráfica registra glaciação ediacarana
em Mato Grosso: Formação Serra Azul. Geologia USP: Série Cientı́-
fica, 8, 65 – 75.
Figueiredo, M., Babinski, M. & Alvarenga, C. J. S. 2011. The Serra
Azul Formation, Paraguay Belt, Brazil. In: Arnaud, E., Halverson,
G. P. & Shields-Zhou, G. (eds) The Geological Record of Neo-
proterozoic Glaciations. Geological Society, London, Memoirs, 36,
499– 502.
Font, E., Nédélec, A., Trindade, R. I. F., Macouin, M. & Charrière,
A. 2006. Chemostratigraphy of the Neoproterozoic Mirassol d’Oeste
cap dolostone (Mato Grosso, Brazil): an alternative model for
Marinoan cap dolostone formation. Earth and Planetary Science
Letters, 250, 89 – 103.
Gaucher, C., Boggiani, P. C., Sprechmann, P., Sial, A. N. & Fair-
child, T. 2003. Integrated correlation of the Vendian to Cambrian
Arroyo del Soldado and Corumbá Groups (Uruguay and Brazil):
palaeogeography, palaeoclimatic and palaeobiologic implications.
Precambrian Research, 120, 241– 278.
Geraldes, M. C., Tassinari, C. C. G. et al. 2008. Isotopic evidence for
the Late Brasiliano (500– 550 Ma) ore-forming mineralization of
the Araés Gold Deposit, Brazil. International Geology Review, 50,
177– 190.
Hahn, G., Hahn, R., Leonardos, O. H., Pflug, H. D. & Walde, D. H. G.
1982. Körperlich erhaltene Scyphozoen – Reste aus dem
Jungprakambrium Brasiliens. Geologica et Palaeontologica, 16,
1 – 18.
Hennies, W. T. 1966. Geologia do centro-oeste matogrossense. PhD
thesis, Escola Politécnica, Universidade de São Paulo, Brazil.
Hoppe, A., Schobbenhaus, C. & Walde, D. H. G. 1987. Precambrian
iron formation in Brazil. In: Appel, P. W. U. & LaBerge, G. L.
(eds) Precambrian Iron Formation. Theophrastus Publications,
Athenas, 347– 390.
Klein, C. & Ladeira, E. A. 2004. Geochemistry and mineralogy of
Neoproterozoic banded iron-formations and some selected, siliceous
manganese formations from the Urucum District, Mato Grosso do
Sul, Brazil. Economic Geology, 99, 1233– 1244.
Litherland, M., Annels, R. N. et al. 1986. The Geology and Mineral
Resources of the Bolivian Precambrian Shield. British Geological
Survey, Overseas Memoir, 9.
Luz, J. S. & Abreu Filho, W. 1978. Aspectos geológico-econômicos da
Formação Araras e do Grupo Alto Paraguai – MT. In: An. 308 Con-
gresso Brasileiro de Geologia, sociedade Brasileira de Geologia,
Recife, 4, 1816–1826.
Maciel, P. 1959. Tilito Cambriano no Estado de Mato Grosso. Boletim da
Sociedade Brasileira de Geologia, 8, 31 –39.
Misi, A., Kaufman, A. J. et al. 2007. Chemostratigraphic correlation of
Neoproterozoic successions in South America. Chemical Geology,
237, 143– 167.
Nogueira, A. C. R & Riccomini, C. 2006. O Grupo Araras (Neoprotero-
zoico) na parte norte da Faixa Paraguai e sul do Craton Amazônico,
Brazil. Revista Brasileira de Geociências, 36, 576–587.
Nogueira, A. C. R., Riccomini, C., Sial, A. N., Moura, C. A. V. &
Fairchild, T. R. 2003. Soft-sediment deformation at the base of
Neoproterozoic Puga cap carbonate (southwestern Amazon craton,
Brazil): confirmation of rapid icehouse to greenhouse transition in
snowball Earth. Geology, 31, 613– 616.
Nogueira, A. C. R., Riccomini, C., Sial, A. N., Moura, C. A. V.,
Trindade, R. I. F. & Fairchild, T. R. 2007. Carbon and strontium
497
isotope fluctuations and paleoceanographic changes in the late Neo-
proterozoic Araras carbonate platform, southern Amazon craton,
Brazil. Chemical Geology, 80, 168– 190.
Piacentini, T. 2008. A Formação Ferrı́fera da Formação
Puga: Avaliação Regional dos Recursos da Serra da Bodoquena,
MS. MS thesis, Universidade de São Paulo, Instituto de Geociências,
Brasil.
Piacentini, T., Boggiani, P. C., Yamamoto, J. K., Freitas, B. T. &
Campanha, G. A. C. 2007. Formação Ferrı́fera associada à sedi-
mentação glaciogênica da Formação Puga (Marinoano) na Serra da
Bodoquena, MS. Revista Brasileira de Geociências, 37, 530–541.
Pimentel, M. M., Fuck, R. A. & Alvarenga, C. J. S. 1996. Post-
Brasiliano (Pan-African) high-K granitic magmatism in central
Brazil: the role of late Precambrian –early Paleozoic extension.
Precambrian Research, 80, 217– 238.
Pinho, F. E. C., Sial, A. N. & Figueiredo, M. F. 2003. Contribution
to the Neoproterozoic C and O isotopic record: carbonate rocks
from the Paraguay Belt, Mato Grosso, Brazil. In: IV South American
Symposium on Isotope Geology, Salvador, Brasil, Short Papers, 1,
386– 389.
Riccomini, C., Nogueira, A. C. R. & Sial, A. N. 2007. Carbon and
oxygen isotope geochemistry of Ediacaran outer platform carbonates,
Paraguay Belt, central Brazil. Anais da Academia Brasileira de
Ciências, 79, 519–527.
Rocha-Campos, A. C. & Hasuı́, Y. 1981. Late Precambrian Jangada
Group and Puga Formation of central western Brazil. In: Hambrey,
M. J. & Harland, W. B. (eds) Earth’s Pre-Pleistocene Glacial
Record. Cambridge University Press, Cambridge, 916– 919.
Tohver, E., D’Agrella-Filho, M. S. & Trindade, R. I. F. 2006. Paleo-
magnetic record of Africa and South America for the 1200–500 Ma
interval, and evaluation of Rodina and Gondwana assemblies.
Precambrian Research, 147, 193– 222.
Trindade, R. I. F., Font, E., D’Agrella-Filho, M. S., Nogueira,
A. C. R. & Riccomini, C. 2003. Low-latitude and multiple geomag-
netic reversals in the Neoproterozoic Puga cap carbonate of Amazo-
nia. Terra Nova, 15, 441– 446.
Trompette, R. 1994. Geology of Western Gondwana (2000 –500 Ma).
Pan-African Brasiliano Aggregation of South America and Africa.
Balkema, Rotterdam.
Tompette, R., Alvarenga, C. J. S. de & Walde, D. 1998. Geological
evolution of the Neoproterozoic Corumbá graben system
(Brazil). Depositional context of the stratified Fe and Mn ores of
the Jacadigo Group. Journal of South American Earth Sciences, 11,
587– 597.
Urban, H., Stribrny, B. & Lippolt, H. J. 1992. Iron and manganese
deposits of the Urucum District, Mato Grosso do Sul, Brazil.
Economic Geology, 87, 1375–1392.
Vieira, A. J. 1965. Geologia do centro-oeste de Mato Grosso. Petrobrás-
DEBSP, Relatório Técnico Interno no. 303, 58.
Walde, D. H. G. & Hagemann, S. G. 2007. The Neoproterozoic
Urucum/Mutúm Fe and Mn deposits in W-Brazil/SE-Bolivia:
assessment of ore deposits models. Zeitschriftdeutschen Gesellschaft
für Geowissenschaften, 158, 45 –55.
Walde, D. H. G., Gierth, E. & Leonardos, O. H. 1981. Stratigraphy and
mineralogy of the manganese ores of Urucum, Mato Grosso, Brazil.
Geologische Rundschau, 70, 1077– 1085.
Walde, D. H. G., Leonardos, O. H., Hahn, G. & Pflug, H. D. 1982.
The first Precambrian megafossils from South America,
Corumbella werneri. Anais da Academia Brasileira de Ciências,
54, 461.
Zaine, M. F. & Fairchild, T. R. 1985. Comparison of Aulophycus Lucia-
noi Beurlen and Sommer from Ladário (MS) and genus Cloudina
germs, Ediacaran of Namibia. Anais da Academia Brasileira de Ciên-
cias, 57, 180.
Zaine, M. F. & Fairchild, T. R. 1987. Novas considerações sobre os
fósseis da Formação Tamengo, Grupo Corumbá, SW do Brasil. In:
Anais Xº Congresso Brasileiro de Paleontologia, Rio de Janeiro,
Brazil, 797– 806.
 Chapter 46

The Serra Azul Formation, Paraguay Belt, Brazil

MILENE F. FIGUEIREDO1 *, MARLY BABINSKI1 & CARLOS J. S. ALVARENGA2

1
Instituto de Geociências, Universidade de São Paulo, São Paulo-SP, 05508-080, Brazil
2
Instituto de Geociências, Universidade de Brası́lia, Brası́lia, DF, 70910-900, Brazil
*Corresponding author (e-mail: milene.figueiredo@gmail.com)

Abstract: A new succession of diamictites and siltstones has been found within the Araras Group and is interpreted to record Ediacaran
glaciation in the northern Paraguay Belt, Brazil. This discontinuous stratigraphic unit, named the Serra Azul Formation (Fm.) (Figueiredo
et al. 2005; Alvarenga et al. 2007), is up to 300 m thick. It lies above dolomites of the Nobres Fm. and below sandstones of the Raizama
Fm. At the stratotype section, the Serra Azul Fm. comprises c. 70 m of massive glacial diamictites, overlain by 200 m of laminated
siltstones and rhythmites that contain sparse intercalations of very fine sandstone lenses.

The Serra Azul Fm. lies between carbonates of the upper Araras
Group and the siliciclastic lower Alto Paraguai Group in the north-
ern Paraguay Belt, southeastern Amazon craton (Fig. 46.1). The
Serra Azul Fm. was first described (Figueiredo et al. 2004) as a
new diamictite level associated with clayey-siltstones and was
recently elevated to formation status (Alvarenga et al. 2007).
These rocks are poorly exposed because they are generally either
covered by talus or strongly weathered. The formation occurs in
elongated areas of low relief between ranges of Raizama Fm. sand-
stones and hills comprising the Araras Fm. carbonates. The best
exposures are found in rills, quarries and road cuts. The stratotype
of the Serra Azul Fm. is exposed on the east –west-trending Serra
Azul Syncline (Fig. 46.1) in Mato Grosso State. Other exposures
along the belt are stratigraphically incomplete, typically preser-
ving only the upper part of the formation.
Structural framework
The Serra Azul Fm. was deposited on a passive margin along
the border of the Amazon Craton, which was folded by the
Brasiliano – Pan-African Orogeny during the Cambrian Period
(Alvarenga & Trompette 1993; Trompette 1994; Trompette et al.
1998; Trindade et al. 2003; Tohver et al. 2010). The stratotype
of the formation occurs along the Serra Azul (Azul Range) in the
northern Paraguay Belt on the southern limb of an openly
folded, east – west-trending asymmetric syncline (Fig. 46.1; Alvar-
enga et al. 2007), located in the external tectonic zone of the
fold-and-thrust belt, where the metamorphism is minimal and the
sedimentary structures are well preserved.
Stratigraphy
The Serra Azul Fm. occurs between the Araras Group below and
the Alto Paraguai Group above (Figs 46.1 & 46.2). It is underlain
by dolomites of the predominantly shallow marine Nobres Fm.
(Araras Group) and is overlain by sandstones of the tidally influ-
enced Raizama Fm. (Alto Paraguay Group).
Dolostones of the Nobres Fm. were first documented by Castel-
neau (1850) and then by Almeida (1964), who attributed them to
the upper Araras Group. It is c. 500 m thick (Hennies 1966) and
overlies limestones of the Guia Fm. (Alvarenga et al. 2011). The
Nobres Fm., from bottom to top, comprises dolosparite with two
interbedded layers of arkoses, a thick silicified carbonate level,
impure dolograinstone with cross-bedding, peloids and upward
increasing intraclasts and oolites, sometimes interbedded with thin
layers of primary dolomudstone. The succession ends with
stromatolitic dolostones, mud cracks and dolomitized evaporites
(Alvarenga et al. 2004; Figueiredo 2006; Nogueira et al. 2007).
Zaine (1991) and Nogueira et al. (2007) interpreted the deposi-
tional environment of the uppermost part of the Nobres Fm. as
peritidal, arid tidal and sabkha.
The Serra Azul Fm. ranges between 250 and 300 m thick in the
region of its type section where it is completely exposed. It is subdi-
vided into two informal members: a lower unit comprising diamic-
tites and an upper unit of pelites with increasing fine sand upsection
(Figueiredo 2006; Alvarenga et al. 2007). The diamictite member is
exposed only around the Serra Azul Syncline and near the town of
Nobres. The clayey-siltstone member occurs over hundreds of kilo-
metres in the northern part of the Paraguay Belt (Fig. 46.1). Lateral
variations in the thickness of the Serra Azul Fm. result mainly from
variations in the thickness of the diamictite member, which does
not crop out in some areas due to non-deposition or erosion prior
to deposition of the upper member. Where the Serra Azul Fm. is
absent, stromatolitic dolomites from the upper Nobres Fm. lie dis-
conformably beneath sandstones of the Raizama Fm.
The Raizama Fm. is c. 1200 m thick (Hennies 1966) and is com-
posed of conglomerates, sandstones and arkoses interbedded with
siltstones and shales (Hennies 1966; Ribeiro Filho et al. 1975).
Shales are more abundant in the lower part of the Raizama Fm.
Structures suggesting a shallow to emergent depositional environ-
ment, such as cross-stratification in sandstones, ripple-waves in silt-
stones and mud cracks in claystones, are common (Ribeiro Filho
et al. 1975; Zaine 1991). Zaine (1991) found a low-diversity ichno-
fossil assemblage in the sandstones from the basal portion of this
formation: (i) Cochlichnus (simple and sinuous caving) between
the crests of ripples and (ii) Lockeia (resting mark) associated with
(iii) Planolites (trackway), all in the same stratum. The contact
with the Serra Azul Fm. is gradational in the Azul Range region
and erosive in other localities (Figueiredo et al. 2008), where the
Nobres Fm. is in sharp contact with the Raizama Fm. (Hennies
1966; Ribeiro Filho et al. 1975; Nogueira et al. 2007). The basal
part of the Raizama Fm. has been interpreted as transgressive depos-
its with marine and fluvial facies (Nogueira et al. 2007).
Glaciogenic deposits and associated strata
Serra Azul Fm.
The diamictite unit is c. 70 m thick and was deposited on top of the
Nobres Fm., but this contact is not exposed (Fig. 46.2). It is com-
posed of massive to crudely stratified diamictites with an abundant
sandy-clayey-silty matrix and sparse clasts of varied composition,

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 499– 502. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.46
500
Fig. 46.1. Geological maps showing the location of Paraguay Belt on the Amazon craton and the occurrence of the Serra Azul Fm. in the northern part of the
Paraguay Belt.
size and shape (Alvarenga et al. 2007). The matrix is commonly
reddish and massive, becoming yellowish and stratified in the
uppermost 60 cm, immediately below the siltstone unit.
Clast lithologies include sandstone, arkose, chert, quartz, quart-
zite, meta-conglomerate, weathered carbonate, claystone, diabase,
granite, gneiss, basalt and rhyolite. The clast size ranges from
granules, pebbles (c. 5%) and cobbles (c. 1%) to boulders (1%),
with a few reaching 30 cm in diameter (Figueiredo et al. 2008).
Their shape also varies from well rounded to highly angular;
some are faceted, polished and striated. Some rounded and
polished clasts have impact marks preserved on the non-striated
surface, while some oblate clasts have parallel sharply abraded
and striated surfaces, suggesting a reworking of fluvial pebbles
by glaciers. Other clasts are bullet-shaped; the major axes of one
large cobble and one boulder have a WNW trend that could
Fig. 46.2. Generalized Paraguay Belt stratigraphy and a detailed stratigraphic
log of the shelfal Serra Azul Fm. in the northern part of the belt.
M. F. FIGUEIREDO ET AL.
indicate a glacial transport direction from the Amazon Craton
(WNW) towards the Paraguay Belt (ESE).
The siltstone unit is about 25 m thick and sharply overlies the
diamictite level. This unit comprises a reddish laminated siltstone
that in places contains sparse granules (2 –3 mm) of quartzite and
quartz (Figueiredo et al. 2008). Towards the top, this reddish silt-
stone grades into rhythmites that are about 150 m thick (Figueiredo
et al. 2008). At the base of this succession, there are alternating
clayey and silty plane-parallel laminae with intercalations of sand-
stone layers (5 –10 cm thick) showing small hummocky lami-
nation that indicates oscillatory flow events. In the Nobres town
region, there is a lens (12 m) of limestone in the basal portion of
this rhythmite succession (Alvarenga et al. 2007). The textures
of the limestone, from bottom to top, include laminated mudstone,
massive mudstone with small-scale cut-and-fill channels filled by
wackestone, laminated mudstone with clotted texture, mudstone
with truncated laminae, and cyclic successions of breccias with
slumping deformation and plane-parallel lamination. Red clayey
films occur between limestone laminae, the upwards increase of
which causes a nodular texture due to the progressive deformation
of the limestone by compaction (Möller & Kvingan 1988).
In the intermediary portion of the rhythmite succession, there
is a constant input of sandy, alternating clayey (2 mm to 3 cm)
and very fine sandy (1 –5 mm) laminae overlain by a massive
medium sandstone layer c. 15 cm thick with very angular quartz
grains. Above this layer, the rhythmites continue, although with
more contribution from very fine sand (laminae of 2 mm to
2 cm) in relation to silt (laminae of 1–10 mm).
The rhythmites grade into a heterolithic succession c. 40 m thick
that is composed of inter-bedded siltstones and sandstones (Figueir-
edo et al. 2008). The basal part has intercalated lenses (0.5–2 cm
thick) of cross-laminated very fine sandstone within plane-parallel
clayey-siltstones. The sandstone lenses become thicker (from
laminae to metric to decimetric-scale beds) and coarser towards
the top, while siltstone beds decrease upwards. At the top of the suc-
cession, the sandstone facies dominates in metre-scale sigmoidal
beds (lobes) interbedded in a heterolithic bed. The first thick sand-
stone bed is parallel laminated, and the upper beds display
massive bedding, cross-stratification and hummocky structures.
Boundary relations with overlying and underlying
non-glacial units
In the Azul Range region (Fig. 46.1), the basal contact of diamic-
tites from the Serra Azul Fm. with the dolomites of the underlying
 THE SERRA AZUL FORMATION, PARAGUAY BELT
Nobres Fm. is not exposed. However, the presence of brecciated,
silicified, stromatolitic and nodular carbonate clasts with charac-
teristics similar to those of the Nobres Fm. suggests an erosive
contact between these two units.
The upper contact between the diamictites and the laminated
siltstones is sharp and conformable (Figueiredo 2006; Alvarenga
et al. 2007). However, it is difficult to define where the last evi-
dence of glacial influence occurs within the above siltstone
member (Figueiredo et al. 2008). The upper silstones grade into
rhythmites. Towards the top of the formation, sandstone inter-
bedded with these rhythmites grades into heterolithic sediments.
The sandstones are cross-stratified and become thicker and
coarser towards the top of the Serra Azul Fm., marking a transi-
tional contact with the Raizama Fm.
Chemostratigraphy
No geochemical data are available for the Serra Azul Fm. Stable
and radiogenic isotope data from the underlying Araras Group
are reviewed elsewhere in this volume (Alvarenga et al. 2011).
To summarize, d13C (carbonate) values in the Nobres Group
vary between c. –2 and 0‰ and the least radiogenic 87Sr/86Sr
ratios in the upper Guia Fm. limestones and Nobres Fm. range
from 0.7078 and 0.7081 (Nogueira et al. 2007; Alvarenga
et al. 2008).
Geochronological constraints
No direct radiometric ages or fossils are available for the Serra
Azula Fm., and precise ages are notably sparse in the Paraguay
Belt. A Pb/Pb whole-rock isochron age of 633 + 25 Ma (Babisnki
et al. 2006) was determined on carbonates of the Mirassol d’Oeste
Fm. at the base of the Araras Group, providing a maximum age
constraint on the Serra Azul Fm. that implies a mid-Ediacaran
age. A minimum age constraint is provided by an encapsulation
40Ar/39Ar metamorphic age of 528 + 26 Ma obtained on clay
fractions from multiple stratigraphic levels within the Paraguay
Belt (Tohver et al. 2010). Additional radiometric data include
K –Ar ages of c. 730 Ma obtained on volcanic clasts recovered
from the diamictites, and Sm– Nd TDM ages ranging from 1.9 to
1.6 Ga obtained on the matrix of the diamictites and pelites of
the Serra Azul Fm. (Figueiredo, unpublished data). These TDM
ages suggest that the rocks of the Ventuari-Tapajós and Sunsás
provinces (Tassinari & Macambira 1999) from the Amazon
craton could be the main source of sediments to the Serra
Azul deposits. The source area of the younger volcanic clasts
(c. 730 Ma), however, has not yet been identified.
Discussion
The massive and poorly sorted characteristics of the Serra Azul
Fm. diamictite indicate deposition from high density flows
(Edwards 1978) without reworking by currents. Furthermore, the
large variety of clast compositions requires a transportation
agent capable of scouring huge areas and mixing many rock
types (Eyles & Miall 1984). In addition, large quantities of striated,
faceted and polished clasts were recovered from the diamictites,
which normally indicates a glacial environment (Eyles 1993).
Two bullet-shaped boulders both orientated along the same trend
were also found that suggest east – SE ice movement, that is,
from the palaeocontinent (Amazon craton) to the ocean (Paraguay
Belt). However, these two boulders alone are insufficient evidence
to classify this diamictite as an ice contact deposit. Other charac-
teristic structures, such as deformed or sheared substrate, eroded
basement and striated pavement, are necessary for such a classifi-
cation, but these were not observed (Figueiredo et al. 2008).
501
For the last 60 cm of the diamictite unit, a crude stratification is
apparent, and the clasts become smaller, suggesting retreat of the
glacier. The laminated siltstones deposited directly on the diamic-
tites could represent further retreat of the glacier, producing distal
facies deposits that are less glacially influenced.
A calcareous lens deposited above the reddish laminated silt-
stone suggests an interruption of clastic deposition and could
mark a maximum flooding surface. The rhythmite (sand and
clay) deposited on top of the carbonate represents a new sequence
of clastic sedimentation. This sequence began as cyclic (rhythmite)
with storm events (hummocky cross stratification). Upsection,
sandstone layers become thicker and coarser and contain tidal
bundles. This progression of facies suggests a progradational
deltaic sequence.
The dolomite –siliciclastic succession of facies in the under-
lying Nobres Fm. is consistent with a shallowing-upward platform
succession (Almeida 1964; Alvarenga et al. 2004; Nogueira et al.
2007) with evolution from a slope and moderately deep platform to
transitional and supratidal environments like arid tidal flats and
sabkhas (Nogueira et al. 2007). Thus, Serra Azul glaciers trans-
gressed an Ediacaran-aged carbonate platform.
The siliciclastic Raizama Fm. that overlies the Serra Azul Fm.
suggests deposition in an epeiric environment during slow and pro-
gressive basin subsidence (Almeida 1964; Nogueira et al. 2007).
The conglomerates record development of fluvial channels with
subaerial exposure, and the sedimentary structures and textures
of the sandstones record a tidally influenced platform (Almeida
1964; Nogueira et al. 2007). These characteristics indicate that
the basal Raizama Fm. was deposited from proximal to distal
facies in a transgressive regime (Nogueira et al. 2007).
We thank the Conselho Nacional de Desenvolvimento Cientı́fico e Tecnológico
(CNPq) for a research grant (Proc. 473614/2004-9) and the PhD scholarship
awarded to the senior author. M. F. Figueiredo thanks the Fundação de Amparo
à Pesquisa do Estado de São Paulo (FAPESP) for a Master’s scholarship (Proc.
04/06225-5). M. Babinski and C. J. S. Alvarenga are CNPq Research Fellows.
This represents a contribution of the IUGS- and UNESCO-funded IGCP (Inter-
national Geoscience Programme) project #512.
References
Almeida, F. F. M. 1964. Geologia do Centro-Oeste Matogrossense. Min-
istério de Minas e Energia, DNPM, Boletim da Divisão de Geologia
Mineral, 215, 1– 137.
Alvarenga, C. J. S. & Trompette, R. 1993. Evolução Tectônica Brasili-
ana da Faixa Paraguai na região de Cuiabá. Revista Brasileira de
Geociências, 23, 18 –30.
Alvarenga, C. J. S., Santos, R. V. & Dantas, E. L. 2004. C-O-Sr iso-
topic stratigraphy of cap carbonates overlying Marinoan-age glacial
diamictites in the Paraguay Belt, Brazil. Precambrian Research,
131, 1– 21.
Alvarenga, C. J. S., Figueiredo, M. F., Babinski, M. & Pinho, F. E. C.
2007. Glacial diamictites of Serra Azul Formation (Ediacaran, Para-
guay Belt): evidence of the Gaskiers glacial event in Brazil. Journal
of South American Earth Science, 23, 236– 241.
Alvarenga, C. J. S., Boggiani, P. C. et al. 2011. Glacially-influenced
sedimentation of the Puga Fm., Cuiabá Group and Jacadigo
Group, and associated carbonates of the Araras and Corumbá
groups, Paraguay Belt, Brazil. In: Arnaud, E., Halverson, G. P.
& Shields-Zhou, G. (eds) The Geological Record of Neo-
proterozoic Glaciations. Geological Society, London, Memoirs,
36, 487– 497.
Babinski, M., Trindade, R. I. F., Alvarenga, C. J. S., Boggiani, P. C.,
Liu, D., Santos, R. V. & Brito Neves, B. B. 2006. Chronology
of Neoproterozoic ice ages in central Brazil. In: V South American
Symposium on Isotope Geology, 2006, Punta del Este, Uruguay,
Short Papers, 1, 303–306.
Castelneau, F. 1850. Expedition dans les parties centrales de l’Améri-
que du Sud, de Rio de Janeiro a Lima, et de Lima au Pará. Histoire
du Voyage. Librairie Editeur, Paris, Tomo II.
502 M. F. FIGUEIREDO ET AL.
Edwards, M. B. 1978. Glacial environments. In: Reading, H. G. (ed.)
Sedimentary Environments and Facies. Elsevier, New York,
416– 438.
Eyles, N. 1993. Earth’s glacial record and its tectonic setting. Earth
Science Reviews, 35, 1– 248.
Eyles, N. & Miall, A. D. 1984. Glacial facies. In: Walker, R. G. (ed.)
Facies Models, 2nd edn. Geoscience Canada, Reprint Series, 1,
15 –38.
Figueiredo, M. F. 2006. Quimioestratigrafia das rochas ediacarianas do
extremo norte da Faixa Paraguai, Mato Grosso. MS thesis, Instituto
de Geociências, Universidade de São Paulo.
Figueiredo, M. F., Babinski, M., Alvarenga, C. J. S. & Pinho, F. E. C.
2004. Diamictites overlying Marinoan-age carbonates of Araras
Formation, Paraguay Belt, Brazil: evidence of a new glaciation?
In: Symposium on Neoproterozoic-Early Paleozoic Events in SW-
Gondwana, IGCP-478, Second Meeting, Brazil, 18– 19.
Figueiredo, M. F., Babinski, M., Alvarenga, C. J. S. & Pinho, F. E. C.
2005. Nova unidade litoestratigráfica: Formação Serra Azul, Faixa
Paraguai, Mato Grosso. IX Simpósio de Geologia do Centro-Oeste,
Brazil, 23 –25.
Figueiredo, M. F., Babinski, M., Alvarenga, C. J. S. & Pinho, F. E. C.
2008. Nova unidade litoestratigráfica registra glaciação ediacarana
em Mato Grosso: Formação Serra Azul. Geologia USP, 8, 65 –75.
Hennies, W. T. 1966. Geologia do Centro-Norte Mato-Grossense.
Unpublished PhD dissertation, Escola Politécnica, Universidade de
São Paulo.
Möller, N. K. & Kvingan, K. 1988. The genesis of nodular limestone in
the Ordovician and Silurian of the Oslo Region (Norway). Sedimen-
tology, 35, 405– 420.
Nogueira, C. R. A., Riccomini, C., Sial, A. N., Moura, C. A. V., Trin-
dade, R. I. F. & Fairchild, T. R. 2007. Carbon and strontium isotope
fluctuations and paleoceanographic changes in the late Neoprotero-
zoic Araras carbonate platform, southern Amazon Craton, Brazil.
Chemical Geology, 237, 168– 190.
Ribeiro Filho, W., Luz, J. S. & Abreu Filho, W. 1975. Relatório Final
do Projeto Serra Azul. Ministério de Minas e Energia, DNPM/
CPRM, 1, 1 –104.
Tassinari, C. C. G. & Macambira, M. J. B. 1999. Geochronological pro-
vinces of the Amazonian Craton. Episodes, 22, 174–182.
Tohver, E., Trindade, R. I. F., Solum, J. G., Hall, C. M., Riccomini, C.
& Nogueira, A. C. 2010. Closing the Clymene ocean and bending a
Brasiliano belt: evidence for the Cambrian formation of Gondwana,
southeast Amazon craton. Geology, 38, 267– 270.
Trindade, R. I. F., Font, E., D’Agrela Filho, M. S., Nogueira, A. C. &
Riccomini, C. 2003. Low-latitude and multiple geomagnetic rever-
sals in the Neoproterozoic Puga cap carbonate, Amazon craton.
Terra Nova, 15, 441–446.
Trompette, R. 1994. Geology of Western Gondwana (2000– 500 Ma).
Pan-African –Brasiliano Aggregation of South America and Africa.
Balkema.
Trompette, R., Alvarenga, C. J. S. & Walde, D. 1998. Geological evol-
ution of the Neoproterozoic Corumbá graben system (Brazil).
Depositional context of the stratified Fe and Mn ores of Jacadigo
Group. Journal of South American Earth Science, 11, 587– 597.
Zaine, M. F. 1991. Análise dos fósseis de parte da Faixa Paraguai
(MS, MT) e seu contexto temporal e paleoambiental. Unpublished
PhD dissertation, Instituto de Geociências, Universidade de São
Paulo.
 Chapter 47

The Bebedouro Formation, Una Group, Bahia (Brazil)

J. T. GUIMARÃES1, A. MISI2*, A. J. PEDREIRA1 & J. M. L. DOMINGUEZ2

1
Geological Survey of Brazil (CPRM) Av. Ulysses Guimarães, 2862 – Sussuarana – Centro Administrativo
da Bahia 41213-000 Salvador-Bahia, Brazil
2
Universidade Federal da Bahia, Centro de Pesquisa em Geofı́sica e Geologia, Instituto de Geociências,
Campus da Federação 40170-290 Salvador-Bahia, Brazil
*Corresponding author (e-mail: aroldo.misi@gmail.com)

Abstract: The Bebedouro Formation (Fm.) is a Neoproterozoic glaciogenic succession at the base of the Una Group on the São
Francisco Craton. The glaciogenic sequence is composed of diamictites, pelites and sandstones with a variety of lithofacies that are
grouped into four associations: (i) ice-contact, (ii) pro-glacial, (iii) ice-rafted and (iv) aeolian (extraglacial). Thus, the Bebedouro Fm.
is interpreted to have been deposited in a shelf marine environment where glacio-proximal sedimentation dominated. It variably overlies
Palaeoproterozoic/Archaean basement and Mesoproterozoic metasedimentary rocks of the Chapada Diamantina Group.
The overlying carbonate succession (Salitre Fm.) is informally subdivided into five mappable units. Unit C, at the base, is composed of
red argillaceous dolostone with typical negative d13C signatures averaging – 5.1‰ (VPDB). Unit B consists of grey laminated limestones
that grade upward into grey dolostone with tepee structures of Unit B1; both have d13C values around 0‰. The overlying Unit A consists
of interbeds of greyish marl, shales and siltstones grading upward into massive black organic-rich limestone with oolitic and pisolitic beds
of Unit A1. Positive d13C signatures averaging þ8.5‰ VPDB characterize the black limestones of Unit A1. 87Sr/86Sr ratios of well-
preserved samples of these carbonate units show values of 0.70745 and 0.70765. These units can be correlated litho- and chemostrati-
graphically with formations of the Bambuı́ Group (São Francisco Basin) to the west.

The Bebedouro Formation crops out in the central part of Bahia
state, Brazil, where it occurs in narrow and apparently discontinu-
ous north – south strips that span a length of 400 km and cover an
area of more than 40 000 km2 (Fig. 47.1). This formation occurs
in several isolated sub-basins (Fig. 47.1) but was probably depos-
ited within a single, large, precursor basin that was subsequently
segmented during tectonic events related to the assembly of
western Gondwana (Teixeira et al. 2007).
The name ‘Bebedouro’ was introduced by Oliveira & Leonardos
(1940) to designate conglomerates occurring in the eastern border
of the Mesoproterozoic basin of the Chapada Diamantina, in Bahia
state. The conglomerates were first described by Derby (1905) but
their glacial origin was proposed by A. I. Oliveira in 1921, as
reported by Moraes Rego (1930), and Williams (1930). Sub-
sequent work by several authors demonstrated their glacial
origin (Mello Jr 1938; Kegel 1959; Brito Neves 1967; Söfner
1973; Montes 1977; Rocha Campos & Hasui 1981; Dominguez
1993). Montes (1977) noted the presence of striated pavements
and dropstones, and proposed a continental glaciation for the
origin of the Bebedouro Fm. Glaciomarine sequences and ‘conti-
nental tillites’ were described by Söfner (1973), Montes (1977)
and Karfunkel & Hoppe (1988), who showed evidence of ice
flow from east to west. Subsequently, Guimarães (1996) proposed
that the deposition of the Bebedouro Fm. occurred in a glaciomar-
ine environment by sub-aqueous debris-flow, turbidity currents,
iceberg melting and locally aeolian processes.
The Bebedouro Fm. correlates with the Jequitaı́ Fm. of the São
Francisco Basin (Bambuı́ Group) to the west and SW (Uhlein et al.
2011), which is also referred to as the Macaubas Group by
some authors.
Structural framework
The Bebedouro Fm. and the overlying carbonate succession of the
Una Group were likely deposited in a basin formed by extension
related to the fragmentation of the Rodinia supercontinent, starting
c. 950 Ma and continuing until c. 600 Ma (Condie 2002). These
events occurred synchronous with the onset of Gondwana assem-
bly (Condie 2002; Cordani et al. 2003; Teixeira et al. 2007). The
epicontinental and passive margin Neoproterozoic basins were
formed over and around the 1.8 Ga stabilized terrains of the São
Francisco Craton (Almeida 1977). They were intensely deformed
and faulted by the events of the Brasiliano orogeny, during which
the most important movements occurred between 650 and 500 Ma.
The present configuration of the craton separates the Una Group
into four geographically isolated basins on the stabilized eastern
part of the São Francisco craton: the Salitre, Irecê, Una-Utinga
and Ituaçu sub-basins (Fig. 47.1). Presumably, these sub-basins
were originally connected, forming a single, large basin that pre-
dated the Brasiliano orogeny. Deformed equivalents of the Una
Group occur in fold belts along the north and east borders of the
cratonic area within the Rio Preto, Riacho do Pontal, Vasa
Barris/Miaba and Rio Pardo groups. Metamorphic grade is very
low or absent in the intra-cratonic sub-basins and low (up to green-
shist facies) in the adjacent fold belts.
Structural lineaments (fractures and faults) trending NNW –
SSE, probably related to extensional movements, cross both
the basement (Palaeoproterozoic and Mesoproterozoic) and the
Neoproterozoic sedimentary cover, suggesting that the extensional
tectonic regime continued during the sedimentation of the Una
Group (Misi et al. 2005). Later orthogonal compression (north –
south and east –west) generated a basin and dome structure
pattern in the sedimentary cover. In some places, as in the
western border of the Irecê sub-basin, the Mesoproterozoic base-
ment is exposed.
Stratigraphy
The Una Group carbonates were deposited in a series of sub-basins
in Bahia state, northeastern São Francisco craton, whereas the
equivalent Bambuı́ Group was deposited in the São Francisco
Basin, located in the states of Minas Gerais, Goiás and along the
western border of Bahia state (Fig. 47.1). The Bebedouro Fm.,
which is less than 200 m thick, lies unconformably over either
Mesoproterozoic metasedimentary rocks of the Chapada Diaman-
tina Group or a Palaeoproterozoic/Archaean medium to high-
grade metamorphic complex. It is overlain by the carbonate
platform succession of the Salitre Formation (Fig. 47.2).

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 503– 508. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.47
504 J. T. GUIMARÃES ET AL.

Localization 43º15’00’’W 40º15’00’’W

65º00’00’W 9º30’00’’S 9º30’00’’S
NOVA
REMANSO

42º00’00’’W

2º00’00’’N 2º00’00’’N

Study
BRAZIL Area
S. Francisco Basin
(Bambuí Group)

São BAHIA
Francisco XIQUE-
Craton XIQUE
Salitre
21º00’00’’S 21º00’00’’S
Sub-basin

42º00’00’’W
IRECÊ

MORRO
Irecê DO CHAPEU
Sub-basin
500 km 0 500 1.000 1.500 km

BA
44º00’00’’S 44º00’00´´S 052

SEABRA
BR
242
65º00’00’’W

Town
Una-Utinga
Syncline ANDARAI Sub-basin
Brachysyncline ITAETÊ
MACAUBAS
Phanerozoic Cover

São Francisco Supergroup

Salitre Formation

Bebedouro Formation

Glacial contact deposits

Proglacial and ice-rafted debris

Aeolian extraglacial deposits Ituaçu

Sub-basin
Espinhaço Supergroup
ITUAÇU
14º10’00’’S 14º10’00’’S
Basement older than 1.8 Ga 43º15’00’’W 40º15’00’’W

Direction of glaciers movement

50 km 25 0 50 km

Fig. 47.1. Geological map showing the Bebedouro Fm. in the central area of Bahia state, Brazil (modified from Guimarães 1996).

Misi & Souto (1975) identified two members in the Bebedouro The Bebedouro Fm.
Fm., within the Irecê Sub-basin: the Lower Member is composed of
green and reddish meta-siltstones and slates with chert and lime- The Bebedouro Fm. is composed of diamictites, sandstones and
stone lenses, with lithic metagreywacke forming the matrix pelites. These lithofacies show sharp, gradational or erosive con-
above the diamicitites (total thickness c. 100 m); the Upper tacts and are generally texturally and mineralogically immature.
Member is a feldspatic sandstone (meta-arkose), up to 80 m The diamictites occur mainly in the eastern margins of the
thick. A more detailed study by Guimarães (1996) that included Una-Utinga and Campinas sub-basins (Fig. 47.1). They contain
the other sub-basins showed that the diamicites are in fact interca- poorly sorted, heterolithic clasts. At least four types of coarse-clast
lated with arkose, greywackes and lithic greywacke. These lithofa- bearing facies have been identified by Guimarães (1996): (i)
cies are described in the following section. matrix-supported massive diamictite, (ii) matrix-supported dia-
The Salitre Fm., overlying the Bebedouro Fm., is more than mictite with parallel stratification, (iii) matrix-supported diamic-
500 m thick and is composed of five informal units (Misi & tites with sigmoidal cross-bedding and (iv) massive clast-
Souto 1975; Misi 1978) (Fig. 47.2), briefly summarized in the supported conglomerate/breccia. In all facies, clasts are domi-
next section. These units are correlative with the Bambuı́ Group nantly but not exclusively angular and range in size from granules
formations of the São Francisco Basin, as shown in Table 47.1. to blocks .2 m in diameter. Clast compositions include granite,
gneiss, pegmatite, schist, phyllite, basic/ultrabasic rocks, volca-
nics, calc-silicates, quartz, grey and green quartzites, sandstones,
Glaciogenic deposits and associated strata argillites, carbonates and chert. Faceted and polished surfaces
are common, and striated clasts have been found.
The descriptions that follow are essentially based on Misi & Souto The matrix is composed of greywacke, lithic greywacke, lithic
(1975), Misi (1978) and Guimarães (1996). sandstone, arkose, sub-arkose and argillite with quartz and rock
 THE BEBEDOURO FORMATION, BRAZIL 505

Fig. 47.2. Lithostratigraphic successions of

the Una Group in the Irecê sub-basin
(Source: Misi & Souto 1975; Misi 1978).

fragments. Clay material and impregnated iron oxides are impor-
tant components of the matrix and clasts, suggesting that the
matrix component is at least partially derived from the comminu-
tion of larger fragments. Generally, the matrix is poorly sorted and
contains fragments of variable sizes and origins. The larger frag-
ments correspond are angular.
The diamictite bodies are lenticular, lobate or tabular at the
outcrop scale and commonly show reworking by storm waves.
Internally they are variably massive or stratified, and sometimes
exhibit slump structures. Normal and inverse grading, clast imbri-
cation, plane-parallel bedding and sigmoidal cross-bedding occur
within stratified bodies.
Guimarães (1996) has suggested that the diamictites of the
Bebedouro Fm. were derived from rapid deposition with limited
transport in confined debris flows (cohesive and non-cohesive),
high concentration turbidity currents, and grain falls in a subaqu-
eous environment (Fig. 47.3).
The Bebedouro sandstones, although widespread within the
Neoproterozoic basins, are subordinate lithotypes. They include
ochre, brown and grey-greenish arkose, sub-arkose, greywacke,
lithic greywacke, lithic sandstone and quartz arenite. Generally
they are interbedded with diamictites and pelites. The framework
of the sandstones includes spherical to more commonly elongate
rock fragments and granules and clasts of quartz and feldspar.
Guimarães (1996) recognized parallel-bedded reddish (Fe-rich)
sandstones with and without dropstones, hummocky cross-bedding
and large-scale tangential cross-bedding. Asymmetric ripples and
normal gradating are also observed in these lithofacies. Lone-
stones, observed in some places, are gravel to block-size clasts
with variable composition and shape. These sandstones represent
wave-reworked fluidized/liquefied and confined gravitational
flows, which evolved to high-concentration turbidity currents as
a result of iceberg melting.
Bimodal, medium to coarse-grained sandstones with large-
scale tangential and trough cross-bedding interpreted to represent
aeolian deposition occur in isolation on the eastern margin of the
Una-Utinga sub-basin, close to the town of Itaetê (Fig. 47.1).
The pelites include variably silicified grey-greenish, red-ochre
and purplish siltstone, greywacke and arkose. These lithotypes
are widespread, found mainly in the Irecê and Ituaçu sub-basins
as well as in the western margins of the Campinas and Una-Utinga
sub-basins (Fig. 47.1). Laminated and massive pelites are present
WEST EAST

BUOYANT SEDIMENTS PLUME

Table 47.1. Lithostratigraphic correlation between the Bambuı́ and
ICEBERGS
Una groups
SL ICE SHEET

Lithotypes Bambuı́ Group Una Group (units

(formations) or formations)

Arkose, siltstone Três Marias –

Siltstone, pelites Serra da Saudade –
Black oolitic limestones Lagoa do Jacaré Unit A1 Log A

Marl, shales Serra Santa Helena Unit A Log B

Dolostone laminated Sete Lagoas Unit B1 Log C
limestones, red dolostone Unit B
Unit C
Diamictite, arkose, Jequitaı́, Macaubas Bebedouro Fig. 47.3. Ice-contact glaciomarine system, Bebedouro Fm. See Figure 47.4
pelitic rocks for schematic representation of logs A, B and C as indicated in this figure
(Source: Guimarães 1996).
506 J. T. GUIMARÃES ET AL.

(a) (b) (c)

SUSPENSION
SUSPENSION DERIVED SEDIMENTS
DERIVED SEDIMENTS

DEBRIS FLOW

HIGH-DENSITY
DEBRIS FALL TURBIDITE SUSPENSION

DERIVED
DEBRIS FLOW DEBRIS FLOW SEDIMENTS

HIGH-DENSITY
TURBIDITE
DEBRIS FALL Vertical Scale
0

DEBRIS FLOW SUSPENSION

DERIVED SEDIMENTS 2m
Fig. 47.4. Schematic representation of
glacial lithofacies associations within the
DEBRIS FLOW
Bebedouro Fm. (logs A, B and C, Fig. 47.4)
(Source: Guimarães 1996). (a) Ice – contact
diamictites; (b) Ice – marginal fan; and (c)
Iceberg melt.

with or without dropstones. Pelite packages range in thickness
from centimetres to 20 m. Reworking by storm waves is com-
monly observed. Compositionally, the pelites contain clay-
minerals, micro-crystalline quartz, iron oxides and carbonate
cement. The larger grains are spherical to elongated and either
angular or sub-rounded. They are composed of quartz, feldspar,
mica and rock fragments. Interspersed dropstones are of diverse
shape, size and composition.
The pelitic rocks are believed to be resedimented sub-aqueous
deposits, formed by cohesive and unconfined gravity flows, low-
concentration turbidity currents, and suspension fallout from
icebergs and hypopycnal plumes (Figs 47.3 & 47.4).
The Salitre Fm.
The carbonate succession overlying the Bebedouro Fm., referred
to as the Salitre Fm. in the available geological maps, is well
13
δ C(%0VPDB)
-15 -10 -5 0 +5 +10 +15
87
Sr/ S r
0.70745
Unit A 1
400 m
Unit A
Unit B 1
Salitre Fm.
exposed in all of the studied sub-basins. It is informally subdivided
into five mappable units within the Una Group. The following
lithofacies, briefly described bottom to top, were originally ident-
ified in the Irecê sub-basin (Misi & Souto 1975) and later in the
other sub-basins (Misi 1979): red argillaceous dolostones (Unit
C), laminated limestone (Unit B), evaporitic dolostone (Unit
B1), marl (Unit A) and black oolitic limestone (Unit A1).
The red dolostone of Unit C overlies the Bebedouro Fm.
(Fig. 47.2) and has a maximum thickness of 70 m. It is overlain
by c. 180 m of grey laminated limestone with rhythmic successions
of centimetre-thick beds of limestone or dolomitic limestone and
argillaceous material (Unit B). This facies grades upward into
dolomitic layers with replaced nodular anhydrite, tepee structures,
microbialaminites and columnar stromatolites that comprise Unit
B1, which has a total thickness of c. 50 m and was observed in
its entirety in drill hole IL 53 (Misi & Kyle 1994). The overlying
unit A is composed of marl, argillites and siltstones and varies
from 0 to 100 m in thickness. This unit is followed above by
more than 150 m of black, organic-rich oolitic and pisololitic
limestone of Unit A1 (Fig. 47.2).
At least two major sedimentary cycles can be identified in the
Salitre Fm. (Misi et al. 2005, 2007). Cycle 1 starts at the base of
86
the Salitre Fm. (Units C and B) and ends in the dolomites with
tepee structures (Unit B1) that representing the exposure surface
of a typical shallowing-upward sequence. Cycle 2 starts with a
transgressive event represented by the marl and shale (Unit A)
Olithic, black
limestone immediately above the dolomitic facies of Unit B1, passing to
oolitic and pisolitic black limestones with trough cross-bedding,
Marl
Dolostone indicating shallow, high-energy sedimentation (Unit A1).

Laminated
0.70765 limestone
Pink dolostone
Unit B Boundary relations with overlying and underlying
Diamictite
0.70780 non-glacial units
200 m

In most locations, the Bebedouro Fm. lies unconformably over

Mesoproterozoic sedimentary rocks of the Morro do Chapeu Fm.
of the Chapada Diamantina Group. On the eastern side of the
Unit C
Una-Utinga sub-basin (Fig. 47.1), relatively small outcrops of the
Bebedouro Fm. directly overlie the Archaean/Palaeoproterozoic
Bebedouro Fm.
terrains of the São Francisco Craton, which there is composed of
0m
gneisses and migmatites.
Fig. 47.5. Stratigraphic variation of d13C and 87Sr/86Sr in carbonates of the The Bebedouro Fm. is typically overlain by red argillaceous
Una Group, Irecê Basin (Isotopic data from Torquato & Misi 1977; and Misi & dolostones of Unit C of the Salitre Fm., which is widely occurring
Veizer 1998). but absent on the western margin of several sub-basins. In these
 THE BEBEDOURO FORMATION, BRAZIL
locations, the Bebedouro Fm. is overlain by laminated limestones
of Unit B.
Chemostratigraphy
Chemostratigraphic studies of the carbonate successions of the
Una Group in the Irecê sub-basin (Torquato & Misi 1977; Misi
& Veizer 1998) show remarkable stratigraphic variation of Sr
and C isotopes. These trends are summarized in Figure 47.5.
In summary, 87Sr/86Sr ratios vary between 0.70745 and 0.71776,
but the best-preserved samples, indicated by the lowest Mn –Sr
ratios (,0.03) and highest total Sr contents (.300 to
c. 2300 ppm) define a much narrower range of 0.70745–
0.70769. The d13C signatures show low values of c. –5‰
(n ¼ 12) in Unit C (Torquato & Misi 1977), values varying
around 0‰ through Units B and B1, and a positive excursion of
up to 9.4‰ in the black and organic-rich carbonates of Unit A1.
Powis (2006), Misi et al. (2007) and Vieira et al. (2007) report
similar isotopic variation in the carbonate successions of the
Bambuı́ Group, in the São Francisco Basin. Negative anomalies
like that found in Unit C are characteristic of many carbonate suc-
cessions immediately above glaciogenic lithotypes around the
world (Knoll et al. 1986; Kaufman et al. 1991; Halverson et al.
2005; Shields 2005).
Geochronological constraints
The ages of the Neoproterozoic successions of the São Francisco
Craton, including the Bebedouro Fm. in Bahia state, are poorly
resolved due to the lack of datable volcanic rocks. Metamorphism
of the rocks, although of low grade, has contributed to a general
lack in confidence of the Rb –Sr and Pb/Pb age determinations
(Macedo 1982; Macedo & Bonhomme 1984; Trindade et al.
2004). The most widely accepted radiometric determination is a
Pb/Pb isochron age of 740 + 27 Ma from six samples of the
basal Sete Lagoas Fm. (Bambuı́ Group) (Babinski et al. 2007).
Pelitic units of the Mesoproterozoic Chapada Diamantina Group
underlying the Bebedouro Fm., dated by Babinski et al. (1993),
yielded a Pb/Pb age of 1140 + 140 Ma, whereas intrusive mafic
rocks within the Chapada Diamantina Group were dated by U –
Pb zircon at 1496 + 3.2 Ma (Guimarães et al. 2005).
Mineralization
In the Irecê sub-basin, Unit B1 hosts massive stratiform and strata-
bound pyrite, sphalerite and galena deposits associated with
medium grey tepee dolostone containing nodules of calcite,
silica and barite (Misi & Souto 1975; Misi et al. 2005). Galena
and sphalerite mineralization also occurs in the Una-Utinga sub-
basin. The sulphide concentrations in the Una Group are associated
NNE – SSW faults (Misi et al. 2005). Economic phosphate deposits
intimately associated with columnar and laminar stromatolites
occur within massive grey to reddish-grey dolostone in a lower
stratigraphic position of the Unit B1 in the Irecê sub-basin (Misi
& Kyle 1994).
Discussion
A glacial origin for the Bebedouro Fm. was originally proposed
by Williams (1930), who described conglomerates and shales
cropping out in the northern Una-Utinga sub-basin. This author
stated that ‘it is difficult to explain the presence of a peculiar
kind of clasts within the pelitic rocks. In this sense, we do not hes-
itate in suggesting a glacial origin [for the conglomerate] which
was probably deposited in calm and deep water within a wide-
spread sea, on which icebergs were floating and liberating clasts
[to the deep sea sediments]’.
507
Evidence for the glacial origin of the Bebedouro Fm., as estab-
lished by many subsequent studies, may be summarized as follows
(Montes 1977; Montes et al. 1985; Guimarães 1996):
† the presence of clasts from different sources and variable forms
and sizes, within pelitic rocks, which are angular to sub-angular
as well as sub-rounded and rounded, some showing striae and
preferential orientation;
† the presence of striated pavements;
† the presence of fragmented and angular forms of minerals like
quartz, microcline, oligoclase and andesine, as observed in thin
sections (garnet crystals with chatter marks were also observed
by Montes et al. 1985);
† the widespread distribution of the Bebedouro Fm. in all the
studied sub-basins.
The lithofacies of the Bebedouro Fm. are thought to be the pro-
ducts of debris flows, turbidity currents, ice-rafting and suspension
mechanisms. The palaeo-flow directions of sediment transport
were from east to west. Locally, there is evidence of sediment
reworking by aeolian processes. Guimarães (1996) grouped these
lithofacies into four genetic associations: glacier contact, pro-
glacial, iceberg meltout and aeolian extraglacial, predominantly
in a proximal glaciomarine environment.
Misi et al. (2007) pointed out that ‘glaciogenic diamictite and
associated lithologies are known to occur in two discrete strati-
graphic positions within Neoproterozoic successions of South
America, and these most-likely correspond to Sturtian and Mar-
inoan (or Varanger) ice ages worldwide’. The Bebedouro Fm.
and the correlated units of the Bambuı́ Group are thought by
many authors to be Sturtian in age (Babinski et al. 2007), that is,
belonging to the older of two Cryogenian glacial epochs. Direct
evidence of a younger (end-Cyrogenian or Marinoan) glaciation
within the Bambuı́ Group or Una Group has remained elusive,
but may be cryptically expressed by the negative d13C anomaly
in dolostones in Unit B1 (Fig. 47.5) and corresponding to the top
of the first of two cycles within the Salitre Fm.
The authors are grateful to the Geological Survey of Brazil (CPRM) and to the
National Research Council of Brazil (CNPq), which provided funding for the
research over the last 10 years. Currently, A. M. receives support from project
no. 486416/2006-2 from CNPq. This represents a contribution of the IUGS-
and UNESCO-funded IGCP (International Geoscience Programme) Project #512.
References
Almeida, F. F. M. 1977. O Cráton do São Francisco. Revista Brasileira de
Geociências, 1, 13 –21.
Babinski, M., Van Schmus, W. R., Chemale, F. Jr, Neves, B. B. B. &
Rocha, A. J. D. 1993. Idade isocrônica Pb –Pb em rochas carbonáti-
cas da Formação Caboclo em Morro do Chapéu, BA. II Simpósio
sobre o Cráton do São Francisco, Sociedade Brasileira de Geologia,
2, Salvador (BA), Anais, 160– 163.
Babinski, M., Vieira, L. C. & Trindade, R. I. F. 2007. Direct dating of
the Sete Lagoas cap carbonate (Bambuı́ Group), Brazil and impli-
cations for the Neoproterozoic glacial events. Terra Nova, 19,
401– 406.
Brito Neves, B. B. 1967. Geologia das Folhas de Upamirim e Morro do
Chapéu-BA. CONESP/SUDENE. Relatório, 17, 35.
Condie, K. C. 2002. The supercontinent cycle: are there two patterns of
cyclicity? Journal of African Earth Sciences, 35, 179– 183.
Cordani, U. G., Brito Neves, B. B. & D’Agrella Filho, M. S. D. 2003.
From Rodinia to Gondwana: a review of the available evidence from
South America. Gondwana Research, 6, 265– 273.
Derby, O. A. 1905. Notas geológicas sobre o Estado da Bahia. Secretaria
de Agricultura, Viaçăo, Industria e Obras. Públicas Boletim, 7, 12–31.
Dominguez, J. M. L. 1993. As coberturas do Cráton do Francisco: uma
abordagem do ponto de vista de análise de bacias. In: Dominguez,
J. M. L. & Misi, A. (eds) O Cráton do São Francisco. Sociedade
Brasileira de Geologia, Núcleo Bahia-Sergipe, 137– 159.
508 J. T. GUIMARÃES ET AL.
Guimarães, J. T. 1996. A Formação Bebedouro no Estado da Bahia;
faciologia, estratigrafia e ambientes de sedimentação. Dissertação
(Mestrado em Geologia), Instituto de Geociências, Universidade
Federal da Bahia (UFBA), Brasil.
Guimarães, J. T., Martins, A. A. M. et al. 2005. Projeto Ibitiara-Rio de
Contas, Ba. CPRM.
Halverson, G. P., Hoffman, P. R., Schrag, D. P., Maloof, A. C. &
Rice, A. H. N. 2005. Toward a Neoproterozoic composite carbon-
isotope record. GSA Bulletin, 117, 1181–1207.
Karfunkel, J. & Hoppe, A. 1988. Late Proterozoic glaciation in
Central-Eastern Brazil: synthesis and model. Palaeogeography,
Palaeoclimatology, Palaeoecology, 65, 1 –21.
Kaufman, A. J., Hayes, J. M., Knoll, A. H. & Germs, G. J. B. 1991.
Isotopic composition of carbonates and organic carbon from upper
Proterozoic successions in Namibia: stratigraphic variations and the
effects of diagenesis and metamorphism. Precambrian Research,
49, 301– 327.
Kegel, W. 1959. Estudos Geológicos na Zona Central da Bahia.
Divisão de Geologia e Mineralogia, Departamento Nacional da
Produção Mineral (DGM-DNPM), Boletim 198, Rio de Janeiro.
Knoll, A. H., Hayes, J. M., Kaufman, A. J., Swett, K. & Lambert, L.
B. 1986. Secular variation in carbon isotope ratios from upper Proter-
ozoic successions of Svalbard and East Greenland. Nature, 321,
832– 838.
Macedo, M. H. F. 1982. Les sistémes isotopiques rubidium-strontium et
potassium-argon dans lês argiles extraités de sediments carbonatés.
Application à la datation du protérozoique sedimentary du Brésil
dans les Etáts de Bahia et Santa Catarina. Thése presentée a
l’Université L. Pasteur pour obtenir le titre de Docteur-Ingénieur,
Strasbourg, France, 119.
Macedo, M. H. F. & Bonhomme, M. G. 1984. Contribuição à cronoestra-
tigrafia das formações Caboclo, Bebedouro e Salitre na Chapada
Diamantina (BA) pelos métodos Rb –Sr e K –Ar. Revista Brasileira
de Geociências, 153– 163.
Mello, J. L. Jr. 1938. Geologia e Hidrologia do Noroeste da Bahia.
Serviço Geológico e Mineralógico (SGM), Boletim 90, Rio de
Janeiro.
Misi, A. 1978. Ciclos de sedimentaupoçao e mineralċzçŏes de Pb –Zn nas
sequências Bambui (Supergrupo São Francisco), Estado da Bahia.
XXX Congresso Brasileriro de Geologia, Sociedae Brasileira de Geo-
logia, Anais, 4, 2548–2561.
Misi, A. 1979. O Grupo Bambuı́ no Estado da Bahia. In: Inda, H. A. V.
(ed.) Geologia e Recursos Minerais do Estado da Bahia. Textos
Básicos. Salvador, SME/COM, 1, 120– 154.
Misi, A. & Souto, P. G. 1975. Controle estratigráfico das mineralizações
de Pb –Zn –F –Ba no Grupo Bambuı́, parte leste da Chapada de Irecê,
BA. Revista Brasileira de Geociências, 5, 30 –45.
Misi, A. & Kyle, J. R. 1994. Upper Proterozoic carbonate stratig-
raphy, diagenesis and stromatolitic phosphorite formation, Irecê
Basin Bahia, Brazil. Journal of Sedimentary Research, 64, 299– 310.
Misi, A. & Veizer, J. 1998. Neoproterozoic carbonate sequences of the
Una Group, Irecê Basin, Brazil: Chemostratigraphy, age and corre-
lations. Precambrian Research, 89, 87– 100.
Misi, A., Iyer, S. S. S. et al. 2005. Sediment-hosted lead-zinc
deposits of the Neoproterozoic Bambui Group and correlative
sequences, São Francisco Craton, Brazil: a review and a
possible metallogenic evolution model. Ore Geology Reviews, 26,
263– 304.
Misi, A., Kaufman, A. J. et al. 2007. Chemostratigraphic correlation of
Neoproterozoic successions in South America. Chemical Geology,
237, 143– 167.
Montes, A. S. L. 1977. O contexto estratigráfico e sedimentológico da
Formação Bebedouro na Bahia. Brası́lia. Tese (Mestrado) Universi-
dade de Brası́lia (UNB).
Montes, A. S. L., Gravenor, C. P. & Montes, M. L. 1985. Glacial sedi-
mentation in the Late Precambrian Bebedouro Formation, Bahia,
Brazil. Sedimentary Geology, 44, 349–358.
Moraes Rego, L. F. 1930. Glaciação eopaleozóica no centro do Brasil.
Anais da Academia Brasileira de Ciências, 2, 109– 112.
Oliveira, A. I. & Leonardos, O. H. 1940. Geologia do Brasil. Comissão
Brasileira dos Centenários de Portugal.
Powis, K. 2006. Stable isotope geochemistry of the Neoproterozoic
Bambuı́ Group at Serra do Ramalho, Bajia, Brazil. PhD
thesis, Ottawa-Carleton Geoscience Center, University of Ottawa,
Canada.
Rocha Campos, A. C. & Hasui, Y. 1981. The Late Precambrian Bebe-
douro Formation, Bahia, Brazil. In: Hambrey, M. J. & Harland,
W. B. (eds) Earth’s Pre-Pleistocene Glacial Record. Cambridge
University Press, Cambridge. (IGCP Project, 38: Pre-Pleistocene
Tillites.)
Shields, G. 2005. Neoproterozoic cap carbonates: a critical appraisal of
existing models and the plumeworld hypothesis. Terra Nova, 17,
299– 310.
Söfner, B. 1973. Observações sobre a estratigrafia do Pré-Cambriano
da Chapada Diamantina Sudeste e da área contı́gua. In: Sociedade
Brasiliera de Geologia, Congresso Brasiliera, 27, Aracajú, Anais,
1, 23 – 33.
Teixeira, J. B. G., Misi, A. & Silva, M. G. 2007. Supercontinent evol-
ution and the Proterozoic metallogeny of South America. Gondwana
Research, 11, 346– 361.
Torquato, J. R. F. & Misi, A. 1977. Medidas isotópicas de carbono e
oxigênio em carbonatos do Grupo Bambuı́ na região centro-norte
do Estado da Bahia. Revista Brasileira de Geociências, 7, 14 –24.
Trindade, R. I. F., D’agrella Filho, M. S., Babinski, M., Font, E. &
Neves, B. B. B. 2004. Paleomagnetism and geochronology of the
Bebedouro cap carbonate: evidence for continental-scale Cambrian
remagnetization in the São Francisco craton, Brazil. Precambrian
Research, 128, 83– 103.
Uhlein, A., Alvarenga, C. J. S., Dardenne, M. A. & Trompette, R. R.
2011. The glaciogenic Jequitaı́ Formation, southeastern Brazil.
In: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The
Geological Record of Neoproterozoic Glaciations. Geological
Society, London, Memoirs, 36, 541–546.
Vieira, L. C., Trindade, R. I. F., Nogueira, A. C. R. & Ader, M. 2007.
Identification of a Sturtian cap carbonate in the Neoproterozoic Sete
Lagoas carbonate platform, Bambui Group, Brazil. Comptes Rendus
Geoscience, 339, 240–258.
Williams, H. E. 1930. Estudos geológicos na Chapada Diamantina,
Estado da Bahia. Boletim do Serviço Geológico e Mineralógico,
Brasil, 44.
 Chapter 48

Neoproterozoic successions of the São Francisco Craton, Brazil: the Bambuı́, Una,
Vazante and Vaza Barris/Miaba groups and their glaciogenic deposits

AROLDO MISI1*, ALAN J. KAUFMAN2, KAREM AZMY3, MARCEL AUGUSTE DARDENNE4,

ALCIDES NÓBREGA SIAL5 & TOLENTINO FLÁVIO DE OLIVEIRA6
1
Metallogenesis Group, Department of Geology and Research Center on Geophysics and Geology (CPGG),
Federal University of Bahia, Instituto de Geociências, Campus da Federação 40170-290, Salvador, Bahia, Brazil
2
Departments of Geology and ESSIC, University of Maryland, College Park, Maryland, USA
3
Department of Geology, Memorial University of Newfoundland, St. John’s, Canada
4
Institute of Geosciences, University of Brasilia, Brasilia (DF), Brazil
5
NEG-LABISE, Department of Geology, Federal University of Pernambuco, Recife (PE), 50670-000, Brazil
6
Votorantim Metais, P.O. Box 03, 38780-000, Vazante, MG, Brazil
*Corresponding author (e-mail: aroldo.misi@gmail.com)

Abstract: The Neoproterozoic successions of the São Francisco Craton are primarily represented by the Bambuı́ and Una groups,
deposited in cratonic epicontinental basins, and by the Vazante and Vaza Barris/Miaba groups, which accumulated on passive
margins on the edges of the craton. The epicontinental basins comprise three megasequences: glaciogenic, carbonate platform
(marine) and dominantly continental siliciclastics. Possible correlative sequences are observed in the passive margin deposits.
At least two major transgressive–regressive sea-level cycles occurred during the evolution of the carbonate megasequence, which
lies above glaciomarine diamictites of probable early Cryogenian (i.e. Sturtian) age. C, O, Sr and S isotope trends from analyses of
well-preserved samples, together with lithostratigraphic observations, provide reasonable correlations for most of the Neoproterozoic suc-
cessions of the São Francisco Craton. The 87Sr/86Sr record of these successions, ranging from 0.70769 to 0.70780, supports the proposed
correlation with the Bambuı́, Una and Vaza/Barris successions, and with the basal units of the Vazante Group. In addition, C-isotope posi-
tive excursions ranging from þ8.7 to þ14‰ and negative excursions from –5.7 to – 7‰ VPDB in the Bambuı́, Una and Vaza-Barris suc-
cessions provide key markers for correlations. The precise ages of the sedimentation in these successions remains a matter of debate, but
organic shales of two units of the Vazante Group have been dated by Re–Os techniques in two different laboratories, both yielding
Mesoproterozoic ages. The Neoproterozoic and Mesoproterozoic successions preserve significant glaciogenic deposits.

The global occurrences of low-latitude Neoproterozoic glacial
sediments and the texturally and isotopically anomalous post-
glacial carbonates deposited during potential ‘Snowball Earth’
events (Hoffman et al. 1998a; Hyde et al. 2000; Hoffman &
Schrag 2002) provide stratigraphic markers that have been used
for inter-continental correlations. However, the timing and
duration of the Neoproterozoic ice ages are in most cases poorly
constrained, due to the lack of absolute ages from appropriate
lithologies. This is the case for the carbonate-dominated
Vazante, Bambuı́ and Una groups (cf. Azmy et al. 2001, 2006;
Misi et al. 2007) on the São Francisco Craton in east central
Brazil (Fig. 48.1), although new chronometric techniques are
revealing surprising results (cf. Babinski et al. 2007).
In the absence of radiometric or biostratigraphic tie-lines in
these Neoproterozoic successions, time-series trends in stable
isotope signatures have been used as long-distance correlation
tools for the cap carbonates and other carbonate lithofacies
throughout the marine sequences (e.g. Knoll et al. 1986; Derry
et al. 1989; Fairchild et al. 1990; Kaufman et al. 1991; Kaufman
& Knoll 1995; Jacobson & Kaufman 1999; Brasier & Shields
2000; Azmy et al. 2001; Cozzi et al. 2004, Halverson et al.
2005; Azmy et al. 2006). However, a priori assumptions of the
ages of some of the Neoproterozoic glacial deposits (i.e. Sturtian
v. Marinoan) in the absence of direct radiometric constraints
(Hoffman & Schrag 2002), and the likelihood of depositional
(and diagenetic) variations in C- and Sr-isotope compositions
complicate the use of chemical stratigraphy for basin-to-basin
comparisons. In addition, correlations need to take into account
the possibility of structural complications due to the tectonic
setting during and after sedimentation.
With these caveats in mind, aspects of the litho- and
sequence-stratigraphy, chemostratigraphy and radiometric ages
of the most representative glaciogenic and carbonate-rich succes-
sions of the São Francisco Craton, and their possible global corre-
lation, are discussed in this chapter.
Structural and geotectonic framework
Most of the glaciogenic successions of the São Francisco Craton
were likely deposited during extensional events related to the frag-
mentation of the Rodinia supercontinent starting around 950 Ma
and extending until c. 600 Ma (Condie 2002). These extensional
events were episodic during the closure of the Pan-African-Brasi-
liano rift (Porada 1989; Brito-Neves et al. 1999; Condie 2002;
Cordani et al. 2003). A relationship between Neoproterozoic gla-
ciation and extensional tectonics has been previously postulated
(e.g. Young 1995).
These sedimentary successions deposited on the São Francisco
Craton were distributed in the following geotectonic settings
(Misi et al. 2007) (Fig. 48.1):
† mixed carbonate and siliciclastic strata deposited on tectoni-
cally stable cratons, including the Bambuı́ and Una groups in
the São Francisco Basin and in the Irecê, Campinas, Una-
Utinga and Ituaçú sub-basins, respectively;
† intensely deformed mixed carbonate and siliciclastic strata in
passive-margin basins surrounding the stable cratons, including
part of the Vazante group (Brası́lia Fold Belt) and the Vaza
Barris/Miaba group (Sergipano Fold Belt), among others sur-
rounding the cratonic area.

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 509– 522. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.48
510 A. MISI ET AL.
Some authors (Chang et al. 1988; Thomaz Filho et al. 1998;
Dardenne 2001; Pimentel et al. 2001) have postulated that the Neo-
proterozoic sediments on the São Francisco Craton accumulated in
foreland basins during compressive events related to the Brasi-
liano–Pan African orogeny. On the other hand, D’el Rey Silva
(1999) demonstrated that sedimentation of the Vaza Baris/Miaba
domains in the Sergipano Belt (NE Brazil) occurred on a passive
continental margin. The similarity of cyclic sedimentation in the
marginal and epicontinental marine basins suggests that sediments
originally spanned the craton. Extensional tectonism associated
with the glacial deposits ‘is indicated by the presence of large poly-
lithic clasts of both basement and cover in the diamictites . . . point-
ing to uplift-erosion of footwall and possibly hanging wall blocks
twice in the evolution of the basin’ (Dardenne 1978a, 1979, 1981;
D’el Rey Silva 1999, p. 463). Alkmim (1996) and Misi (1999)
demonstrated the existence of widespread aligned structures with
orientation 108N–208W intersecting both the older Proterozoic
basement and the Neoproterozoic sedimentary cover (Fig. 48.1).
Stratigraphy
The Neoproterozoic successions of the São Francisco Craton are
part of the São Francisco Supergroup. They may be subdivided
Fig. 48.1. The São Francisco Craton in
NE Brazil with the Neoproterozoic cover
(São Francisco Supergroup) and the
folded belts surrounding the
cratonic area.
into three megasequences (glaciogenic, carbonate platform and
dominantly continental siliciclastic units) separated by erosional
unconformities (Misi 2001; Misi et al. 2007). Each megasequence
may have other smaller sequence boundaries that, when associated
with high-resolution chemostratigraphic data, may provide key
event markers for stratigraphic correlations. These megasequences
are briefly described below (Fig. 48.2).
Glaciogenic megasequence
Siliciclastic rocks at the base of the successions are dominated
by conglomerate, metagreywacke, diamictite, pelite and quartzite.
These deposits are interpreted as having a continental glacial to
glacial –marine origin. Striated pavements and dropstones as
well as faceted and striated clasts in most of the lithofacies have
been described (Söfner 1973; Karfunkel & Hoppe 1988; Montes
1977; Guimarães 1996; Uhlein et al. 2004). These lithofacies
unconformably overlie the Palaeoproterozoic/Archaean basement
complex and the Mesoproterozoic Espinhaço Supergroup. They
constitute the Jequitaı́ Fm. in the Bambuı́ Group (states of Minas
Gerais, Bahia and Goiás), and the Bebedouro Fm. in the Una
Group (state of Bahia). In the passive margin basins, glacial dia-
mictite is present in the Panelinha Fm. (Rio Pardo Group) and
 NEOPROTEROZOIC SUCCESSIONS OF THE SÃO FRANCISCO CRATON 511

Fig. 48.2. Megasequences of the

Neoproterozoic successions of the São
Francisco Supergroup (modified from Misi
et al. 2007).

the Macaúbas Group, both of the Araçuaı́ Fold Belt, the Palestina
Fm. in the Vaza Barris/Miaba Group of the Sergipano Fold Belt,
and the Santo Antonio do Bonito Fm. in the Vazante Group of
the Brasilia Fold Belt (Fig. 48.2).
In addition, Brody et al. (2004), Olcott et al. (2005) and Azmy
et al. (2006) document the presence of abundant dropstones in the
black shale of the Serra do Poço Verde and Lapa formations higher
up in the Vazante Group. The Serra do Poço Verde shale is sand-
wiched between carbonate breccia, which may also be glacial in
origin (Olcott et al. 2005), and contains glendonite, a carbonate
mineral after ikaite that only forms in sediments at frigid tempera-
tures. Based on recent geochronological data (Re – Os in black
shales, Azmy et al. 2008; Geboy et al. 2009), this possible glacio-
genic event is considered by some authors as Mesoproterozoic in
age, as will be discussed later.
(northern Bahia) and by part of the Vazante Group (western
Minas Gerais), include possibly equivalent strata, although the tec-
tonic setting differs in each of these regions. The stratigraphic
thicknesses in these marginal basins are greater than in the intra-
cratonic basins, and parts were intensely deformed during the
Brasiliano – Pan African orogeny.
At least two transgressive –regressive cycles are clearly rep-
resented in the carbonate platform sediments deposited in the intra-
cratonic basins (Dardenne 1978b, 1979, 2000; Misi 1978; Misi
et al. 2007). The first cycle ends with an extensive sub-aerial
exposure surface associated with tepee structures, replaced nodular
anhydrite, dissolution breccias, and laminated and columnar stro-
matolites (Fig. 48.2).

Table 48.1. Correlations between the Bambuı́ formations and the Una
Carbonate platform megasequence (marine) Group units

The carbonate platform facies are composed of limestone, dolo-

Group
stone, marl, shale and siltstone. Internal to the São Francisco Lithotypes BAMBUÍ UNA ,
craton, the Bambuı́ Group (Dardenne 1978b, 1979, 2000) is subdi- ,
(Formations)* † (Units) ( Form.)‡ §
vided into the following formations: Sete Lagoas (dolarenite with
teepee structures at the top, laminated limestones and red argillac- Arkose, siltstone Três Marias –
eous dolostone at the base), Serra de Santa Helena (marl and inter-
calated limestone), Lagoa do Jacaré (black, organic-rich oolitic Siltstone, pelite and intercalated limestone Serra da Saudade –

limestone) and Serra da Saudade (siltstone, pelites and intercalated

limestone). In the Irecê and Una-Utinga basins of the Una Group, Black organic -rich calcarenite and calcilutite Lagoa do Jacaré Unit A1
with oolitic and pisolitic limestones with
the dominant carbonate succession has been subdivided into five
Salitre Formation

interbedded pelite and marl

informal units that correlate with those of the Bambuı́ Group
(Misi & Souto 1975; Misi 1978; Misi & Veizer 1998): Unit C Grey marl, pelite and siltstone with black Serra de Santa Unit A
limestones Helena
(base of Sete Lagoas Fm.), Unit B (mid section of Sete Lagoas
Fm.), Unit B1 (top of Sete Lagoas Fm.), Unit A (Serra de Sta. Clear grey dolarenite w/stromatolites and Sete Lagoas 3 Unit B1
tepee structures (Upper)
Helena Fm.) and Unit A1 (Lagoa do Jacaré Fm.) See Table 48.1 Sete Lagoas 2 Unit B
Laminated limestone and shale (Middle)
for a brief description of these facies.
Red argillaceous dolostone (Lower) Sete Lagoas 1 Unit C
The passive margin successions bordering the craton, which are
represented by the Vaza Barris/Miaba Group (NE of the São
Diamictite, arkose, pelitic rocks Jequitaí Bebedouro
Francisco Craton, in Bahia and Sergipe states), the Rio Pardo
*Dardenne (1978), †Misi (1978), ‡Misi & Souto (1975) and §Branner (1911).
Group (Bahia), the Rio Preto and Riacho do Pontal groups
512 A. MISI ET AL.
Dominant continental siliciclastic megasequence (molasses)
Continental siliciclastic rocks composed of arkose, siltstone, phyl-
lite and conglomerate constitute the Três Marias Fm. of the
Bambuı́ Group. The origin of these deposits is related to post-
tectonic molassic sedimentation along the borders of the cratonic
area (Brito Neves & Cordani 1991). Whereas Brito Neves &
Cordani (1991) claimed that the contact between the Três Marias
Fm. and the Bambuı́ Group is an erosional unconformity, others
have postulated a gradational contact between the Três Marias
Fm. and the Serra da Saudade Fm. (Dardenne 1978b, 1979,
2000; Martins-Neto, pers. comm; cf. Misi 2001).
Characteristics and distribution of the glaciogenic
deposits and associated strata
The Neoproterozoic glaciogenic deposits of the São Francico Craton
are widespread and found in the basal sections of the epicontinental
basins of the Bambuı́ Group (São Francisco Basin; see also Uhlein
et al. 2011) and the Una Group (Irecê, Una-Utinga and Campinas
sub-basins; see also Guimarães et al. 2011) and in the basal and
middle parts of the passive margin basin of the Vaza Barris/
Miaba Group (Sergipano Fold Belt). The glaciogenic deposits of
the Vazante Group (Brasilia Fold Belt) occur in the basal and
upper parts of the sequence, but their ages are still a matter of
debate, as will be discussed in subsequent sections. Figure 48.1
shows the distribution of these basins in the studied area.
Bambuı́ and Una Groups
The Bambuı́ Group in the São Francisco Basin and the Una Group
in the Irecê, Una-Utinga and Campinas sub-basins (Fig. 48.1) were
deposited in epicontinental, shallow-marine environments.
Remarkable lithostratigraphic and chemostratigraphic similarities
support detailed correlation between these isolated sub-basins,
despite their present discontinuity.
Deposition in this likely once contiguous basin began with
dominantly glacial and glaciomarine sedimentation in the Jequitaı́
Fm. (Bambuı́ Group) and the Bebedouro Fm. (Una Group), as
represented by glacial diamictites, which contain faceted and
striated clasts, as well as dropstones in finely laminated pelites.
In some regions, this diamictite contains basement clasts in a
fine-grained Fe-rich mudstone, and in the Sete Lagoas region,
it is overlain by a carbonate lithofacies containing centimetre-
scale sea-floor precipitates. Discontinuous aeolian sandstone with
variable thickness occurs above the Bebedouro Fm. diamictite
(Guimarães 1996). These units were deposited unconformably
over the Palaeoproterozoic and Archaean basement, as well as
the Mesoproterozoic sedimentary rocks of the Espinhaço Super-
group (Fig. 48.1).
The carbonate platform marine sedimentation that follows was
deposited unconformably on the glaciogenic and glaciomarine
lithotypes. The lithostratigraphic units from the top to bottom
include formations with similar characteristics in the Bambuı́
and Una groups, as noted by Dardenne (1978b, 1979, 2000) and
Misi et al. (2007) (Table 48.1).
The thickness of the glaciogenic megasequence ranges from 0 to
150 m, while the thickness of the carbonate-rich interval is variable,
probably controlled by a basement fault system active during sedi-
mentation (Dardenne 1978, 1979, 2000; Misi et al. 2007). Seismic
surveys from the central part of the São Francisco Basin indicate
that the carbonate platform is up to 1000 m thick (Teixeira et al.
1993). In contrast, it reaches only 600 m in thickness in the Sete
Lagoas area (Pedrosa Soares et al. 1994), 400 m in the Serra do
Ramalho and Januaria-Itacarambi areas (Dardenne 1978b, 1979,
2000; Misi 1979), and 600 m in the Irecê Basin (Misi 1979).
Vazante Group
The Vazante Group (Dardenne 2000, 2001) consists mainly of stro-
matolitic carbonate muds and rhythmites deposited on a shallow
marine platform (Dardenne 2001) that extended more than
300 km north–south along the western margin of the craton. The
succession now lies in the external zone of the Brası́lia Fold Belt
(Fig. 48.3) in the São Francisco Basin (SFB). The stratigraphy of
the marginal marine sediments of the Vazante Group (Fig. 48.4)
has been studied in detail and refined by several authors (e.g. Dard-
enne 1978, 1979, 2000, 2001; Madalosso 1979; Karfunkel & Hoppe
1988; Fairchild et al. 1996; Azmy et al. 2001; Misi 2001; Misi et al.
2007). The Vazante sequence, from bottom to top, includes the
Santo Antonio do Bonito, Rocinha, Lagamar, Serra do Garrote,
Serra do Poço Verde, Serra do Calcario and Serra da Lapa-
Velosinho formations (Fig. 48.4). In the eastern part of the basin,
carbonate, diamictite and shale of the Vazante Group are generally
well preserved and little metamorphosed. To the west near the Bra-
silia Fold Belt, however, the sediments are highly deformed and
have experienced amphibolite to granulite facies metamorphism
(Dardenne 1978a; Fuck et al. 1994). Earlier studies indicated that
the Vazante Group sediments accumulated in a passive margin
setting (e.g. Dardenne 1979; Campos Neto 1984a, b; Fuck et al.
1994), which is consistent with the lack of volcanic input to the
sequence. The northern and southern segments were separated by
a regional WSW–ESE lineament situated on the same latitude of
the Federal District (DF, Fig. 48.1). They show distinct geotectonic
characteristics: intensely deformed in the southern segments with
tectonic transport of great magnitude (from west to east), while in
the northern segment the sedimentary units are less deformed and
the stratigraphic organization remains well preserved, probably
due to the location of this segment at an upper crustal level of
the granite-gneiss basement, ‘which acted as a rigid block in front
of the compressive trend of the Brasiliano Cycle’ (Dardenne
2000, p. 234).
Recent geochronological studies suggest that at least part of
what is referred to as the Vazante Group may predate the Bambuı́
Group (Geboy 2006; Rodrigues 2007; Azmy et al. 2008; Rodri-
gues et al. 2008).
The Vazante Group rests on a glaciogenic unit (D), which con-
stitutes the uppermost part of the Santo Antonio do Bonito Fm.
(Fig. 48.4). This diamictite contains faceted and striated cobbles
and is overlain by the Cubatão Fm., a thick stromatolitic dolostone.
New observations of dropstones in organic-rich shale sand-
wiched between previously recognized dolomite breccia in the
Serra do Poço Verde Fm. suggest the presence of a second glacial
horizon (D II) near the top of the succession (Olcott et al. 2005;
Azmy et al. 2006). The carbonate diamictites were found in
exploration drill holes in a known geographic extension of over
150 km in an approximate north–south direction, between the
cities of Unai and Lagamar. In most of the area they are associated
with black shales and debris flows with fist-sized dropstones. The
preserved thickness of this unit ranges from 20 m to .100 m, and
dolostone clasts vary from from millimetres to several metres
in diameter.
This unit can be broken down into two distinct levels: one in the
upper Morro do Calcário Formation, frequently in unconformable
contact with the Lapa Fm., and a lower one in the Lower Poço
Verde Fm. (DNPM hole 1-PSB01 from 275.50 to 304.50 m).
Thin section observations revealed the presence of the unusual
cold-water carbonate ‘glendonite’ in some of the shale samples
(Olcott et al. 2005), which may support the syn-glacial interpret-
ation of this unit. A sharp drop in d13C values is recorded in
bedded dolomite above the shale horizon in one core (Azmy
et al. 2001), and possible cap-carbonate lithofacies (red dolo-
stones) have been recognized in field exposures near Vazante
mine. A regional unconformity separating the Morro do Calcário
Fm. from the overlying Lapa Fm. occurs at the top of the upper
diamictite (D II) (Misi et al. 2005).
 NEOPROTEROZOIC SUCCESSIONS OF THE SÃO FRANCISCO CRATON
Coupled with a negative d13C excursion in overlying carbonate
and marl, Brody et al. (2004) and Azmy et al. (2006) interpret basal
Lapa sediments as a post-glacial cap-carbonate lithofacies. Alter-
natively, the Morro do Calcário interval of carbonate breccia has
been interpreted as debris-flow deposit along the western flank
of a stromatolitic platform, and thus may be intra-formational in
origin and not directly related to glaciation (Dardenne 1979,
2000, 2006, 2007).
Vaza Barris/Miaba Group
The Sergipano belt was divided by D’el Rey Silva (1999) into three
tectonic domains: the cratonic Estância domain, the external Vaza
Barris domain and the intermediate Macururé domain. These
domains form a continuous wedge of craton, platform and basinal
deposits developed along a continental margin (Fig. 48.5). The
intermediate Macururé domain (max. thickness, 13 km) is com-
posed of carbonate and siliciclastic sedimentary rocks with associ-
ated metavolcanics. The Vaza Barris domain (1– 4 km thick)
includes marine platform siliciclastic and carbonate sediments
with minor volcanic horizons, comprising the Miaba, Simão
Dias and Vaza Barris groups. The Estancia domain (1 –3 km
thick) is composed of continental to shallow marine sediments.
D’el Rey Silva (1999) described two sedimentary cycles in these
domains, each including two megasequences: Cycle 1 is
513
Fig. 48.3. Simplified regional geological
map of the Brasilia Belt showing the
distribution of the Vazante Group, based on
Marini et al. (1984), Dardenne (2000),
Pimentel et al. (2001) and Valeriano et al.
(2004).
represented by a lower siliciclastic megasequence (Juetê, Itabaiana
and Ribeirópolis formations) and an upper carbonate megase-
quence (Acauã and Jacoca formations). According to D’el
Rey Silva (1999), ‘the lateral correlation between the formations
of each megasequence in Cycle 1 is constrained by their same
stratigraphic position, by the strong similarity of the rock types,
and by the fact that the Simão Dias Group forms a blanket of sili-
ciclastics that is continuous across the craton-fold belt boundary’
(Fig. 48.6). Cycle 2 is similarly composed of a lower siliciclastic
interval, including the Simão Dias Group and the Palestina Fm.
and an upper carbonate interval, represented by the Olhos
d’Água Fm.
Carbonates of the Jacoca Fm., at the basal section of the Vaza
Barris/Miaba Group, are in sharp contact with the Ribeiropolis
diamictite below. This unit is 0– 300 m thick and locally reaches
thicknesses of c. 500 m. It consists of diamictite with dropstones,
metagreywacke and quartz-sericite phyllite overlying quartzites
of the Itabaiana Fm. D’el Rey Silva (1995, 1999) has reported
volcanics within this formation. A typical section of the Jacoca
Fm. starts with a thick layer of laminated grey to pink dolostone
containing pyrite and chalcopyrite, followed up-section by a 3–
15-m-thick layer of laminated dolostone, and dark grey to black
phyllites, followed by c. 10-m-thick massive dolostone (Sial
et al. 2009). This unit is overlain by a 40-m-thick heterolithic
sequence of grey limestones and dark to black phyllites and
finally by grey dolostones.
514 A. MISI ET AL.

(Bambuı́ Group) formations on the São Francisco craton

Formation
(Tables 48.1 & 48.2).

(m ne x.
Group

) ss
ic pro
Member

Th Ap
k
Chemostratigraphy

High-resolution chemostratigraphic studies have been conducted

Serra da Lapa Velosinho
Lapa

650 in the Irecê Sub-basin (Fig. 48.1), Una Group (Torquato & Misi
organic-rich shales 1977; Misi & Veizer 1998), in the São Francisco Basin, Serra do
D II Ramalho area (Powis 2006; Misi et al. 2007), in the Brasilia
Fold Belt, Vazante Group (Azmy et al. 2001, 2006) and in the Ser-
Morro do
Calcário

gipano Fold Belt (Sial et al. 2000, 2003, 2005, 2006a, b). Kawa-
Upper Pamplona shita (1998) analysed Sr, C and O isotopes in the carbonate
300
sections at a lower stratigraphic resolution. The following review
is mostly based on the recent synthesis by Misi et al. (2007).

Irecê Basin
Rocinha Lagamar Serra do Garrote Serra do Poço Verde

Middle Pamplona 400

One of the earliest isotopic studies of a post-glacial cap carbonate
was performed in the Irecê Basin (Una Group) by Torquato & Misi
Lower Pamplona 200 (1977), although at the time the unit was considered lacustrine in
V A Z A N T E

origin. These authors analysed 85 carbonate samples for C and

Upper Morro do Pinheiro 500 O isotopes through two complete sections of the Una Group. The
results reveal a d13C trend from consistently negative values in
Lower Morro do Pinheiro 500 the red argillaceous dolostone (Unit C, c. –6‰) immediately
above diamictite of the Bebedouro Fm. to values c. 0‰ up-section
(Units B and B1) and continuing to highly positive values (þ8‰,
Unit A1). The negative values in the dolostone overlying the Bebe-
douro Fm. diamictite are similar to post-glacial trends worldwide
and may support the glacial interpretation for these deposits.
>1000 This trend has been reproduced by Misi & Veizer (1998), who ana-
lysed well-preserved micro-samples that were first characterized
by petrographic and geochemical techniques. In addition, these
authors determined 87Sr/86Sr ratios on the best-preserved carbon-
ates. The least radiogenic 87Sr/86Sr value of 0.70780 from Unit B
Sumidouro is consistent with a sedimentation age between 750 and 600 Ma
250 (cf. Jacobsen & Kaufman 1999) (Fig. 48.7).
Arrependido
São Francisco Basin
1000
The carbonates of the Bambuı́ Group in the São Francisco Basin
have been the subject of many chemostratigraphic studies (e.g.
arkosic sandstone Iyer et al. 1995; Kiang 1997; Kawashita 1998; Santos et al.
St. Antonio
do Bonito

D 2000, 2004; Kaufman et al. 2001). The most detailed chemostrati-

250 graphic study developed in the Bambuı́ Group thus far comes from
high-resolution profiles from the Serra do Ramalho area, western
Bahia (Powis 2006; Misi et al. 2007). Chemostratigraphic
studies have also been conducted by Iyer et al. (1995), Kaufman
Fig. 48.4. Stratigraphic units of the Vazante Group. et al. (2001) and Vieira et al. (2007) in the Sete Lagoas area of
the São Francisco Basin. As in the Irecê Basin, the d13C compo-
sition of carbonate rocks in the Serra do Ramalho and in the Sete
The Olhos d’Água Fm. (200 – 1300 m) is composed of inter- Lagoas area increase up-section from c. –4‰ at the base of the
bedded limestones that become organic-rich towards the top. Sete Lagoas Fm. overlying glacial diamictite of the Jequitaı́ Fm.
The upper section consists of c. 50 m of limestone –pelite inter- (assumed to be correlative to the Bebedouro Fm.) and rising stea-
calations that are overlain by laminated limestones (40 m thick) dily up to c. þ14‰ near the top of the succession in Lagoa do
culminating with a 30-m-thick layer of organic-rich limestone. Jacaré Fm. The best-preserved samples (Mn/Sr  0.02) show a
The lower section of laminated limestones is in sharp contact narrow range of 87Sr/86Sr values between 0.7074 and 0.7075
with the diamictites of the Palestina Fm., which is composed of (Fig. 48.7).
metagreywackes with clasts up to 30 cm in diameter. The clasts
are of gneiss, granite, quartz, quartzite, phyllite and metacarbonate
supported in a quartz –sericite matrix. Lenses of Fe-cemented Brasilia Fold Belt (Vazante Group)
quartzite unconformably overlie the siliciclastic sediments of the
Simão Dias Group (Fig. 48.6). Petrographic investigations showed that some of the Vazante
The Palestina Fm. is interpreted as a glaciogenic deposit. At a carbonates were affected by dolomitization, but examination of
lower stratigraphic level, the equivalent Juetê and Ribeirópolis for- thin sections shows that these rocks have very fine-grained and
mations (D’el Rey Silva 1999) are also interpreted as glaciogenic fabric-retentive dolomicrites (Azmy et al. 2001, 2006). Three gen-
and may correlate with the Bebedouro (Una Group) and Jequitaı́ erations of secondary dolomitic cements have been observed,
 NEOPROTEROZOIC SUCCESSIONS OF THE SÃO FRANCISCO CRATON
including fibrous, equant and late fracture-filling phases (cf. Azmy
et al. 2001). No significant increase in crystal size was observed
associated with dolomitization, suggesting that the original sedi-
ment did not suffer from extensive and/or repeated meteoric
alteration.
The trace element compositions of the Vazante carbonates,
particularly those of the fabric retentive micrites, suggest signifi-
cant geochemical preservation (Azmy et al. 2001, 2006). The
C-isotope profiles reveal a negative shift up to 8‰ in dolomite
beds above the dropstone-laden shale of the Serra do Poço
Verde Fm. (Fig. 48.8) and near the base of the Lapa Fm.
(cf. Azmy et al. 2006; Misi et al. 2007). The least radiogenic
Sr-isotope values associated with the negative d13C shifts are
0.7061 in the Serra do Poço Verde Fm. and 0.7068 in the Lapa
Fm. (Fig. 48.8).
515
Fig. 48.5. Simplified geological map of
the Sergipano Belt, modified from D’el
Rey Silva (1999). Basement domes:
1, Itabaiana; 2, Simão Dias; 3, Jirau do
Ponciano. Main thrust and strike–slip
fault: MF, Macururé fault; BMJF, Belo
Monte-Jeremoabo fault; SMAF, São
Miguel do Aleixo fault; IF, Itaporangã
fault.
Sergipano Fold Belt (Vaza Barris Group)
Chemostratigraphic studies have been performed on some carbon-
ate units of the Vaza Barris Group by Kawashita (1998) and by Sial
et al. (2000, 2003, 2005, 2006a, b). The first author determined
least-altered 87Sr/86Sr ratios of 0.7078 in black limestones of the
Olhos d’Água Fm. In addition Sial et al. (2000) report d13C low
values of – 4.7‰ immediately above diamictite of the Palestina
Fm. These numbers increase up-section to values as high as
þ10‰ in the upper Olhos d’Água Fm. Sial et al. (2006a) also
report negative excursions as low as – 5% in the lower carbonates
of the Acauã Fm. (equivalent to the Jacoca Fm.; Fig. 48.9) immedi-
ately above the diamictites of the Ribeirópolis Fm., suggesting that
at least two important glaciogenic events are represented in the
Vaza Barris Group. The results support a possible correlation of
Fig. 48.6. Summary stratigraphy of the
Itabaiana dome area, based on D’el Rey
Silva (1999). Megasequences: LSM, lower
siliciclastic; LCM, lower carbonate
(megasequences of Cycle I); USM, upper
siliciclastic; UCM, upper carbonate
(megasequences of Cycle II). Formations:
JU, Juetê A, Acauã; I, Itabaiana; R,
Ribeirópolis; JC, Jacoca; LP, Laegarto-
Palmares; Ja, Jacaré; FP, Frei Paulo; P,
Palestina; OA, Olthos d’Água Formation.
516 A. MISI ET AL.

Table 48.2. Correlations between intracratonic and passive-margin basins Geochronological constraints
Brasiliano Fold Belt Intracratonic Basins Sergipano Fold Belt
Passive-margin Basin Passive-margin Basin Despite the absence of reliable absolute geochronological age
Vazante group Bambuí group Una group Vaza barrís/miaba gr.
estimates in the platform carbonate successions of the São Fran-
(Formations) (Formations) (Units) (Form.) (Formations) cisco craton, there is a general agreement in considering a
Três Marias – –
middle Cryogenian age for the Bambuı́ and Una groups. The
Mesoproterozoic Units

Serra da Lapa
least radiogenic Sr isotope values of 0.7074– 0.7077 and S (from
(overthrusted terrain)

Morro do Calcário Serra sa Saudade – – stratform barite), with a mean value of c. þ30 per mil CDT in
Serra do Poço Verde Lagoa do Jacaré A1 the Una Group, indicate that sedimentation likely occurred

Salitre Formation
Olhos d’Água
Serra de Santa Helena A between c. 780 Ma and 600 Ma (Misi & Veizer 1998; Misi et al.
Serra da Garrote
Sete Lagoas 3 B1 Palestina 1999). More recently, Babinski & Kaufman (2003) and Babinski
Simão Dias Group
Lagamar
Sete Lagoas 2 B et al. (2007) determined new Pb/Pb carbonate ages from well-
Jacoca
Rocinha Sete Lagoas 1 C preserved post-glacial sea-floor cements in the Sete Lagoas
Santo Antônio do Bonito Jequitaí Bebedouro Ribeirópolis Fm. of the Bambuı́ Group, suggest a depositional age of
740 + 22 Ma (MSDW ¼ 0.66). Notably, almost the same age
(730 Ma, U –Pb in zircon) has been recorded by Brito-Neves
(pers comm.) in acidic tuffs intercalated in the Ribeirópolis
the Vaza Barris Group with the carbonate platform successions of diamictites of the Vaza Barris/Miaba Group, in the Sergipano
the Bambuı́ and Una Groups, as sustained by some authors (e.g. Belt. According to D’el Rey Silva (1999), sedimentation in the Ser-
Misi 1979; Teixeira & Figueiredo 1991; Trompette 1994; D’el gipano Belt lasted from c. 1.0 Ga (U –Pb zircon in metavolcanics,
Rey Silva 1999). The diamictites of the Ribeirópolis Fm. may be Brito Neves et al. 1993) to possibly 0.65 Ga.
equivalents of the Bebedouro and the Jequitaı́ formations (respect- On the other hand, the global stratigraphic correlation of the
ively, from the Una and Bambuı́ groups), but no equivalent glacial Vazante Group in the São Francisco Basin has been a challenging
deposits correlating with the Palestina Fm. have been clearly issue. The resetting of the radiogenic clock by the major
demonstrated from these epicontinental successions (Fig. 48.9 & Brası́lian –Pan African orogeny (c. 550– 600 Ma) (Trindade
Table 48.2). et al. 2004) and the scarcity of volcanic ash layers makes absolute

Fig. 48.7. Proposed lithostratigraphic and

chemostratigraphic correlation between the
Bambuı́ and Una Groups, respectively in the
São Francisco Basin (Serra do Ramalho
area) and in the Irecê Basin (Misi & Veizer
1998; Powis 2006; Misi et al. 2007).
 NEOPROTEROZOIC SUCCESSIONS OF THE SÃO FRANCISCO CRATON
Fig. 48.8. Chemostratigraphy of the Vazante Group.
geochronological age control of this succession problematic. The
abundance of Conophyton metula Kirichenko in the Lagamar
Fm. (Fig. 48.4) suggests an age estimate between 1350 Ma and
950 Ma for the sedimentation of this part of the Vazante Group
(Cloud & Dardenne 1973). Nevertheless, Misi et al. (1997) and
Sanches et al. (2007) analysed carbonate fluorapatite and fine-
grained organic-rich limestone (micrite) from the lower Rocinha
Formation for their Sr-isotope composition. 87Sr/86Sr values
range from 0.70760 to 0.70790, similar to those reported for the
Bambuı́, Una and Vaza Barris/Miaba Groups. All seven samples
had high total Sr (.1300 ppm) and low Mn– Sr (0.1 –0.01), indi-
cating a high degree of preservation of the original seawater signal.
Sm –Nd studies of provenance (Pimentel et al. 2001) on fine-
grained sediments of the Bambuı́, Vazante and Paranoá groups
have furnished TDM model age intervals of 1.9 –1.4 Ga, 2.3–
1.7 Ga and 2.3 –1.9 Ga for the respective groups. According to
these authors, the provenance results suggest that the Vazante
Group constitutes a transitional sequence between the Paranoá
and Bambuı́ groups.
Recent Re – Os investigations of organic-rich shales of the upper
Vazante Group (Geboy 2006; Azmy et al. 2008) suggest a signifi-
cantly older, Mesoproterozoic age for the Morro do Calcário and
Serra do Garrote units, with ages of c. 1 Ga and older. However,
U –Pb determinations on detrital zircons (Dardenne et al. 2003,
2005; Dardenne 2007a; Rodrigues 2008; Rodrigues et al. 2008)
indicated maximum depositional ages for the Santo Antonio do
Bonito Fm. (diamictites at the basal section of the Vazante
Group, Fig. 48.4) of 988 + 15 Ma. Rodrigues et al. (2008) and
Rodrigues (2008) also report a significant Neoproterozoic zircon
population from the Rocinha Fm. in which the youngest grains
have been dated c. 925 Ma. Nonetheless, in the same study the
authors identified provenance patterns for the conglomerates at
517
the basal portion of the Lagamar Fm. with a maximum age of
1.95 Ga. consistent with the c. 1.35 Ga Re – Os ages obtained by
Geboy (2006) from organic shales of the same unit.
Mineralization
Mineral deposits associated with Late Precambrian glacial and
carbonate sediments of the São Francisco craton include the
following:
† Fe – Mn deposits near Porteirinha (Minas Gerais state), which
are intimately associated with glacial sedimentation of the
Macaúbas Group in the Araçuaı́ belt, and likely related to exha-
lative processes (Dardenne & Schobbenhaus 2001).
† Phosphorite deposits interbedded with carbonates in the lower
Vazante Group at Coromandel, Rocinha and Lagamar (Dard-
enne et al. 1986, 1997; Da Rocha-Araújo et al. 1992; Sanches
et al. 2007), at the base of the Sete Lagoas Fm. in the
Bambuı́ Group at Campos Belos (Dardenne et al. 1986), in
the lower half of the Serra da Saudade Fm. in the Bambuı́
Group at Cedro do Abaeté (Dardenne et al. 1986; Lima et al.
2005), and in stromatolitic beds of the Salitre Fm. in the Irecê
basin (Misi & Kyle 1994).
† Fluorite, barite and Pb –Zn deposits associated with the upper
part of the first carbonate megacycle of the Bambuı́ Group
near Januária-Itacarambi, Montalvânia and Serra do Ramalho
(Dardenne 1979; Misi 1979; Dardenne & Freitas-Silva 1999).
† Pb– Zn sulphide deposits associated with brecciated dolomite
in the middle Vazante Group at Morro Agudo mine (Hitzman
et al. 1995; Misi et al. 1998, 1999; Dardenne & Freitas-Silva
1999; Bettencourt et al. 2001; Monteiro 2002).
† Zn silicate deposits associated with major fault structures at the
Vazante mine, Minas Gerais (Dardenne & Freitas-Silva 1999;
Bettencourt et al. 2001; Monteiro 2002; Misi et al. 2005).
Discussion
Broadly speaking, the Neoproterozoic Era experienced at least three
widespread ice ages, commonly referred to as the Gaskiers, Mar-
inoan and Sturtian events, which are radiometrically constrained
to be c. 585 Ma (Newfoundland: Bowring et al. 2003; Halverson
et al. 2005), c. 630 Ma (South China: Condon et al. 2005; Zhang
et al. 2005) and c. 750 Ma (Namibia: Hoffman et al. 1996), respect-
ively. Some researchers suggest that these singular glacial events
may have been diachronous, spanning over tens of millions of
years, or alternatively split into several discrete ice ages (e.g.
Kaufman et al. 1997; Kendall et al. 2006). In most cases, the
glacial diamictites are overlain by texturally and isotopically enig-
matic cap carbonates with strong negative C-isotope anomalies
(e.g. Kaufman et al. 1997; Hoffman et al. 1998b), which have
also been recorded in syn- and immediately pre-glacial carbonates
(e.g. Kaufman et al. 1997; Kennedy et al. 2001; Halverson et al.
2002). In contrast, S-isotope compositions in the cap carbonates pre-
serve remarkably positive d34S anomalies (e.g. Hurtgen et al. 2002).
In all the studied Neoproterozoic successions from the São Fran-
cisco craton, at least one horizon of glacial diamictite (generally
recognized on the basis of faceted and striated clasts of hetero-
geneous composition in a fine-grained matrix) has been recog-
nized. Most often, these glacial horizons occur at the base of the
succession. These ice age deposits are overlain by cap carbonates,
although in some cases a thin siliciclastic wedge, perhaps record-
ing glacial outwash facies during meltback (Bailey & Peters 1998),
separates the diamictite from carbonate lithofacies. Chemostrati-
graphic studies (including C, S and Sr isotopes) and Pb/Pb
dating of carbonates from these glaciogenic successions provide
tools for possible correlations with Neoproterozoic counterparts
around the world.
518 A. MISI ET AL.
Perhaps of greatest interest are the correlation of Bambuı́,
Una, Vazante and Vaza Barris/Miaba group deposits on the São
Francisco Craton with Neoproterozoic equivalents on the Congo
Craton in Namibia. If we accept that throughout much of the
Mesoproterozoic and Neoproterozoic eras these cratons were
joined, the broadly equivalent sedimentary successions should
reveal striking similarities. However, whereas the well-studied
Otavi Group, in Namibia, hosts at least two cap carbonates atop
glacial sediments (Kaufman et al. 1991; Hoffmann & Prave
1996; Hoffman et al. 1998b), only one ice age deposit is currently
recognized in the Bambuı́ and equivalent Una groups. Diamictites
from these units containing basement clasts in a fine-grained
Fe-rich mudstone and overlained by a cap-carbonate lithofacies
with centimetre-scale sea-floor precipitates, are strikingly similar
to those of the younger Maieberg Fm. cap carbonate in Namibia.
The Sete Lagoas precipitates are preserved in limestone with nega-
tive d13C values and high Sr abundances. Sr-isotope compositions
of these lithologies are consistently around 0.7073, which is an
exact match with the precipitate interval in the Maieberg Fm., sup-
porting this intracontinental correlation. Notably, Pb/Pb dating of
the Sete Lagoas precipitates yielded an age of 740 + 22 Ma
(11-point isochron; MSWD ¼ 0.66; Babinski et al. 2007). Accep-
tance of this date and the correlation with the precipitates in the
Maieberg Fm. associates both of these post-glacial deposits with
the Sturtian ice age. This assignment is at odds with the view of
the Maieberg Fm. as the archetypical Marinoan cap carbonate
(Hoffman & Schrag 2002), and the c. 630 Ma age assigned to
the underlying Ghaub diamictite by the correlation of this unit
with an ash-bearing diamictite in the deformed Swakop Group of
central Namibia (Hoffmann et al. 2004). Independent chemostrati-
graphic evidence for the equivalence of the dated Swakop Group
Fig. 48.9. Chemostratigraphy of the Vaza
Barris/Miaba Groups (modified from Sial
et al. 2000).
diamictite with the Otavi Group is missing and we regard it
is possible that these glacial deposits instead correlate with
possibly younger glacial deposits of the Witvlei Group on the
Kalahari Craton in southern Namibia (Kaufman et al. 1997;
Saylor et al. 1998).
On the western margin of the Sao Francisco Craton in Brazil,
the basal glacial deposit of the Vazante Group (Santo Antonio
do Bonito Fm.) has long been considered as an equivalent to the
basal diamictite of the Bambuı́ Group beneath the Sete Lagoas
Fm. Overlying strata in both successions are variably mineralized
(phosphate and Pb– Zn) and contain thick carbonate intervals.
However, C-isotope compositions of the carbonates differ signifi-
cantly. The d13C profile for Vazante carbonates is generally near
zero, and shows only minor excursions to negative values that
are interpreted to reflect cap-carbonate lithofacies (Azmy et al.
2001, 2006). In contrast, the strongly negative d13C values of the
Sete Lagoas cap carbonate are followed by remarkably positive
values (upwards to þ16‰) in overlying strata (Misi et al. 2007).
Thus the equivalence of these successions is suspect, although
Sr-isotope compositions of lower Vazante phosphorites are quite
similar to those in well-preserved limestones of the Bambuı́ and
Una groups (Misi & Veizer 1998). Further complicating the inter-
basinal correlation is the presence of a second Vazante Group dia-
mictite (D II) in the Serra do Poço Verde Fm. (Fig. 48.2: Olcott
et al. 2005; Azmy et al. 2006), and the possibility that the over-
lying Lapa Fm. was deposited during oceanic transgression after
a third possible glacial event (Brody et al. 2004; Azmy et al. 2006).
It has been suggested previously that the shale and marly car-
bonate of the Lapa Fm. may be equivalent to the Rasthof Fm. in
northern Namibia (the older of the two Otavi Group cap carbon-
ates). This assignment, however, was based on comparable
 NEOPROTEROZOIC SUCCESSIONS OF THE SÃO FRANCISCO CRATON
negative d13C excursions in both units, and on a single Sr-isotope
value (c. 0.7068) from a dolomite cement in the basal Lapa For-
mation (Azmy et al. 2001, 2006) that matches with the lowest
values recorded in Sr-rich Rasthof limestones. However, the
Sr-isotope equivalence of these units is problematic and, further-
more, new Re –Os age constraints for shale of the Lapa and
older organic-rich lithologies in the Vazante Group suggest that
this succession may be far older than previously considered.
Re –Os age determinations on Lapa Fm. shale suggest an age of
close to one billion years (Azmy et al. 2008), and similar measure-
ments of organic-rich shale in the underlying Serra do Poço Verde
and Serra do Garrote formations yield ages of c. 1.1 and 1.3 Ga
(Geboy 2006; Geboy et al. 2006, 2009). A Mesoproterozoic
age for this succession is consistent with the relative invariance
of C-isotope trends in the Vazante Group, which are comparable
with the Paranoá Group stratigraphically beneath Bambuı́
sediments in central and eastern Brazil (Santos et al. 2000).
This older age is similarly suggested from Nd-isotope studies of
detrital grains in the Vazante Group (Pimentel et al. 2001) and
the appearance of Conophyton metula Kirichenko stromatolites
(Cloud & Dardenne 1973), which are also known from the
Paranoá Group.
The Re –Os ages provided by two different laboratories would
push the known record of Late Precambrian glaciation into the
Mesoproterozoic Era, some 500 million years earlier than pre-
viously considered. Further tests of the Re – Os ages, as well as
time-series elemental, isotopic and biomarker studies (cf. Olcott
et al. 2005) of the Vazante Group, which are currently in progress,
should shed light on the depositional environment of the glacio-
genic Vazante Group and provide better resolution of these contro-
versial age estimates.
During the last ten years CNPq has provided funds for research on the Neoproter-
ozoic basins of the São Francisco Craton. Currently, A. M. receives support from
project no. 486416/2006-2. A. J. K. acknowledges the NSF and NASA for
funding for work in the Proterozoic successions of Brazil, as well as the Deutsche
Forschungsgemeinschaft and Westfälische Wilhelms-Universität Münster for
sabbatical funding in 2007–2008. A. N. S acknowledges support from the CNPq
(grants PROSUL/CNPq 490136/2006-0 and 475657/2006-3) and PRONEX
APQ-047-1.07/06 that have defrayed costs with fieldwork in Sergipe and
Bahia. We are grateful to the directors of Votorantim Metais for supporting
fieldwork at the Vazante area. This represents a contribution of the IUGS- and
UNESCO-funded IGCP (International Geoscience Programme) Project #512.
References
Alkmim, F. F., Brito Neves, B. B. & Alves, J. A. C. 1993. Arcabouço
Tectônico do Cráton do São Francisco: uma revisão. In: Dominguez,
J. M. L. & Misi, A. (eds) O Craton do São Francisco SBG, SGM,
CNPq, 45 –62.
Alkmim, F. F., Chemale, F., Jr. & Endo, I. 1996. A deformação das
coberturas proterozóicas do Craton do São Francisco e o seu signifi-
cado tectônico. Revista da Escola de Minas de Ouro Preto, 49,
22 – 38.
Amaral, G. & Kawashita, K. 1967. Determinações da idade do Grupo
Bambuı́ pelo método Rb– Sr. 21st Congresso Brasileiro de Geologia.
Curitiba, Anais, 214– 217.
Azmy, K., Veizer, J., Misi, A., Oliveira, T. F., Sanches, A. L. & Dard-
enne, M. A. 2001. Dolomitization and isotope stratigraphy of the
Vazante Formation, São Francisco Basin, Brazil. Precambrian
Research, 112, 303– 329.
Azmy, K., Kaufman, A. J., Misi, A. & Oliveira, T. F. 2006. Isotope
stratigraphy of the Lapa Formation, São Francisco Basin, Brazil:
implications for Late Neoproterozoic glacial events in South
America. Precambrian Research, 149, 231– 248.
Azmy, K., Kendall, B., Creaser, R. A., Heaman, L. & de Oliveira,
T. F. 2008. Global correlation of the Vazante Group São Francisco
Basin, Brazil: Re– Os and U– Pb radiometric age constraints. Pre-
cambrian Research, 164, 160172.
519
Babinski, M. & Kaufman, A. J. 2003. First direct dating of a Neoproter-
ozoic post-glaciogenic cap carbonate. IV South American Symposium
on Isotope Geology, Short Papers 1, 321– 323.
Babinski, M., Vieira, L. C. & Trindade, R. I. F. 2007. Direct dating of
the Sete Lagoas cap carbonate (Bambuı́ Group, Brazil) and impli-
cations for the Neoproteozoic glacial events. Terra Nova, 19,
401– 406.
Bailey, C. M. & Peters, S. E. 1998. Glacially influenced sedimentation in
the late Neoproterozoic Mechum River Formation, Blue Ridge pro-
vince, Virginia. Geology, 26, 623– 626.
Bettencourt, J. S., Monteiro, L. V. S., Bello, R. M. S., Oliveira, T.
F. & Juliani, C. 2001. Metalogênese do Zinco e chumbo na
região de Vazante-Paracatu, Minas Gerais. In: Pinto, C. P. &
Martins-Neto, M. A. (eds) Bacia do São Francisco: Geologia e
recursos minerais. SBG-Núcleo de Minas Gerais, Belo Horizonte,
161– 198.
Bowring, S., Myrow, P., Landing, E., Ramezani, J. & Grotzinger, J.
2003. Geochronological constraints on terminal Proterozoic events
and the rise of metazoans. Geophysical Research Abstracts (EGS,
Nice), 5, 219.
Brasier, M. D. & Shields, G. 2000. Neoproterozoic chemostratigraphy
and correlation of the Port Askaig glaciation, Dalradian Supergroup
of Scotland. Journal of the Geological Society of London, 157,
909– 914.
Brito Neves, B. B., Van Schmus, W. R., Babinski, M. & Sabin, T. 1993.
O evento de magmatismo de 1,0 Ga nas faixas mó veis ao norte do
Craáton São Francisco. Sociedade Brasileira de Geologia, Simpósio
o Cráton do São Francisco 2, Salvador, Anais, 243– 245.
Brito-Neves, B. B., Campos-Neto, M. C. & Fuck, R. A. 1999.
From Rodinia to Western Gondwana: an approach to the Brasiliano–
Pan African Cycle and orogenic collage. Episodes, 22, 155– 166.
Brody, K. B., Kaufman, A. J., Eigenbrode, J .L. & Cody, G. D. 2004.
Biomarker geochemistry of a post-glacial Neoproterozoic succession
in Brazil (Abstract), Geological Society of America, Abstracts with
Programs, 36, 477.
Campos-Neto, M. C. 1984a. Litoestratigrafia, relações estratigráficas e
evolução paleogeográfica dos grupos Canastra e Paranoá (região
Vazante-Lagamar, MG). Revista Brasileira de Geociências, 14,
81– 91.
Campos Neto, M. C. 1984b. Geometria e fases de dobramentos brasilia-
nos superpostos no oeste de Minas Gerais. Revista Brasileira de Geo-
ciência, 14, 60– 68.
Chang, H. K., Miranda, F. P. & Alkmim, F. F. 1988. Considerações
sobre a evolução tectônica da Bacia do São Francisco. In: 35th Con-
gresso Brasileiro de Geologia (Belém), Anais, 5, 2076–2090.
Cloud, P. E. & Dardenne, M. A. 1973. Proterozoic age of the Bambuı́
Group in Brazil. Geological Society of America Bulletin, 84,
1673–1676.
Condie, K. C. 2002. The supercontinent cycle: are there two patterns of
cyclicity? Journal of African Earth Sciences, 35, 179– 183.
Cordani, U. G., Brito-Neves, B. B. & D’Agrella-Filho, M. S. 2003.
From Rodinia to Gondwana: a review of the available evidence
from South América. Gondwana Research, 6, 275– 283.
Condon, D., Zhu, M., Bowring, S., Wang, W., Yang, A. & Jin, Y. 2005.
U–Pb ages from the Neoproterozoic Doushantuo Formation, China.
Science, 308, 95– 97.
Cozzi, A., Allen, P. & Grotzinger, J. 2004. Understanding carbonate
ramp dynamics using d13C profiles: examples from the Neoprotero-
zoic Buah Formation of Oman. Terra Nova, 16, 62 –67.
DA ROCHA ARAUJO, P. R., FLICOTEAUX, R., PARRON, C. & TROMPETTE, R.
1992. Phosphorites of Rocinha mine, Patos de Minas (Minas Gerais
Brasil): genesis and evolution of a Middle Proterozoic Deposit tecto-
nized by the Brasiliano Orogeny. Economic Geology, 87, 332– 351.
Dardenne, M. A. 1978a. Zonação tectônica na borda ocidental do Cráton
São Francisco. 30th Congresso Brasileiro de Geologia, Recife, SBG,
1, 299– 308.
Dardenne, M. A. 1978b. Sı́ntese sobre a estratigrafia do Grupo Bambuı́
no Brasil Central. 30th Congresso Brasileiro de Geologia, SBG, 2,
597– 610.
Dardenne, M. A. 1979. Les minéralisations de plomb, zinc, fluor du Pro-
térozoı̈que Supérieur dans le Brésil Central. Thèse de Doctorat
d’État, Université de Paris VI. Coromandel. Dissertação de Mestrado.
520 A. MISI ET AL.
Dardenne, M. A. 1981. Os grupos Paranoá e Bambuı́ na Faixa Dobrada
Brası́lia. An. Simp. Cráton São Francisco e suas Faixas Marginais,
Salvador, SBG, 140–157.
Dardenne, M. A. 2000. The Brası́lia Fold Belt. In: Cordani, U. G.,
Milani, E. J., Thomaz-Filho, A. & Campos, D. A. (eds) Tectonic
Evolution of South America. International Geological Congress, 31,
Rio de Janeiro, 231– 263.
Dardenne, M. A. 2001. Lithostratigraphic sedimentary sequence of the
Vazante Group. In: Misi, A. & Teixeira, J. B. G. (eds) Proterozoic
Base Metal Deposits of Africa and South America. Proceedings of
the 1st Field Workshop IGCP 450, CNPq/UNESCO/IUGS, Belo
Horizonte and Paracatu (MG), Brazil, 48 –50.
Dardenne, M. A. 2006. A glaciação neoproterozóica do Grupo Vazante
na Faixa Brası́lia: discussão e alternativas. Congresso Brasileiro de
Geologia, 43, SBG, Aracaju, Resumos.
Dardenne, M. A. 2007a. The pseudo-diamictites of the Vazante Group.
Symposium IGCP 478, Stellenbosch, Abstracts.
Dardenne, M. A. 2007b. Lithostratigraphy of the Vazante and Bambuı́
groups in the São Francisco Craton and the Brası́lia Fold Belt.
IGCP 478, Stellenbosch, Abstracts.
Dardenne, M. A. & Freitas-Silva, F. H. 1999. Modelos genéticos dos
depósitos Pb-Zn nos grupos Bambuı́ e Vazante. Workshop: Depósitos
Minerais Brasileiros de Metais-Base. Salvador, CAPES-PADCT-
ADIMB, 86– 93.
Dardenne, M. A. & Schobbenhaus, C. 2001. Metalogênese do Brasil.
Editora UnB, Brası́lia, 392.
Dardenne, M. A., Trompette, R., Magalhães, L. F. & Soares, L. A.
1986. Proterozoic and Cambrian phosphorites-regional review:
Brazil. In: Cook, P. J. & Shergold, J. H. (eds) Phosphate Deposit
of the World. Proterozoic and Cambrian Phosphorites. Cambridge
University Press, 116–131.
Dardenne, M. A., Freitas-Silva, F. H., Nogueira, G. M. S. & Souza,
J. F. C. 1997. Depósitos de fosfato de Rocinha e Lagamar, Minas
Gerais. In: Schobbenhaus, C., Queiroz, E. T. & Coelho,
C. E. S. (eds) Principais Depósitos Minerais do Brasil. DNPM/
CPRM, IVC, 113– 122.
Dardenne, M. A., Pimentel, M. M. & Alvarenga, C. J. S. 2003.
Provenance of conglomerates of the Bambuı́, Jequitaı́, Vazante and
Ibiá groups, implications for the evolution of the Brası́lia Belt.
In: Sociedade Brasileira de Geologia, Simpósı́o Nacional de Estudos
Tectônicos, 9, International Symposium on Tectonics, 3, Armação de
Búzios, 2003. Boletim de Resumos, 47–50.
D’el-Rey Silva, L. J. H. 1995. Tectonic evolution of the Sergipano Belt,
northeastern Brazil. Revista Brasileira de Geociências, 25, 315– 332.
D’el Rey Silva, L. J. H. 1999. Basin infilling in the southern-central
part of the Sergipano Belt (NE Brazil) and implications for the
evolution of Pan-African/Brasiliano cratons and Neoproterozoic
sedimentary cover. Journal of South American Earth Sciences, 12,
453– 470.
Derry, L. A., Keto, L. S., Jacobsen, S. B., Knoll, A. H. & Swett, K.
1989. Sr isotopic variations of upper Proterozoic carbonates from
East Greenland and Svalbard. Geochimica Cosmochimica Acta, 53,
2331– 2339.
Fairchild, I. J., Marshall, J. D. & Bertrand-Safati, J. 1990. Strati-
graphic shifts in carbon isotopes from Proterozoic stromatolitic
carbonates (Mauritania): influence of primary mineralogy and dia-
genesis. American Journal of Science, 290A, 46 –79.
Fairchild, T. R., Schopf, J. W. et al. 1996. Recent discoveries of
Proterozoic microfossils in south-central Brazil. Precambrian
Research, 80, 125– 152.
Fuck, R. A., Pimentel, M. M. & Silva, J. H. D. 1994. Compartimentação
tectônica na porção oriental da Provı́ncia Tocantins. Anais 388
Congresso Brasileiro de Geologia, Camboriú, SBG, 1, 215– 216.
Geboy, N. J. 2006. Rhenium– osmium age determinations of glaciogenic
shales from the Mesoproterozoic Vazante Formation, Brazil. MS
thesis, University of Maryland, 99.
Geboy, N. J., Kaufman, A. J. & Walker, R. J. 2006. A stable isotope and
Re –Os study of organic-rich mudstones across a Proterozoic glacial
cycle in Brazil. Geological Society of America Annual Meeting
(Philadelphia), Abstracts, 38, 125.
Geboy, N. J., Kaufman, A. J. et al. 2009. Evidence of Mesoproterozoic
Ice Ages in Brazil. Internal report, unpublished.
Guimarães, J. T. 1996. A Formação Bebedouro no Estado da Bahia;
faciologia, estratigrafia e ambientes de sedimentação. Dissertação
(Mestrado em Geologia), Instituto de Geociências, Universidade
Federal da Bahia (UFBA), Brasil.
Guimarães, J. T., Misi, A., Pedreira, A. J. & Dominguez, J. M. L. 2011.
The Bebedouro Formation, Una Group, Bahia (Brazil). In: Arnaud,
E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological
Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 503–508.
Halverson, G. P., Hoffman, P. F., Schrag, D. P. & Kaufman, A. J.
2002. A major perturbation of the carbon cycle before the Ghaub
glaciation (Neoproterozoic) in Namibia: prelude to snowball Earth.
Geophysics, Geochemistry, Geosystems, 3, doi: 10.1029/2001G
C000244.
Halverson, G. P., Hoffman, P. F. & Schrag, D. P. 2005. Toward a
Neoproterozoic composite carbon isotope record. Geological
Society of America Bulletin, 117, 1181– 1207.
Hitzmann, M. W., Thorman, C. H., Romagna, G., Oliveira, T. F.,
Dardenne, M. A. & Drew, L. J. 1995. The Morro Agudo Zn –Pb
deposit, Minas Gerais, Brazil: a proterozoic Irish-type carbonate-
hosted SEDEX – replacement deposit. Annual Meeting, New
Orleans, GSA, Abstract, 408.
Hoffman, P. F. & Schrag, D. P. 2002. The snowball Earth hypothesis:
testing the limits of global change. Terra Nova, 14, 129–155.
Hoffman, P. F., Hawkins, D. P., Isachsen, C. E. & Bowring, S. A. 1996.
Precise U–Pb zircon ages for early Damaran magmatism in the
Summas Mountains and Welwitschia inlier, northern Damara belt,
Namibia. Geological Survey of Namibia Communications, 11,
47 – 52.
Hoffman, P. F., Kaufman, A. J., Halverson, G. P. & Schrag, D. P.
1998a. A Neoproterozoic snowball Earth. Science, 281, 1342– 1346.
Hoffman, P. F., Kaufman, A. J. & Halverson, G. P. 1998b. Comings
and goings of global glaciation on a Neoproterozoic tropical platform
in Namibia. GSA Today, 8, 1– 9.
Hoffmann, K.-H. & Prave, A. R. 1996. A preliminary note on a revised
subdivision and regional correlation of the Otavi Group based on gla-
ciogenic diamictites and associated cap dolostones. Communications
of the Geological Survey of Namibia, 11, 77 – 82.
Hoffmann, K.-H., Condon, D. J., Bowring, S. A. & Crowley, J. L.
2004. A U– Pb zircon date from the Neoproterozoic Ghaub For-
mation, Namibia: Constraints on Marinoan glaciation. Geology, 32,
817– 820.
Hurtgen, M. T., Arthur, M. A., Suits, N. S. & Kaufman, A. J. 2002.
The sulfur isotopic composition of Neoproterozoic seawater sulfate:
implications for a snowball earth? Earth Planet. Sci. Lett. 203,
413– 430.
Hyde, W. T., Crowley, T. J., Baum, S. K. & Peltier, W. R. 2000.
Neoproterozoic ‘snowball Earth’ simulations with a coupled
climate/icesheet model. Nature, 405, 425– 429.
Iyer, S. S., Babinski, M., Krouse, H. R. & Chemale, F., Jr. 1995. Highly
13
C enriched carbonate and organic matter in the Neoproterozoic
sediments of the Bambuı́ Group, Brazil. Precambrian Research, 73,
271– 282.
Jacobsen, S. B. & Kaufman, A. J. 1999. The Sr, C and O
isotope evolution of Neoproterozoic seawater. Chemical Geology,
161, 37– 57.
Karfunkel, J. & Hoppe, A. 1988. Late Proterozoic glaciation in
Central-Eastern Brazil: synthesis and model. Paleogeography, Paleo-
climatology, Paleoecology, 65, 1– 21.
Kaufman, A. J. & Knoll, A. H. 1995. Neoproterozic variations in the C-
isotopic composition of seawater: stratigraphic and biogeochemical
implications. Precambrian Research, 73, 27 – 49.
Kaufman, A. J., Hayes, J. M., Knoll, A. H. & Germs, G. J. B. 1991.
Isotopic compositions of carbonates and organic carbon from upper
Proterozoic successions in Namibia: stratigraphic variation and the
effect of diagenesis and metamorphism. Precambrian Research, 49,
301– 327.
Kaufman, A. J., Knoll, A. H. & Narbonne, G. M. 1997. Isotopes, ice
ages, and terminal Proterozoic earth history. Proceedings of the
National Academy of Science USA, 94, 6600–6605.
Kaufman, A. J., Varni, M. A., Misi, A. & Brito-Neves, B. B. 2001.
Anomalous d34S signatures in trace sulfate from a potential cap
 NEOPROTEROZOIC SUCCESSIONS OF THE SÃO FRANCISCO CRATON
carbonate in the Neoproterozoic Bambuı́ Group, Brazil. In: Misi, A.
& Teixeira, J. B. G. (organizers) Proterozoic Sediment-Hosted
Base Metal Deposits of Western Gondwana. I Field Workshop,
Beolo Horizonte and Paracatu (Minas Gerais), Brazil, 62 –65.
Kawashita, K. 1998. Rochas carbonáticas neoproterozóicas da América
do Sul: idades e inferências quimioestratigráficas. Full professor
thesis, University of São Paulo, Brazil, 126.
Kiang, C. H. 1997. Isótopos estáveis (C, H, O) e 87Sr/86Sr: implicações na
estratigrafia e na paleo-circulação de fluidos na Bacia do São Fran-
cisco. Full Professor Thesis, UNESP, São Paulo, Brazil.
Knoll, A. H., Hayes, J. M., Kaufman, A. J., Swett, K. & Lambert, L.
B. 1986. Secular variation in carbon isotope ratios from upper
Proterozoic successions of Svalbard and East Greenland. Nature,
321, 832– 838.
Kendall, B., Creaser, R. A. & Selby, D. 2006. Re– Os geochronology
of post-glacial black shales in Australia: constraints on the timing of
‘Sturtian’ glaciation. Geology, 34, 729–732.
Kennedy, M. J., Christie-Blick, N. & Sohl, L. E. 2001. Are Proterozoic
cap carbonates and isotopic excursions a record of gas hydrate
destabilization following Earth’s coldest intervals? Geology, 29,
443– 446.
Lima, O. N. B., Uhlein, A. & Britto, W. 2005. Geologia dos depósitos
fosfáticos do Grupo Bambuı́ na Serra da Saudade, Cedro do Abaeté,
Minas Gerais. In: Short Papers-III Simpósio sobre o Craton do São
Francisco, Salvador, SBG, 320–323.
Madalosso, A. 1979. Stratigraphy and sedimentation of the Bambuı́
Group in Paracatu region, minas Gerais, Brazil. MA thesis, Univer-
sity of Missouri, USA, 127.
Marini, O. J., Fuck, R. A., Danni, J. C., Dardenne, M. A., Loguercio,
S. O. C. & Ramalho, R. 1984. As faixas de dobramentos Brası́lia,
Uruaçu e Paraguai-Araguaia e o Maciço Mediano de Goiás. In:
Schobbenhaus, C., Campos, D. A., Derze, G. R. & Asmus, H. E.
(eds) Geologia do Brasil. Texto Explicativo do Mapa Geológico do
Brasil e da Área Oceânica Adjacente Incluindo Depósitos Minerais.
Ministério da Minas e Energias/Departamento Nacional da
Produção Mineral, Brası́lia, 251– 303.
Misi, A. 1978. Ciclos de sedimentação e mineralizações de Pb –Zn nas
sequências Bambui (Supergrupo São Francisco), Estado de Bahia.
XXX Congresso Brasileiro de Geologia. Sociedade Brasileira de Geo-
logia, Anais, 4, 2548–2561.
Misi, A. 1979. O Grupo Bambuı́ no Estado da Bahia. In: Inda, H. V. (ed.)
Geologia e Recursos Minerais do Estato da Bahia. Textos Básicos,
SME/CPM, Salvador, 1, 120– 154.
Misi, A. 1999. Um modelo de evolução metalogenética para os depósitos
de zinco e chumbo hospedados em sedimentos proterozóicos de
cobertura do Craton do São Francisco. Full Professror Thesis,
Federal University of Bahia (Brazil).
Misi, A. 2001. Estratigrafia isotópica das sequências do Supergrupo São
Francisco, coberturas neoproterozóicas do Cráton do São Francisco.
Idade e correlações. In: Pinto, C. P. & Martins-Neto, M. A.
(eds) Bacia do São Francisco, Geologia e Recursos Naturais, SBG/
MG, Belo Horizonte, 67– 92.
Misi, A. & Kyle, J. R. 1994. Upper Proterozoic carbonate
stratigraphy, diagenesis, and stromatolitic phosphorite formation,
Irecê Basin, Bahia, Brazil. Journal of Sedimentary Research, 64,
299– 310.
Misi, A. & Souto, P. G. 1975. Controle estratigráfico das mineralizações
de Pb –Zn –F –Ba do Grupo Bambui, parte leste da Chapada de Irecê
(Bahia). Revista Brasileria de Geosciências, 5, 30 – 45.
Misi, A. & Veizer, J. 1998. Neoproterozoic carbonate sequences of the
Una Group, Irecê Basin, Brasil: chemostratigraphy, age and corre-
lations. Precambrian Research, 89, 87 –100.
Misi, A., Veizer, J., Kawashita, K. & Dardenne, M. A. 1997. The age
of the Neoproterozoic carbonate platform sedimentation based on
87
Sr/86Sr determinations, Bambuı́ and Una Groups, Brazil. I South
American Symposium on Isotope Geology, Campos do Jordão, São
Paulo, Brazil. Extented Abstracts, 199– 200.
Misi, A., Iyer, S. S. S. et al. 1999. Geological and isotopic constraints
on the metallogenic evolution of the Proterozoic sediment-hosted
Pb – Zn (Ag) deposits of Brazil. Gondwana Research, 2, 47 – 65.
Misi, A., Iyer, S. S. S. et al. 2005. Sediment-hosted lead– zinc deposits of
the Neoproterozoic Bambui Group and correlative sequences, São
521
Francisco Craton, Brazil: a review and a possible metallogenic evol-
ution model. Ore Geology Reviews, 3, 263–304.
Misi, A., Kaufman, A. J. et al. 2007. Chemostratigraphic correlation of
Neoproterozoic successions in South America. Chemical Geology,
237, 161–185.
Montes, A. S. L. 1997. O contexto estratigráfico e sedimentológico da
Formação Bebedouro na Bahia. Um possı́vel portador de diamantes.
MSc thesis, University of Brası́lia, Brazil.
Monteiro, L. V. S. 2002. Modelamento metalogenético dos depósitos
de Vazante, Fagundes e Ambrósia associados ao Grupo Vazante,
Minas Gerais. Tese Doutorado, Universidade de São Paulo-
USP, 317.
Olcott, A. N., Sessions, A. L., Corsetti, F. A., Kaufman, A. J. & de
Oliviera, T. F. 2005. Direct evidence for significant primary
production during widespread Neoproterozoic glaciation. Science,
310, 471–474.
Pedrosa-Soares, A. C., Dardenne, M. A., Hasui, Y., Castro, F. D. C.,
Carvalho, M. V. A. & Reis, A. C. 1994. Mapa Gológico do Estdo de
Minas gerais e Nota Explicativa. Secretaria de Recursos Minerais,
Hı́dricos e Energéticos, Companhia Mineradora de Minas Gerais
(COMIG), mapas e texto, 97.
Pimentel, M. M., Dardenne, M. A., Viana, M. G., Costa, S. M., Gioia,
L., Junges, S. & Seer, H. J. 2001. Nd isotopes and the provenance
of sediments from the Neoproterozoic Brası́lia Belt, Central Brazil:
geodynamic implications. Journal of South American Earth Sciences,
14, 571– 585.
Porada, H. 1989. Pan-African rifting and orogenesis in southern to
equatorial Africa and Eastern Brazil. Precambrian Research, 44,
103– 136.
Powis, K. 2006. Stable isotope geochemistry of the Neoproterozoic
Bambuı́ Group at Serra do Ramalho, Bahia, Brazil. PhD thesis,
Ottawa-Carleton Geoscience Center, University of Ottawa, Canada,
197.
Rodrigues, J. B. 2007. Estudos de proveniência de sedimentos da região
Centro-Oeste da Faixa Brası́lia-uma abordagem geocronológica.
Exame de Qualificação Doutorado, Universidade de Brası́lia-UnB,
Brası́lia, 50.
Rodrigues, J. B. 2008. Proveniência de sedimentos dos grupos Canastra,
lbiá, Vazante e Bambuı́, Um estudo de zircões detrı́ticos e ldades
Modelo Sm –Nd. Dr. Thesis, University of Brasilia, Brasilia DF.
Rodrigues, J. B., Pimentel, M. M., Buhn, B., Dardenne, M. A. &
Alvarenga, C. J. S. 2008. Provenance of the Vazante Group–
Preliminary data. VI South American Symposium on Isotope
Geology, Bariloche, Argentina. Extented Abstracts, CD-ROM.
Sanches, A. L., Misi, A., Kaufman, A. J. & Azmy, K. 2007. As sucessões
carbonáticas neoproterozóicas do Craton do São Francisco e os
depósitos de fosfato: correlações e fosfogênese, Revista Brasileira
de Geociências, 37, 1034–1046.
Santos, R. V., de Alvarenga, C. J. S., Dardenne, M. A., Sial, A. N. &
Ferreira, V. F. 2000. Carbon and oxygen isotopes across
Meso-Neoproterozoic limestones from Central Brazil: Bambuı́ and
Paranoá groups. Precambrian Research, 104, 107–122.
Santos, R. V., Alvarenga, C. J. S. et al. 2004. Carbon isotopes of
Mesoproterozoic-Neoproterozoic sequences from Southern São
Francisco Craton and Araçuaı́ Belt, Brazil: palaeogeographic impli-
cations. Journal of South American Earth Sciences, 18, 27 –39.
Saylor, B. Z., Kaufman, A. J., Grotzinger, J. P. & Urban, F. 1998. A
composite reference section for terminal Proterozoic strata of
southern Namibia. Journal of Sedimentary Research, 68, 1223– 1235.
Sial, A. N., Ferreira, V. P., Almeida, A. R., Romano, A. W., Parente,
C., Da Costa, M. L. & Santos, V. H. 2000. Carbon isotope
fluctuations in Precambrian carbonate sequences of several
localities in Brazil. Anais Academia Brasileira de Ciências, 72,
540– 557.
Sial, A. N., Ferreira, V. P., Moura, C. V. A. & Santos, V. H. 2003.
C-, O- and Sr-isotope stratigraphy of the Sturtian Jacoca and Olho
D’Agua Formations, state of Sergipe, Northeastern Brazil. Short
papers. IV South American Symposium on Isotope Geology. Salvador,
Bahia, 394–397.
Sial, A. N., Ferreira, V. P. et al. 2005. Two Neoproterozoic cap carbon-
ates in the states of Sergipe and Bahia, Northeastern Brazil: C- and
Sr-isotopes and mercury as paleoclimatic tracer. X Congresso
522 A. MISI ET AL.
Brasileiro de Geoquı́mica e II Simpósio de Geoquı́mica de Paı́ses do
Mercosul, Porto de Galinhas. Short paper CD-ROM.
Sial, A. N., Ferreira, V. P. et al. 2006a. Chemostratigraphy
of two Neoproterozoic cap carbonates from the Sergipano
belt (northeastern Brazil). Short Papers, V South American
Symposium on Isotope Geology (V SSAGI). Punta del Este,
Uruguay, 314–317.
Sial, A. N., Ferreira, V. P. et al. 2006b. C- and Sr-isotopes and mercury
as paleoclimatic tracer in two Neoproterozoic cap carbonates in
northeastern Brazil. Snowball Earth Conference. Ascona, Suiça,
100– 101.
Sial, A. N., Dardenne, M. A. et al. 2009. The São Francisco Paleocon-
tinent. In: Neoproterozoic-Cambrian Tectonics, Global Change and
Evolution: A Focus on Southwestern Gondwana. Developments in
Precambrian Geology, 16, 31– 69.
Söfner, B. 1973. Observações sobre a estratigrafia do Pré-Cambriano
da Chapada Diamantina Sudeste e da área contı́gua. In: SBG, 27th
Congresso Brasileiro de Geologia. Aracajú, Anais, 1, 23 –33.
Souza, J. C. F. 1997. Litoestratigrafia e sedimentologia da Formação
Vazante na região de Coromandel. MSc thesis, Universidade de
Brası́lia.
Teixeira, L. B., Martins, M. & Braun, O. P. G. 1993. Evolução geo-
lógica da Bacia São Francisco com base em sı́smica de reflexão e
métodos potenciais. II Simpósio sobre o Craton do São Francisco.
SBG, Núcleo Bahia, Anais, 179–181.
Teixeira, W. & Figueiredo, M. C. H. 1991. An outline of Early Proter-
ozoic crustal evolution in the São Francisco Craton, Brazil: a review.
Precambrian Research, 53, 1 –22.
Thomaz-Filho, A., Kawashita, K. & Cordani, U. G. 1998. A
origem do Grupo Bambuı́ no contexto da evolução geotectônica e
de idades radiométricas. Anais Academia Brasileira de Ciências,
70, 527– 548.
Torquato, J. R. F. & Misi, A. 1977. Medidas isotópicas de carbono
e oxigênio em carbonatos do Grupo Bambui na região centro-
norte do Estado da Bahia. Revista Brasileira de Geociências, 7,
14 –24.
Trindade, R. I. F., D’Agrella-Filho, M., Babinski, M. & Brito Neves,
B. B. 2004. Paleomagnetism and geochronology of the Bebedouro
cap carbonate: evidence for continental-scale Cambrian remagnetiza-
tion in the São Francisco craton, Brazil. Precambrian Research, 128,
83 – 103.
Trompette, R. 1994. Geology of Western Gondwana (2000– 500 Ma).
A.A. Balkema, Rotterdam, 350.
Uhlein, A., Alvarenga, C. J. S., Trompette, R., Dupont, H. S. J. B.,
Egydio-Silva, M., Cukrov, N. & Lima, O. N. B. 2004. Glaciação
neoproterozóica sobre o Craton do São Francisco e faixas dobradas
adjacentes. In: Mantesso-Neto, V., Bartorelli, A., Carneiro,
D. R. C. & Brito Neves, B. B. (eds) Geologia do Continente
Sul-Americano: Evolução da obra de Fernando Flávio de Almeida.
Beca Ed., São Paulo, 539– 553.
Uhlein, A., de Alvarenga, C. J. S., Dardenne, M. A. & Trompette, R.
R. 2011. The glaciogenic Jequitaı́ Formation, southeastern Brazil.
In: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The
Geological Record of Neoproterozoic Glaciations. Geological
Society, London, Memoirs, 36, 541–546.
Valeriano, C. M., Dardenne, M. A., Fonseca, M. A., Simões, L. S. A.
& Seer, H. J. 2004. A Evolução Tectônica da Faixa Brası́lia. In:
Mantesso-Neto, V., Bartorelli, A., Carneiro, D. R. C. &
Brito Neves, B. B. (eds) Geologia do Continente Sul-Americano:
Evolução da obra de Fernando Flávio de Almeida. Beca Edta, São
Paulo, 575–593.
Viera, L. C., Almeida, R. P., Trindade, R. I. F., Nogueira, A. C. R. &
Janikian, L. 2007. A Formação Sete Lagoas em sua área tipo: fácies,
estratigrafia e sistemas deposicionais. Revista Brasileira de Geo-
sciências, 37, 1020– 1033.
Young, G. M. 1995. Are Neoproterozoic glacial deposits preserved on the
margins of Laurentia related to the fragmentation of two superconti-
nents? Geology, 23, 153– 156.
Zhang, S., Jiang, G., Zhang, J., Song, B., Kennedy, M. J. & Christie-
Blick, N. 2005. U– Pb sensitive high-resolution ion microprobe ages
from the Doushantuo Formation in south China: constraints on late
Neoproterozoic glaciations: Geology, 33, 473–476.
 Chapter 49

The Neoproterozoic Macaúbas Group, Araçuaı́ orogen, SE Brazil

A. C. PEDROSA-SOARES1*, MARLY BABINSKI2, CARLOS NOCE1, MAXIMILIANO MARTINS1,

GLÁUCIA QUEIROGA1 & FRANCISCO VILELA1
1
Universidade Federal de Minas Gerais, Instituto de Geociências-CPMTC, Campus Pampulha,
31270-901 Belo Horizonte, MG, Brazil
2
Universidade de São Paulo, Instituto de Geociências, Rua do Lago 562, Cidade Universitária 05508-080 São Paulo, SP, Brazil
*Corresponding author (e-mail: pedrosa@pesquisador.cnpq.br)

Abstract: The Neoproterozoic Macaúbas Group records accumulation in the precursor basin of the Araçuaı́ orogen, located on the
eastern margin of the São Francisco craton (SE Brazil). The Macaúbas basin evolved from a late Tonian continental rift to a passive
margin that lasted at least until c. 660 Ma. It was orogenically inverted during the late Neoproterozoic Brasiliano event. The Macaúbas
Group includes the pre-glacial Matão, Duas Barras and Rio Peixe Bravo formations, the glaciogenic Serra do Catuni, Nova Aurora and
Lower Chapada Acauã formations, and the post-glacial Upper Chapada Acauã and Ribeirão da Folha formations. In the central sector of
the Araçuaı́ orogen, the oldest glaciogenic unit of the group, the Serra do Catuni Formation, overlies the sandstone-conglomerate package
of the Duas Barras Formation. The Serra do Catuni Formation consists of massive diamictite with minor sandstone and rare pelite, depos-
ited mostly in a proximal glaciomarine environment. This unit passes upward and eastward into the Lower Chapada Acauã Formation, a
thick succession of stratified diamictite, graded sandstone, pelite, transitional basalt and rare carbonate. This distal glaciomarine unit is
covered by the diamictite-free Upper Chapada Acauã Formation, which passes eastward into the Ribeirão da Folha Formation, which
includes fine-grained turbidite, pelite and ocean-floor rocks. In the northern sector of the Araçuaı́ orogen, the sandstone-pelite succession
of the pre-glacial Rio Peixe Bravo Formation is covered by the Nova Aurora Formation, the glaciomarine unit rich in diamictite and Fe-
rich diamictite, with minor graded sandstone and rare pelite. The Nova Aurora Formation is covered by the sandstone-pelite package
of the Upper Chapada Acauã Formation. The pre-glacial and glaciogenic successions record the continental rift to transitional stages
of the Macaúbas basin. The post-glacial succession represents proximal and distal passive margin to ocean floor environments. The
Serra do Catuni Formation seems to be a proximal glaciomarine equivalent of the Jequitaı́ glacio-terrestrial deposits located on the
São Francisco craton.

The Macaúbas Group represents the precursor basin of the
Neoproterozoic Araçuaı́ orogen, located in southeastern Brazil
(Fig. 49.1a). This orogen extends from the eastern border of the
São Francisco craton to the Atlantic Ocean (Pedrosa-Soares
et al. 2001, 2008).
Since the beginning of the 1930s the name Macaúbas has
appeared in the international geological literature to refer to a sedi-
mentary unit that includes Precambrian glaciogenic rocks (Moraes
& Guimarães 1931). The first field studies were published by
Moraes (1929), who described the succession of ‘conglomeratic
phyllites, phyllites, quartzites and mica schists cropping out in
the Macaúbas River valley and Catuni Ridge’, located to the
north of the city of Diamantina (Fig. 49.2). Moraes & Guimarães
(1930) named that succession the Macaúbas Fm. and interpreted
the ‘conglomeratic phyllites’ as metamorphosed glaciogenic
rocks. In 1932, Moraes noted the very extensive distribution of
the Macaúbas Fm. after identifying a thick pile of ‘highly meta-
morphosed conglomeratic phyllonites’ associated with quartzite,
schist and rare carbonate layers along the western valleys of the
Jequitinhonha and Rio Pardo rivers, and many of their tributaries
(Fig. 49.2). Moraes (1932) was also the first to suggest a correlation
between the Macaúbas Fm. and the glaciogenic Jequitaı́ Fm.
located on the São Francisco craton (Fig. 49.1a). Since Schöll
(1972) and Pflug & Schöll (1975), the designation Macaúbas
Group (instead of formation) has been used. Almeida (1977) con-
sidered the Macaúbas Group as the main fill of the ‘Alpine-type
geosyncline’ that became the Araçuaı́ fold belt after the Brasiliano
orogenic event in late Neoproterozoic time.
Based on its first definition (Moraes 1929, 1932; Moraes & Gui-
marães 1930, 1931), the name Macaúbas Group is here exclusively
applied to refer to the extensive unit with metadiamictite-bearing
formations that underwent regional deformation and metamorph-
ism within the Araçuaı́ orogen (Figs 49.2 & 49.3). This group
was thrust over the pelite-carbonate cratonic cover (Bambuı́
Group) and does not show any direct field relation with the Jequitaı́
and Carrancas diamictites located on the southern São Francisco
craton (Figs 49.1a & 49.2), although their correlation is often
assumed (Uhlein et al. 2011).
Several names such as conglomeratic phyllite, conglomeratic
phyllonite, paraconglomerate, conglomeratic greywacke, mixtite
and tillite were applied to the Macaúbas metadiamictites until
the 1970s, reflecting disagreement (and misunderstandings) over
the glaciogenic interpretation, evolution of geological knowledge,
and standardization of nomenclature. Indeed, the diamictite-
bearing units preserved on the São Francisco craton record the
best evidence of glaciation, despite the relatively small area over
which they occur. The Jequitaı́ Fm. includes glacio-terrestrial sedi-
ments and structures preserved from orogenic deformation (e.g.
Isotta et al. 1969; Viveiros & Walde 1976; Walde 1976; Rocha-
Campos & Hasui 1981; Gravenor & Monteiro 1983; Karfunkel &
Hoppe 1988; Martins-Neto et al. 1999; Uhlein et al. 1999, 2011;
Karfunkel et al. 2002; Martins-Neto & Hercos 2002), and has a
maximum sedimentation age around 880 Ma (U – Pb LA-ICPMS,
detrital zircon; Rodrigues 2008). The Carrancas diamictite
occurs in a few outcrops that lack clear evidence of glacial
origin, but it is overlain by the c. 740 Ma carbonate typical of
post-glacial rocks within the cratonic cover (Schöll 1972; Babinski
et al. 2007).
On the other hand, most of the Macaúbas metadiamictites rep-
resent sub-aqueous deposits that underwent strong deformation
and metamorphism, hampering the recognition of solid evidence
of glaciation. Nevertheless, the areas of occurrence, regional stra-
tigraphy, Neoproterozoic age and glaciogenic nature of the Macaú-
bas metadiamictites seem to be definitively demonstrated by the
data and correlations presented by Hettich (1975, 1977), Karfunkel
& Karfunkel (1976, 1977), Hettich & Karfunkel (1978), Viveiros

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 523– 534. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.49
524 A. C. PEDROSA-SOARES ET AL.
Fig. 49.1. (a) Location of the Araçuaı́ orogen in relation to the São Francisco
craton, showing the distribution of the Macaúbas Group and Neoproterozoic
cratonic covers. (b) The Araçuaı́–West Congo orogenic system in relation to
the São Francisco-Congo craton.
et al. (1979), Karfunkel et al. (1984, 1985), Karfunkel & Hoppe
(1988), Mourão & Pedrosa-Soares (1992), Pedrosa-Soares et al.
(1992, 1998, 2000, 2008), Grossi-Sad et al. (1997a), Noce et al.
(1997), Uhlein et al. (1998, 1999, 2007), Babinski et al. (2005,
2011), Gradim et al. (2005) and Martins (2006).
In this chapter, we present an updated synthesis on the Macaú-
bas Group with special emphasis on the glaciogenic units that are
considered to represent at least one Cryogenian glacial event.
Structural framework
The correlation between the Araçuaı́ orogen and its counterpart
located in southwestern Africa, the West Congo belt (Fig. 49.1b),
is well-established in the geological literature (e.g. Porada 1989;
Pedrosa-Soares et al. 1992, 2008; Trompette 1994; Cordani
et al. 2003; Alkmim et al. 2006). The Araçuaı́– West Congo oro-
genic system, as well as its precursor basin, developed inside an
embayment of the São Francisco –Congo craton (Fig. 49.1b), char-
acterizing a Neoproterozoic confined orogen (Pedrosa-Soares et al.
2001, 2008; Rogers & Santosh 2004). That precursor basin
evolved from a Tonian continental rift to an inland-sea basin (a
large gulf) that was partially floored by Cryogenian oceanic crust
along its southern sector, and was inverted during the late Neo-
proterozoic Brasiliano –Pan African orogenesis, resulting in the
Araçuaı́ and West Congo belts (Pedrosa-Soares et al. 2001,
2007, 2008; Tack et al. 2001; Alkmim et al. 2006, 2007).
This system was split up in two quite different but complemen-
tary counterparts, after the Atlantic Ocean opening in Cretaceous
time (Pedrosa-Soares et al. 2008). The West Congo belt inherited
the thick package of bimodal volcanic rocks of the early Tonian
continental rift, as well as rift and passive margin successions
from Tonian to late Cryogenian age, but no Neoproterozoic ophio-
lite sliver (Tack et al. 2001; Frimmel et al. 2006; Pedrosa-Soares
et al. 2008). The Araçuaı́ orogen inherited the basin sector,
lacking in rift-related magmatic rocks, but including the remnants
of Neoproterozoic oceanic lithosphere and the extensive Macaúbas
Group that includes rift, transitional and passive margin sediments
from late Tonian to late Cryogenian age (Pedrosa-Soares et al.
2007, 2008). Neoproterozoic diamictites occur in both counter-
parts of the Araçuaı́ –West Congo orogenic system; those
located in the West Congo belt are reviewed in Tait et al. (2011).
The Macaúbas Group is the most complete record of the precur-
sor basin within the Araçuaı́ orogen because it shows evidence of
the rift and passive margin stages. Normal faults and clastic dykes
associated with the sedimentation of the lower Macaúbas Group,
as well as late Tonian mafic dykes and anorogenic granites, the
main evidence of the rift-related extensional tectonics (Karfunkel
et al. 1976, 1977, 1984; Machado et al. 1989; Guimarães et al.
1993; Grossi-Sad et al. 1997a; Uhlein et al. 1988, 1999; Martins-
Neto et al. 2001; Martins-Neto & Hercos 2002; Knauer et al. 2006;
Martins 2006; Martins et al. 2008; Silva et al. 2008). Transitional
mafic volcanism occurred in the late continental rift stage (Gradim
et al. 2005). Transgressive sedimentary successions free of
diamictite and ophiolite slivers record the passive margin stage
and associated oceanic spreading that lasted at least until
c. 660 Ma (Pedrosa-Soares et al. 1998, 2001, 2008; Queiroga
et al. 2007).
The Macaúbas Group was deformed, metamorphosed and
intruded by orogenic granites during the Brasiliano orogeny that
lasted from c. 630 Ma (onset of the pre-collisional stage) to c.
490 Ma (end of the post-collisional stage) in the Araçuaı́ orogen
(Pedrosa-Soares et al. 2001, 2008; Silva et al. 2005).
The Macaúbas Group occurs in the west-verging fold-and-thrust
belt that characterizes the external (western) tectonic domain of the
Araçuaı́ orogen (Figs 49.1 & 49.2). Along that belt, the major
structures of the first deformational phase are thrust faults and
ductile shear zones, some of them representing the tectonic inver-
sion of normal faults related to the rift stage of the Macaúbas basin,
and west-verging tight asymmetric folds. These structures are
associated with an east-dipping regional foliation and a stretching
lineation (very well marked by stretched clasts in metadiamictites).
The kinematic indicators, such as fold vergence, S-C foliation and
asymmetric sigmoidal features (e.g. stretched sigma clasts in dia-
mictites) of this first deformational phase, record tectonic transport
towards the São Francisco craton. The regional metamorphism
related to the first deformational phase increases from west to
east so that the Macaúbas Group records the chlorite, biotite,
garnet, staurolite, kyanite and sillimanite zones of an intermediate
 NEOPROTEROZOIC MACAÚBAS GROUP 525

Fig. 49.2. Simplified geological map of the Araçuaı́ orogen (modified from Pedrosa-Soares et al. 2007, 2008). Section A– B shows structural and metamorphic features
of the Macaúbas Group.
526 A. C. PEDROSA-SOARES ET AL.

Fig. 49.3. Sketched map showing the distribution of the different formations of the Macaúbas Group in the external (western) tectonic domain of the Araçuaı́ orogen,
along with the regional stratigraphic scheme (modified from Pedrosa-Soares et al. 2007).

pressure metamorphic regime (Dossin & Dardenne 1984a;
Pedrosa-Soares et al. 1984, 1992, 1993, 1996, 2001, 2008; Alkmim
et al. 2006; Uhlein & Chaves 1989; Grossi-Sad et al. 1997a;
Uhlein et al. 1998; Marshak et al. 2006).
A west-dipping crenulation cleavage associated with asym-
metric folds and normal faults characterizes the second deforma-
tional phase (Pedrosa-Soares et al. 1993; Mourão & Grossi-Sad
1997; Oliveira et al. 1997a; Pedrosa-Soares & Grossi-Sad 1997;
Tupinambá & Grossi-Sad 1997; Alkmim et al. 2006). These struc-
tures have been associated with the gravitational collapse of the
Araçuaı́ orogen (Marshak et al. 2006). Two NE- and NW-trending,
high-angle-dip regional fracture systems associated with open
megafolds (flexures) characterize the third deformational
phase (Pedrosa-Soares et al. 1992; Grossi-Sad et al. 1997a),
which may also be related to the gravitational collapse of
the orogen.
 NEOPROTEROZOIC MACAÚBAS GROUP
Fig. 49.4. General stratigraphic scheme for the central sector (17–188S) of the
Macaúbas Group (not to scale; estimated thicknesses are referred to in the text
and Fig. 49.5; modified from Pedrosa-Soares et al. 2008).
In fact, all primary rocks of the Macaúbas Group were modified
by regional deformation and metamorphism of variable intensity.
However, the general low metamorphic grade (greenschist to inter-
mediate amphibolite facies) and the preservation of sedimentary
and volcanic structures in low strain zones allow the interpretation
of the pre-metamorphic lithology and other features, such that the
protoliths can be confidently described.
Stratigraphy
The regional stratigraphic scheme adopted here resulted from the
Espinhaço Project (a 1:100 000 scale geological mapping pro-
gramme that largely covered the Macaúbas Group; Grossi-Sad
et al. 1997a; Noce et al. 1997), with modifications inserted by
Lima et al. (2002), Pedrosa-Soares et al. (2007, 2008) and
Martins et al. (2008). The Macaúbas Group is subdivided into
seven formations, from oldest to youngest Matão, Duas Barras,
Rio Peixe Bravo, Serra do Catuni, Nova Aurora, Chapada Acauã
and Ribeirão da Folha (Fig. 49.3).
Two uplifted basement blocks divide the distribution of the
Macaúbas Group into three sectors with variably complete strati-
graphic records: the northern sector, related to the Porteirinha
block; the central sector, located between 178 and 188S; and the
southern sector, related to the Guanhães block (Fig. 49.3).
The most complete package of the Macaúbas Group occurs in
the central sector where the lowermost units are the diamictite-free
Matão and Duas Barras formations that are covered by the most
proximal and oldest diamictite-bearing unit of the group, the
Serra do Catuni Fm. (Figs 49.3 & 49.4). This formation passes
upward and eastward into the diamictite-bearing Lower Chapada
Acauã Fm., which, in turn, is covered by the diamictite-free
Upper Chapada Acauã Fm. This unit passes eastwards into the
diamictite-free Ribeirão da Folha Fm. (Fig. 49.4).
In the northern sector, only the Serra do Catuni Fm. occurs
between the São Francisco craton boundary and the western edge
of the Porteirinha block (Fig. 49.3). To the east of the Porteirinha
block the diamictite-free Rio Peixe Bravo Fm. is overlain by the
diamictite-bearing Nova Aurora Fm., which is partially capped
by the Upper Chapada Acauã Fm.
In the southern sector, the stratigraphic record of the Macaúbas
Group is very incomplete. The Serra do Catuni Formation,
locally overlain by strata that probably correlate to the Chapada
Acauã Formation, occurs along a narrow zone in the western
thrust front of the Araçuaı́ orogen (Figs 49.2 & 49.3). To the east
527
of the Guanhães block, only the Ribeirão da Folha Fm. occurs
(Figs 49.3 & 49.4).
The Matão, Duas Barras and Rio Peixe Bravo formations rep-
resent the oldest deposits of the Macaúbas Group (Figs 49.3 &
49.4). The Matão Fm. consists of breccia and conglomerate rich
in sandstone pebbles and cobbles, covered by sandstone with con-
glomerate lenses. Its maximum thickness is 200 m. An erosive
unconformity and normal faults define the contact between the
Matão Fm. and the basement (locally represented by eolian sand-
stone of the Statherian Espinhaço Supergroup). The Matão Fm.
records sedimentation under unstable tectonic conditions related
to the early rift stage of the Macaúbas basin (Martins 2006;
Martins et al. 2008). The Duas Barras Fm. consists of sandstone
and conglomeratic sandstone with variable contents of mica, feld-
spar, iron oxide and/or lithic fragments, quartz sandstone, con-
glomerate and rare pelite. It shows fluvial and shallow marine
sedimentary structures and bimodal (NW –SE and SE –NW)
palaeocurrent sets. The maximum thickness is approximately
100 m (Noce 1997a; Grossi-Sad et al. 1997b; Martins 2006).
The Rio Peixe Bravo Fm. consists of micaceous, ferruginous
and/or feldspathic sandstone, pelite locally rich in hematite and/
or graphite, and rare conglomerate, with a maximum thickness
of around 700 m (Viveiros et al. 1979; Mourão & Grossi-Sad
1997; Oliveira et al. 1997a; Roque et al. 1997; Knauer et al.
2006). The Duas Barras and Rio Peixe Bravo formations, which
lack evidence of glacial influence, represent fluvial to marine sedi-
mentation during the continental rift stage of the Macaúbas basin
(Noce et al. 1997; Uhlein et al. 1998, 2007; Martins-Neto et al.
2001; Martins 2006; Pedrosa-Soares et al. 2007, 2008).
The Serra do Catuni Fm., the oldest diamictitic unit of the
Macaúbas Group, is a package of massive diamictite with minor
sandstone and rare pelitic rhythmite lenses (Figs 49.3 & 49.4).
This formation is locally rich in chaotic boulders, and contains
faceted and striated flat-iron-shaped cobbles and pebbles (Karfun-
kel & Karfunkel 1976, 1977; Karfunkel et al. 1984; Karfunkel &
Hoppe 1988; Pedrosa-Soares et al. 1992, 2008; Noce et al. 1997;
Uhlein et al. 1999, 2007; Martins-Neto et al. 2001; Martins-Neto &
Hercos 2002; Martins 2006).
The Nova Aurora Fm. mainly comprises stratified diamictite,
ferruginous diamictite-bearing and diamictitic Fe formation, with
minor intercalations of sandstone with fining-up graded bedding
and rare pelite (Viveiros et al. 1979; Pedrosa-Soares et al. 1992;
Mourão & Grossi-Sad 1997; Noce et al. 1997; Oliveira et al.
1997a; Pedrosa-Soares & Oliveira 1997; Roque et al. 1997;
Uhlein et al. 1998, 1999; Drumond 2000).
The Chapada Acauã Fm. includes a lower diamictitic unit and
an upper diamictite-free unit (Figs 49.3 & 49.4). The Lower
Chapada Acauã Fm. essentially consists of stratified diamictite,
graded sandstone, pelite and mafic volcanic rocks (Hettich 1977;
Mourão & Pedrosa-Soares 1992; Pedrosa-Soares et al. 1992,
2007, 2008; Mourão & Grossi-Sad 1997; Oliveira et al. 1997a, b;
Tupinambá & Grossi-Sad 1997; Uhlein et al. 1999; Gradim et al.
2005; Martins 2006). At least one thick carbonate lens (c. 20 m)
occurs intercalated in a pelite layer with isolated outsized clasts at
the top of the Lower Chapada Acauã Fm. (Pedrosa-Soares & Grossi-
Sad 1997). The mafic volcanic rocks, metamorphosed to green-
schists, show pillow structures and other features of sub-aqueous
flows (Schrank et al. 1978; Chula et al. 1996; Uhlein et al. 1998;
Gradim et al. 2005). They have tholeiitic basalt protoliths with a
dominant within-plate signature (Gradim et al. 2005), Sm –Nd
TDM model age of c. 1.5 Ga, and inherited zircon grains with det-
rital features and U – Pb SHRIMP ages from Archaean to Late
Mesoproterozoic (Babinski et al. 2005). However, samples with
oceanic signature and slightly positive 1Nd(900 Ma) of þ0.23,
together with the inherited zircon grains, suggest a transitional
mafic magma that migrated through a thinned continental crust
(Babinski et al. 2005; Gradim et al. 2005). Accordingly, these
greenschists provide strong evidence of volcanism in an exten-
sional marine basin floored by thin continental crust, during the
528 A. C. PEDROSA-SOARES ET AL.
Fig. 49.5. Composite stratigraphic columns for the Serra do Catuni, Nova
Aurora and Lower Chapada Acauã formations (Macaúbas Group). The Serra do
Catuni column is based on sections described by Martins (2006) along the
Macaúbas River valley (Fig. 49.2). The Nova Aurora column synthesizes
field and drill holes data from the region limited by a white rectangle in
Figure 49.3. The column for the Lower Chapada Acauã Fm. is based on field
data from the eastern escarpments of the Acauã Plateau (Fig. 49.3) and medium
valley of the Araçuaı́ River (Fig. 49.2), as well as on data from the Macaúbas
River valley.
transitional phase from the late rift to early passive margin stages
of the Macaúbas basin.
Both the Nova Aurora and Lower Chapada Acauã formations
show cyclic deposition from coarse- to fine-grained sediments,
fining-up graded bedding, erosional contacts between cycles and
load structures (Mourão & Pedrosa-Soares 1992; Pedrosa-Soares
et al. 1992; Noce et al. 1997; Pedrosa-Soares & Oliveira 1997;
Uhlein et al. 1999, 2007; Martins-Neto et al. 2001). The features
of these formations, as well as of the Serra do Catuni Fm., are
described and interpreted in subsequent sections.
The Upper Chapada Acauã Fm. is a succession of interbedded
sandstone and pelite, free of diamictite, interpreted as a post-
glacial unit deposited in a shelf environment during the passive
margin stage of the Macaúbas basin (Pedrosa-Soares et al.
1992; Mourão & Grossi-Sad 1997; Noce et al. 1997; Oliveira
et al. 1997a; Pedrosa-Soares & Grossi-Sad 1997; Martins-Neto
et al. 2001).
The Ribeirão da Folha Fm. is free of diamictite and includes
distal passive margin and ocean floor deposits (Figs 49.2, 49.3
& 49.4). The proximal part of the Ribeirão da Folha Fm. is a suc-
cession of fine-grained turbidites and very rare limestone lenses.
It overlies the diamictite-turbidite unit of the Lower Chapada
Acauã Fm. and is interpreted as a deep-water correlative of the
shelf sandstone-pelite pile of the Upper Chapada Acauã Fm.
(Pedrosa-Soares et al. 1992, 2007, 2008; Pedrosa-Soares &
Grossi-Sad 1997). The distal part of the Ribeirão da Folha For-
mation is locally associated with thrust slices of oceanic meta-
mafic and meta-ultramafic rocks, constituting a tectonically
dismembered ophiolite complex (Pedrosa-Soares et al. 1992,
1998, 2001, 2008; Queiroga et al. 2007). The distal Ribeirão
da Folha Fm. is rich in peraluminous and carbonaceous pelites
with intercalations of volcanic-exhalative deposits (sulphide-
bearing cherts, massive sulphide bodies, and banded iron for-
mations of oxide, silicate and sulphide types) and mafic volcanic
rocks with ocean-floor geochemical and isotopic (1Nd(660 Ma) of
c. þ4) signatures (Pedrosa-Soares et al. 1998, 2001, 2008; Queir-
oga et al. 2006, 2007).
Glaciogenic deposits and associated strata
The Macaúbas Group includes three diamictite-bearing units that
have been interpreted as glaciogenic deposits: the Serra do
Catuni, Nova Aurora and Lower Chapada Acauã formations
(Figs 49.3, 49.4 & 49.5). The rock-assemblage included in the
Serra do Catuni Fm. was mapped and/or studied in detail by Kar-
funkel & Karfunkel (1976, 1977), Dossin & Dardenne (1984a),
Karfunkel et al. (1984), Pedrosa-Soares et al. (1984, 1992, 2007,
2008), Uhlein & Chaves (1989), Fogaça (1997), Grossi-Sad et al.
(1997b), Guimarães (1997), Guimarães et al. (1997), Knauer &
Grossi-Sad (1997), Mourão et al. (1997), Noce (1997a, b), Oliveira
et al. (1997b), Tupinambá & Grossi-Sad (1997), Uhlein et al. (1998,
1999, 2007), Knauer et al. (2006) and Martins (2006). The Serra do
Catuni Fm. is a very extensive and rather homogeneous unit that
persistently occurs for over than 400 km along the western
border of the Araçuaı́ orogen (Fig. 49.3). This formation is rich
in massive diamictite, the thickness of which ranges from a few
tens of metres up to 300 m (Fig. 49.5). The following description
is based on sections in the Macaúbas River valley (Fig. 49.2),
where the Serra do Catuni Fm. is reasonably well preserved
from deformation and erosion.
The Serra do Catuni Fm. overlies the Duas Barras Formation
and contains a 250-m-thick pile of massive diamictite with
lenses of massive sandstone (Martins 2006). The diamictite has
an abundant (60 –80% in volume), poorly sorted, muddy-
sandstone matrix composed of detrital quartz, K-feldspar and
carbonate, with a metamorphic foliation marked mainly by fine-
grained muscovite. The clasts, ranging in size from granules to
boulders, are also poorly sorted in texture and composition
 NEOPROTEROZOIC MACAÚBAS GROUP
(milky quartz, sandstone, granitoid, carbonate and mafic rock).
Faceted and striated flat-iron-shaped cobbles and pebbles can
also be found in the Serra do Catuni diamictite. Lenses of
massive sandstone up to 2 m thick appear mainly in the upper
part of the section. This sandstone is poorly sorted and consists
of quartz with minor K-feldspar, carbonate and iron oxides
(Martins 2006).
The rock-assemblage included in the Nova Aurora Fm.
was mapped and/or studied in detail by Viveiros et al. (1979),
Pedrosa-Soares et al. (1984, 1992, 2000, 2007, 2008), Mourão &
Grossi-Sad (1997), Oliveira et al. (1997a), Pedrosa-Soares &
Oliveira (1997), Roque et al. (1997), Uhlein et al. (1998, 1999),
Drumond (2000) and Knauer et al. (2006). The Nova Aurora
Fm. is well exposed in road cuts to the west of Salinas town
(Fig. 49.3), and part of its stratigraphic pile is also known in drill
holes for iron ore prospecting (Viveiros et al. 1979; Uhlein et al.
1999; Vilela 2010). Viveiros et al. (1979) estimated the
minimum thickness for the Nova Aurora Fm. to be c. 1200 m
based on detailed sections and data from drill holes. Again, the
strong regional deformation hampers any accurate thickness evalu-
ation for the whole unit, which has independently been estimated
to be thicker than 2 km (e.g. Grossi-Sad et al. 1997a; Uhlein
et al. 1998, 1999).
The following description is based mainly on detailed studies
carried out by Viveiros et al. (1979), Pedrosa-Soares et al.
(1984, 1992), Pedrosa-Soares & Oliveira (1997) and Vilela
(2010) in the region west of Salinas (white rectangle in
Fig. 49.3). The Nova Aurora Fm. is essentially a package of dia-
mictites with minor intercalations of graded sandstone and
scarce pelite, with the most striking feature being the great
amount of Fe-rich diamictite. The lower Nova Aurora Fm. is
dominated by Fe-rich diamictite followed by non-ferruginous
diamictite (Fig. 49.5). Despite the Fe content, the diamictites
vary considerably in grain size, matrix composition and clast/
matrix ratio. They are metamorphosed to greenschist facies and
their clasts are mainly of milky to blue quartz, metasandstone,
carbonate, granitoid, gneiss and metapelite. The Fe-rich
diamictite package consists of ferruginous diamictite (hematite þ
magnetite .5% in volume) and diamictitic Fe formation (Fe
.15 wt%), with minor intercalations of sulphide-bearing
diamictite, ferruginous and non-ferruginous sandstone, hema-
tite – magnetite schist and pelite with scattered outsized clasts.
The foliated matrix of the Fe-rich diamictite consists of quartz,
hematite, magnetite, muscovite, biotite and chlorite. The amount
of magnetite and the total Fe content (in excess of 40% in
places) increases in shear zones. A thin carbonate intercalation,
recently found in a drill hole, occurs in the upper section of the
Fe-rich diamictite. Layers of sulphide-bearing diamictite occur at
the base and top of the Fe-rich diamictite package. The sulphides
are mainly pyrite and pyrrhotite. The upper package consists of
stratified diamictite with minor layers and lenses of fining-up
graded sandstone and thin intercalations of pelite. This diamictite
has a foliated matrix composed mainly of quartz, biotite, musco-
vite, garnet and carbonate. The arenaceous intercalations vary in
composition from immature (micaceous, feldspathic, ferruginous)
to quartz sandstone, and commonly show load structures. Ero-
sional contacts can be observed between diamictite and sandstone
intercalations.
The rock-assemblage included in the Lower Chapada Acauã Fm.
was mapped and/or studied in detail by Hettich (1975, 1977),
Karfunkel & Karfunkel (1976, 1977), Schrank et al. (1978),
Uhlein & Chaves (1989), Mourão & Pedrosa-Soares (1992),
Pedrosa-Soares et al. (1992, 1993, 2007, 2008), Chula et al.
(1996), Grossi-Sad et al. (1997b), Mourão & Grossi-Sad (1997),
Noce (1997b), Oliveira et al. (1997a), Pedrosa-Soares & Grossi-
Sad (1997), Tupinambá & Grossi-Sad (1997), Uhlein et al.
(1998, 1999, 2007), Gradim et al. (2005) and Martins (2006).
The Lower Chapada Acauã Fm. is well exposed along the
escarpments of the Acauã Plateau (Fig 49.3) and waterways in
529
the southwestern Araçuaı́ and Jequitinhonha valleys (Fig. 49.2).
Detailed sections carried out along creeks of the medium valley
of the Araçuaı́ River suggest a local minimum thickness of
800 m for the Lower Chapada Acauã Fm. (Mourão & Pedrosa-
Soares 1992; Pedrosa-Soares & Grossi-Sad 1997). However, the
strong regional deformation hampers any accurate thickness
evaluation for the whole unit, which seems to be thicker than
2 km (e.g. Grossi-Sad et al. 1997a; Uhlein et al. 1998, 1999).
The following description is based on detailed studies carried
out in the middle valley of the Araçuaı́ River, the type-area
of the Lower Chapada Acauã Fm. (Mourão & Pedrosa-Soares
1992; Pedrosa-Soares et al. 1992; Pedrosa-Soares & Grossi-Sad
1997), and in the Macaúbas River valley (Martins 2006). The
Lower Chapada Acauã Fm. has many vertical and lateral variations
of lithofacies. It is essentially composed of fining-up cycles of
stratified diamictite, graded sandstone and pelitic rhythmite
(Fig. 49.5). The diamictite layers show great variations in clast/
matrix ratios and the foliated matrix consists mainly of quartz,
feldspar, muscovite, chlorite, biotite and garnet (according to the
metamorphic zone). The clasts vary from granules to boulders
and are composed of milky and blue quartz, metasandstone, car-
bonate, metapelite, granitoid, gneiss and meta-mafic rock. The dia-
mictite layers envelope lenses of cross-laminated sandstone,
ranging in length from decimetres to a few metres and up to 1 m
thick. At the top of the cycles the diamictite gives way to sandstone
and pelite. Erosional contacts between cycles are common. The
sandstone layers and lenses show fining-up graded bedding
and vary in composition from wacke-rich in carbonate + feldspar
+ lithoclasts + micas to quartz sandstone. Conglomerate +
conglomeratic sandstone can be present at the base of the sand-
stone layers and lenses. The pelitic rhythmite layers locally show
isolated outsized clasts, mainly cobbles and boulders. These
layers have mineralogical compositions similar to the diamictite
matrix, with variable contents of metamorphic minerals, such as
quartz, micas and garnet, in alternating laminae and bands.
Boundary relations with overlying and underlying
non-glacial units
The Serra do Catuni Fn. was thrust over the Neoproterozoic pelite-
carbonate cratonic cover (Fig. 49.2), outlining a clear limit
between the Araçuaı́ orogen and São Francisco craton (Almeida
1977; Uhlein et al. 1998; Alkmim et al. 2006; Pedrosa-Soares
et al. 2008).
A regional disconformity occurs between the diamictite-
free Duas Barras Fm. and the overlying Serra do Catuni Fm.,
the oldest diamictitic unit of the Macaúbas Group (Karfunkel
& Hoppe 1988; Grossi-Sad et al. 1997a; Uhlein et al. 1999,
2007; Martins-Neto et al. 2001; Martins 2006). Locally, the
Serra do Catuni diamictite shows erosional contact with
the Duas Barras Fm., but normal and tectonic (sheared) contacts
are also observed (Grossi-Sad et al. 1997b; Noce 1997a;
Martins 2006).
The Serra do Catuni Fm. gradually passes upwards and east-
wards to the interlayered succession of diamictite, sandstone,
pelite and mafic volcanic rocks that characterizes the Lower
Chapada Acauã Fm. (Fig. 49.4). Intraformational erosive contacts
between diamictite and sand-pelite turbidites of both the Lower
Chapada Acauã and Nova Aurora formations can be locally
observed (Pedrosa-Soares et al. 1992, 2008; Pedrosa-Soares &
Oliveira 1997; Pedrosa-Soares & Grossi-Sad 1997; Uhlein et al.
1998, 1999).
In the Araçuaı́ River valley and eastern escarpments of the
Acauã Plateau (Figs 49.2 & 49.3), the Lower Chapada Acauã
Fm. gradually passes upwards to the diamictite-free package of
the Upper Chapada Acauã Fm. To the east, this sandstone-
pelite unit transitions to the fine-grained turbidite package of
530 A. C. PEDROSA-SOARES ET AL.
the proximal Ribeirão da Folha Fm. (Pedrosa-Soares et al. 1992,
2008; Pedrosa-Soares & Grossi-Sad 1997; Uhlein et al.
1999, 2007).
In the northern sector of the Macaúbas Group, the diamictite-
free Upper Chapada Acauã Formation conformably covers the
Nova Aurora Formation (Viveiros et al. 1979; Roque et al.
1997; Mourão & Grossi-Sad 1997; Oliveira et al. 1997b; Pedrosa-
Soares & Oliveira 1997).
Chemostratigraphy
No chemostratigraphic study is available for the Macaúbas Group,
but there are isotopic data for older carbonate units located in the
Araçuaı́ orogen and for carbonate clasts extracted from the Serra
do Catuni diamictite. Santos et al. (2004) presented C-isotopic
data for metadolomite and metamarlstone samples of the Domin-
gas Fm., formerly interpreted as the oldest unit of the Macaúbas
Group (Grossi-Sad et al. 1997b; Noce 1997a). In fact, the
shallow marine dolomite-pelite package of the Domingas Fm. is
unconformably covered by a fluvial sandstone-conglomerate
succession of the Duas Barras Fm. (age , 900 + 21 Ma, see
‘Geochronological constraints’) and contains the stromatolite
Conophyton metula Kirichenko that suggests a Mesoproterozoic
age (Schöll 1976). The stromatolitic dolomite samples of the Dom-
ingas Formation yielded d13CPDB values from þ0.4 to þ0.7‰,
similar to other Mesoproterozoic dolostones in the region (the
Rio Pardo Grande Fm.; Santos et al. 2004). Other metadolomite
samples of the Domingas Fm. yielded d13CPDB values from þ0.2
to –1.9‰ and one sample of metamarlstone is as low as –4.0‰
(Santos et al. 2004).
Martins (2006) reported d13CPDB values ranging – 7.2 to 0‰
obtained on 14 carbonate pebbles extracted from the Serra do
Catuni diamictite. These data are similar to those obtained from
carbonate clasts of the Jequitaı́ diamictite (d13CPDB from – 0.6 to
–3.1‰) and from some samples of the Domingas Fm. by Santos
et al. (2004).
Other characteristics
Manganese deposits have been mined in the southern and central
sectors of the Macaúbas Group. They consist of manganese
oxides formed by Cenozoic lateritic alteration and supergene con-
centration on pelitic horizons of the Chapada Acauã Fm. (Dossin &
Dardenne 1984b).
The thick carbonate lens of the Lower Chapada Acauã Fm.
(Figs 49.4 & 49.5) has been sporadically mined (Pedrosa-Soares
& Grossi-Sad 1997).
The lower part of the Nova Aurora Fm. is rich in ferruginous
diamictite and diamictitic Fe formation that form deposits up to
200 m thick, with Fe content up to 40 wt% (Viveiros et al. 1979;
Vilela 2010). These deposits are widely capped by thick lateritic
soils on the plateaus that characterize the northern sector of the
Macaúbas Group. However, new aeromagnetic data suggest
significant Fe-rich diamictite deposits that are the target of
current exploration.
Martins (2006) suggested that the Serra do Catuni diamictite is
the source for diamonds found in alluvial deposits along
watercourses of the Macaúbas River valley (Fig. 49.2).
Palaeolatitude and palaeogeography
No palaeomagnetic data are available for the Macaúbas Group.
However, palaeomagnetic data and palaeogeographic settings for
the São Francisco–Congo craton in late Mesoproterozoic and
Neoproterozoic times have been discussed in several papers
(e.g. Cordani et al. 2003; D’Agrella-Filho et al. 2004).
Geochronological constraints
Although there are abundant data from detrital zircon grains
extracted from rocks of the Matão, Duas Barras, Serra do Catuni
and Chapada Acauã formations, and from intrusions related to dis-
tinct extensional stages of the precursor basin of the Araçuaı́
orogen, these data yield few useful constraints on the timing of
glacial sedimentation in the Macaúbas Group. The best maximum
age constraint on the onset of glacial sedimentation in the Macaú-
bas Group comes from a detrital study of the Duas Barras Fm., the
lowermost stratigraphic unit, where the youngest zircon has an age
of 900 + 21 Ma (Babinski et al. 2011). The youngest concordant
zircon grain in the Upper Chapada Acauã Fm. provides a
maximum depositional age of 864 + 30 Ma (Pedrosa-Soares
et al. 2000).
The Pedro Lessa mafic dykes and Salto da Divisa granite
(Fig. 49.2) are considered to be related to the continental rift
stage of the precursor basin of the Araçuaı́ orogen (Pedrosa-Soares
et al. 2008; Silva et al. 2008). The meta-dolerite dyke swarm of
Pedro Lessa was dated at 906 + 2 Ma (zircon and baddeleyite,
U –Pb TIMS; Machado et al. 1989). The fluorite-bearing anoro-
genic Salto da Divisa granite yielded a zircon U –Pb SHRIMP
age of 875 + 9 Ma (Silva et al. 2008).
The maximum age of deep sea sedimentation for the distal
Ribeirão da Folha Fm. is constrained by the age of 660 + 29 Ma
(U –Pb LA-ICPMS), obtained for the magmatic crystallization of
euhedral zircon crystals extracted from meta-plagiogranite
patchs found in ophiolitic meta-mafic rocks (Queiroga et al.
2007). The entire Macaúbas must be older than the regional meta-
morphism, which is dated at c. 580 Ma (Pedrosa-Soares et al. 2001,
2008; Silva et al. 2005).
Discussion
The Macaúbas Group includes three widespread diamictitic units,
the Serra do Catuni, Nova Aurora and Lower Chapada Acauã for-
mations (Figs 49.3, 49.4 & 49.5). The Serra do Catuni Fm. has been
interpreted as a glaciogenic unit deposited mostly in a proximal
marine environment based on multiple observations. First, it com-
prises massive diamictite occurring for over 400 km along the
proximal region of the Macaúbas basin (Fig. 49.3). Second, it con-
tains zones rich in chaotic boulders, faceted and striated cobbles
and pebbles, and only scarce finer-grained sediments. Third, it is
correlated with the glacio-terrestrial deposits of the Jequitaı́ Fm.
located on the little deformed São Francisco cratonic cover
sequence to the west.
The Nova Aurora and Lower Chapada Acauã formations also
occur in extensive areas, mainly to the east of the Serra do
Catuni Fm., along a more distal region of the Macaúbas basin
(Fig. 49.3). They essentially consist of stratified diamictite, sand-
stone and pelite, and show sedimentary features such as cyclic
deposition from coarse- to fine-grained sediments, graded bedding,
erosional contacts between cycles and load structures that are evi-
dence of debris flow and turbiditic sedimentation in a distal glacio-
marine environment. The clear relationship between the Serra do
Catuni and Lower Chapada Acauã formations also points to a tran-
sition from proximal to distal glaciomarine conditions, respect-
ively (Figs 49.4 & 49.5). The Nova Aurora Fm., which overlies
the pre-glacial Rio Peixe Bravo Fm., can be interpreted as a
distal correlative of the Serra do Catuni Fm. and, to some extent,
a lateral equivalent of the Lower Chapada Acauã Fm.
Based on the occurrence of glaciogenic diamictites, the Macaú-
bas Group is subdivided into a pre-glacial diamictite-free succes-
sion (Matão, Duas Barras and Rio Peixe Bravo formations), a
glaciogenic succession (Serra do Catuni, Nova Aurora and Lower
Chapada Acauã formations) and a post-glacial succession
(Upper Chapada Acauã and Ribeirão da Folha formations).
These successions are related to different evolutionary stages of
 NEOPROTEROZOIC MACAÚBAS GROUP
the Macaúbas basin (Fig. 49.3). The pre-glacial and glaciogenic
successions have been ascribed to the continental rift to transi-
tional stages of the Macaúbas basin based on multiple lines of
evidence:
† the predominance of immature sandstone-conglomeratic sedi-
mentation with bimodal (NW – SE and SE –NW) palaeocurrent
sets in the pre-glacial succession;
† syn-sedimentary normal faulting;
† clastic dykes related to the rudite-sandstone sedimentation of
the pre-glacial succession and Serra do Catuni Fm.;
† the huge ferruginous diamictite deposits of the Nova Aurora
Fm.;
† the transitional geochemical and isotopic signatures of the
Chapada Acauã submarine basalts that also carry inherited
zircon grains from continental sources.
The post-glacial succession represents the passive margin stage of
the Macaúbas basin, when sea-floor spreading continued at least
until c. 660 Ma.
The onset of a rifting stage in the region is constrained by
the U –Pb ages of the Pedro Lessa mafic dykes (c. 906 Ma;
Fig. 49.2) and Salto da Divisa anorogenic granites (c. 875 Ma;
see ‘Geochronological constraints’). Indeed, these anorogenic
magmatic episodes provide solid evidence that an active continen-
tal rift developed during the late Tonian (900 –850 Ma) in this
region (Fig. 49.2), and it is certain that at least the pre-glacial
strata were deposited during this rift stage. The Serra do Catuni
Fm. and the cratonic Jequitaı́ Fm. have also been assigned to the
continental rift stage (Pedrosa-Soares et al. 1992, 2008; Uhlein
et al. 1998, 1999; Martins-Neto et al. 2001; Martins-Neto &
Hercos 2002). However, if the regional disconformity between
the pre-glacial and glaciogenic successions represents a significant
hiatus (Fig. 49.3; see ‘Boundary relations’), the Serra do Catuni
Fm. could be early Cryogenian in age (Uhlein et al. 2007).
With regards to the age of glaciation, the cratonic Jequitaı́ and
Carrancas diamictites located in the southern sector of the
nearby São Francisco craton (Fig. 49.1a) should be considered.
Although the Macaúbas, Jequitaı́ and Carrancas diamictites show
no direct field relation, they have been correlated with each other
since the 1930s, owing to their shared glacial origin. The
maximum sedimentation ages for the Jequitaı́ Fm. (c. 880 Ma;
Rodrigues 2008) and the oldest Macaúbas diamictites (c. 900 Ma;
see ‘Geochronological constraints’) are similar, but they do not
provide any precise information on the glaciation age. On the
other hand, the Carrancas diamictite can be correlated to an early
Cryogenian glacial event because it is covered by the c. 740 Ma
cap carbonate of the cratonic cover (Babinski et al. 2007, 2011).
In order to fully understand the timing and nature of glaciation
on the São Francisco craton, the entire Araçuaı́ –West Congo oro-
genic system and its precursor basin should also be considered
(Fig. 49.1b). Because these now rifted margins were juxtaposed
prior to full opening of the Macaúbas basin, the Brazilian and
African counterparts of the precursor basin must have lain in
similar palaeoclimatic zones (e.g. Cordani et al. 2003; Pisarevsky
et al. 2003). In the West Congo belt, the Lower Mixtite Fm. is the
diamictite unit that has been related to the early Cryogenian glacial
event (Trompette 1994; Tack et al. 2001; Frimmel et al. 2006).
Pedrosa-Soares et al. (2008) correlated the Lower Mixtite and
Lower Chapada Acauã formations because they include glacio-
marine diamictites with intercalations of transitional submarine
basalts, representing a late rift stage. These authors also correlated
the Lower Chapada Acauã and Lower Mixtite formations with the
Carrancas diamictite, and ascribed them to the early Cryogenian
glacial event. The Upper Mixtite Fm. in the West Congo Belt,
which is ascribed to the late Cryogenian glacial event (Frimmel
et al. 2006), does not appear to have a counterpart in the Macaúbas
Group (Pedrosa-Soares et al. 2008).
531
In conclusion, it seems that the Macaúbas Group records at least
the early Cryogenian glacial event. The Serra do Catuni Fm. and its
probable correlative, the Jequitaı́ Fm., were likely deposited during
an early phase of that glacial event. The Lower Chapada Acauã
Fm. and its probable equivalents, the Carrancas diamictite and
the Lower Mixtite Fm. then represents a late phase of the early
Cryogenian glacial event. Indeed, the very extensive and thick dia-
mictitic package of the Macaúbas Group might contain equivalents
of all the Neoproterozoic glaciogenic units preserved on the São
Francisco craton. The challenging question is how many discrete
glacial epochs do they represent?
The authors acknowledge financial support provided by CNPq (Conselho Nacio-
nal de Desenvolvimento Cientı́fico e Tecnológico), FINEP/CT-Infra (Financia-
dora de Estudos e Projetos), FAPEMIG (Fundação de Amparo à Pesquisa de
Minas Gerais) and FAPESP (Fundação de Amparo à Pesquisa do Estado de
São Paulo). We also thank W. Altermann, M. A. Martins-Neto and an anonymous
reviewer, as well as Emmanuelle Arnaud, for corrections and suggestions. We
dedicate this work to our beloved friend and coauthor – the late Carlos Maurı́cio
Noce – who contributed very significantly to understanding the Macaúbas basin.
This represents a contribution of the IUGS- and UNESCO-funded IGCP (Inter-
national Geoscience Programme) Project #512.
References
Alkmim, F. F., Marshak, S., Pedrosa-Soares, A. C., Peres, G. G.,
Cruz, S. C. & Whittington, A. 2006. Kinematic evolution of the
Araçuaı́-West Congo orogen in Brazil and Africa: nutcracker tec-
tonics during the Neoproterozoic assembly of Gondwana. Precam-
brian Research, 149, 43 –63.
Alkmim, F. F., Pedrosa-Soares, A. C., Noce, C. M. & Cruz, S. C. P.
2007. Sobre a evolução tectônica do Orógeno Araçuaı́-Congo
Ocidental. Geonomos, 15, 35 – 43.
Almeida, F. F. M. 1977. O Cráton do São Francisco. Revista Brasileira de
Geociências, 7, 349–364.
Babinski, M., Gradim, R. J., Pedrosa-Soares, A. C., Alkmim, F. F.,
Noce, C. M. & Liu, D. 2005. Geocronologia U– Pb (SHRIMP) e
Sm–Nd de xistos verdes basálticos do Orógeno Araçuaı́: implicações
para a idade do Grupo Macaúbas. Revista Brasileira de Geociências,
35 (4-supplement), 77– 81.
Babinski, M., Vieira, L. C. & Trindade, R. I. F. 2007. Direct dating of the
Sete Lagoas cap carbonate (Bambuı́ Group, Brazil) and implications
for the Neoproterozoic glacial events. Terra Nova, 19, 401 –406.
Babinski, M., Pedrosa-Soares, A. C., Trindade, R. I. F., Martins, M.,
Noce, C. M. & Liu, D. 2011. Neoproterozoic glacial deposits from
the Araçuaı́ orogen, Brazil: age, provenance and correlations with
the São Francisco craton and West Congo belt. Gondwana Research,
doi: 10.1016/j.gr.2011.04.008.
Chula, A. M., Knauer, L. G. & Almeida-Abreu, P. A. 1996. Estratigra-
fia do Supergrupo Espinhaço na região de Planalto de Minas, Diaman-
tina, MG. Geonomos, 3, 69 –81.
Cordani, U. G., Brito-Neves, B. B., D’agrella-Filho, M. S. & Trin-
dade, R. I. F. 2003. Tearing-up Rodinia: the Neoproterozoic paleo-
geography of South American cratonic fragments. Terra Nova, 15,
343– 349.
D’Agrella-Filho, M. S., Pacca, I. I. G., Trindade, R. I. F., Teixeira,
W., Raposo, M. I. B. & Onstott, T. C. 2004. Paleomagnetism and
40
Ar39Ar ages of mafic dykes from Salvador, Brazil: new constraints
on the São Francisco craton APW path between 1080 and 1010 Ma.
Precambrian Research, 132, 55– 77.
Dossin, I. A. & Dardenne, M. A. 1984a. Geologia da borda ocidental da
Serra do Cipó, Minas Gerais. In: Congresso Brasileiro de Geologia
33, Rio de Janeiro, Anais, 7. Sociedade Brasileira de Geologia,
3104–3117.
Dossin, I. A. & Dardenne, M. A. 1984b. Os depósitos supergênicos de
manganês da borda ocidental da Serra do Cipó, MG (Quadrı́cula
Inhame). In: Congresso Brasileiro de Geologia 33, Rio de Janeiro,
Anais, 3. Sociedade Brasileira de Geologia, 1129–1143.
Drumond, J. B. V. 2000. Folhas Cordeiros, Belo Campo e Curral de
Dentro. CODEMIG, Projeto Leste, Belo Horizonte.
532 A. C. PEDROSA-SOARES ET AL.
Fogaça, A. C. C. 1997. Geologia da Folha Diamantina. In: Grossi-Sad,
J. H., Lobato, L. M., Pedrosa-Soares, A. C. & Soares-Filho,
B. S. (eds) Projeto Espinhaço. CODEMIG, Belo Horizonte,
1575– 1666.
Frimmel, H. E., Tack, L., Basei, M., Nutman, A. & Boven, A. 2006.
Provenance and chemostratigraphy of the Neoproterozoic West
Congolian Group in the Democratic Republic of Congo. Journal
of African Earth Sciences, 46, 221–239.
Gradim, R. J., Alkmim, F. F., Pedrosa-Soares, A. C., Babinski, M. &
Noce, C. M. 2005. Xistos verdes do Alto Araçuaı́, Minas Gerais:
vulcanismo básico do rifte neoproterozóico Macaúbas. Revista
Brasileira de Geociências, 35 (4-suplemento), 59 –69.
Gravenor, C. P. & Monteiro, R. 1983. Ice-thrust features in the Proter-
ozoic Macaúbas Group, Jequitaı́ area, Minas Gerais, Brazil. Journal
of Geology, 91, 113– 116.
Grossi-Sad, J. H., Lobato, L. M., Pedrosa-Soares, A. C. &
Soares-Filho, B. S. 1997a. Projeto Espinhaço. CODEMIG, Belo
Horizonte.
Grossi-Sad, J. H., Roque, N. C., Knauer, L. G., Noce, C. M. &
Fonseca, E. 1997b. Geologia da Folha Carbonita. In: Grossi-Sad,
J. H., Lobato, L. M., Pedrosa-Soares, A. C. & Soares-Filho,
B. S. (eds) Projeto Espinhaço. CODEMIG, Belo Horizonte,
1251–1372.
Guimarães, M. L. V. 1997. Geologia da Folha Botumirim. In: Grossi-
Sad, J. H., Lobato, L. M., Pedrosa-Soares, A. C. & Soares-Filho,
B. S. (eds) Projeto Espinhaço. CODEMIG, Belo Horizonte,
543– 610.
Guimarães, M. L., Crocco-Rodrigues, F. A., Abreu, F. R., Oliveira,
O. A. B. & Greco, F. M. 1993. Geologia do bloco Itacambira-Monte
Azul entre Barrocão e Porteirinha, MG. Boletim do Núcleo Minas
Gerais-Sociedade Brasileira de Geologia, 12, 74– 78.
Guimarães, M. L. V., Grossi-Sad, J. H. & Fonseca, E. 1997. Geologia da
Folha Francisco Sá. In: Grossi-Sad, J. H., Lobato, L. M., Pedrosa-
Soares, A. C. & Soares-Filho, B. S. (eds) Projeto Espinhaço.
CODEMIG, Belo Horizonte, 223– 313.
Hettich, M. 1975. Zur stratigraphie und Genese des Macaúbas nordlich
der Serra Negra, Espinhaço-Zone, Minas Gerais, Brasilien. Geolo-
gisches Jahrbuch, 14, 47 –85.
Hettich, M. 1977. A glaciação proterozóica no centro-norte de Minas
Gerais. Revista Brasileira de Geociências, 7, 87– 101.
Hettich, M. & Karfunkel, J. 1978. Um esker, um varvito e seixos
estriados no Grupo Macaúbas, norte de Minas Gerais. Revista
Escola de Minas de Ouro Preto, 34, 5– 8.
Isotta, C. A. L., Rocha-Campos, A. C. & Yoshida, R. 1969. Striated
pavement of the Upper Precambrian glaciation in Brazil. Nature,
222, 466– 468.
Karfunkel, J. & Karfunkel, B. 1976. Estudos petro-faciológicos do
Grupo Macaúbas na porção mediana da Serra do Espinhaço, Minas
Gerais. In: Congresso Brasileiro de Geologia 29, Ouro Preto,
Anais, 2. Sociedade Brasileira de Geologia, 179 – 188.
Karfunkel, J. & Karfunkel, B. 1977. Fazielle Entwicklung der mittle-
ren Espinhaço-Zone mit besonderer Berücksichtigung des Tillit-
Problems (Minas Gerais, Brasilien). Geologisches Jahrbuch, 24,
3 –91.
Karfunkel, J. & Hoppe, A. 1988. Late Precambrian glaciation in central-
eastern Brazil: synthesis and model. Palaeogeography, Palaeoclima-
tology, Palaeoecology, 65, 1– 21.
Karfunkel, J., Moreira, P. C., Ribeiro, M. C. & Franco, A. L. 1984.
Aspectos genéticos e deposicionais do Grupo Macaúbas na região
da barragem do Paraúna e sua importância na contribuição para um
modelo paleogeográfico e geotectônico. In: Congresso Brasileiro
de Geologia 33, Rio de Janeiro, Anais, 7. Sociedade Brasileira de
Geologia, 3091– 3103.
Karfunkel, J., Pedrosa-Soares, A. C. & Dossin, I. A. 1985. O Grupo
Macaúbas em Minas Gerais: revisão dos conhecimentos. Boletim
do Núcleo Minas Gerais-Sociedade Brasileira de Geologia, 5,
45 –59.
Karfunkel, J., Hoppe, A. & Noce, C. M. 2002. Serra da Água Fria e
vizinhancas, MG: vestı́gios de glaciação neoproterozóica. In: Schob-
benhaus, C., Campos, D. A., Queiroz, E. T., Winge, M. &
Berbert-Born, M. (eds) Sı́tios Geológicos e Paleontológicos do
Brasil, 1. DNPM, Brası́lia, Brazil, 165– 173.
Knauer, L. G. & Grossi-Sad, J. H. 1997. Geologia da Folha Presidente
Kubtischek. In: Grossi-Sad, J. H., Lobato, L. M., Pedrosa-Soares,
A. C. & Soares-Filho, B. S. (eds) Projeto Espinhaço. CODEMIG,
Belo Horizonte, 1902– 2056.
Knauer, L. G., Silva, L. L., Souza, F. B. B., Silva, L. R. & Carmo, R. C.
2006. Folha Monte Azul 1:100.000. CPRM, Programa Geologia do
Brasil, Brası́lia.
Lima, S. A. A., Martins-Neto, M. A., Pedrosa-Soares, A. C., Cordani,
U. G. & Nutman, A. 2002. A Formação Salinas na área-tipo, NE de
Minas Gerais: uma proposta de revisão da estratigrafia da Faixa
Araçuaı́ com base em evidências sedimentares, metamórficas e
idades U– Pb SHRIMP. Revista Brasileira de Geociências, 32,
491– 500.
Machado, N., Schrank, A., Abreu, F. R., Knauer, L. G. &
Almeida-Abreu, P. A. 1989. Resultados preliminares da geocrono-
logia U– Pb na Serra do Espinhaço Meridional. Boletim do Núcleo
Minas Gerais-Sociedade Brasileira de Geologia, 10, 171–174.
Marshak, S., Alkmim, F. F., Whittington, A. & Pedrosa-Soares,
A. C. 2006. Extensional collapse in the Neoproterozoic Araçuaı́
orogen, eastern Brazil: a setting for reactivation of asymmetric crenu-
lation cleavage. Journal of Structural Geology, 28, 129–147.
Martins, M. S. 2006. Geologia dos diamantes e carbonados aluvionares
da bacia do Rio Macaúbas, MG. PhD thesis, Instituto de Geociências,
Universidade Federal de Minas Gerais, Belo Horizonte.
Martins, M. S., Karfunkel, J., Noce, C. M., Babinski, M., Pedrosa-
Soares, A. C., Sial, A. N. & Lyu, D. 2008. A seqüência pré-glacial
do Grupo Macaúbas na área-tipo e o registro da abertura do rifte
Araçuaı́. Revista Brasileira de Geociências, 38, 768– 779.
Martins-Neto, M. A. & Hercos, C. M. 2002. Sedimentation and tectonic
setting of Early Neoproterozoic glacial deposits in southeastern
Brazil. In: Altermann, W. & Corcoran, P. L. (eds) Precambrian
Sedimentary Environments: A Modern Approach to Ancient Deposi-
tional Systems. International Association of Sedimentologists,
Special Publications, 33, 383– 403.
Martins-Neto, M. A., Gomes, N. S., Hercos, C. M. & Reis, L. A. 1999.
Fácies glácio-continentais (outwash plain) na Megasseqüencia
Macaúbas (Serra da Água Fria, MG) e seu contexto geotectônico.
Revista Brasileira de Geociências, 29, 179–188.
Martins-Neto, M. A., Pedrosa-Soares, A. C. & Lima, S. A. A. 2001.
Tectono-sedimentary evolution of sedimentary basins from Late
Paleoproterozoic to Late Neoproterozoic in the São Francisco
craton and Araçuaı́ fold belt, eastern Brazil. Sedimentary Geology,
141/142, 343–370.
Moraes, L. J. 1929. Geologia da região diamantina de Minas Gerais. In:
Relatório Anual do Diretor 1928, Rio de Janeiro. Serviço Geológico
e Mineralógico, 29 – 34.
Moraes, L. J. 1932. Área ocupada pela Formação Macaúbas no
norte de Minas Gerais. Anais Academia Brasileira de Ciências, 4,
111– 114.
Moraes, L. J. & Guimarães, D. 1930. Geologia da região diamantı́fera do
norte de Minas Gerais. Anais Academia Brasileira de Ciências, 2,
153– 186.
Moraes, L. J. & Guimarães, D. 1931. The diamond-bearing
region of Northern Minas Gerais, Brazil. Economic Geology, 26,
502– 530.
Mourão, M. A. A. & Pedrosa-Soares, A. C. 1992. Evidências de sedi-
mentação turbidı́tica no Grupo Macaúbas, Faixa Araçuaı́. Revista
Escola de Minas de Ouro Preto, 45, 94– 96.
Mourão, M. A. A. & Grossi-Sad, J. H. 1997. Geologia da Folha Padre
Carvalho. In: Grossi-Sad, J. H., Lobato, L. M., Pedrosa-Soares,
A. C. & Soares-Filho, B. S. (eds) Projeto Espinhaço. CODEMIG,
Belo Horizonte, 316– 418.
Mourão, M. A. A., Grossi-Sad, J. H. & Fonseca, E. 1997. Geologia da
Folha Janaúba. In: Grossi-Sad, J. H., Lobato, L. M., Pedrosa-
Soares, A. C. & Soares-Filho, B. S. (eds) Projeto Espinhaço.
CODEMIG, Belo Horizonte, 11 – 123.
Noce, C. M. 1997a. Geologia da Folha Curimataı́. In: Grossi-Sad, J. H.,
Lobato, L. M., Pedrosa-Soares, A. C. & Soares-Filho, B. S. (eds)
Projeto Espinhaço. CODEMIG, Belo Horizonte, 1207– 1250.
Noce, C. M. 1997b. Geologia da Folha Itacambira. In: Grossi-Sad, J. H.,
Lobato, L. M., Pedrosa-Soares, A. C. & Soares-Filho, B. S. (eds)
Projeto Espinhaço. CODEMIG, Belo Horizonte, 853–926.
 NEOPROTEROZOIC MACAÚBAS GROUP
Noce, C. M., Pedrosa-Soares, A. C. et al. 1997. Nova subdivisão estra-
tigráfica regional do Grupo Macaúbas na Faixa Araçuaı́: o registro de
uma bacia neoproterozóica. Boletim do Núcleo Minas Gerais-
Sociedade Brasileira de Geologia, 14, 29 –31.
Oliveira, M. J. R., Grossi-Sad, J. H., Romano, A. W. & Lobato, L. M.
1997a. Geologia da Folha Grão Mogol. In: Grossi-Sad, J. H.,
Lobato, L. M., Pedrosa-Soares, A. C. & Soares-Filho, B. S.
(eds) Projeto Espinhaço. CODEMIG, Belo Horizonte, 611–642.
Oliveira, M. J. R., Fogaça, A. C. C. & Fonseca, E. 1997b. Geologia da
Folha Baldim. In: Grossi-Sad, J. H., Lobato, L. M., Pedrosa-
Soares, A. C. & Soares-Filho, B. S. (eds) Projeto Espinhaço.
CODEMIG, Belo Horizonte, 2437– 2531.
Pedrosa-Soares, A. C. & Grossi-Sad, J. H. 1997. Geologia da Folha
Minas Novas. In: Grossi-Sad, J. H., Lobato, L. M., Pedrosa-
Soares, A. C. & Soares-Filho, B. S. (eds) Projeto Espinhaço.
CODEMIG, Belo Horizonte, 925– 1053.
Pedrosa-Soares, A. C. & Oliveira, M. J. R. 1997. Geologia da Folha
Salinas. In: Grossi-Sad, J. H., Lobato, L. M., Pedrosa-Soares,
A. C. & Soares-Filho, B. S. (eds) Projeto Espinhaço. CODEMIG,
Belo Horizonte, 419– 541.
Pedrosa-Soares, A. C., Leonardos, O. H. & Correia-Neves, J. M.
1984. Aspectos metamórficos de seqüências supracrustais da Faixa
Araçuaı́ em Minas Gerais. In: Congresso Brasileiro de Geologia
33, Rio de Janeiro, Anais, 6. Sociedade Brasileira de Geologia,
3056– 3065.
Pedrosa-Soares, A. C., Noce, C. M., Vidal, P., Monteiro, R. & Leo-
nardos, O. H. 1992. Toward a new tectonic model for the Late Pro-
terzoic Araçuaı́ (SE Brazil) — West Congolian (SW Africa) Belt.
Journal of South American Earth Sciences, 6, 33 –47.
Pedrosa-Soares, A. C., Baars, F. J., Lobato, L. M., Magni, M. C. V. &
Faria, L. F. 1993. Arquitetura tectono-metamórfica do setor central
da Faixa Araçuai. Boletim do Núcleo Minas Gerais-Sociedade Brasi-
leira de Geologia, 12, 176– 182.
Pedrosa-Soares, A. C., Leonardos, O. H., Ferreira, J. C. & Reis, L. B.
1996. Duplo regime metamórfico na Faixa Araçuaı́: re-
interpretação à luz de novos dados. In: Congresso Brasileiro
de Geologia 39, Salvador, Anais, 6. Sociedade Brasileira de Geolo-
gia, 5– 8.
Pedrosa-Soares, A. C., Vidal, P., Leonardos, O. H. & Brito-Neves,
B. B. 1998. Neoproterozoic oceanic remnants in eastern Brazil:
further evidence and refutation of an exclusively ensialic evolution
for the Araçuaı́-West Congo orogen. Geology, 26, 519– 522.
Pedrosa-Soares, A. C., Cordani, U. & Nutman, A. 2000. Constraining
the age of Neoproterozoic glaciation in eastern Brazil: first U –Pb
(SHRIMP) data from detrital zircons. Revista Brasileira de Geociên-
cias, 30, 58 – 61.
Pedrosa-Soares, A. C., Noce, C. M., Wiedemann, C. M. & Pinto, C. P.
2001. The Araçuaı́-West Congo orogen in Brazil: an overview of a
confined orogen formed during Gondwanland assembly. Precam-
brian Research, 110, 307–323.
Pedrosa-Soares, A. C., Noce, C. M., Alkmim, F. F., Silva, L. C.,
Babinski, M., Cordani, U. & Castañeda, C. 2007. Orógeno
Araçuaı́: sı́ntese do conhecimento 30 anos após Almeida 1977. Geo-
nomos, 15, 1– 16.
Pedrosa-Soares, A. C., Alkmim, F. F., Tack, L., Noce, C. M., Babinski,
M., Silva, L. C. & Martins-Neto, M. A. 2008. Similarities and
differences between the Brazilian and African counterparts of the
Neoproterozoic Araçuaı́-West Congo orogen. In: Pankhurst, R. J.,
Trouw, R. A. J., Brito Neves, B. B. & De Wit, M. J. (eds) West
Gondwana: Pre-Cenozoic Correlations Across the South Atlantic
Region. Geological Society, London, Special Publications, 294,
153– 172.
Pflug, R. & Schöll, W. U. 1975. Proterozoic glaciation in Eastern
Brazil: a review. Geologische Rundschau, 64, 287– 299.
Pisarevsky, S. A., Wingate, M. T. D., Powell, C. Mca., Johnson, S. &
Evans, D. A. D. 2003. Models of Rodinia assembly and fragmentation.
In: Yoshida, M., Windley, B. & Dasgupta, S. (eds) Proterozoic East
Gondwana: Supercontinent Assembly and Breakup. Geological
Society, London, Special Publications, 206, 35–55.
Porada, H. 1989. Pan-African rifting and orogenesis in southern to
equatorial Africa and Eastern Brazil. Precambrian Research, 44,
103–136.
533
Queiroga, G. N., Pedrosa-Soares, A. C., Quéméneur, J. & Casta-
ñeda, C. 2006. A unidade metassedimentar do ofiolito de Ribeirão
da Folha, Orógeno Araçuaı́, Minas Gerais: petrografia, geotermobar-
ometria e calcografia. Geonomos, 14, 9 –12.
Queiroga, G. N., Pedrosa-Soares, A. C. et al. 2007. Age of the
Ribeirão da Folha ophiolite, Araçuaı́ Orogen: the U– Pb zircon
dating of a plagiogranite. Geonomos, 15, 61 – 65.
Rocha-Campos, A. C. & Hasui, Y. 1981. Tillites of the Macaúbas
Group (Proterozoic) in central Minas Gerais and southern Bahia,
Brazil. In: Hambrey, M. J. & Harland, W. B. (eds) Earths’s
Pre-Pleistocene Glacial Record. Cambridge University Press,
933– 939.
Rodrigues, J. B. 2008. Proveniência de sedimentos dos grupos Canastra,
Ibiá, Vazante e Bambuı́: um estudo de zircões detrı́ticos e idades-
modelo Sm –Nd. PhD thesis, Instituto de Geociências, Universidade
de Brası́lia, Brazil.
Rogers, J. W. & Santosh, M. 2004. Continents and Supercontinents.
Oxford University Press.
Roque, N. C., Grossi-Sad, J. H., Noce, C. M. & Fonseca, E. 1997. Geo-
logia da Folha Rio Pardo de Minas. In: Grossi-Sad, J. H., Lobato,
L. M., Pedrosa-Soares, A. C. & Soares-Filho, B. S. (eds) Projeto
Espinhaço. CODEMIG, Belo Horizonte, 126–221.
Santos, R. V., Alvarenga, C. et al. 2004. Carbon isotopes of Mesopro-
terozoic –Neoproterozoic sequences from southern São Francisco
craton and Araçuaı́ Belt: paleogeographic implications. Journal of
South American Earth Sciences, 18, 27 –39.
Schöll, W. U. 1972. Der südwestliche Randbereich der Espinhaço
Zone, Minas Gerais, Brasilien. Geologische Rundschau, 61,
201– 216.
Schöll, W. U. 1976. Estromatólitos Conophyton em dolomitos do Grupo
Macaúbas. In: Congresso Brasileiro de Geologia 29, Ouro Preto,
Anais, 2. Sociedade Brasileira de Geologia, 67– 73.
Schrank, A., Dourado, B. V. & Biondi, J. C. 1978. Estudo preliminar
dos metavulcanitos do Grupo Macaúbas na região do Alto Jequitin-
honha, MG. In: Congresso Brasileiro de Geologia 30, Recife,
Anais, 3. Sociedade Brasileira de Geologia, 1323–1335.
Silva, L. C., Mcnaughton, N. J., Armstrong, R., Hartmann, L. &
Fletcher, I. 2005. The Neoproterozoic Mantiqueira Province and
its African connections. Precambrian Research, 136, 203– 240.
Silva, L. C., Pedrosa-Soares, A. C. & Teixeira, L. R. 2008. Tonian
rift-related, A-type continental plutonism in the Araçuaı́ orogen,
Eastern Brazil: new evidences for the breakup stage of the São Fran-
cisco– Congo Paleocontinent. Gondwana Research, 13, 527– 537.
Tack, L., Wingate, M. T .D., Liégeois, J. P., Fernandez-Alonso, M.
& Deblond, A. 2001. Early Neoproterozoic magmatism
(1000–910 Ma) of the Zadinian and Mayumbian Groups
(Bas-Congo): onset of Rodinian rifting at the western edge of
the Congo craton. Precambrian Research, 110, 277–306.
Tait, J., Delpomdor, F., Préat, A., Straathof, G., Kanda, V. & Tack,
L. 2011. Neoproterozoic Sequences of the West Congo and Lindi/
Ubangi Supergroups in the Congo Craton, central Africa. In:
Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geo-
logical Record of Neoproterozoic Glaciations. Geological Society,
London, Memoirs, 36, 185–194.
Trompette, R. 1994. Geology of Western Gondwana (2000 –500 Ma).
Pan-African–Brasiliano aggregation of South America and Africa.
A. A. Balkema, Rotterdam.
Tupinambá, M. & Grossi-Sad, J. H. 1997. Geologia da Folha Rio Ver-
melho. In: Grossi-Sad, J. H., Lobato, L. M., Pedrosa-Soares,
A. C. & Soares-Filho, B. S. (eds) Projeto Espinhaço. CODEMIG,
Belo Horizonte, 1667–1806.
Uhlein, A. & Chaves, M. S. C. 1989. Geologia da borda nordeste da Serra
do Espinhaço (região de Mendanha a São Gonçalo do Rio Preto, MG).
Boletim do Núcleo Minas Gerais-Sociedade Brasileira de Geologia,
10, 175– 179.
Uhlein, A., Trompette, R. & Egydio-Silva, M. 1998. Proterozoic
rifting and closure, SE border of the São Francisco craton, Brazil.
Journal of South American Earth Sciences, 11, 191– 203.
Uhlein, A., Trompette, R. & Alvarenga, C. 1999. Neoproterozoic
glacial and gravitational sedimentation on a continental rifted
margin: the Jequitaı́-Macaúbas sequence (Minas Gerais, Brazil).
Journal of South American Earth Sciences, 12, 435– 451.
534 A. C. PEDROSA-SOARES ET AL.
Uhlein, A., Trompette, R., Egydio-Silva, M. & Vauchez, A. 2007. A
glaciação Sturtiana (750 Ma), a estrutura do rifte Macaúbas-Santo
Onofre e a estratigrafia do Grupo Macaúbas, Faixa Araçuaı́. Geono-
mos, 15, 45 – 60.
Uhlein, A., de Alvarenga, C. J. S., Dardenne, M. A. & Tromp-
ette, R. R. 2011. The Glaciogenic Jequitaı́ Formation,
Southeastern Brazil. In: Arnaud, E., Halverson, G. P. &
Shields-Zhou, G. (eds) The Geological Record of Neoprotero-
zoic Glaciations. Geological Society, London, Memoirs, 36,
541– 546.
Vilela, F. T. 2010. Caracterização de metadiamictitos ferruginosos da
Formação Nova Aurora (Grupo Macaúbas, Orógeno Araçuail) a
oeste de Salinas, ME. MSc thesis, Instituto de Geosciências, Univer-
sidade Federal de Minas Gerais, Belo Horizonte.
Viveiros, J. F. M. & Walde, D. 1976. Geologia da Serra do Cabral, Minas
Gerais, Brasil. Münsterische Forschungshefte Geologie und Palaeon-
tologie, 38/39, 15 –27.
Viveiros, J. F. M., Sá, E. L., Vilela, O. V., Santos, O. M., Moreira,
J. M. P., Holder-Neto, F. & Vieira, V. S. 1979. Geologia dos vales
dos rios Peixe Bravo e Alto Vacaria, norte de Minas Gerais. Boletim
do Núcleo Minas Gerais-Sociedade Brasileira de Geologia, 1, 75– 87.
Walde, D. 1976. Neue Hinweise für eine proterozoische Vereisung in
Ostbrasilien. Münsterische Forschungshefte Geologie und Palaeon-
tologie, 38/39, 47 –59.
 Chapter 50

Moema laminites: a newly recognized Neoproterozoic (?) glaciogenic

unit, São Francisco Basin, Brazil

ANTONIO C. ROCHA-CAMPOS1*, BENJAMIN B. DE BRITO NEVES1, MARLY BABINSKI1, PAULO R. DOS

SANTOS1, SÔNIA M.B. DE OLIVEIRA1 & A. ROMANO2
1
Instituto de Geociências, Universidade São Paulo, Rua do Lago, 562, Cidade Universitária, São Paulo, SP, Brazil
2
Instituto de Geociências, Universidade Federal de Minas Gerais, Belo Horizonte, MG, Brazil
*Corresponding author (e-mail: acrcampo@usp.br)

Abstract: A recently identified diamictite and silt-clay laminite, which discordantly overlie the Archaean basement and underlie the
Neoproterozoic Bambui Group, have been informally named as the Moema Laminites. They are preserved at the southwestern
margin of the São Francisco Basin in southeastern Brazil, and are widely distributed in the central-western Minas Gerais state. They
crop out discontinuously over at least 140 km along a north–south direction. The nomenclature of and stratigraphic relationships
between the Moema Laminites and other isolated Neoproterozic occurrences of similar rocks are in a state of flux. Two exceptionally
good exposures of the Moema Laminites show good evidence for deposition under glacial conditions. At the Formiga locality, a
single glacial advance is registered by a deformation tillite, while overlying laminite records deposition in a post-glacial, probably
marine basin following deglaciation. At the SAFFRAN quarry, striations on a bedding plane may have been caused by floating sea-
ice that just touched the bottom of the basin. Much additional work is needed to establish relationships between the Moema Laminites
and other similar occurrences. If these and Moema Laminites are shown to be Cryogenian glacial deposits, the area covered by the
Cryogenian glaciations in the São Francisco basin is much larger than formerly believed.

The name Moema Laminites is being used informally to designate Stratigraphy

a newly identified, possibly Neoproterozoic unit, made up of dia-
mictite and silty-clayey laminites. The strata, so far known at
reconnaissance level, crop out extensively in the central-western
State of Minas Gerais (Fig. 50.1), below carbonate assigned to
the Sete Lagoas Formation (Fm.) of the Bambuı́ Group (Neoproter-
ozoic) (Fig. 50.2).
The Moema Laminites crop out discontinuously over a large
portion of the southern São Francisco Basin, from the latitude of
the town of Onça do Pitangui (198430 S, 448W), southward for c.
140 km, at least as far as the town of Cristais (208520 S, 45830 W)
(Fig. 50.1).
Two exceptionally good exposures of the strata are found at a
road cut of BR-354, near the town of Formiga, 1 km north of the
intersection with road MG-050, in the southern part of the study
area, and at the SAFFRAN quarry (named after the owner
company), some 100 km north of the first locality, in the vicinity
of Onça do Pitangui, 10 km north of road BR-262.
The thickest exposure of the unit is in the vicinity of the town of
Moema (Fig. 50.1), but because the name is provisional, no formal
type-section has yet been designated.
Structural framework
The Moema Laminites are part of the sedimentary cover of the
intracratonic São Francisco Basin, in the central-western State of
Minas Gerais. They occur immediately below the Neoproterozoic
Bambui Group along the southwestern margin of the basin.
In this area, the horizontal to gently folded platformal cover of
the basin fills a depression in the cratonic foreland, located west
of the Sete Lagoas High (Fig. 50.1). Deposition of the Moema sedi-
ments was contemporaneous with the sag phase of tectonic evol-
ution of the basin (Alkmin 2004).
Outcrops of the Moema Laminites are, in general, preserved in
topographic lows on the irregular surface of the cratonic basement.
Strata are essentially horizontal but have been slightly affected by
normal faults. Near Onça do Pitangui they are part of an east –west
trending syncline with plunges up to 308E.
Rocks of the unit are essentially non-metamorphic.
The sparcity of outcrops (mostly road cuts), moderate topography
and deep weathering make it difficult to log a complete represen-
tative stratigraphic section. A regional field reconnaissance was
performed along several roads in the area between the towns of
Cristais and Bom Despacho, in order to evaluate the distribution
and stratigraphic relations of the studied unit (Fig. 50.1). This
was complemented by more detailed stratigraphic and sedimento-
logical study of critical outcrops.
Laterally, the thickness of the diamictite varies from 30 cm on
the Formiga outcrop to, more commonly, c. 5 m, and may be
missing in places. In other places, the unit is represented only by
remains of the basal diamictite, the upper laminites having been
eroded away. The thickness of the overlying laminites reaches at
least 60 m in the area of the town of Onça do Pitangui (Fig. 50.1).
Strata below and above the lower nonconformity are deeply
weathered. Kaolinitized regolith often reaches considerable thick-
ness on top of Archaean basement in the study area and may record
a significant and intense phase of pre-Neoproterozoic weathering,
superposed by widespread recent alteration.
The stratigraphic relations of the Moema Laminites are similar
to that of other isolated outcrops of conglomerate/diamictite and
associated rocks that are found below carbonate rocks of the Neo-
proterozoic Sete Lagoas Fm. (Bambuı́ Group). These are known
from several places, fringing the western margin of the São Fran-
cisco Basin (e.g. Member/Facies Carrancas, Costa et al. 1961;
Facies Samburá, Miranda 1953, and so on) and are traditionally
correlated with the Jequitaı́ Fm. The latter unit, of well established
glacial origin (Isotta et al. 1969), crops out some 400 km north of
the study area (Fig. 50.1; see Dardenne et al. (1978) and Rocha-
Campos & Hasui (1981) for a fuller account of these occurrences).
At present, there are inconsistencies in the interpretation of the
nomenclature, stratigraphic relations and depositional setting of
the strata lying below the Sete Lagoas Fm. (Bambuı́ Group),
including the Moema Laminites.
Pierre Muzzi Magalhães, for example, in his Masters disser-
tation of 1989, recognizes a Conglomeratic Facies (polymitic)
and a Pellitic Facies (argillite and siltstone), below carbonate

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 535– 540. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.50
536 A. C. ROCHA-CAMPOS ET AL.
rocks of the Sete Lagoas Fm., that are probably equivalent to the
Moema Laminites, as a basal unit of the Bambuı̀ Group
(Fig. 50.2). A glacial influence during deposition of these rocks
was, however, not recognized.
On recent geological maps from the southern part of the São
Francisco Basin (SEME/COMIG/ CPRM, 2003; CPRM, 2007),
rocks considered by us to be part of the Moema Lamites have
been called the Serra de Santa Helena Fm. and Samburá Fm.
The former is a pelitic unit lying above the Sete Lagoas Fm.,
and the latter corresponds to one of the infra-Bambuı́ Group dia-
mictite occurrences. As pointed out by Dardenne & Castro
(2000), the Samburá diamictite is a fault-controlled, fan-delta
deposit that occurs at different levels of the Bambui Group.
Under these circumstances, the use of a provisional name seems
more advisable.
Glaciogenic deposits and associated strata
The sedimentary characteristics of the Moema Laminites are based
on the best exposures in the region: the Formiga area and the
SAFFRAN quarry.
Fig. 50.1. Geological map of the São Francisco
Basin, Southeastern Brazil, showing location of
the best exposures of the Moema Laminites.
Squares: 1, Formiga outcrop; 2, SAFFRAN
quarry. SFC, São Francisco Craton.
Formiga area
A cut on the northeastern side of the road BR-364 exposes a lower,
discontinuous, thin diamictite of variable thickness (30 –40 cm),
resting non-conformably on topographic highs of Archaean ortho-
gneiss or on its regolith, followed with a sharp contact by 15 m of
laminites. The smooth surface of the gneiss bears fine, poorly pre-
served striae (Fig. 50.3).
Isolated or semi-detached round boulders of up to 1.5 m lie on
top of the orthogneiss and regolith and seem to correspond to core-
stones (Fig. 50.3). Deep weathering of local rocks has altered their
colours, thus enhancing internal features.
The diamictite corresponds to a layer of tectonically deformed
regolith of the local basement, containing many centimetre-
to-decimetre clasts, predominantly of the local gneiss, but also
of two different lithologies (fine and coarse quartz schist), in a
sandy-silty-clayey matrix. Deformation features recognized in
the diamictite include boudins, stretched clasts, attenuated and
transposed folds and clast alignment. Small thrust faults cut the
top of the gneissic basement (Fig. 50.3).
The contact between the deformed bed and overlying laminites
seems to be sharp. Two subunits are recognized, a lower one of
 MOEMA LAMINITES 537

Thin laminations are present in the lower subunit. Normal

Sandstones graded bedding was observed in centimetre-thick laminae of the
Serra de Santa Helena Fm.

upper subunit, as was evidence of local disharmonic, soft sediment

Mudstones and Siltstones
folding.
Stromatolitic structures Unequivocal evidence of dropstones was not found in the
laminites.
Dolomitic calcarenite

Calcarenite with parallel

SAFFRAN quarry
and cross bedding
Bambuí Group

Calcilutite with parallel At least 60 m of thin (cm) silty-clayey laminites of kaolinitic com-
and cross bedding position and interbedded, irregular, laterally discontinuous, ferru-
Sete Lagoas Fm.

Siltstone
200 m
Basement 0
Fig. 50.2. Stratigraphy of the Bambuı́ Group in Southern São Francisco Basin.
Conglomeratic (not shown) and pellitic facies (siltstone) at the base of the
Bambuı́ Group is thought to correspond to the Moema Laminites (Muzzi
Magalhães, pers. comm. 1989).
dark laminites, 2 m thick, overlain with a sharp contact by a section
of light laminites, 12 m thick (Fig. 50.3).
The lower interval seems to fill a wide and shallow depression
on top of the deformed bed and orthogneiss. Laminae are deformed
around corestones of weathered gneiss.
ginous (hematite) bands are exposed on the extraction faces of the
quarry (Fig. 50.4). A diamictite bed occurs under the laminites
nearby, but its contact is not exposed. The upper boundary with
the laminites was not seen in this area.
The section is made up of a lower interval where ferruginous
beds are more common and an upper thicker, whiteish one,
where they are rare (Fig. 50.4).
The two intervals differ also in their sedimentary structures.
Whereas the lower interval is characterized by intercalation of
thin silt-clay lenses (starved ripples?) and normal graded and non-
graded laminae, the whiteish clay bands of the upper interval are
mostly thinly laminated.
A bedding plane at the bottom of the quarry contains two areas
bearing sets of parallel, few-metres-long, fine striae and furrows
(1 –3 cm wide). The features are rectilinear or curved and their
orientation varies from 2898N to 3398N. On the same surface,
there are a few examples of striated areas surrounded on both
sides by more elevated non-striated ones on the same bedding
plane. Rare features, such as masses of sediment laterally covering
furrows (Rocha-Campos et al. 1994), suggest that the bed was not
consolidated when the striae were produced.

Fig. 50.3. Partial diagrams (a and b) of the

Formiga outcrop. Deformation, thought to be
glaciotectonic in origin includes stretched
clasts, attenuated folds, boudins and
comminution (A –C). 1, Orthogneiss
(Archaean); 2, regolith; 3, deformed bed
(regolith and lower dark laminites); 4, upper
light laminites. Note lower dark laminites (unit
3) in (a) are undeformed on the left side of the
outcrop.
538 A. C. ROCHA-CAMPOS ET AL.

Top

//
Explanation
2 Laminites

Ferruginous layer
1
Interbedded laminites
0 and ferruginous layers
.m
Fractures/faults
Fig. 50.4. Stratigraphic log of laminites at the
Base Striae SAFFRAN quarry. showing distribution of
ferruginous layer (in black).

Boundary relations with overlying and underlying
non-glacial units
Rocks of the Moema Laminites rest non-conformably on an
erosive surface formed on top of predominantly granitic terranes
of the Archaean basement of the São Francisco Basin (Fig.
50.2). The upper contact with the overlying Sete Lagoas Fm.
carbonates was not directely observed in the studied area.
Chemostratigraphy
contribute to the understanding of their origin. This involved deter-
mination of major (FRX) and trace elements (FRX and ICP-MS),
REE (ICP-MS), content of Fe2þ and identification of mineralogi-
cal phases by X-ray diffraction. REE data were normalized to
NASC (North American Shale Composite).
Clastic components of the laminites (silt-clay and kaolinite)
present a homogeneous composition and chemical signature of
more abundant weathering-resistant elements (Cr, Ga, Hf, Rb Sc,
Ta, Th and Zr), in comparison with the ferruginous layers (up to
48% hematite). Both clastic and ferruginous layers depict a nega-
tive anomaly for Eu and a positive anomaly for Ce.

There are no known carbonate rocks associated directly with the

Moema Laminites and, as such, no chemostratigraphic studies of Other characteristics
C, O or Sr have been undertaken to date.
Eight samples of laminites from the two areas studied were geo- Laminites from the SAFFRAN Quarry are being exploited as a
chemically analysed to better characterize the two rocks and to source of kaolinite for the ceramic industry.
 MOEMA LAMINITES
Palaeolatitude and palaeogeography
There are no palaeomagnetic data from the Moema Laminites.
Palaeomagnetic studies carried out previously on carbonate units
of the São Francisco Basin revealed that a continental-scale fluid
event could have been responsible for the resetting of the palaeo-
magnetic, as well as the isotopic system, at c. 520 Ma (D́Agrella
Filho et al. 2000; Trindade et al. 2004). Recently, however,
Babinski et al. (2007) referred to good-quality palaeomagnetic
poles obtained from rocks of the Congo –São Francisco craton,
penecontemporaneous with the Sete Lagoas Fm., that would
place the southern São Francisco craton at low-intermediate
palaeolatitudes (20– 308).
Geochronological constraints
No direct radiometric data are yet available to constrain the deposi-
tional age of the Moema Laminites. However, the regional strati-
graphic position of other units with similar lithologies below
carbonate beds of the Sete Lagoas Fm. along all sections exam-
ined, and evidence of a glacial influence in the Moema Laminites,
are suggestive of correlation between the Moema Laminites and
the Jequitaı́ Fm., which has a cap carbonate dated at 740 + 22 Ma
(Babinski et al. 2007).
Although the Moema Laminites were deposited on top of the
Archaean basement of the São Francisco Basin at both localities
discussed (Teixeira et al. 2000), Sm –Nd TDM ages yielded by
rocks from the Formiga area range from 1.8 to 1.98 Ga. The
results indicate that besides the basement, other younger, probably
Palaeoproterozoic terranes, may also have contributed as a source
of sediments.
A similar picture was obtained for samples of laminites and ferru-
ginous bands in the SAFFRAN Quarry. The 147Sm– 144Nd of these
rocks are usually very fractionated, with values higher than 0.14,
especially for the ferruginous bands, where they can reach values
of 0.18. Samples with 147Sm– 144Nd close to 0.12 (normal crustal
value) yielded Sm–Nd TDM ages of 1.6–1.9 Ga, again suggesting a
much younger source area for the deposits, besides the Archaean
basement. On the basis of these data, the age of deposition of the
Moema Laminites cannot be older than Palaeoproterozoic.
Discussion
Several erosional and depositional features at the Formiga and
SAFFRAN quarry outcrops support the existence of glacial influ-
ence in this part of the São Francisco Basin.
At Formiga they include (i) a smooth surface of round topo-
graphic highs of orthogneiss, bearing striae and (ii) the presence
in the diamictite and gneissic basement of deformation features
(Fig. 50.3).
These structures are interpreted as having resulted from ductile
deformation of the basement rocks (regolith) under a simple shear
regimen. Small thrusts affecting the orthogneiss point to brittle be-
haviour of the rock. Vergence of the features indicates a direction
of strain oriented from west to east.
Deformation seems to have varied in intensity along the upper
surface of the basement, as demonstrated by changes in thickness
of the deformed bed, size and shapes of clasts and distribution of
deformation features. As a rule, regolith was deformed on topo-
graphic highs of the orthogneissic substrate, but not in depressions
between them.
Such features are commonly found associated with sub-glacial,
glacial –tectonic deformation of Recent and Pleistocene glacial
deposits (e.g. Bennett & Glasser 1997; Hart 1998), which results
in the generation of deformation tills. The same explanation
is applicable to the diamictite at the Formiga outcrop. Glacial
displacement caused fracturing and small-scale thrusting of the
substrate and abrasion, resulting in smooth round surfaces on top
539
of basement highs. Deformation seems to have been associated
with lodgement of clasts.
The glacier may have moved towards the east over the weath-
ered, deformable bed, in a glacial terrestrial setting that was sub-
sequently inundated following glacier retreat and deglaciation. A
second possibility is that a grounded glacier advanced into a pre-
existing water-filled basin, where laminites had accumulated.
The fact that the laminites were not involved in the deformation
and that the glacier caused deformation of the regolith favours
the first hypothesis (Fig. 50.3).
Deposition of the laminites was probably through settling of silt
and clay, from suspended plumes fed by the glacier, as well as by
turbidity or other types of density currents during the deglaciation
phase. Instability of the unconsolidated deposit led to frequent
soft-sediment deformation. Lack (or rarity ?) of dropstones in
the laminites may indicate that the glacial terminus was far
removed from the basin margin or the intervention of mechanisms
that either prevent the occurrence of icebergs or result in their fast
evacuation, as described from present glacial estuarine environ-
ments (Dowdeswell et al. 2000).
Sedimentation processes identified in the two localities studied
seem to have been grossly similar, as shown by textures and sedi-
mentary structures present.
No ice-contact features were found at the SAFFRAN quarry.
Interpretation of glacial influence in the area relies only on the pre-
sumed origin of the striated laminites by erosion of floating ice
masses touching the bottom of the sedimentary basin.
Striae on bedding plane of laminites at the SAFFRAN quarry
and characteristics such as bending or change of direction are com-
patible with their origin through scratching of unconsolidated
bottom sediments by free-floating ice masses (Woodworth-Lynas
& Dowdeswell 1994; Rocha-Campos et al. 1994), either free
sea-lake ice or icebergs.
The nature of the water body in which the silty-clayey laminites
accumulated (either a lake or sea) is difficult to ascertain. The
extensive distribution of the pellitic strata in the São Francisco
Basin seems to favour a marine setting.
Geochemical data indicate that Fe from ferruginous layers of
the laminites originated from weathering of basic rocks, under
more reducing conditions than the present ones and not from the
Late Proterozoic and Archaean banded iron formations (BIFs)
that crop out in Central Minas Gerais State. Source for kaolinite
was probably deeply weathered feldspar-rich Archaean gneissic
terrane.
Geochemical data also suggest a strong fluviatile contribution in
the transportation of sedimentary particles to the depositional site
(Graf et al. 1994; Kato et al. 1998). These features denote differ-
ences in source areas for the particles and in palaeoenvironmental
setting between the Formiga outcrop and SAFFRAN Quarry areas.
A more terrestrially influenced (less saline?) environment could
then be envisaged for laminites at the quarry.
Data from the literature and from this study concur in showing
that the Moema Laminites differ in stratigraphic position and litho-
logical composition from the pelitic Serra de Santa Helena Fm.,
situated above the basal carbonates (Sete Lagoas Fm.) of the
Bambuı́ Group, in which they have been included in recent geo-
logical maps of the area.
Additional fieldwork is needed to clarify the relations of the
Moema Laminites with other isolated occurrences of infra-Bambuı́
Group diamictite outcrops, including the Jequitaı́ Fm., known in
the São Francisco Basin (Dardenne et al. 1978; Rocha-Campos
& Hasui 1981), and to establish their mode of origin.
Possible correlation among them would significantly enlarge the
area known to have been affected by the Early Cryogenian glacia-
tion in Southeastern Brazil.
We thank F. M. Canile for her help with the preparation of this text and figures and
I. McReath for kindly revising the English language. The original text was much
improved by reviews from G. Young and E. Arnaud. This represents a
540 A. C. ROCHA-CAMPOS ET AL.
contribution of the IUGS- and UNESCO-funded IGCP (International Geoscience
Programme) Project #512.
References
Alkmin, F. F. 2004. O que faz de um cráton um cráton? O Cráton do São
Francisco e as revelações almedianas ao delimitá-lo. In: Mantesso
Neto, V., Bartorelli, A., Carneiro, C., dal Ré, & Brito
Neves, B. B. de (organizers) Geologia do Continente Sul-Americano.
Beca, 19 –35.
Alkmin, F. F. & Marshak, S. 1998. Transamazonian Orogeny in the
Southern São Francisco Craton Region, Brazil: evidence for Paleo-
proterozoic collision and collapse in the Quadrilátero Ferrı́fero.
Precambrian Research, 99, 29 –58.
Babinski, M., Vieira, L. C. & Trindade, R. I. F. 2007. Direct dating of
the Sete Lagoas cap carbonate (Bambuı́ Group, Brazil) and impli-
cations for the Neoproterozoic glacial events. Terra Nova, 19,
401– 406.
Bennet, M. R. & Glasser, N. F. 1997. Glacial Geology, Ice sheets and
Landforms. Wiley.
Costa, M. T. da & Branco, J. J. R. 1961 (eds). Introdução, Roteiro para a
Excursão Belo Horizonte-Brası́lia. Contribuição ao XIV Congresso
Brasileiro de Geologia, Belo Horizonte, Instituto de Pesquisas Radio-
ativas, 15, 9– 25.
CPRM – Serviço Geológico do Brasil, 2007. Mapa Geológico, Folha
Guapé, 1:100.000.
D’Agrella Filho, M. S., Babinski, M., Trindade, R. I. F., Van
Schmus, W. R. & Ernesto, M. 2000. Simultaneous remagnetization
and U– Pb isotope resetting in Neoproterozoic carbonates of the São
Francisco craton, Brazil. Precambrian Research, 99, 179– 196.
Dardenne, M. A. & Castro, P. T. A. 2000. The sedimentology, stratigra-
phy and tectonic context of the São Franscisco Group at the southwest
boundary of the São Francisco craton, Brazil. Revista Brasileira de
Geociências, 30, 439–441.
Dardenne, M. A., Faria, A., Magalhães, L. F. & Soares, L. A. 1978. O
tilito da base do Grupo Bambuı́ na borda ocidental do cráton do São
Francisco. Sociedade Brasileira de Geologia, Núcleo Centro-Oeste,
Boletim Informativo, 7/8, 85 –97.
Dowdeswell, J. A., Whittington, R. J., Jennings, A. E., Andrews,
J. T., Mackensen, A. & Marienfeld, F. 2000. An origin for lami-
nated glacimarine sediments through sea-ice build-up and suppressed
iceberg rafting. Sedimentology, 47, 557– 576.
Graff, J. L., Jr., O’Connor, E. A. & Van Leewen, P. 1994. Rare earth
element evidence of origin and depositional environment of Late Pro-
terozoic ironstone beds and manganese-oxide deposits, SW Brazil
and SE Bolivia. Journal of South American Earth Sciences, 7,
115– 133.
Hart, J. 1998. The deforming bed/debris rich basal ice continuum
and its implications for the formation of glacial landforms (flutes)
and sediments (melt-out till). Quaternary Science Reviews, 17,
737– 754.
Isotta, C. A. L., Rocha-Campos, A. C. & Yoshida, R. 1969. Striated
pavement of the Upper Pre-Cambrian glaciation in Brazil. Nature,
222, 466– 468.
Kato, Y., Ohta, I., Tsunematsu, T., Watanabe, Y., Isozaki, Y., Mar-
uyama, S. & Imai, N. 1998. Rare earth element variation in mid-
Archean banded iron formations: implications for the chemistry of
ocean and continent and plate tectonics. Geochimica et Cosmochi-
mica Acta, 62, 3475– 3497.
Miranda, J. 1953. Folha Bambuı́-Seção de Geologia. Departamento
Nacional da Produção Mineral, Divisão de Geologia e Mineralogia,
Relatório Anual do Diretor, 1925, 25 – 27.
Rocha-Campos, A. C. & Hasui, Y. 1981. Proterozoic diamictites of
western Minas Gerais and east Goiás, central Brazil. In: Hambrey,
M. J. & Harland, W. B. (eds) Earth’s Pre-Pleistocene Glacial
Record. Cambridge University Press, Cambridge, 920–923.
Rocha-Campos, A. C., Santos, P. R dos & Canuto, J. R. 1994. Ice
scouring structures in Late Paleozoic rhythmites, Paraná Basin,
Brazil. In: Deynoux, M., Miller, J. M. G., Domack, E., Eyles,
N., Fairchild, I. J. & Young, G. M. (eds) Earth’s Glacial Record.
Cambridge University Press, Cambridge, 234– 240.
SEME (Secretaria de Estado de Minas e Energia)/COMIG (Companhia
Mineradora de Minas Gerais)/CPRM (Serviço Geológico do
Brasil), 2003. Projeto São Francisco, Geologia e Hidrologia, Text
and Maps, 1:250.000 (PDF and CD-ROM).
Teixeira, W., Sabaté, P., Barbosa, J., Noce, C. M. & Carneiro, M. A.
2000. Archean and Paleoproterozoic tectonic evolution of the
São Francisco Craton. In: Cordani, U. G., Milani, E. J.,
Thomaz Filho, A. & Campos, D. A. (eds) Tectonic Evolution of
South America. Sociedade Brasileira de Geologia, São Paulo,
101– 137.
Trindade, R. I. F., D’Agrella-Filho, M. S., Babinski, M., Font, E. &
Brito Neves, B. B. 2004. Paleomagnetism and geochronology of the
Bebedouro cap carbonate: evidence for continental-scale Cambrian
remagnetization in the São Francisco craton, Brazil. Precambrian
Research, 128, 83– 103.
Woodworth-Lynas, C. M. T. & Dowdeswell, J. A. 1994. Soft-
sediment striated surfaces and massive diamicton facies produced
by floating ice. In: Deynoux, M., Miller, J. M. G., Domack, E.,
Eyles, N., Fairchild, I. J. & Young, G. M. (eds) Earth’s Glacial
Record. Cambridge University Press, Cambridge, 241–259.
 Chapter 51

The glaciogenic Jequitaı́ Formation, southeastern Brazil

A. UHLEIN1*, C. J. S. ALVARENGA2, M. A. DARDENNE2 & R. R. TROMPETTE3

1
Universidade Federal de Minas Gerais – UFMG, Dept. de Geologia, Av. Antônio Carlos, 6627, Campus Pampulha,
Belo Horizonte, Minas Gerais, 31270-901, Brazil
2
Universidade de Brası́lia – UnB, Inst. de Geociências, Campus, Asa Norte, 70910-900, Brası́lia (DF), Brazil
3
35 rue Pascal, 75013, Paris, France
*Corresponding author (e-mail: uhlein@netuno.lcc.ufmg.br)

Abstract: Glaciogenic deposits of the Jequitaı́ Formation (Fm.) are well exposed along the margins of the Serra do Cabral on the São
Francisco Craton, southeastern Brazil. The Jequitaı́ Formation is thin (0–150 m thick), lenticular and overlies the Espinhaço Supergroup
on a discrete unconformity. Sandstones show subglacial erosional structures such as grooved and striated pavements oriented ENE–
WSW. The Jequitaı́ Fm. consists of massive and stratified diamictites with granules, pebbles and boulders of gneiss, granite, quartzite
and carbonate. At the base, the diamictites are massive, whereas the upper part contains many alternating beds of clast-rich and -poor
diamictites. They also contain discontinuous, fine-grained sandstones and a few laminated siltstone–mudstone intercalations. This dia-
mictite association indicates glaciomarine sedimentation. The Jequitaı́ Fm. covers the São Francisco cratonic domain and its equivalent
extends eastward over the Araçuaı́ fold belt where it is part of the metasedimentary Macaúbas Group, a thick Neoproterozoic unit with
metadiamictites, quartzites and schists. The diamictite–turbidite association of the Macaúbas Group was deposited on the border of the
Pan-African– Brasiliano rift as gravity flows.

Three distinct glaciogenic units are present in the Neoproterozoic
of Brazil (Alvarenga & Trompette 1992; Nogueira et al. 2003;
Alvarenga et al. 2007). The older and more widespread unit is
probably of middle Cryogenian age (665 –710? Ma). It is well
known on the São Francisco Craton where it is represented by
the Jequitaı́ Fm. at the base of the Bambuı́ Group in the São Fran-
cisco basin (Rocha-Campos & Hasui 1981; Karfunkel & Hoppe
1988; Trompette 1994) and as the Bebedoura Fm. in the Una
Group, which occurs in a series of segmented sub-basins in the
NE of the craton (Guimarães et al. 2011).
The Jequitaı́ Fm. occurs in the Serra do Cabral region
(Fig. 51.1), at the transition between the São Francisco craton
and the external zone of the Araçuaı́ fold belt, in southeastern
Brazil (Minas Gerais State). Along the Brası́lia fold belt at the
western margin of the São Francisco Craton (Goiás State), these
glaciogenic rocks are also exposed in the Cristalina area. The
Jequitaı́ Fm. constitutes the base of the São Francisco Supergroup,
a cover sequence of the São Francisco craton. It crops out at
the lower part of the Bambuı́ Group along the margins of the
Serra do Cabral region, overlying the Palaeo- to Mesoproterozoic
Espinhaço Supergroup. The Jequitaı́ Fm. was recognized early on
as being of glacial origin, first by Branner (1919) and Moraes &
Guimarães (1930). Subsequently, Isotta et al. (1969) discovered
a striated pavement and confirmed the glacial origin.
Structural framework
Brasiliano – Pan-African orogenic belts surround the southern São
Francisco craton: the Brası́lia fold belt to the west and the Araçuaı́
fold belt to the east (Brito Neves et al. 1999; Alkmim et al. 2001).
The Jequitaı́ Fm. occurs on the São Francisco craton, with the type
section in the Serra do Cabral region (Fig. 51.1). Meso- to
Neoproterozoic rocks form the sedimentary cover of the São Fran-
cisco Craton, the São Francisco basin, a sub-horizontal sequence
that was deformed at 560 –550 Ma near the cratonic margins
(Uhlein et al. 1998; Alkmim et al. 2006). The Jequitaı́ Fm. covers
the cratonic domain and extends eastward into the Araçuaı́ fold
belt and the metasedimentary Macaúbas Group (Fig. 51.2), a
thick Neoproterozoic unit with metadiamictites, quartzites and
schists (Karfunkel & Hoppe 1988; Uhlein et al. 1999; Pedrosa
Soares et al. 2001, 2011). The Jequitaı́ Fm. crops out in the Serra
do Cabral Anticlinorium, extending eastward under the Bambuı́
Group and reappearing much thicker due to thrust duplication in
the Araçuaı́ fold belt (Figs 51.1 & 51.2).
Tectonism and metamorphism related to the Brasiliano orogeny
increases in intensity eastward, reaching epizonal metamorphism.
The Araçuaı́ fold belt (Macaúbas Group) shows west-vergent
thrusts and folds, and greenschist- to amphibolites-facies meta-
morphism (Fig. 51.2). Over the cratonic domain, rocks of the
Mesoproterozoic Espinhaço Supergroup are overlain by glacio-
genic rocks of the Jequitaı́ Fm. in the Serra do Cabral region
(Walde 1978; Uhlein 1991), where they are exposed in large
brachy-anticline structures. To the north, near the town of Jequitaı́,
the Serra da Água Fria region displays an anticline structure with
sandstones overlain by Jequitaı́ Fm. diamictites and siltstones
and carbonates of the Bambuı́ Group (Figs 51.1–51.3).
Stratigraphy
The São Francisco Basin includes the Espinhaço Supergroup and
the unconformably overlying São Francisco Supergroup, with
the Jequitaı́ Fm. at the base overlain by marine siliciclastic and
carbonate sediments of the Bambuı́ Group. The Espinhaço Super-
group consists of metaquartzites and metasiltstones probably
deposited c.1500 Ma (Uhlein et al. 1998; Martins Neto 2000).
They are cut by Neoproterozoic tholeiitic mafic dykes that do
not cut the overlying São Francisco Group.
The Jequitaı́ Fm. is thin (0 –150 m thick) and lenticular and
overlies the Espinhaço Supergroup, on a discrete unconformity,
along the margins of the Serra do Cabral. To the north, at the
Serra da Água Fria, between the towns of Jequitaı́ and Francisco
Dumond, the basal sandstones show striated pavements (Isotta et al.
1969; Karfunkel & Karfunkel 1976; Rocha-Campos & Hasui
1981; Karfunkel & Hoppe 1988). Grooves and numerous striae
are oriented ENE – WSW (1008).
The overlying Bambuı́ Group can be subdivided into two major
units in the southeastern part of the São Francisco basin (Dardenne
1978; Martins 1999). The lower unit is characterized by lenticular

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 541– 546. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.51
542 A. UHLEIN ET AL.
carbonates alternating with fine clastics sediments. The limestones
and dolomites are commonly oolitic, oncolitic and stromatolitic,
and show abundant cross-bedding. The upper unit, the Três Marias
Fm., consists of green or brown fine-grained arkoses interlayered
with greenish argillaceous siltstones (Uhlein et al. 2004).
Three lithostratigraphic units are recognized in the Araçuaı́ fold
belt: a polycyclic basement, the Palaeo – Mesoproterozoic Espin-
haço Supergroup, and the Neoproterozoic Macaúbas Group. The
latter consists of metadiamictites with pebbles and boulders, some
of which are striated and polished (Hettich 1977; Karfunkel &
Hoppe 1988; Uhlein et al. 1998, 1999; Pedrosa Soares et al.
2001). The lateral continuity between the Jequitaı́ Fm. into the
Macaúbas Group (Fig. 51.2) was originally suggested by Karfun-
kel & Karfunkel (1976), Karfunkel & Hoppe (1988) and Uhlein
et al. (1998, 1999) based on the geographical distribution of
facies, the variation of thicknesses, and the rare indication pro-
vided by glacial abrasion features on the craton. The Macaúbas
Group contains metadiamictites grading vertically and laterally
Fig. 51.1. Simplified geological map
showing location of the Jequitaı́ Fm. on the
São Francisco Craton. Modified from
Uhlein et al. (1999) and Pedrosa Soares
et al. (2001).
eastward to rhythmic facies, with metamorphosed sandstone, silt-
stone and shale intercalations (Fig. 51.2).
Glaciogenic deposits and associated strata
The Jequitaı́ Fm., near its type location, is 0 –150 m thick and
comprises a diamictite-sandstone-pelite facies association (Figs
51.3 & 51.4). It consists of massive diamictites, as well as very
rare and narrow centimetre- to metre-scale alternations of sand-
stones and pelites (Uhlein et al. 1999; Martins-Neto & Hercos
2002). The diamictites contain granules to large boulders, which
are angular to sub-rounded and composed of sandstones, carbon-
ates, pelites, granites and gneisses, vein quartz and volcanic rocks.
Carbonate clasts are more common towards the base, whereas
crystalline basement clasts dominate towards the top of the
Jequitaı́ Fm. (Karfunkel & Hoppe 1988). Coarse material was
probably derived from the underlying basement and the Espinhaço
Fig. 51.2. Schematic geological
cross-section of the (a) São Francisco Basin
and (b) Araçuaı́ Fold Belt. Modified from
Uhlein et al. (1999).
 THE GLACIOGENIC JEQUITAÍ FORMATION
Fig. 51.3. Geological map of the Jequitaı́ region, Minas Gerais State, with
locations of the studied sections. Modified from Cukrov et al. ( 2005).
See Figure 51.1 for location.
Supergroup, which contains a few limestone and dolostone beds.
The diamictites are massive, sheet-like to lenticular bodies, with
a high variation in clast/matrix ratio. In the lower portions, the
diamictites are massive and show very little variation in clast
size or clast/matrix ratio. The upper portions contain many alter-
nating beds of clast-rich and clast-poor diamictite (Fig. 51.4).
Clasts float in a texturally and compositionally immature sand-
stone –siltstone matrix with clay minerals, calcite and sericite.
Some diamictite clasts are scratched, polished and faceted (Isotta
et al. 1969; Karfunkel & Hoppe 1988).
The stratified diamictites also contains discontinous sand-
stones and a few intercalated laminated pelites. The sandstones
are massive (facies Sm in Fig. 51.4), graded or horizontally strati-
fied, fine- to coarse-grained, sheet-like to lenticular bodies, poorly
sorted, and overall upward fining. They are interpreted as
543
Fig. 51.4. Measured vertical sections in Jequitaı́ Fm., Serra da Água Fria,
around Jequitaı́ town, Minas Gerais State. See Figure 51.3 for locations of
the sections.
subaqueous mass-flow deposits and turbidity currents. Cross-
stratified sandstones (facies Sc in Fig. 51.4) are fine- to coarse-
grained. The diamictites also contain spherical or ovoidal,
discontinous, 10–12 m-thick, fine- to medium-grained, massive
sandstones intercalations variably interpreted as palaeo-eskers by
Karfunkel & Karfunkel (1976) and as large rafts of the Espinhaço
substratum (Martins-Neto & Hercos 2002). The pelites are com-
posed of millimetre-scale siltstone and claystone alternations
(rhythmites) or structureless mudstones, intercalated with the dia-
mictites (Fig. 51.4), and related to low-density turbidity currents or
vertical accretion.
In the Serra da Água Fria, 10 km SE of Jequitaı́, diamictites of
the Jequitaı́ Fm. rest directly on scratched, grooved and polished
pavements developed on sub-horizontal sandstones, interpreted
as basement rocks belonging to the Espinhaço Supergroup (Isotta
et al. 1969; Karfunkel & Hoppe 1988; Uhlein et al. 1999). The
erosional features vary from very fine scratches to grooves up
20 cm wide and 5 cm deep (Isotta et al. 1969). They show V-
and U-shaped grooves, and crescent fractures have been observed.
These features, which have an azimuth of 1008 indicate roughly
east – west ice palaeoflow. These glacial abrasion scars are inter-
preted as land-based and the diamictites are interpreted as tillites
resting directly on striated pavements (Isotta et al. 1969; Rocha-
Campos & Hasui 1981; Karfunkel & Hoppe 1988). Striated and
faceted clasts, varvites, inter-tillite striated pavements and ice-
thrust features have been identified (Hettich & Karfunkel 1978;
Gravenor & Monteiro 1983). These observations indicate that
the glaciers that deposited the tillites were continental ice sheets
(Isotta et al. 1969; Karfunkel & Karfunkel 1976; Rocha-Campos
& Hasui 1981; Karfunkel & Hoppe 1988).
Rocha-Campos et al. (1996) suggested a soft-sediment origin
for glacial striae and grooves found on top of the sandstones of
the Serra da Água Fria, near Jequitaı́. The main lines of evidence
are (i) striae inside grooves laterally covered by slumped plough
ridges; (ii) clasts inside furrows partially embedded in the sand-
stone; (iii) sinuosity of furrows; (iv) crescent-like skip marks and
ridges transverse to furrows; (v) occurrence of striated surfaces
on two bedding planes that are separated by a 25-cm-thick bed
544 A. UHLEIN ET AL.
of quartzite; (vi) contiguous striated and ripple-marked areas on
the same bedding plane; ripple marks a few centimetres below
the striated surface are, however, not deformed. This
re-interpretation indicates that the striated sandstones may be
part of the Jequitaı́ Fm. rather than the Espinhaço Supergroup.
At the same time, the terrestrial setting of the Jequitaı́ Fm. has
been questioned. Uhlein et al. (1999) pointed to the scarcity of
outwash deposits around Jequitaı́ to support reinterpretation of
the monotonous, massive, clast-poor diamictites as representing
glaciomarine sedimentation. Rocha-Campos et al. (1996) re-
interpreted the glacial abrasion marks as having been formed in
the fluctuating grounding zone of a marine ice sheet or tongue
just grazing the sediment surface. Finally, Martins-Neto &
Hercos (2002) describe wide grooves with inner steps that yield
the palaeo-direction of glacier movement (2858 azimuth) on the
western border of the Serra da Água Fria, suggesting a more
complex dispersion pattern.
The metadiamictites of the Macaúbas Group in the Araçuai
fold belt contain clasts from granule to boulder size in a
dominantly muddy and/or sandy matrix. The clasts are mainly
quartzites, granitoids and gneisses, limestones, quartz and
schists. Clasts are commonly tectonically stretched. Clast-rich
and clast-poor beds of diamictites alternate and commonly
show different contents of sandy and muddy matrix. The diamic-
tite facies are interpreted as subaqueous debris flows deposited
along basin margins during a period of tectonic activity. Two
main sequences of metadiamictites, 300–2000 m thick, are sep-
arated by an interval of metarhythmites. The latter consist of
parallel-bedded, centimetre- to metre-scale schists, metasilt-
stones, calc-silicate rocks and quartzites, which in places show
normal grading. Hence, the Macaúbas Group is interpreted as
resedimented glacial material deposited by subaqueous debris
flow and turbidity currents (Uhlein et al. 1999). From west to
east, a depositional model can be proposed for the Jequitaı́ –
Macaúbas units: a cratonic diamictite sequence interpreted as
glaciomarine (Jequitaı́ Fm.) grading to a diamictite –turbidite
sequence deposited on the border of a Neoproterozoic rift and
subsequently passive margin basin, the precursor to the
Araçuaı́ fold belt (Uhlein et al. 1998, 1999; Pedrosa Soares
et al. 2001, 2011).
Boundary relations with overlying and underlying
non-glacial units
In the Serra do Cabral region, the lower contact is a very slight
angular unconformity and an erosional disconformity separating
the Jequitaı́ Fm. from the underlying Espinhaço Supergroup
(Walde 1978; Uhlein 1991). Palaeotopographical relief of 300 to
400 m is developed on this contact. The lenticular character of
the Jequitaı́ Fm. along the margins of the Serra do Cabral can be
related to infilling of palaeotopography. Block tilting, indicated
by angular relationships across the Serra do Cabral, reflects Neo-
proterozoic extensional tectonics related to Rodinia break-up
(Uhlein et al. 1999; Martins-Neto & Hercos 2002). The Jequitaı́
Fm. is conformably overlain by metasiltstone of the Bambuı́
Group in the Serra do Cabral region. A hiatus and a discontinuity
between the Jequitaı́ Fm. and the Bambuı́ Group can be recognized
(Dardenne 1978; Martins 1999).
Chemostratigraphy
The lack of biostratigraphic controls and suitable minerals for
radiometric age measurements make chemostratigraphy a poten-
tial alternative for the establishment of a reliable isotope strati-
graphic profile for Neoproterozoic sequences. Meanwhile, in the
Serra do Cabral region, the glaciomarine diamictites of the Jequitaı́
Fm. are overlain by siltstones and shales of the Bambuı́ Group,
which represent deep-marine platform deposits. Thus it is not poss-
ible to recognize a capping carbonate overlying the glacial rocks of
the Jequitaı́ Fm. in the Serra do Cabral and Serra da Água
Fria regions.
C- and O-isotopic studies have been carried out in Sete Lagoas
Fm., basal unit of the Bambuı́ Group in southern São Francisco
Craton (Santos et al. 2000, 2004; Vieira et al. 2005, 2007),
around the cities of Sete Lagoas and Belo Horizonte. The Sete
Lagoas Fm. shows sea-floor precipitates with aragonitic pseudo-
morphs (Peryt et al. 1990) and negative d13C values (c. – 4‰),
which permit the characterization of a basal carbonate unit of the
Bambuı́ Group as a capping carbonate (Vieira et al. 2007). The
Sete Lagoas Fm. displays two upward-shallowing megacycles,
with very high d13C (10‰) in the second megacycle (Vieira
et al. 2007). Sr-isotopic compositions of least altered carbonates
in the Bambguı́ Group range from 0.7074 to 07.7076 (reviewed
in Misi et al. 2007).
Mineralization and other characteristics
The Macaúbas Group contains economic ferruginous and manga-
niferous deposits in the Araçuaı́ Fold Belt. The Fe deposits
form the Riacho dos Poções Member of the Nova Aurora Fm.,
between Peixe Bravo and Vacaria Rivers (Viveiros et al.
1978). Manganiferous levels originate from the lateritic weather-
ing of rhythmites and schists. Ferruginous diamictites, hematite
quartzites and banded iron formations (BIFs), rhythmites, and
hematite schists form the ferruginous facies of the Macaúbas
Group. They were deposited in the transition zone between a
glaciated continental domain and a relatively deep marine
environment with turbidity currents. Locally, mafic lavas with
pillow structures are intercalated in the Macaúbas Group,
suggesting that hydrothermal activity in the basin may have
been important.
Palaeolatitude and palaeogeography
Palaeomagnetic studies on basic dykes in the eastern São Francisco
Craton (D’Agrella Filho et al. 1990) place the southern part of the
craton between 408 and 608 latitude at c. 1.0 Ga, but there are
no available data to constrain palaeolatitudes during deposition
of the Jequitaı́ Fm. or Bambuı́ Group. The Jequitaı́ glaciation
occurred on the elevated Rodinia supercontinent, where larger
areas may have been covered by a continental ice sheet or
several ice caps (Trompette 1994).
Geochronological constraints
Radiometric age determinations on the São Francisco craton
provide few concrete age constraints on glaciogenic diamictites
of the Jequitaı́ Fm. and equivalent units. Rb –Sr dating of clay
minerals yields ages between 932 + 30 Ma and 570 + 7 Ma
(Macedo & Bonhomme 1984) and their geological interpretation
remains difficult. Badelleyite from mafic rocks intruded into the
Espinhaço Supergroup yield an age of 906 + 2 Ma (Machado
et al. 1989), while the youngest concordant detrital zircon in the
Macaúbas Group is 864 + 30 Ma (Pedrosa Soares et al. 2000).
Rocks inferred to represent related oceanic crust in the Macaúbas
Group have produced a Sm– Nd age of 816 + 72 Ma (Pedrosa
Soares et al. 2001) and a 660 + 29 Ma U –Pb age by LA-
ICPMS of a plagiogranite (Pedrosa Soares et al. 2011). The Sete
Lagoas Fm., which overlies the Jequitaı́ Fm. on the São Francisco
craton, has been dated at 740 + 22 Ma by the Pb/Pb carbonate
method (Babinski et al. 2007).
 THE GLACIOGENIC JEQUITAÍ FORMATION
Discussion
Based on the different characteristics of the diamictites and associ-
ated sediments, two distinct depositional environments can be
recognized: (i) a glaciomarine environment on the São Francisco
palaeocontinent, with striated bedrock pavements and a 0 –150-
m-thick diamictite-sandstone-pelite association (i.e. the Jequitaı́
Fm.) and (ii) a deepwater environment dominanted by debris
flows (metadiamictites) and turbidites (quartzites, metarhythmites)
represented by the Macaúbas Group in the Araçuaı́ fold belt, a
Neoproterozoic orogen adjacent to the São Francisco craton.
The Jequitaı́ Fm. was initially interpreted as glaciocontinental
and the diamictites regarded as tillites resting directly on striated
pavements (Isotta et al. 1969; Karfunkel & Karfunkel 1976;
Rocha-Campos & Hasui 1981; Karfunkel & Hoppe 1988).
However, the massive clast-poor diamictites and the stratified
diamictites of the Jequitaı́ Fm. have been reinterpreted to represent
glaciomarine deposits because of its significant thickness
(0 –150 m), fine-grained sediment intercalations, and absence of
outwash facies.
In the Jequitaı́ area, clast-poor diamictites overlying the striated
pavement suggests two phases for the record of the glacial event.
The first phase, probably on the continent, produced ice erosion
(striated pavement); during the second phase, clast-poor diamictite
was deposited by gravity flows during ice retreat and sea-level rise
(Cukrov et al. 2005) at a time of active extension (Uhlein et al.
1998, 1999; Martins-Neto & Hercos 2002).
The lateral transition eastward to the Macaúbas Group is based
on the geographical distribution of facies, thickness changes and
the presence of glacial abrasion features (Karfunkel & Hoppe
1988; Uhlein et al. 1998, 1999). The Macaúbas Group in the
Araçuaı́ fold belt is composed of metadiamictites, quartzites,
rhythmites, ferruginous diamictites, BIF and schists. It consists
of resedimented glacial material deposited by subaqueous debris
flows and turbidity currents. The Macaúbas Group represents syn-
glacial, rift-to-drift sedimentation related to the break-up of the
Rodinia supercontinent. A rift system opened and was filled by
thick gravitational sedimentation, (i.e. debris flows and turbidites;
Uhlein et al. 1999; Pedrosa Soares et al. 2001). The elevated rift
shoulders with ice caps provided glacial material for subaqueous
resedimentation. Tectonic activity on the edge of the rift zone
may have triggered slumps that developed into debris flow and
turbidity currents.
We acknowledge financial support from the CNPq-Brazilian National Research
Council (grant 301732/2005-1 PQ) and FAPEMIG-Research Support Foundation
of Minas Gerais State (grants CRA-1321/98 and CRA-80772/05), Brazil.
F. J. Baars and L. M. Lobato helped improve the English of the original manu-
script. This represents a contribution of the IUGS- and UNESCO-funded IGCP
(International Geoscience Programme) Project #512.
References
Alkmim, F. F., Marshak, S. & Fonseca, M. A. 2001. Assembling West
Gondwana in the Neoproterozoic: clues from the São Francisco
craton region, Brazil. Geology, 29, 319– 322.
Alkmim, F. F., Marshak, S., Pedrosa Soares, A. C., Peres, G. G., Cruz,
S. & Whittington, A. 2006. Kinematic evolution of the Araçuaı́—
West Congo orogen in Brazil and África: Nutcracker tectonics during
the Neoproterozoic assembly of Gondwana. Precambrian Research,
149, 43– 64.
Alvarenga, C. J. S. & Trompette, R. 1992. Glacially influenced sedi-
mentation in the Later Proterozoic of the Paraguay belt (Mato
Grosso, Brazil). Palaeogeography, Palaeoclimatology, Palaeoecol-
ogy, 92, 85– 105.
Alvarenga, C. J. S. de, Figueiredo, M. F., Babinski, M. & Pinho,
F. E. C. 2007. Glacial diamictites of Serra Azul Formation (Edia-
caran, Paraguai belt): evidence of the Gaskiers glacial event in
Brazil. Journal of South American Earth Sciences, 23, 236– 241.
545
Babinski, M., Vieira, L. C. & Trindade, R. I. F. 2007. Direct dating
of the Sete Lagoas cap carbonate (Bambuı́ Group, Brazil) and
implications for the Neoproterozoic glacial events. Terra Nova, 19,
1– 6.
Branner, J. C. 1919. Outlines of the geology of Brazil to accompany the
Geological Map of Brazil. Geological Society of American Bulletin,
30, 189– 338.
Brito Neves, B. B., Campos Neto, M. C. & Fuck, R. A. 1999. From
Rodinia to Western Gondwana: an approach to the Brasiliano-
Panafrican Cycle and orogenic collage. Episodes, 22, 155– 166.
Cukrov, N., Alvarenga, C. J. S. de & Uhlein, A. 2005. Litofácies da
glaciação neoproterozóica na porção sul do Cráton do São Francisco:
exemplos de Jequitaı́, MG e Cristalina, GO. Revista Brasileira de
Geociências, 35, 69 –76.
D’Agrella Filho, M. S., Pacca, I. G., Teixeira, W., Onstott, T. C. &
Renne, P. R. 1990. Paleomagnetic evidence for the evolution
of Meso to Neoproterozoic glaciogenic rocks in central-eastern
Brazil. Palaeogeography, Palaeoclimatology, Palaeoecology, 80,
255– 265.
Dardenne, M. A. 1978. Sı́ntese sobre a estratigrafia do Grupo Bambuı́
no Brasil Central. Anais XXX Congresso Brasileiro de Geologia,
SBG, Recife, 2, 597–610.
Guimarães, J. T., Misi, A., Pedreira, A. J. & Dominguez, J. M. L. 2011.
The Bebedouro Formation, Una Group, Bahia (Brazil). In: Arnaud,
E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological
Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 503–508.
Gravenor, C. P. & Monteiro, R. L. B. P. 1983. Ice-thrust features and a
possible intertillite pavement in the Proterozoic Macaúbas Group,
Jequitaı́ area, Minas Gerais, Brazil. Journal of Geology, 91, 113– 116.
Hettich, M. 1977. A glaciação proterozóica no centro-norte de Minas
Gerais. Revista Brasileira Geociências, 7, 87– 101.
Hettich, M. & Karfunkel, J. 1978. Um esker, um varvito e seixos
estriados no Grupo Macaúbas—norte de Minas Gerais. Revista da
Escola de Minas de Ouro Preto, REM, 34, 5– 8.
Isotta, C. A. L., Rocha-Campos, A. C. & Yoshida, R. 1969. Striated
pavement of the Upper-Precambrian glaciation in Brazil. Nature,
222, 466–468.
Karfunkel, B. & Karfunkel, J. 1976. Estudos petro-faciológicos do
Grupo Macaúbas na porção mediana da Serra do Espinhaço, MG.
XIX Congresso Brasileiro de Geologia, 2, 179–188.
Karfunkel, J. & Hoppe, A. 1988. Late proterozoic glaciation in central-
eastern Brazil: synthesis and model. Palaeogeography, Palaeoclima-
tology, Palaeoecology, 65, 1 –21.
Macedo, M. H. F. & Bonhomme, M. G. 1984. Contribuição à cronoestra-
tigrafia das Formações Caboclo, Bebedouro e Salitre na Chapada
Diamantina (BA) pelos métodos Rb– Sr e K– Ar. Revista Brasileira
de Geociências, 14, 153– 163.
Machado, N., Schrank, A., Abreu, F. R. de, Knauer, L. G. & Abreu,
P. A. A. 1989. Resultados preliminares da geocronologia U/Pb
na Serra do Espinhaço Meridional. An 5 Simp. Geologia Núcleo
Minas Gerais – 1 Simp. Núcleo Brası́lia, Belo Horizonte, 171– 174.
Martins, M. 1999. Análise estratigráfica das sequências mesoproterozói-
cas (borda oeste) e Neoproterozóicas da bacia do São Francisco.
MSc thesis, UFRGS.
Martins-Neto, M. A. 2000. Tectonics and sedimentation in a Paleo-
Mesoproterozoic rift-sag basin (Espinhaço Basin, southeastern
Brazil). Precambrian Research, 103, 147–173.
Martins-Neto, M. A. & Hercos, C. M. 2002. Sedimentation and tectonic
setting of Early Neoproterozoic glacial deposits in south-eastern
Brazil. International Association of Sedimentology, Special Publi-
cations, 33, 383–403.
Misi, A., Kaufman, A. J. et al. 2007. Chemostratigraphic correlation
of Neoproterozoic successions in South America. Chemical Geology,
237, 143–167.
Moraes, L. J. & Guimarães, D. 1930. Geologia da região norte de Minas
Gerais. Anais Academia Brasileira de Ciências, 2, 153–186.
Nogueira, A. C. R., Riccomini, C., Sial, A. N., Moura, C. A. V. &
Fairchild, T. R. 2003. Soft-sediment deformation at the base of
the Neoproterozoic Puga cap carbonate (southwestern Amazon
craton, Brazil): confirmation of rapid icehouse to greenhouse tran-
sition in snowball Earth. Geology, 31, 613– 616.
546 A. UHLEIN ET AL.
Pedrosa Soares, A. C., Cordani, U. G. & Nutman, A. 2000. Constrain-
ing the age of Neoproterozoic glaciation in Eastern Brazil: first U –Pb
(SHRIMP) data for detrital zircons. Revista Brasileira de Geociên-
cias, 30, 58 –61.
Pedrosa Soares, A. C, Noce, C. M., Vidal, P., Wiedemann, C. M. &
Piva-Pinto, C. 2001. The Araçuaı́—West Congo Orogen in Brazil:
an overview of a confined orogen formed during Gondwanaland
assembly. Precambrian Research, 110, 307– 323.
Pedrosa-Soares, A. C., Babinski, M., Noce, C., Martins, M., Queir-
oga, G. & Vilela, F. 2011. The Neoproterozoic Macaúbas Group
(Araçuaı́ orogen, SE Brazil) with emphasis on the diamictite for-
mations. In: Arnaud, E., Halverson, G. P. & Shields-Zhou,
G. (eds) The Geological Record of Neoproterozoic Glaciations. Geo-
logical Society, London, Memoirs, 36, 523– 534.
Peryt, T. M., Hoppe, A., Bechstadt, T., Koster, J., Pierre, C. J. &
Richter, D. K. 1990. Late Proterozoic aragonitic cement crusts,
Bambuı́ Group, Minas Gerais, Brazil. Sedimentology, 37, 279– 286.
Pimentel, M. M., Alvarenga, C. J. S. de & Armstrong, R. 2002.
Proveniência da Formação Jequitaı́, Brasil Central, com base em
dados U/Pb Shrimp em zircões detrı́ticos. Anais XLI Congresso
Brasileiro de Geologia, SBG, 503, João Pessoa (PB).
Rocha-Campos, A. C. & Hasui, Y. 1981. Tillites of the Macaúbas Group
(Proterozoic) in central Minas Gerais and Southern Bahia, Brazil. In:
Hambrey, M. J. & Harland, W. B. (eds) Earth’s Pre-Pleistocene
Glacial Record. Cambridge University Press, Cambridge, 924– 927.
Rocha-Campos, A. C., Young, G. M. & Santos, P. R. 1996. Re-examin-
ation of a striated pavement near Jequitaı́, MG: implications for pro-
terozoin stratigraphy and glacial geology. Anais Academia Brasileira
de Ciências, 68, 593.
Santos, R. V., Alvarenga, C. J. S. de, Dardenne, M. A., Sial, A. N. &
Ferreira, V. P. 2000. Carbon and oxigen isotope profiles across
Meso-Neoproterozoic limestones from central Brazil: Bambuı́ and
Paranoá Groups. Precambrian Research, 104, 107– 122.
Santos, R. V., Alvarenga, C. J. S. de et al. 2004. Carbon isotopes
of Mesoproterozoic – Neoproterozoic sequences from Southern São
Francisco craton and Araçuaı́ belt, Brazil: paleogeographic impli-
cations. Journal of South American Earth Sciences, 18, 27– 39.
Trompette, R. 1994. Geology of Western Gondwana (2000– 500 Ma).
Pan-African –Brasiliano Aggregation of South America and Africa.
Balkema, Rotterdam.
Uhlein, A. 1991. Transição cráton – faixa dobrada: exemplo do cráton
do São Francisco e da Faixa Araçuaı́ (Ciclo Brasiliano) no Estado
de Minas Gerais. Aspectos estratigráficos e estruturais. PhD thesis,
USP, São Paulo.
Uhlein, A., Trompette, R. & Egydio-Silva, M. 1998. Proterozoic
rifting and closure, SE border of the São Francisco craton, Brazil.
Journal of South American Earth Sciences, 11, 191– 203.
Uhlein, A., Trompette, R. & Alvarenga, C. J. S. de 1999. Neoproter-
ozoic glacial and gravitational sedimentation on a continental rifted
margin: the Jequitaı́ –Macaúbas sequence (Minas Gerais, Brazil).
Journal of South American Earth Sciences, 12, 435– 451.
Uhlein, A., Lima, O. N. B., Fantinel, L. M. & Baptista, M. C. 2004.
Estratigrafia e evolução geológica do Grupo Bambuı́, Minas Gerais.
Roteiro de Excursão. Excursão 2. Congresso Brasileiro de Geologia.
CD-Rom.
Vieira, L. C., Trindade, R. I. F. & Nogueira, A. C. R. 2005. Quimio-
estratigrafia da Formação Sete Lagoas, Grupo Bambuı́, Minas Gerais.
III Simposio Cráton do São Francisco, Salvador (BA), 299–302.
Vieira, L. C., Trindade, R. I. F., Nogueira, A. C. R. & Ader, M. 2007.
Identification of a Sturtian cap carbonate in the Neoproterozoic Sete
Lagoas carbonate platform, Bambuı́ Group, Brazil. Comptes Rendus
Geoscience, 339, 240–258.
Viveiros, J. F. M., Sá, E. L., Vilela, O. V., Santos, O. M. & Moreira,
J. M. P. 1978. Geologia dos Vales do rios Peixe Bravo e Vacaria,
Norte de Minas Gerais. XXX Congresso Brasileiro de Geologia,
Recife, 1, 243–254.
Walde, D. H. G. 1978. Desenvolvimento faciológico do Precambriano
entre a Serra Mineira e a Serra do Cabral (região sudoeste da Serra
do Espinhaço, Minas Gerais). Anais XXX Congresso Brasileiro de
Geologia, Recife, 2, 711–725.
 Chapter 52

The Playa Hermosa Formation, Playa Verde Basin, Uruguay

P. J. PAZOS1,2*, A. E. RAPALINI1,2, L. SÁNCHEZ BETTUCCI3 & O. R. TÓFALO1

1
Departamento de Ciencias Geológicas, FCEN, Universidad de Buenos Aires, Ciudad Universitaria,
Buenos Aires 1428, Argentina
2
Consejo Nacional de Investigaciones Cientı́ficas y Técnicas (CONICET)
3
Facultad de Ciencias, Universidad de la República, Iguá 4225, 11400 Montevideo, Uruguay
*Corresponding author (e-mail: pazos@gl.fcen.uba.ar)

Abstract: The Playa Hermosa Formation (Fm.) (Playa Verde Basin) is a volcano-sedimentary unit that crops out in the extreme south of
the Dom Feliciano Belt in Uruguay. This formation has traditionally been interpreted as non-glacial in origin, but recently it has been
suggested that the lower part at least may be glacially influenced. The stratigraphic position of the Playa Hermosa Fm. and its correlation
with other Neoproterozoic units of Uruguay, Argentina and southern Brazil remains in dispute. An age of c. 580 Ma is indicated by a
hornblende 39Ar/40Ar age on magmatism regarded as coeval with syn-sedimentary volcanism. Preliminary palaeomagnetic data
suggest a primary remanence and a mean geomagnetic pole consistent with the proposed apparent polar wander path for the Rı́o de
La Plata craton. In combination, these data suggest mid-Ediacaran glacial sedimentation in low to intermediate latitudes on the Rı́o
de la Plata craton. However, this conclusion needs to be reinforced by more thorough studies of the age and origin of the Playa
Hermosa Fm.

The Ediacaran Playa Hermosa Formation (Masquelin & Sánchez
Bettucci 1993), regarded as part of the Las Ventanas Formation
by Blanco & Gaucher (2005) and Blanco et al. (2009), is com-
monly omitted in regional correlations and palaeoclimatic
interpretations (e.g. Eyles & Janusczack 2004). This formation is
exposed in southern Uruguay, close to the city of Piriápolis
(between 348490 4100 S, 0558190 0000 W and 348500 0500 S,
0558180 4500 W; see Figs 52.1 & 52.2) and the best outcrops are
found along the coast. It was originally interpreted by Masquelin
& Sánchez Bettucci (1993) as a shallow turbiditic sequence associ-
ated with coarse-grained beds deposited in distal fans and canyons
sourced from a delta system. Subsequently, Sánchez Bettucci &
Pazos (1996) ascribed the Playa Hermosa Fm. to the base of the
latest Proterozoic –Ordovician Playa Verde basin. The upper part
of the lower member of the Playa Hermosa Fm. was deposited con-
temporaneously with volcanic flows and intrusions of the middle
Ediacaran Sierra de Las Animas Complex (Sánchez Bettucci
1998; Sánchez Bettucci et al. 2009), which provide the most accu-
rate time constraints available for its deposition.
A glacial influence in the deposition of the Playa Hermosa was
originally proposed by Pazos et al. (1998) and later reinforced by
Pazos et al. (2003), although this interpretation was initially not
accepted (e.g. Gaucher et al. 2003, 2005). Pazos et al. (2008) pub-
lished an exhaustive analysis of the available literature on the
Playa Hermosa Fm. and concluded that a glaciogenic origin for
the lower member of the Playa Hermosa Fm. is unequivocal,
although these sediments appear to be dominantly subaqueously
reworked glacial debris. Only rainout processes, evidenced by
dropstones and coarse-grained rhythmite intervals, are regarded
as directly related to ice transport. Correlation between this
formation and other glaciogenic and suspected glaciogenic units
in the Rı́o de La Plata craton was critically analysed by Pazos
et al. (2008).
Structural framework
The Uruguayan Precambrian basement is separated into four main
tectonic units (Fig. 52.1). From west to east, three units are part of
the Rı́o de La Plata craton: (i) the Palaeoproterozoic Piedra Alta
terrane (c. 2.1 Ga) of low- to medium-grade metamorphic orogenic
belts, virtually unaffected by younger orogenies, (ii) the Nico
Pérez terrane represented by Palaeoproterozoic medium- to high-
grade metamorphic orogenic belts affected by the Neoproterozoic
Brasiliano – Pan-African orogenic cycle, and (iii) the Neoprotero-
zoic to Early Palaeozoic Dom Feliciano belt that developed on
the eastern border of the Nico Pérez terrane. The fourth and east-
ernmost basement unit in Uruguay is the Punta del Este tectono-
stratigraphic terrane (also known as Cuchilla Dionisio terrane),
which has been interpreted as a fragment of the Kalahari craton
accreted to the Rı́o de La Plata craton c. 540 Ma and left behind
after the opening of the South Atlantic (Sánchez Bettucci et al.
2010). Evidence of the suture is seen in the Sierra Ballena shear
zone (Figs 52.1 & 52.2), a continental-scale structure (Fernandes
& Koester 1999; Basei et al. 2005).
The Playa Hermosa Fm. is located in the southwestern extreme
of the Dom Feliciano belt. Sánchez Bettucci & Ramos (1999)
suggested that the Playa Verde basin originated during post-
orogenic collapse. In contrast, Gaucher et al. (2003) interpreted
this succession as the remnant of a passive margin.
The Playa Hermosa Fm. is associated with bimodal (basalt-
rhyolite) volcanism rich in sub-alkaline subvolcanic (syenite)
intrusions. This magmatism has been related to the post-orogenic
phase of the Rio Doce Orogeny (Brasiliano Cycle) that developed
between c. 590 and 510 Ma (Campos Netto & Figueiredo 1995).
The Playa Hermosa Fm. records the opening of the Playa Verde
basin in a transtensional tectonic setting (Pazos et al. 2003;
Pecoits et al. 2008).
The Playa Hermosa Fm. is exposed in a homocline. Its base
(348510 S, 0558180 W) is underlain by tonalitic gneisses of the
1735 + 32 Ma Palaeoproterozoic Campanero Unit (Sánchez
Bettucci et al. 2004, 2009; Mallmann et al. 2007). Its upper
contact, however, is covered variably by modern beach sands
and Quaternary sediments (Pazos et al. 2003). Tectonic defor-
mation is characterized by extension-related stretching and fractur-
ing. The succession lacks regional metamorphism, despite lying
adjacent to the conspicuous tectonic boundary between the Dom
Feliciano belt and the Punta del Este terrane (i.e. the Sierra
Ballena Shear Zone, Oyhantçabal et al. 2007). This major shear
zone was apparently active after deposition of the Playa Hermosa

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 547– 553. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.52
548 P. J. PAZOS ET AL.
sediments, recording docking of the Punta del Este terrane in the
Early Cambrian. Gaucher et al. (2007) proposed that the Nico
Pérez terrane (Fig. 52.1) was displaced some 500 km northward
during the Cambrian, along the Sarandı́ del Yı́ major shear zone
that separates this terrane from the Piedra Alta terrane to the
west. However, the geotectonic setting and the detailed geological
evolution of these rocks is still far from resolved.
Stratigraphy
The Playa Verde basin (Sánchez Bettucci & Pazos 1996) infill con-
stitutes the Maldonado Group (Pecoits et al. 2008), which is sub-
divided into the Playa Hermosa, the Las Ventanas and the San
Carlos formations (Fig. 52.3). The Las Ventanas and San Carlos
formations are discussed in a separate chapter (Pecoits et al.
2011). The glacially influenced Playa Hermosa Fm. comprises
two members. The lower member contains conglomerate, fine-
grained sandstone, pelite and rhythmically laminated fine-grained
sediments resembling varves. Coarse-grained lithologies are
locally deformed by loading and other syn-sedimentary defor-
mation (Pazos et al. 2003). The upper member is composed of
conglomerate, rare sandstone, diamictite beds and volcanic rocks
that intruded and brecciated the sediments and produced hyalo-
clastic features indicative of syn-sedimentary volcanism (Sánchez
Bettucci et al. 2009). The Las Ventanas Fm. is characterized by
conglomerate, sandstone and rhythmites containing out-sized
clasts (Pecoits et al. 2008). The San Carlos Fm. is a succession
of fluvial metaconglomeratic-sandstone, quartzite, feldspar sand-
stone and subordinate metapelite (Sanchez Bettucci & Pazos
1996) and rhyolithic flows (Pecoits et al. 2008). This unit is
found to the east of the Sierra Ballena shear zone (see Fig. 52.2).
Fig. 52.1. Location of the studied area and
morphotectonic divisions of the eastern part
of the Rı́o de la Plata Craton. Modified from
Pazos et al. (2008). SYSZ, Sarandı́ del Yı́
shear zone; MAFMSZ, Marı́a Albina-Fraile
Muerto shear zone; SBSZ, Sierra Ballena
shear zone; CSZ, Cordillera shear zone.
The San Carlos and the Las Ventanas formations were originally
assigned to the Ordovician by Sánchez Bettucci (1998) based on
the occurrence of basalt and syenite clasts in the Las Ventanas
Fm. that were thought to be sourced from the Ediacaran to Cam-
brian Sierra de Las Ánimas Complex (Sánchez Bettucci 1998;
Sánchez Bettucci & Rapalini 2002). Later, Pecoits et al. (2008)
suggested that conglomerate units and rhythmite intervals of the
Las Ventanas Fm. are Neoproterozoic with evidence of glacial
influence. This succession was later folded during the Early
Palaeozoic. Both Pazos et al. (2008) and Pecoits et al. (2008) cor-
related the lower member of the Playa Hermosa Fm. with the fine-
grained deposits of the Las Ventanas Fm. based on the similarity
of rainout deposits (Fig. 52.3).
Glaciogenic deposits and associated strata
The sedimentology of the Playa Hermosa Fm. has been studied by
Sánchez Bettucci & Pazos (1996), Pazos et al. (1998, 2003) and
Fambrini et al. (2003). The Playa Hermosa is divided into two
members, with the lower one containing deposits thought to be gla-
ciogenic. In the logged section of the lower member analysed by
Pazos et al. (2003), two facies associations (FA I and II) were
defined (Fig. 52.4). The lowermost, FA I, is composed of rhyth-
mite, contorted conglomerate and rare normally graded sandstone
and pelite beds. The conglomerates have variable matrix content
and include sub-rounded and angular pebbles of felsic magmatic
rocks as well as intrabasinal rhythmite clasts that are disaggregated
and deformed. Massive diamictites contain abundant rip-up
rhythmite clasts, as well as some angular feldspar clasts. Evidence
for syn-sedimentary deformation is seen in folded stringers of
gravels. Conglomerate beds have irregular, sometimes loaded,
 THE PLAYA HERMOSA FORMATION
basal contact, are lenticular at outcrop scale, and have clasts
forming aggregates or clusters with clast-supported fabric.
The most distinctive lithology of FA I is the fine- to coarse-
grained rhythmite. The coarse rhythmite consists of couplets of
gravel and sand/mud layers. These rhythmite units contain large,
out-sized, bullet-shaped boulders of quartzite. The intervals of
549
Fig. 52.2. Geology of the southern area of
the Dom Feliciano Belt. Modified from
Pazos et al. (2008).
fine-grained rhythmite units rarely occur in FA I and are composed
of dark and light green silt and clay couplets.
FA II (Fig. 52.4) consists of fine-grained rhythmites, some with
brecciated tops, normally graded and rippled sandstones, some
containing lenticular, linsen and moderate- to high-angle ripple
drift cross-lamination. Diamictite beds are usually sharp based,
550 P. J. PAZOS ET AL.

Fig. 52.3. Comparative stratigraphic chart of the Playa Verde basin, Arroyo del
Soldado Group and Tandilia System. Taken and modified from Gaucher et al.
(2003), Pecoits et al. (2008) and Pazos & Rapalini (2011). Triangles indicate
glaciogenic intervals. The brick symbol corresponds to carbonates. Grey boxes
are siliciclastic units.

although some units are interbedded with normal graded sandstone

and dark claystone beds. All palaeocurrents measured from ripples
indicate palaeoflow towards the NE. Horizons of completely disag-
gregated mixtures of gravel, sand and mud with relict rhythmite
clasts are also present (Bm(i) in Fig. 52.4). Rare out-sized clasts
of felsic magmatic origin occur. Close to the top, metre-thick con-
volute bedding occurs in fine-grained rhythmitic deposits below a
second interval of distorted conglomerate. This distorted conglom-
erate deposit is overlain by extremely thin, tectonically deformed
rhythmite beds.

Boundary relations with overlying and underlying

non-glacial units

The basal contact of the Playa Hermosa Fm. is a disconformity

overlying 1.7 Ga tonalitic gneiss (Sánchez Bettucci et al. 2009).
The contact between the lower glaciogenic member and the
upper member and the contact at the top of the Playa Hermosa
Fm. are covered by beach sands.

Chemostratigraphy
No chemostratigraphic studies have been carried out in the Playa
Hermosa Fm. or other units within the Playa Verde Basin.
Fig. 52.4. Logged section of the lower member of the Playa Hermosa Fm.
Modified from Pazos et al. (2003). Cl, claystone; Fs, fine-grained sandstone;
Ms, medium-grained sandstone; Cs, coarse-grained sandstone; Fcg,
fine-grained conglomerate; Mcg, medium-grained conglomerate.

Palaeolatitude and palaeogeography
Preliminary palaeomagnetic data on fine-grained sandstone units
of the Playa Hermosa Fm. were reported by Sánchez Bettucci &
Rapalini (2002). This study included just two sites in the lower
part of this succession near Piriápolis (Fig. 52.1). The mean geo-
magnetic pole suggests a low palaeolatitude (12.7 þ 9.5/ –8.18S)
for deposition of this succession, although these data must be con-
firmed with further analysis before any palaeogeographical
interpretation is verified. The entire succession is exposed as a
homoclinal sequence tilted some 408 towards WNW, which does
not permit the performance of any fold or tilt test to date the
remanence. Detailed alternating field (AF) and thermal cleaning
and principal component analysis (Kirschvink 1980) assure a
reliable isolation and definition of the magnetic components.
Isothermal remanent acquisition curves confirm that magnetite is
the likely carrier of the remanence in this formation. Lack of
resemblance of the characteristic directions (both in situ and
after untilting) with post-Ordovician expected directions for
South America, coupled with the tectonic stability of the region
after the Ordovician suggests a possible primary origin for the
characteristic remanence of the Playa Hermosa Fm. However,
the very small number of sites (two) and samples (six) make this
‘palaeomagnetic pole’ a simple mean geomagnetic pole with no
certainty that non-dipole components of the Earth’s magnetic
field have been fully compensated. Under such conditions, ‘true’
palaeolatitudes from palaeomagnetic data are not certain.
Despite this uncertainty, the geomagnetic pole for the Playa
Hermosa Fm. is consistent with the distribution of other Neoproter-
ozoic palaeomagnetic poles for the Rı́o de La Plata craton
 THE PLAYA HERMOSA FORMATION
(Sánchez Bettucci & Rapalini 2002; Rapalini 2006; Pazos & Rapa-
lini 2011). Comparison of this pole with the apparent polar wander
path suggests an age of c. 590 Ma. This apparent polar wander path
also indicates that the Rı́o de La Plata craton was located at inter-
mediate southern latitudes during the Ediacaran and Cambrian
(Fig. 52.5).
Palaeogeographic reconstruction of the basin is highly uncertain
because the Playa Hermosa Fm. only crops out as a narrow belt on
the coast, with very little obtainable information on the geometry
of the original basin. The systematic palaeocurrent pattern from
the SW and quartzite composition of the outsized clasts (drop-
stones) suggests that the Tandilia System and its eastern pro-
longation (Rapela et al. 2007; Pazos & Rapalini 2011) may have
been the source area. Mylonitized magmatic clasts within the
Playa Hermosa Fm. also indicate provenance from an intensely
tectonized region.
Geochronological constraints
The age of the Playa Hermosa Fm. is not constrained by any direct
radiometric dates. However, the interaction of magmatism with
unconsolidated sediments in the upper part of the lower member
of the Playa Hermosa Fm. (Pazos et al. 2008) strongly suggests
that deposition was contemporaneous with this magmatic event.
The characteristics of this interaction were described by Sánchez
Bettucci et al. (2009). The bimodal magmatism corresponds to
the first cycle of the Sierra de Las Animas complex and has been
loosely constrained between 615 and 550 Ma based on K –Ar
and Rb – Sr ages (Sánchez Bettucci & Linares 1996; Sánchez
Betucci & Rapalini 2002) from samples taken from different
units of this complex, including flows intercalated in the
551
sedimentary succession. Oyhantçabal et al. (2007) obtained an
age of 579 + 2 Ma (39Ar/40Ar in hornblende) from a syenitic
body (Pan de Azúcar pluton) that is part of this magmatic cycle
and is exposed 12 km from the Playa Hermosa outcrops. This
age and the geological relationships suggest a roughly similar
age for the deposition of the Playa Hermosa sediments. The pres-
ence of boulders of conglomerate and sandstone lithologies of the
lower member of the Playa Hermosa Fm., enclosed in quartz –
syenite breccia of the Sierra de Las Animas complex (Sánchez
Bettucci et al. 2009), also supports this interpretation.
Discussion
The lower member of the Playa Hermosa Fm. is attributable to a
glacially influenced environment that records high rates of sedi-
mentation in a tectonically active depositional system. It is separ-
ated into two facies associations. FA I is thought to record a
pro-glacial depositional setting that grades upward into the more
distal FA II (Pazos et al. 2003). Identification of glacial influence
is based on a combination of sedimentary facies and features,
including ice-rafted debris and rhythmites forming couplets of
angular grains (coarse layer) immersed in a fine-grained matrix
and dark clay. These couplets are common in Carboniferous degla-
cial to post-glacial transitions in Argentina (see Pazos et al. 2008
for references) and in the Quaternary. A large out-sized, elongate
clast of quartzite that disrupted laminations was interpreted by
Pazos et al. (2003) as the most robust evidence for ice-rafting pro-
cesses. A debris flow origin is ruled out due to the lateral disconti-
nuity of disruption to bedding around the dropstone and the
well-laminated nature of the couplets hosting the clast. Very fine-
grained rhythmite units present in the FA II are similar to varves
Fig. 52.5. Palaeogeographic reconstruction
for the Rı́o de la Plata Craton and other
blocks during the Ediacaran–Cambrian
periods. (a) 600 Ma, (b) 575 Ma,
(c) 550 Ma. AM, Amazonia; AR, Arabia;
Bal, Baltica; C, Congo; K, Kalahari;
Lau, Laurentia; PA, Pampia; RP, Rı́o de la
Plata; SF, San Francisco; WA, Western
Africa; WN, Western Nile (or Sahara
metacraton). Modified from Sánchez
Bettucci & Rapalini (2002).
552 P. J. PAZOS ET AL.
and suggest a cyclic control on deposition. No tillite deposits have
been reported and for this reason, the succession was considered as
only glacially influenced by Pazos et al. (2008), consistent with
earlier interpretations of a pro-glacial depositional setting (Pazos
et al. 2003; Pecoits et al. 2008) for the lower member of the
Playa Hermosa Fm.
Gravity-driven deposits, such as normally graded sandstone
beds with ripple drift cross-lamination and giant convolutions, as
well as irregular conglomerate beds, suggest that deposition was
abrupt on an unstable slope related to a pro-glacial environment.
Top-brecciated rhythmite beds and syn-depositional folding and
slumping are interpreted to record syn-sedimentary seismicity.
It is difficult to discern between a marine and non-marine deposi-
tional setting for the Playa Hermosa Fm. Fambrini et al. (2003)
argued for a marine environment based on micro-hummocky struc-
tures. However, these structures seem to represent ripple-drift cross-
lamination rather than storm-wave processes, and hence do not dis-
tinguish between marine and lacustrine environments (Pazos et al.
2003). Sánchez Bettucci et al. (2009) described orange palagonite
glass in the volcanism–sediment interaction that is a common altera-
tion product where basalts extrude into the marine environment.
The relationship between the Playa Hermosa Fm. and other units
within the Maldonado Group (Playa Verde basin; Fig. 52.3), in
particular the purportedly glaciogenic Las Ventanas Fm. (Pecoits
et al. 2011), remains subject to debate. Several other Ediacaran-
aged successions occur on the Rı́o de la Plata craton (Fig. 52.3),
including the Cerro do Buggio Fm. (Camaquã basin) and the
Arroyo del Soldado Group in the Ribeira belt and the Sierras
Bayas Group and equivalent units in the Tandilia System (Pazos
& Rapalini 2011). Possible correlations of the Playa Hermosa
Fm. with other Ediacaran units cropping out in Brazil have been
discussed in detail by Pazos et al. (2003, 2008). In summary, no
conclusive correlations can yet be made due to highly fragmentary
exposures and insufficient radiometric control.
We are grateful to M. A. S. Basei for a constructive review and especially to
E. Arnaud for her thorough review and editorial improvements. This is a contri-
bution R-39 to the Instituto Don Pablo Groeber. This represents a contribution of
the IUGS- and UNESCO-funded IGCP (International Geoscience Programme)
Project #512.
References
Basei, M. A. S., Frimmel, H. E., Nutman, A. P., Preciozzi, F. & Jacob,
J. 2005. A connection between the Neoproterozoic Dom Feliciano
(Brazil/Uruguay) and Gariep (Namibia/South Africa) orogenic
belts – evidence from a reconnaissance provenance study. Precam-
brian Research, 139, 195–221.
Blanco, G. & Gaucher, C. 2005. Estratigrafı́a, paleontologı́a y edad de
la Formación Las Ventanas (Neoproterozoico, Uruguay). Latin Amer-
ican Journal of Sedimentology and Basin Analysis, 12, 115–131.
Blanco, G., Rajesh, H. M., Gaucher, C., Germs, G. J. B. & Chemale,
F., Jr. 2009. Provenance of the Arroyo del Soldado Group (Ediacaran
to Cambrian, Uruguay): implications for the paleogeographic evolu-
tion of southwestern Gondwana. Precambrian Research, 171, 57 – 73.
Campos Neto, M. C. & Figueiredo, M. C. 1995. The Rı́o Doce Orogeny,
Southeastern Brazil. Journal of South American Earth Sciences, 8,
143– 162.
Eyles, N. & Januszczak, N. 2004. ‘Zipper-rift’: a tectonic model for
Neoproterozoic glaciations during the break up of Rodinia after
750 Ma. Earth-Science Reviews, 65, 1– 73.
Fambrini, G. L., Paes-de-Almeida, R., Riccomini, C. & Fragoso-
Cesar, A. R. S. 2003. Tempestitos com Influência Glacial da Forma-
ção Playa Hermosa (Neoproterozóico), Piriápolis, Uruguai. Revista
Brasileira de Geociências, 33, 1 –12.
Fernandes, L. A. D. & Koester, E. 1999. An overview of the Neoproter-
ozoic Dorsal de Canguç u strike-slip shear zone and its role in the
tectonic evolution of the continental crust in southern Brazil. Journal
of African Earth Sciences, 29, 3– 24.
Gaucher, C., Boggiani, P. C., Sprechmann, P., Sial, A. N. & Fair-
child, T. 2003. Integrated correlation of the Vendian to Cambrian
Arroyo del Soldado and Corumbã Groups (Uruguay and Brazil)
palaeogeographic, palaeoclimatic and palaeobiologic implications.
Precambrian Research, 120, 241– 278.
Gaucher, C., Poiré, D., Peral, L. & Chiglino, L. 2005. Litoestratigra-
fı́a, Bioestratigrafı́a y correlaciones de las sucesiones sedimentarias
del Neoproterozoico-Cámbrico del Cratón del Rı́o de La Plata
(Uruguay y Argentina). Latin American Journal of Sedimentology
and Basin Analysis, 12, 145– 160.
Gaucher, C., Poiré, D., Finney, S. C., Valéncia, V., Blanco, G.,
Pamoukaghlian, K. & Gómez Peral, L. 2007. Zircones detrı́ticos
de secuencias neoproterozoicas de Uruguay y Argentina: Inferencias
sobre la evolución paleogeográfica del craton del Rı́o de La Plata. V
Congreso Uruguayo de Geologı́a, Montevideo, Uruguay. CD-Rom
(extended paper, 25).
Kirschvink, J. L. 1980. The least-squares and plane and the analysis of
palaeomagnetic data. Geophysical Journal of the Royal Astronomical
Society, 62, 699– 718.
Mallmann, G., Chemale, F., Jr., Avila, J. N., Kawashita, K. & Arm-
strong, R. A. 2007. Isotope geochemistry and geochronology of the
Nico Pérez Terrane, Rı́o de La Plata Craton, Uruguay. Gondwana
Research, 12, 489– 508.
Masquelin, H. & Sánchez-Bettucci, L. 1993. Propuesta de evolución
tectono-sedimentaria para la fosa tardi-Brasliana en la región
de Piriápolis, Uruguay. Revista Brasileira de Geociencias, 23,
313– 322.
Oyhantçabal, P., Siegesmund, S., Wemmer, K., Frei, R. & Layer, P.
2007. Post-collisional transition from calc-alkaline to alkaline mag-
matism during transcurrent deformation in the southernmost Dom
Feliciano Belt (Braziliano –Pan-African, Uruguay). Lithos, 98,
141– 159.
Pazos, P. J. & Rapalini, A. 2011. The controversial stratigraphy of the
glacial deposits in the Tandilia System, Argentina. In: Arnaud, E.,
Halverson, G. P. & Shields-Zhou, G. (eds) The Geological
Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 565–569.
Pazos, P. J., Tofalo, R. & Sánchez Bettucci, L. 1998. Procesos
sedimentarios e indicadores paleoclimáticos en la sección inferior
de la Formación Playa Hermosa, Cuenca Playa Verde, Piriápolis,
Uruguay. II Congreso Uruguayo de Geologı́a, Punta del Este,
Uruguay, 64 –69.
Pazos, P. J., Sánchez Bettucci, L. & Tófalo, R. O. 2003. The record of
the Varanger glaciation at the Rı́o de La Plata craton, Vendian-
Cambrian of Uruguay. Gondwana Research, 6, 65 –78.
Pazos, P. J., Sánchez Bettucci, L. & Loureiro, J. 2008. The Neopro-
terozoic glacial record in the Rı́o de La Plata Craton: a critical reap-
praisal. In: Pankhurst, R. J., Trouw, R. A. J., de Brito Neves,
B. B. & de Witt, M. J. (eds) West Gondwana Pre-Cenozoic Corre-
lations Across the South Atlantic Region. Geological Society,
London, Special Publications, 294, 343–364.
Pecoits, E., Gingras, M., Aubet, N. & Konhauser, K. 2008. Ediacaran
in Uruguay: palaeoclimatic and palaeobiological implications. Sedi-
mentology, 55, 689– 721.
Pecoits, E., Gingras, M. K. & Konhauser, K. O. 2011. Las Ventanas
and San Carlos formations, Maldonado Group, Uruguay. In:
Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geo-
logical Record of Neoproterozoic Glaciations. Geological Society,
London, Memoirs, 36, 555–564.
Rapalini, A. E. 2006. New Late Proterozoic paleomagnetic pole for the
Rı́o de La Plata craton: implications for Gondwana. Precambrian
Research, 147, 223– 233.
Rapela, C. W., Pankhurst, R. J. et al. 2007. The Rı́o de La Plata craton
and the assembly of SW Gondwana. Earth Science Reviews, 83,
49 – 82.
Sánchez Bettucci, L. 1998. Evolución tectónica del Cinturón Dom
Feliciano en la región Minas – Piriápolis, Uruguay. PhD thesis,
Universidad de Buenos Aires, Argentina.
Sánchez Bettucci, L. & Linares, E. 1996. Primeras Edades en Basaltos
del Complejo Sierra de Ánimas, Uruguay. XIII Congreso Geológico
Argentino and III Congreso de Exploración de Hidrocarburos I,
Buenos Aires, Argentina, I, 399–404.
 THE PLAYA HERMOSA FORMATION
Sánchez Bettucci, L. & Pazos, P. 1996. Análisis Paleoambiental y
Marco tectónico en la Cuenca Playa Verde, Piriápolis, Uruguay.
XIII Congreso Geológico Argentino, Buenos Aires, Argentina, I,
405– 412.
Sánchez Bettucci, L. & Ramos, V. A. 1999. Aspectos Geológicos de las
rocas metavolcánicas y metasedimentarias del Grupo Lavalleja,
Sudeste de Uruguay. Revista Brasilera de Geociencias, 29, 557– 570.
Sánchez Bettucci, L. & Rapalini, A. E. 2002. Palaeomagnetism of the
Sierra de Las Animas Complex, Southern Uruguay: Its Implications
in the Assembly of Western Gondwana. Precambrian Research,
118, 243– 265.
553
Sánchez Bettucci, L., Oyhantçabal, P., Preciozzi, F., Loureiro, J.,
Ramos, V. A. & Basei, M. A. S. 2004. Mineralizations of the Laval-
leja Group (Uruguay), a Neoproterozoic volcano– sedimentary
sequence. Gondwana Research, 7, 745–751.
Sánchez Bettucci, L., Koukharsky, M., Pazos, P. J. & Stareczek, S.
2009. Neoproterozoic subaqueous extrusive –intrusive rocks in the
Playa Hermosa Formation in Uruguay: regional and stratigraphic
significance. Gondwana Research, 16, 134– 144.
Sánchez Bettucci, L., Peel, E. & Masquelin, H. 2010. Neoproterozoic
tectonic synthesis of Uruguay. International Geology Review, 52,
51– 78.
 Chapter 53

Las Ventanas and San Carlos formations, Maldonado Group, Uruguay

ERNESTO PECOITS*, MURRAY K. GINGRAS & KURT O. KONHAUSER

Department of Earth and Atmospheric Sciences, University of Alberta, 1-26 Earth Sciences Building, Edmonton, AB, T6G 2E3, Canada
*Corresponding author (e-mail: epecoits@ualberta.ca)

Abstract: Together with the Playa Hermosa Formation (Fm.), the Las Ventanas and San Carlos formations constitute the Maldonado
Group, which is better developed in the southeastern part of Uruguay and covers an area of c. 200 km2. The total thickness of both units
(i.e. Las Ventanas and San Carlos formations) reaches c. 1500 m, and comprises mafic and acidic volcanic rocks, pyroclastic rocks, dia-
mictite, sandstone, conglomerate and pelite. Structurally, the Maldonado Group is extensively deformed, although variably, throughout
the region. Strike–slip faults, westward-verging detachment faults, and folds with axis sub-parallel to the strike–slip planes are common
features. The presence of pumpellyite, prehnite, chlorite and epidote in mafic rocks indicates very low- to low-grade metamorphic con-
ditions. The Las Ventanas Fm. is characterized by basal conglomerate, diamictite, sandstone and siltstone that pass upwards into fine-
grained rhythmites (pelite), and is the thickest unit of the Maldonado Group (c. 1250 m). The San Carlos Formation (c. 250 m) comprises
fine-grained conglomerate, sandstone and mudstone towards the top. Both units lie on an angular unconformity above Palaeo- and Neo-
proterozoic basement and are overlain unconformably by late Ediacaran– lowermost Cambrian units. Reliable palaeomagnetic data indi-
cate that the Maldonado Group accumulated at high palaeolatitudes; however, the palaeogeographical evolution of the Rı́o de la Plata
Craton during the Neoproterozoic remains conjectural. Radiometric data from intrusive bodies and cross-cutting strike– slip faults
place the minimum age of the group at c. 565 Ma, whereas basement volcanic rocks dated at 590 + 2 Ma interbedded with meta-sand-
stones hosting detrital zircons c. 600 million years old provide the best constraint on the maximum age of deposition. Given the absence
of carbonate rocks, no chemostratigraphic studies (e.g. C, O, Sr) are available.
The Las Ventanas and San Carlos formations are largely interpreted as units within a thick glacially influenced fan-delta sedimentary
system formed during the early Ediacaran in a strike–slip basin. Based on stratigraphic and sedimentological characteristics it has been
suggested that this succession, containing glacially influenced diamictite and dropstones, records a glacial period that occurred sometime
between c. 570 and 590 million years ago. Ongoing research is focused on establishing the precise age of deposition of the Maldonado
Group and on reconstructing the tectonic evolution of the basin. Further palaeomagnetic studies will be especially useful for determining
the palaeogeography of the Rı́o de la Plata Craton during the Ediacaran and establishing its relationships with neighbouring strata hosting
similar successions.

The Maldonado Group was formally erected by Pecoits et al.
(2005) to include the Playa Hermosa (Pazos et al. 2011) and Las
Ventanas formations, which are better exposed near the towns of
Piriápolis and Pan de Azúcar (Figs 53.1 & 53.2). The San Carlos
Fm. was informally included. Subsequent work showed that the
succession continues to the SW and NE of Minas and Melo
cities (Fig. 53.1). The group reaches a maximum thickness of
c. 1500 m and covers an area of c. 200 km2. It comprises acidic
and basic volcanic rocks, pyroclastic and sedimentary strata gener-
ated in a tectonically active and glacially influenced basin (for a
recent review see Pecoits et al. 2008).
Midot (1984) originally suggested the Las Ventanas Fm.
included conglomerate, sandstone and pelite outcropping at De
Las Ventanas Hill and in the surrounding areas (Fig. 53.3). This
unit was considered by Midot (1984) and various other authors
to be Ordovician in age (e.g. Bossi & Navarro 1991; Masquelin
& Sánchez 1993; Pazos et al. 2003). This assumption was
mainly founded on the inferred development of alluvial fans
sourced from the Cambrian Sierra de las Ánimas Complex,
located westward (Fig. 53.2). However, detailed mapping and stra-
tigraphic analysis of the Las Ventanas Fm. led to its redefinition
as a Neoproterozoic volcanic/sedimentary succession (Pecoits
2003a). The sections exposed in the northern and southern parts
of the type area were designated as the stratotype and parastrato-
type of the unit, respectively (Pecoits 2003a). The former is
located near Paso del Molino, where 1200 m of Las Ventanas
strata are continuously exposed. The parastratotype is situated
near the Burgueño Quarry and Apolonia Mine, where the uncon-
formable contact between the Las Ventanas Fm. and the basement
(Lavalleja Group) is exposed (Fig. 53.3).
The San Carlos Fm. was erected by Masquelin (1990), who
documented that the unit consists of conglomerate, sandstone
and pelite. The depositional environment was likely lacustrine or
fluvio-lacustrine. The stratotype of the formation is located 6 km
south of San Carlos town (Fig. 53.2), where 220 m of San Carlos
strata are exposed with the base and top of formation not visible
(Pecoits et al. 2008). The sedimentary facies and volcanic associ-
ation of the San Carlos Fm. are similar to those of the middle Las
Ventanas Fm. Likewise, palynological macerations carried out in
the pelites of both units reveal the occurrence of similar microbiota
(Pecoits et al. 2005). These observations led Pecoits et al. (2005) to
propose a correlation between the San Carlos and Las Ventanas
formations. Whether both units were deposited in the same
basin, and subsequently dismantled by the displacement of the
Sierra Ballena Shear Zone (Fig. 53.2), or were developed within
different depocentres remains uncertain.
The first evidence of glacial influence in the Las Ventanas Fm.
was recorded by Pecoits (2003a), comprising faceted, outsized
clasts in finely laminated rhythmites that were interpreted as drop-
stones. Recently, Gaucher et al. (2008) reported glacial diamictite
with associated dropstones occurring in laminated siltstone to the
south of Minas (Fig. 53.2). Additionally, glacial diamictite and
fine-grained rhythmites containing striated dropstones are well
exposed c. 15 km NW (El Perdido area) of this locality (Pecoits
et al. 2008). No glacial evidence has yet been recorded in the
San Carlos Fm.
Structural framework
In Uruguay, a significant extensional and synkinematic magmatic
event corresponding to the final stages of the SW-Gondwana
assembly occurred during the Neoproterozoic– lowermost Cam-
brian (Bossi & Campal 1992; Pecoits 2003b; Oyhantçabal 2005).
From a structural perspective, the Sierra Ballena Shear Zone
(Figs 53.1 & 53.2) constitutes the largest remnant of the

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 555– 564. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.53
556 E. PECOITS ET AL.

50° W
WA
SOUTH 3
AMERICA
A SF

C
Florianópolis
Z

ne
RP

Zo
K

r
ea
Sh
(a) (b) Serra Azul Fm

o
in
rc
N

Ge
54° W

r
ajo
BRAZIL Porto Alegre

M
30° S

Rivera 2
ARGENTINA

Paysandú Melo
e
on
r Z
a
She

an
ce
Ballena

O
URUGUAY
ic
ra

nt

Fig. 53.1. (a) Distribution of cratonic

Sier

Minas 34° S
tla

blocks of west Gondwana: A, Amazonian;

A

1 C, Congo; K, Kaoko; SF, São Francisco;

RP, Rı́o de la Plata; WA, West African; Z,
Buenos Montevideo Zaire. (b) Regional map showing the
Aires Fig. 53.2 location of the units discussed in the text
(see (c) for more detail). (c) Distribution of
volcano-sedimentary early Ediacaran units
of Uruguay and southeastern Brazil: 1, Las
(c) 58° W 54° W 0 100 200 Km Ventanas-San Carlos formations; 2,
Camaquã Basin; 3, Itajaı́ Basin.

Brasilian –Pan African Orogeny (c. 700– 500 Ma). This high-
strain transcurrent structure, which was operative primarily
between c. 600 and 580 Ma, contributed significantly to the basin-
fill architecture of the early Ediacaran units (Oyhantçabal 2005). In
this regard, the Las Ventanas and San Carlos formations were
deposited in a strike –slip basin, as indicated by (i) the diverse
depositional facies and their abrupt lateral changes; (ii) apparent
migration of the primary depocentre towards the south; (iii) the
subparallel trend of the basin with respect to its strike –slip
margins; and (iv) synchronous timing with regional shearing
(Sierra Ballena Shear Zone). Following the development of the
widespread transcurrent system, a gravitational orogenic collapse
characterized by high-angle normal faulting and accompanied by
marine transgression occurred during the late Ediacaran –Early
Cambrian (Pecoits et al. 2008).
A similar geotectonic evolution is observed in the associated
magmatism, which was initiated with highly fractionated
calc-alkaline granite (c. 584 Ma), followed by mildly alkaline
granite and shoshonitic volcanics (c. 575 Ma), and concluded
with peralkaline intrusions and volcanics (c. 540 –520 Ma)
(Oyhantçabal et al. 2007). Therefore, the Ediacaran –Early Cam-
brian in Uruguay is characterized by a transition from a back-arc
basin (underlying Lavalleja Group), followed by a strike –
slip-related basin (Maldonado Group), to a foreland basin
(overlying Arroyo del Soldado Group), where strike– slip shear
zones of crustal scale played a major role in the evolution of
the orogen.
At the outcrop scale, the Las Ventanas and San Carlos for-
mations show evidence of both brittle and ductile deformation.
Small- and large-scale strike –slip faults, westward-verging
detachment faults, and folds with axis sub-parallel to the strike –
slip planes are common features (Fig. 53.3). Axial plane slaty
and sporadic millimetre-spaced fracture cleavages are present in
fine-grained facies (pelite). The basic volcanic and pyroclastic
rocks show abundant chlorite and epidote as well as pumpellyite
and prehnite, demonstrating very low- to low-grade metamorphic
conditions (Pecoits 2003a).
Stratigraphy
The Las Ventanas and San Carlos formations lie on an angular
unconformity above a crystalline basement of undetermined
age and the Lavalleja Group (Figs 53.4 & 53.5). Relatively well-
dated basement granitoids are represented by the Campanero
Complex and the Cerro Olivo Complex, with ages of c. 1750 Ma
and 1006 + 37 Ma, respectively (Table 53.1; Oyhantçabal
2005 and references therein). The lithostratigraphy of the Lavalleja
 LAS VENTANAS AND SAN CARLOS FORMATIONS 557

Fig. 53.2. Simplified geological map showing the distribution of the Las Ventanas and San Carlos formations (Maldonado Group) and selected age determinations of
southeastern Uruguay (see Table 53.1 and text for explanation). Sources of geological information: Bossi & Navarro (1991), Pecoits et al. (2005), Oyhantçabal (2005) and
references therein. SYPSZ, Sarandı́ del Yı́-Piriápolis Shear Zone; SBSZ, Sierra Ballena Shear Zone.

Group, although poorly known, is different from that of the Las
Ventanas and San Carlos formations. According to Midot
(1984), the Lavalleja Group is a volcanosedimentary succession
dominated by immature fine-grained siliciclastics, marl, basalt
and limestone towards the top. This limestone hosts columnar stro-
matolites assignable to Conophyton (Poiré et al. 2005), and
although their occurrence extended from the early Proterozoic to
the Ediacaran, preliminary radiometric studies suggest an early
Ediacaran age for this unit. SHRIMP U – Pb detrital zircon analyses
from the Lavalleja Group display ages between 3.4 and 0.6 Ga
(Basei et al. 2008). Likewise, interbedded basalt shows a crystal-
lization age of 590 + 2 Ma (U–Pb SHRIMP; Mallmann et al.
2007), indicating that the deposition must have occurred c. 590
million years ago.
The whole succession (i.e. the Las Ventanas and San Carlos for-
mations) can be divided (from base to top) into three informal
intervals: (i) volcanic and pyroclastic deposits, (ii) conglomerate-
dominated lithofacies and (iii) pelite-dominated lithofacies. Vol-
canic and pyroclastic rocks including basalt, mafic hyaloclastic
breccias and subaqueous tuff, as well as rhyolite, acidic volcano-
clastic and pyroclastic rocks, have been recognized as part of the
Maldonado Group (Fig. 53.4a). This bimodal volcanism has long
been thought to represent part of the Sierra de las Ánimas
Complex (e.g. Sánchez & Rapalini 2002 and references therein).
The latter, however, shows geochemical signatures, radiometric
ages and structural features indicating anorogenic magmatism,
which was extruded after the main deformational phase that
affected the Maldonado Group (Oyhantçabal 2005). In fact, the
Sierra de las Ánimas Complex systematically displays radiometric
ages younger than those of the volcanics assigned to the Maldo-
nado Group (see ‘Geochronological constraints’) and it displays
neither ductile deformation nor metamorphism. Field relationships
show that the Sierra de las Ánimas Complex intrudes the Las
Ventanas Fm., providing definite evidence of the older age of the
Las Ventanas.
Conglomerate-dominated lithofacies (proximal facies associ-
ation of Pecoits 2003a) dominate the basal part of the Las Ventanas
Fm., including clast-supported conglomerate and breccia, diamic-
tite, massive sandstone and conglomerate –sandstone couplets.
Upwards, pelites are abundant with occasional conglomerate
beds (pelites-dominated lithofacies or distal facies association
of Pecoits 2003a). This lithofacies includes laminated siltstone
and sandstone-pelite rhythmites, and massive sandstone and
conglomerate. Likewise, the San Carlos Fm. consists of basal
conglomerate- and upper pelite-dominated lithofacies, but the
clast size in the lower conglomerate never reaches that of the
Las Ventanas Fm. In this sense, two possible explanations can
be drawn. First, the San Carlos Fm. represents a lateral equivalent
558 E. PECOITS ET AL.

Fig. 53.3. Geological map of the type area of the Las Ventanas Fm. The inset shows the location of the stratotype (1) and parastratotype (2) (modified from Pecoits
et al. 2008).
 LAS VENTANAS AND SAN CARLOS FORMATIONS 559

Fig. 53.4. (a) Stratigraphic column of the Las Ventanas Fm. at its stratotype (point 1, Fig. 53.3). (b) Stratigraphic section for the lower Las Ventanas Fm. at its
parastratotype (point 2, Fig. 53.3) (modified from Pecoits et al. 2008). K-Pg, potassium and plagioclase.

to the middle and uppermost part of the Las Ventanas Fm.,
where coarse-grained conglomerate is rare (see ‘Glaciogenic
deposits and associated strata’). Second, both units, although
potentially contemporaneous (see below), were deposited in
different basins.
The Las Ventanas is thought to be overlain by the Arroyo del
Soldado Group, a thick (3000 m) mixed siliciclastic –carbonate
succession, mainly represented by intercalating conglomerate,
sandstone, siltstone, thick carbonate, Fe-formation, black- and
iron-rich shale and chert. It contains a rich fossil assemblage
560
Fig. 53.5. Simplified stratigraphic column of the San Carlos Fm. at its
stratotype, 6 km SE of San Carlos town (see Fig. 53.2) (modified from Pecoits
et al. 2008).
composed of organic-walled microfossils and small shelly fauna,
including the index fossil Cloudina riemkeae (Gaucher et al.
2004). The distinct lithological differences between Las Ventanas
and San Carlos formations with the Arroyo del Soldado Group
have been explained by different prevailing climatic conditions
and tectonic settings. Whereas an evolution towards tropical con-
ditions (Gaucher et al. 2004) and a marine transgression in a fore-
land setting (Pecoits et al. 2005; Basei et al. 2008) have been
E. PECOITS ET AL.
proposed for the latter, a glacially influenced system in a strike –
slip setting was suggested for the Las Ventanas and San Carlos
formations (Pecoits et al. 2005, 2008). The presence of abundant
organic-walled microfossils and shelly fauna in the Arroyo del
Soldado Group points to high biological productivity and an
elevated nutrient supply, possibly related to increased weathering
during warmer conditions (Gaucher et al. 2004). Based on avail-
able geochronology, chemostratigraphy and biostratigraphy,
Pecoits et al. (2008) proposed a maximum depositional age of
c. 560 Ma for the group, which is younger than earlier suggestions
(Gaucher et al. 2004).
Glaciogenic deposits and associated strata
Palaeoenvironmental interpretations indicate that the Las Venta-
nas –San Carlos system records sheet flood-dominated fan-delta
deposits in a glacially influenced setting. Direct evidence of
ancient glacial activity comes from the basal and uppermost
facies of the Las Ventanas Fm., where some deposits have been
described and interpreted as ice-rafted diamictite with striated
and faceted clasts and rhythmite-hosting dropstones (Pecoits
2003a; Pecoits et al. 2008).
Diamictites are mainly coarse-grained and matrix-supported
lithofacies with a massive structure displaying normal and
occasionally reversed grading. Clasts are rounded to angular and
range from granule to boulder size. Compositionally, the diamic-
tite dominantly contains extrabasinal clasts (rhyolite, basalt,
granitoid, gabbro, quartzite). Two types of diamictite can be dis-
tinguished: sub-rounded pebbles and cobbles in a massive
muddy matrix, and angular to sub-rounded cobbles to boulders
in a massive to horizontally laminated silty/clayey matrix.
These deposits have a polymodal texture and form relatively
thick (2 –10 m) tabular beds with faceted and occasionally
striated clasts.
Although common in the uppermost part of the succession, fine-
grained rhythmites are also found interbedded with diamictite at
the base of the Las Ventanas Fm. The ,3-m-thick beds are charac-
terized by a millimetre- to centimetre-intercalation of silty and
sandy material with clay and also by the presence of out-sized
clasts deforming the layering. Based on the diverse composition
of these large clasts (granite, basalt, gabbro, etc.), and on the pres-
ence of pre-depositional foliation, an interpretation of the clasts as
volcanic bombs or other ballistic/pyroclastic material is discarded.
These lithofacies (i.e. rhythmites with out-sized clasts and diamic-
tite) were first described at the parastratotype section of the unit
(Burgueño Quarry and Apolonia Mine; Fig. 53.3), but later discov-
ery in other localities (e.g. NE Minas and Melo; Fig. 53.1)
suggesting that they are more extensive than originally thought.
The type section of the Las Ventanas Fm. is largely dominated
by conglomerate, sandstone and finely laminated siltstone. It begins
with a 690-m-thick fining- and thinning-upward cycle (Fig. 53.4a).
Conglomerate, sandstone and laminated siltstone dominate the
lowermost, medial and uppermost sub-cycles, respectively. The
conglomerate is typically granitic and clast-supported with
arkosic sandstone present at the top of each sub-cycle. The follow-
ing changes occur up-section within the lower major cycle: (i) bed
thickness progressively decreases, from metre-scale to a few
millimetres (laminae); (ii) average grain size decreases from
pebbles to silt; (iii) the proportion of granitic clasts becomes
smaller; and (iv) planar parallel stratification and lamination
become a common feature in the siltstone but are absent in the
lower and middle part of the cycle. This finely laminated siltstone
shows similar features to those described in the basal part of the
unit. Here, the lithofacies is considerably thicker, but the out-sized
clasts (dropstones) identified in it are smaller and rarely reach more
than 10 cm.
The formation passes up-section into a second major cycle that
is nearly 560 m thick, and is composed of minor subcycles of
 LAS VENTANAS AND SAN CARLOS FORMATIONS 561

Table 53.1. Summary of geochronological data available from southeastern Uruguay (see Fig. 53.2)

Stratigraphic unit Rock type Method Age (Ma)

Sierra de las Ánimas Complex Porphyres Rb–Sr 520 + 5

Syenite Ar–Ar 579 + 1.5
A8 del Soldado Group* Depositional age: c. 560– 530 Ma
El Renegado Granite Rb–Sr 559 + 28
Puntas del Pan de Azúcar Lineament ‘Mylonite’ K–ArMusc 572 + 7
Aiguá Batholith Granite Rb–Sr 587 + 16
Solı́s de Mataojo Tonalite U–Pb 584 + 13
Aguas Blancas Mylonite K–ArMusc 594 + 13
Las Ventanas and San Carlos formations* Depositional age: c. 590– 570 Ma
Las Flores Trachybasalt K–Ar 615 + 30
Quartz-sericite schist U–Pb(SHRIMP) 600–3400
Metabasalt U–Pb(SHRIMP) 590 + 2
Lavalleja Group*
Metarhyolite U–Pb 624 + 14
Metarhyolite U–Pb 667 + 4
Cerro Olivo Complex Orthogneiss U–Pb 1006 + 37
Zanja del Tigre Fm.* Meta-sandstone U–Pb(SHRIMP) 1800–3400
Campanero Complex Orthogneiss U–Pb 1735 + 2
Orthogneiss U–Pb 1754 + 7

Source: Oyhantçabal (2005) and references therein, Mallmann et al. (2007), Basei et al. (2008), Pecoits et al. (2008). *Volcanic-sedimentary units.

sandstone and fine-grained conglomerate. The sandstone has a Chemostratigraphy

tabular geometry, is massive in appearance, and occasionally has
non-erosive basal contacts. The conglomerate is clast-supported,
polymictic and has a modal grain size of 3–10 cm. The clast com-
position is variable, and includes rhyolite (32%), granite (2%),
quartz (12%), basic volcanic rocks (11%), alkaline feldspar (10%),
plagioclase feldspar (8%) and schist (5%). Clasts, either in con-
glomerate or sandstone, are fresh and show no signs of chemical
weathering.
Boundary relations with overlying and underlying
non-glacial units
The Las Ventanas Fm. rests unconformably on the Lavalleja Group.
Palaeoproterozoic orthogneiss (Campanero Complex) and Monzo-
granite (La Nativa) are of unknown age (Pecoits et al. 2008). A non-
conformity separates the San Carlos Fm. from Palaeoproterozoic
ortho- and paragneiss (Cerro Olivo Complex; Fig. 53.2). Particularly
important is the relationship with the Lavalleja Group. This unit is
largely dominated by basalt and immature fine-grained siliciclastic
rocks (Midot 1984). An evolution towards warm climate and
stable tectono-magmatic conditions is evidenced by thick stromato-
litic limestones developed in the uppermost part of the unit (Poiré
et al. 2005), upon which the Las Ventanas Fm. uncomformably
lies. However, as discussed below, the hiatus between both units
is poorly constrained. Despite this, the transition from the Lavalleja
Group to the Las Ventanas Fm. is not only indicated by an angular
unconformity but also by a strong change in climatic and
tectono-magmatic activity.
The relationship and nature of the contact between Las Ventanas
and San Carlos formations with the overlying late Ediacaran Arroyo
del Soldado Group is not firmly established. The best locality to
study such transition is to the north of Minas (Fig. 53.2) where
both units are closely exposed. Detailed geological mapping of the
area indicates stratal juxtaposition due to tectonic shortening in a
zone of major thrusting. In other words, both units would be separ-
ated by major structural discontinuities (i.e. thrusts oriented SW–
NE), and no conformable contacts have been recorded. The
contact between the units – although not well exposed – suggests
the presence of an angular unconformity separating the uppermost
fine-grained rhythmites of the Las Ventanas Fm. and the sandstone
of the basal Arroyo del Soldado Group (Yerbal Fm.).
Owing to the lack of carbonate rocks directly associated with the
Las Ventanas or San Carlos formations, no chemostratigraphic
studies (e.g. C, O, Sr) have been performed.
Other characteristics
No evidence of mineral deposits was found in any of the units
described here.
Palaeolatitude and palaeogeography
The location and kinematic history of the blocks involved during
the assembly of West-Gondwana in the late Neoproterozoic is
poorly known (Rapela et al. 2007). In particular, the palaeogeogra-
phical position of the Rı́o de la Plata Craton (Fig. 53.1 inset) during
the Ediacaran is highly disputed (e.g. Cordani et al. 2000). In this
regard, no palaeomagnetic studies have been performed either in
the sedimentary rocks of Las Ventanas Fm. or in the San Carlos
Fm. Preliminary mean geomagnetic poles were only obtained
from sedimentary rocks of the Playa Hermosa Fm., volcanics
from the Las Ventanas Fm., and volcanics and intrusives from
the Sierra de las Ánimas Complex (Sánchez & Rapalini 2002).
According to the same authors, the new data support the Apparent
Polar Wander Path (APWP) previously suggested for the entire
Gondwana since c. 550 Ma, indicating that the Rı́o de la Plata
Craton was indeed at that time part of the supercontinent. Further-
more, it was suggested that a mean geomagnetic pole obtained
from the Playa Hermosa Fm. (12.7 þ 9.5/–8.18) meant that
another example of Neoproterozoic low-latitude glaciation is
evident in Uruguay.
The most reliable palaeomagnetic pole for early Ediacaran units
of the Rı́o de la Plata Craton is derived from the Campo Alegre
lavas (Sánchez & Rapalini 2002). The Campo Alegre Fm., dated
by the U –Pb method at 595 + 5 Ma (Citroni et al. 1999), is
located in the Itajaı́ Basin, SE Brazil (Fig. 53.1). Palaeomagnetic
reconstructions indicate a moderate palaeolatitude of 33.3 +
9.58S (D’Agrella & Pacca 1988), in marked contrast to the low
palaeolatitudes (12.7 þ 9.5/ –8.18) proposed for the Playa Hermosa
Fm. (Sánchez & Rapalini 2002).
562 E. PECOITS ET AL.
In contrast to the scarce database for the early Ediacaran units,
the APWP for Gondwana since 550 Ma is better known. Since
550 Ma, poles for the Rı́o de la Plata Craton and other Gondwanan
continents have tended to form a single APWP ranging from
c. 308S in the late Ediacaran towards lower palaeolatitudes during
the Lower Cambrian (Meert & Van der Voo 1996).
Geochronological constraints
Since the definition of both units, the age of the Las Ventanas and
San Carlos formations was considered Ordovician (e.g. Midot
1984; Masquelin & Sánchez 1993; Pazos et al. 2003). This
assumption was challenged by Pecoits et al. (2008) in reporting
cross-cutting relationships with several intrusive Ediacaran bodies
and major faults that indicated a minimum depositional age of
c. 570 Ma. The age of deposition of the Las Ventanas and San
Carlos formations is now relatively well constrained to c. 590–
575 Ma by radiometric data based on K – Ar, Rb –Sr and U –Pb
dating techniques on basement rocks, interbedded basalt,
intrusive syenite, granitic and trachytic dykes, and cross-cutting
faults (Table 53.1).
This inference is supported by the 590 + 2 Ma age (SHRIMP
U –Pb) obtained for a metabasalt (Mallmann et al. 2007), the det-
rital zircons from the Lavalleja Group in the basement of the Las
Ventanas Fm., which show U– Pb (SHRIMP) ages between 3.4
and 0.6 Ga (Basei et al. 2008), and an intrusive syenite that
yields Ar/Ar ages of 579 +1.5 Ma (Oyhantçabal et al. 2007). Fur-
thermore, the ages are corroborated by (i) basic volcanics inter-
bedded with sedimentary rocks of the Las Ventanas Fm., which
display ages between 615 + 30 and 565 + 30 Ma (K –Ar
method; Sánchez & Linares 1996); (ii) Rb –Sr ages from intrusive
granite with an age of 559 + 28 Ma (Preciozzi et al. 1993) and tra-
chyte of the Sierra de Las Ánimas Complex, which intrude and
overly the Las Ventanas Fm., dated between 520–530 Ma (Bossi
et al. 1993; Linares & Sánchez 1997); (iii) basic dykes cross-
cutting the Las Ventanas Fm. to the south of Minas yielding a
K –Ar age of 485 +12.5 Ma (Poiré et al. 2005); (iv) the last reac-
tivation of the Puntas del Pan de Azúcar Lineament, which cross-
cuts the Las Ventanas Formation (Fig. 53.3), and occurs at
572 + 7 Ma (K –Ar in syn-kinematic muscovites) (Bossi &
Campal 1992); and (v) the San Carlos Fm., which is intensively
deformed by the Sierra Ballena Shear Zone, in which the third
and last deformation phase occurred at c. 550 –500 Ma (Oyhantça-
bal 2005). Finally, the transcurrent tectonics that occurred during
the early Ediacaran, which is closely related to the generation
of strike –slip basins recorded for example by the Las Ventanas
and San Carlos formations, is also associated with a voluminous
syn-kinematic magmatism (Pecoits 2003a; Pecoits et al. 2005).
Radiometric studies performed in all these bodies yield ages sys-
tematically between 570 and 590 Ma (Rb–Sr and U–Pb methods)
(Bossi et al. 1993; Hartmann et al. 2002; Oyhantçabal 2005).
According to Gaucher et al. (2008), the acritarch assemblage
recovered from the Las Ventanas Fm. indicates a depositional
age between 635 and 582 Ma, supporting previous data (Pecoits
2003a, b). This assemblage, however, comprises and is dominated
by individuals with no stratigraphic value, such as Leiosphaeridia
and others of doubtful origin (e.g. Soldadophycus) (Butterfield,
pers. comm. 2008).
Discussion
The facies associations point to the development of sheet flood-
dominated alluvial fans (Blair & McPherson 1994) intercalated
with minor lake deposits in a glacially influenced, transtensional
tectonic setting. The proximal facies association comprises
massive and horizontally stratified clast-supported conglomerate
and rare breccia, massive sandstone, conglomerate – sandstone
couplets and diamictite, while the distal facies includes massive
and normally graded sandstone, pebbly sandstone, laminated
siltstone, and fine-grained massive and graded conglomerate.
The proximal facies association was interpreted by Pecoits
(2003a) as a subaerial alluvial fan in which debris-flow deposits
(diamictite) and sheet flood deposits (stratified conglomerate and
sandstone) constitute the dominant facies. The subaerial alluvial
fan succession is characterized by upward-coarsening and
upward-thickening trends resulting from fan progradation. The
restricted occurrence of debris flow beds and the comparatively
high roundness of the clastic fraction indicate that the preserved
succession represents middle and outer regions of the alluvial-fan
complex. The distal facies association is thought to represent a sub-
marine delta subenvironment with sediment gravity-flow deposits
occasionally interbedded with turbidites (massive and graded
conglomerate and sandstone) and suspension fallout deposits
(laminated siltstone). Although some conglomeratic levels are
interpreted to represent shoreline deposits along the distal fan,
no evidence of wave reworking has been observed (Pecoits 2003a).
The proximal facies association offers evidence of sedimen-
tation under arid climatic conditions, as shown by exceptionally
fresh well-rounded clasts (e.g. basalt) in conglomerate and sand-
stone. Glacial sedimentary evidence comes from the distal facies
association. Therein, outsized clasts within finely laminated silt-
stone have been recorded and interpreted as dropstones. This litho-
facies has been described at the base and top of the Las Ventanas
Fm. In the first case, the laminated strata are ‘sandwiched’ between
massive and bedded diamictite, mostly containing extrabasinal
clasts. Faceted, striated and bullet-shaped clasts are consistent
with glacial transport and suggest a glacial influence during the
deposition of laminated siltstone and diamictite facies. In contrast,
the laminated siltstone described at the top of the unit overlies
laminated siltstone and fine-grained sandstone, which are inter-
preted as turbiditic deposits (Pecoits et al. 2008). Here, the gla-
cially influenced laminated siltstone is differentiated from the
turbidites by the lack of turbidity current structures, finer grain
size and numerous dropstones with impact-induced deformation
of underlying laminae.
Despite the lack of evidence for glacially influenced sedimen-
tation in the San Carlos Fm., the structural and geochronological
framework, stratigraphy and fossil content support the premise
that the San Carlos Fm. is correlative with the middle– upper
part of the Las Ventanas Fm. (Figs 53.4a & 53.5). This would
explain the absence of the basal and uppermost glacially influ-
enced facies described for the Las Ventanas Fm.
Only one systematic palaeomagnetic study has been performed
in glacial Ediacaran units and associated rocks of Uruguay
(Sánchez & Rapalini 2002); however, none of the sampled sites
corresponds to the Las Ventanas and San Carlos formations.
Unfortunately, all the samples from this work were collected
near the border of the Dom Feliciano belt, in an area affected by
intense Neoproterozoic –Cambrian tectono-magmatic activity,
and thus probably affected by widespread remagnetization (e.g.
Rapalini & Sánchez 2008). The two palaeopoles obtained from
volcanic and hypabyssal rocks of the Sierra de las Ánimas
Complex are, from geochronological and structural points of
view, poorly constrained. Much of the radiochronology is based
on the K – Ar method, which usually provides a minimum age,
and recent dating on the same lithologies using more precise
methods has yielded ages as much as 30 Ma older (Oyhantçabal
et al. 2007). Although the early Ediacaran units of Uruguay are
extensively deformed, the interbedded basalt samples for palaeo-
magnetism were not corrected with respect to the palaeohorizontal.
Field relationships have extensively shown that even the youngest
rocks of the Sierra de las Ánimas Complex (i.e. c. 525 million years
old), although not folded, are tilted. Therefore the integrity and
usability of these palaeomagnetic data are problematic.
Recent palaeogeographical reconstructions locate the Rı́o de
la Plata Craton at high palaeolatitudes c. 580 million years ago
 LAS VENTANAS AND SAN CARLOS FORMATIONS
(Trinidade & Macouin 2007). Inclination data from deposits
slightly older than the Gaskiers equivalent in the Avalon Terrane
(Newfoundland) similarly indicate a palaeolatitude of 358S during
the early Ediacaran (c. 608 Ma, U –Pb zircon age; Myrow &
Kaufman 1999). However, palaeogeographical models between
590 and 560 Ma (i.e. when Gaskiers deposits and their possible
Ediacaran correlatives, Squantum, Loch na Cille, and Moelv,
were formed) are controversial due to the ambiguous results pre-
sented by the Laurentian palaeopoles (Trinidade & Macouin
2007 and references therein). For instance, both low and high lati-
tudes for Laurentia at c. 580 Ma have been proposed. If the latter
configuration is confirmed, the glacial strata observed in Laurentia,
Baltica, Cadomia, Avalonia and Rı́o de la Plata cratons are compa-
tible with a palaeoclimatic scenario similar to the Phanerozoic gla-
ciations rather than ‘snowball’ conditions.
Although early proposals promoted a ‘Marinoan’ age for Las
Ventanas and San Carlos formations (Pecoits et al. 2005), an Edia-
caran event seems to be a more reasonable alternative (Pecoits
et al. 2008) based on the radiometric constraints. Indeed, this
would explain the absence of thick ‘cap carbonate’ facies immedi-
ately overlying these deposits as is distinctive of Ediacaran glacial
deposits. Alternatively, the absence of cap carbonates might be due
to deposition in a highly active tectonic setting, characterized by
high rates of subsidence and high accumulation rates of siliciclas-
tic sediments, lack of preservation because they were eroded; or,
they may simply not have been found yet. However, the typical
facies of the Las Ventanas and San Carlos formations indicates
an active participation of the hydrological cycle that is incompati-
ble with the ‘Snowball Earth’ model for Cryogenian glaciations
(e.g. Hoffman & Schrag 2002).
Recently, a similar glacial succession (Tacuarı́ Fm.) was
described in NE Uruguay (Veroslavsky et al. 2006). These depos-
its were long-considered a classic example of the Carboniferous –
Permian glaciation in Gondwana (e.g. Bossi & Navarro 1991).
According to Veroslavsky et al. (2006) and mainly based on a
very similar fossil content to that described for the Las Ventanas
Fm., this unit was tentatively assigned to the Neoproterozoic.
Ongoing research using radiometric dating on cross-cutting grani-
tic dykes (U–Pb TIMS) and detrital zircons (LA-ICP-MS) has
confirmed the Ediacaran age of the succession. The obtained
ages constrain the deposition of the unit between 590 and
570 Ma. This is in agreement with the age proposed for the Las
Ventanas Fm. (570 –590 Ma) and would suggest regional glacial
conditions.
Sedimentological, tectonic and magmatic evidence, supported
by radiometric ages, suggests some similarities between the Mal-
donado Group and other successions in Brazil. In this regard, the
Las Ventanas and San Carlos formations have been correlated
with the Bom Jardim (c. 592–573 Ma) and Cerro do Bugio
(573 –559 Ma) allogroups of the Camaquã Basin located in Rio
Grande do Sul, southern Brazil (Pecoits 2003b; Fig. 53.1). The
Bom Jardim Allogroup is composed of basic to intermediate
volcanic rocks, alluvial conglomerate and turbidites (Paim et al.
2000). SHRIMP age dating of the volcanics yielded an age of
c. 580 Ma (Paim et al. 2000). The Cerro do Bugio Allogroup con-
sists of acidic and basic rocks, alluvial conglomerate, rhythmites
(sandstone-pelite) and pelite. Geochronological studies on acidic
rocks yielded a U – Pb age of 573 + 8 Ma (Paim et al. 2000).
Both the Bom Jardim and Cerro do Bugio allogroups were
deformed by sinistral transcurrent displacement dated to c. 570 Ma
and were subsequently intruded by granitic bodies dated at
559 + 7 Ma and 565 + 14 to 561 + 6 Ma (Paim et al. 2000 and
references therein). Evidence of a seasonal glacial influence
has been suggested by Eerola (2001, 2006) for the Bom Jardim
Allogroup.
Recently, Alvarenga et al. (2007) reported glacial deposits in the
Ediacaran Serra Azul Fm. in the Paraguay belt, Brazil (Alvarenga
et al. 2011). The Paraguay belt is located on the south-eastern edge
of the Amazon Craton (Fig. 53.1), which in conjunction with the
563
Rı́o de la Plata Craton was probably already amalgamated into a
single crustal block by the Ediacaran (e.g. Cordani et al. 2000).
Unlike the Las Ventanas –San Carlos formations and their Brazi-
lian correlatives Cerro do Bugio –Bom Jardim allogroups, the
Serra Azul Fm. was deposited on a passive margin (Alvarenga
et al. 2007), showing no comparable tectono-magmatic activity
but probably similar palaeolatitude (see above). The correlation
between Ediacaran (c. 590–570 Ma) glacially influenced succes-
sions of Brazil and Uruguay strengthen the notion of a post Cryo-
genian glaciation and suggests that these deposits are distributed
more extensively than previously recognized in South America.
Future efforts focused on the sedimentological constraints of
these and other successions (e.g. Itajaı́ Basin in Brazil) are required
to determine the glacial influence on the Brazilian units.
The authors wish to thank Natalie Aubet who actively participated in the
fieldwork and in the preparation of the figures. Constructive comments were pro-
vided by the reviewers Ruben Rieu, Pablo Pazos and editor Emmanuelle Arnaud.
The research was partially funded by the Natural Sciences and Engineering
Research Council of Canada Discovery Grants to KOK and MKG. This is a con-
tribution to IGCP 512 ‘Neoproterozoic Ice Ages”.
References
Alvarenga, C. J. S. de., Figueiredo, M. F., Babinski, M. & Pinho,
F. E. C. 2007. Glacial diamictites of Serra Azul formation (Ediacaran,
Paraguay belt): evidence of the Gaskier event in Brazil. Journal of
South American Earth Sciences, 23, 236– 241.
Alvarenga, C. J. S., Boggiani, P. C. et al. 2011. Glacially-influenced
sedimentation of the Puga Formation, Cuiabá Group and Jacadigo
Group, and associated carbonates of the Araras and Corumbá
groups, Paraguay Belt, Brazil. In: Arnaud, E., Halverson, G. P.
& Shields-Zhou, G. (eds) The Geological Record of Neopro-
terozoic Glaciations. Geological Society, London, Memoirs, 36,
487– 497.
Basei, M. A. S., Frimmel, H. E., Nutman, A. P. & Preciozzi, F. 2008.
West Gondwana amalgamation based on detrital zircon ages from
Neoproterozoic Ribeira and Dom Feliciano Belts of South America
and comparison with coeval sequences from SW Africa. In:
Pankhurst, R. J., Trouw, R. A. J., de Brito Neves, B. B. & de
Wit, M. J. (eds) West Gondwana: Pre-Cenozoic Correlations
Across the South Atlantic Region. Geological Society, London,
Special Publications, 294, 239–256.
Blair, T. C. & McPherson, J. G. 1994. Alluvial fans processes and
forms. In: Abrahams, A. D. & Parsons, A. J. (eds) Geomorphology
of Desert Environments. Chapman and Hall, London, 354–402.
Bossi, J. & Navarro, R. 1991. Geologı́a del Uruguay. Departamento de
publicaciones de la Universidad de la República, Montevideo.
Bossi, J. & Campal, N. 1992. Magmatismo y tectónica transcurrente
durante el Paleozoico Inferior en Uruguay. In: Gutiérrez-Marco,
J. G., Saavedra, J. & Rabano, I. (eds) Paleozoico Inferior de Iber-
oaméric. Mérida, Universidad de Extremadura, 343– 356.
Bossi, J., Cingolani, C., Lambias, E., Varela, R. & Campal, N. 1993.
Caracterı́sticas del magmatismo post-orogénico finibrasiliano en el
Uruguay: formaciones Sierra de Rı́os y Sierra de Ánimas. Revista
Brasileira de Geociências, 23, 282– 288.
Citroni, S. B., Basei, M. A. S., Sato, K. & Siga, O., Jr. 1999. Petrogen-
esis of the Campo Alegre Basin magmatism, based on geochemical
and isotopic data. In: 2nd South-American symposium on isotope
geology. Extended abstracts, Córdoba, Argentina, 1, 174– 177.
Cordani, U. G., Milani, E. J., Thomaz Filho, A. & Campos, D. A. (eds)
2000. Proceedings of the 31st International Geological Congress on
Tectonic Evolution of South America. Rio de Janeiro, Brazil.
D’Agrella, M. S. F. & Pacca, I. G. 1988. Paleomagnetism of the Itajaı́,
Castro and Bom Jardim Groups from Southern Brazil. Geophysical
Journal, 93, 365–376.
Eerloa, T. 2001. Climate change at the Neoproterozoic – Cambrian tran-
sition. In: Zhuravlev, A. Y. & Riding, R. (eds) The Ecology of the
Cambrian Radiation. Perspectives in Paleobiology and Earth
History. Columbia University Press, 90– 106.
564 E. PECOITS ET AL.
Eerola, T. 2006. Myöhäisproterotsooiset ilmastonmuutokset Tutkimuk-
sia Etelä-Brasiliassa. Geologi, 58, 164– 174.
Gaucher, C., Sial, A. N., Blanco, G. & Sprechmann, P. 2004. Che-
mostratigraphy of the lower Arroyo del Soldado Group (Vendian,
Uruguay) and palaeoclimatic implications. Gondwana Research, 7,
715– 730.
Gaucher, C., Blanco, G., Chiglino, L., Poire, D. & Germs, G. J. B.
2008. Acritarchs of the Las Ventanas Formation (Ediacaran,
Uruguay): implications for the timing of coeval rifting and
glacial events in western Gondwana. Gondwana Research, 13,
488– 501.
Hartmann, L., Santos, O., Bossi, J., Campal, N., Schipilov, A. &
McNaughton, N. J. 2002. Zircon and Titanite U– Pb SHRIMP geo-
chronology of Neoproterozoic felsic magmatism on the eastern
border of the Rı́o de la Plata Craton, Uruguay. Journal of South Amer-
ican Earth Sciences, 15, 229– 236.
Hoffman, P. F. & Schrag, D. P. 2002. The Snowball Earth hypo-
thesis: testing the limits of global change. Terra Nova, 14,
129– 155.
Linares, E. & Sánchez Bettucci, L. 1997. Edades Rb/Sr y K/Ar del
cerro Pan de Azúcar, Piriápolis, Uruguay. South American Sym-
posium on Isotope Geology, San Pablo, 1, 176– 180.
Mallmann, G., Chemale, F., Jr., A ´ vila, J. N., Kawashita, K. &
Armstrong, R. A. 2007. Isotope geochemistry and geochronology
of the Nico Pérez Terrane, Rı́o de la Plata Craton, Uruguay. Gond-
wana Research, 12, 489– 508.
Masquelin, H. 1990. Análisis estructural de las zonas de cizalla en las
migmatitas de Punta del Este – Uruguay. Acta Geologica Leopolden-
sia, 30, 139 –158.
Masquelin, H. & Sánchez, L. 1993. Propuesta de evolución tectono-
sedimentaria para la fosa tardi-brasiliana en la región de Piriápolis,
Uruguay. Revista Brasileira de Geociências, 23, 313– 322.
Meert, J. G. & Van der Voo, R. 1996. Paleomagnetic and 40Ar– 39Ar
study of the Sinyai dolerite, Kenya: implications for Gondwana
assembly. Journal of Geology, 104, 131– 142.
Midot, D. 1984. Etude geologique et diagnostic metallogénique pour
l’exploration du secteur de Minas (Uruguay). Diplome de Docteur
de 3è Cycle, Universitè Pierre et Marie Curie.
Myrow, P. M. & Kaufman, A. J. 1999. A newly discovered cap carbonate
above Varanger-age glacial deposits in Newfoundland, Canada.
Journal of Sedimentary Research, 6, 784– 793.
Oyhantçabal, P. 2005. The Sierra Ballena Shear Zone: kinematics,
timing and its significance for the geotectonic evolution of southeast
Uruguay. PhD thesis, Georg-August-Universität zu Göttingen.
Oyhantçabal, P., Siegesmund, S., Wemmer, K., Frei, R. & Layer, P.
2007. Post-collisional transition from calc-alkaline to alkaline
magmatism during transcurrent deformation in the southernmost
Dom Feliciano Belt (Braziliano –Pan-African, Uruguay). Lithos, 98,
141– 159.
Paim, P. S. G., Chemale, F., Jr. & Lopes, R. da 2000. A Bacia do
Camaquã. In: Holz, M. & De Ros, L. F. (eds) A geologı́a do Rio
Grande do Sul. CIGO-Universidade Federal Rio Grande do Sul,
Rio Grande do Sul, 231– 274.
Pazos, P., Tófalo, R. & Sánchez Bettucci, L. 2003. The record of the
Varanger Glaciation at the Rio de la Plata Craton, Vendian-Cambian
of Uruguay. Gondwana Research, 6, 65 – 77.
Pazos, P. J., Rapalini, A. E., Sánchez-Bettucci, L. & Tófalo, O. R.
2011. The Playa Hermosa Formation, Playa Verde Basin, Uruguay.
In: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The
Geological Record of Neoproterozoic Glaciations. Geological
Society, London, Memoirs, 36, 547–553.
Pecoits, E. 2003a. Sedimentologı́a y consideraciones estratigráficas de la
Formación Las Ventanas en su área tipo, departamento de Maldo-
nado, Uruguay. Revista de la Sociedad Uruguaya de Geologı́a,
Special Issue, 1, 124–140.
Pecoits, E. 2003b. Age and preliminary correlation of the Las
Ventanas Formation and Bom Jardim-Cerro do Bugio allogroups
(Vendian, Uruguay and Brazil). In: Frimmel, H. E. (ed.) 3rd Inter-
national Colloquium Vendian-Cambrian of W-Gondwana. University
of Cape Town, Cape Town, South Africa, 32 – 34.
Pecoits, E., Aubet, N., Oyhantçabal, P. & Sánchez Bettucci, L.
2005. Estratigrafı́a de sucesiones sedimentarias y volcanosedimentar-
ias Neoproterozoicas del Uruguay. Revista de la Sociedad Uruguaya
de Geologı́a, 11, 18 –27.
Pecoits, E., Gingras, M., Aubet, N. & Konhauser, K. 2008. Ediacaran
in Uruguay: palaeoclimatic and palaeobiologic implications. Sedi-
mentology, 55, 689– 719.
Poiré, D. G., González, P. D., Canalicchio, J. M., Repetto, F. G. &
Canessa, N. D. 2005. Estratigrafı́a del Grupo Mina Verdún, Proter-
ozoico de Minas, Uruguay. Latin American Journal of Sedimentology
and Basin Analysis, 12, 125– 143.
Preciozzi, F., Masquelin, H. & Sánchez Bettucci, L. (eds) 1993.
Geologı́a de la Porción Sur del Cinturón Cuchilla de Dionisio. In:
1st Simp. Int. Neoproterozoico – Cámbrico de La Cuenca del
Plata. Guı́a de Excursiones, La Paloma, Uruguay.
Rapalini, A. E. & Sánchez, L. B. 2008. Widespread remagnetization of
late Proterozoic sedimentary units of Uruguay and the apparent polar
wander path for the Rio de La Plata craton. Geophysical Journal
International, 174, 55– 74.
Rapela, C. W., Pankhurst, R. J. et al. 2007. The Rı́o de la Plata craton
and the assembly of SW Gondwana. Earth-Science Reviews, 83,
49 – 82.
Sánchez, L. & Linares, E. 1996. Primeras edades en basaltos del
Complejo Sierra de las Ánimas, Uruguay. In: 13th Congreso Geoló-
gico Argentino y 3rd Congreso de Exploración de Hidrocarburos,
Actas, La Plata, Argentina, I, 399– 404.
Sánchez, L. & Rapalini, A. E. 2002. Paleomagnetism of the Sierra de Las
Animas Complex, southern Uruguay: its implications in the assembly
of western Gondwana. Precambrian Research, 118, 243–265.
Trindade, R. I. F. & Macouin, M. 2007. Palaeolatitude of glacial depos-
its and palaeogeography of Neoproterozoic ice ages. Comptes Rendus
Geoscience, 339, 200–211.
Veroslavsky, G., de Santa Ana, H. & Daners, G. 2006. Tacuarı́
Formation (Nov. Nom.): lithostratigraphy, facies, environment, age
and geological significance (Cerro Largo-Uruguay). Revista de la
Sociedad Uruguaya Geologı́a, 13, 23 –35.
 Chapter 54

The controversial stratigraphy of the glacial deposits in the Tandilia System, Argentina

P. J. PAZOS1,2* & A. RAPALINI1,2

1
Departamento de Ciencias Geológicas, FCEN, Ciudad Universitaria, Buenos Aires 1428, Argentina
2
Consejo Nacional de Investigaciones Cientı́ficas y Técnicas (CONICET)
*Corresponding author (e-mail: pazos@gl.fcen.uba.ar)

Abstract: In the Ediacaran–Cambrian Tandilia System of central Argentina, the glacial origin of the thin (10 m) Sierra del Volcán For-
mation (Fm.) has been recognized for many years (Spalletti & del Valle 1984), being the first undisputed glacial deposits recorded in the
Rı́o de la Plata craton. It consists of three units: (i) a basal polymictic diamictite with a kaolin-rich matrix, (ii) a middle pelite with
heterolithic levels, undulatory stratification, symmetric ripples, and outsized clasts previously interpreted as dropstones, and (iii) an
upper polymictic diamictite with subtle normal grading. Here, this poorly age-constrained formation is reviewed from a stratigraphic
and regional context and compared with other putative glaciogenic intervals at the base of the Cerro Largo Fm. and in the Punta
Mogotes borehole.

The Tandilia System consists of an Ediacaran – Ordovician
package of sedimentary rocks that crops out in a 300-km-long,
NW –SE-trending belt south of Buenos Aires, Argentina
(Fig. 54.1). The Tandilia System comprises the Sierras Bayas
Group and the unconformably overlying Cerro Negro and Balcarce
formations. Whereas the stratigraphic and tectonic evolution of the
Tandilia System has received some recent attention (e.g. Gómez
Peral et al. 2007), its glaciogenic record remains poorly described
in the international literature. The glaciogenic origin of the Sierra
del Volcán Fm., which occurs in the southern part of the Tandilia
belt, has long been accepted. However, its age and stratigraphic
position, reviewed recently by Pazos et al. (2008), have remained
ambiguous, having been variably assigned to the Precambrian,
Cambrian and Ordovician. A putative glacial diamictite occurs at
the base of the Cerro Largo Fm. in the lower Sierras Bayas
Group and crops out near the town of Olavarrı́a in the northwestern
part of the Tandilia belt (Fig. 54.1). The relationship between this
diamictite and the Sierra del Volcán Fm., and more generally the
stratigraphic framework for the Tandilia System, is poorly under-
stood, but a relatively complete section preserved in the Punta
Mogotes borehole in the SE of the belt (Marchese & di Paola
1975) sheds light on the Tandilia stratigraphy.
Structural framework
The Tandilia system (Dalla Salda et al. 1988; Cingolani & Dalla
Salda 2000) crops out in a 300-km-long chain of low hills extend-
ing in a NW –SE direction from the town of Olavarrı́a in the centre
of the Buenos Aires province of Mar del Plata on the Atlantic coast
(Fig. 54.1). It rests on a Palaeoproterozoic basement assigned to
the Buenos Aires Complex (Marchese & di Paola 1975). It is
unmetamorphosed (Gómez Peral et al. 2007) and only lightly
folded locally, such as near Olavarrı́a in the NW part of the belt
(Massabie & Nestiero 2005). A low-angle unconformity occurs
between the Cambrian Cerro Negro Fm. and the overlying Ordovi-
cian Balcarce Fm. Glaciogenic deposits of the Sierra del Volcán
Fm. are completely undisturbed, cropping out with subhorizontal
bedding and no metamorphic overprint.
The Sierras Bayas Group records sedimentation in a shallow
marine basin with strong tidal influence and reworking (see
Poiré & Spalletti 2005). The Balcarce Fm. was deposited during
a subsequent sedimentary cycle in a more extensive basin. It over-
lies palaeohighs of basement or the Sierras Bayas Group rocks,
with a very low angle of unconformity in places (Pazos et al.
2008). The last confirmed evidence of magmatism in the Tandilia
System is a series of diabase sills (Los Barrientos) dated by
Rapela et al. (1974) as lower Ordovician and probably associated
with the Ordovician Famatinian magmatic cycle in the western
Proto-Andean continental margin of Gondwana (Pankhurst &
Rapela 1998).
Teruggi & Kilmurray (1975) proposed that the present landscape
of the Tandilia system is due to Miocene Andean reactivation of
basement faults. However, from a detailed geomorphological
study, Demoulin et al. (2005) recognized two regional planation
surfaces, which suggests that most of the present landscape is sig-
nificantly older (Palaeogene or even Cretaceous), and that the
region has experienced slow denudation rates.
Stratigraphy
The main geological units within the Tandilia belt are a Palaeopro-
terozoic igneous and metamorphic basement (The Buenos Aires
Complex) and a Neoproterozoic to Early Palaeozoic cover com-
prised of Ediacaran (?) calcareous and clastic sediments of the
Sierras Bayas Group, the clastic (Cambrian) Cerro Negro Fm.,
and quartzite of the Ordovician Balcarce Fm., which gently over-
laps the underlying units and marks an extensive transgression
across the basin.
The Buenos Aires Complex includes various igneous and meta-
morphic suites. Sm–Nd model ages averaging 2620 + 80 Ma on
several magmatic units suggest a Late Archaean history (Pankhurst
et al. 2003). Hartmann et al. (2002) recognized orogenic peaks
around 2.16 Ga and 2.08 Ga in Uruguay (eastern part of the craton).
The Sierras Bayas Group corresponds approximately to what was
originally described as the La Tinta Fm. (Iñiguez Rodriguez et al.
1989; Marchese & di Paola 1975) and the Sierras Bayas Fm.
(Borrelo 1962). This group, according to Poiré & Spalletti (2005),
presents a simple stratigraphy and covers the basement rocks of
the Buenos Aires Complex (Fig. 54.2). The basal Villa Mónica
Fm. (52–70 m thick) is composed of quartz-arenite, pelite, con-
glomerate and cherty dolomite beds with stromatolites (Poiré &
Spalletti 2005). It is overlain by the Cerro Largo Fm., which is c.
40 m thick and includes a diamictite containing mainly chert
clasts and slumped beds at the base. Poiré & Spalletti (2005) inter-
pret this diamictite as glaciogenic in origin. The diamictite passes
upward into multicolour pelite, glauconitic sandstone units and
whitish quartzite beds containing hummocky, planar and trough
cross-stratification. The succession ends with a shallowing-upward

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 565– 569. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.54
566 P. J. PAZOS & A. RAPALINI

Fig. 54.1. The Tandilia System and

outcrops of the Sierra del Volcán Fm. at del
Volcán Hill. The Punta Mogotes Borehole
is located just south of Mar del Plata.

claystone cycle (Gómez Peral et al. 2007). The Olavarrı́a Fm.,
which follows upwards, is composed of red, fine-grained rippled
sandstone, pelite and heterolithic sediments deposited in a
shallow, marginal marine environment judging from the occurrence
of bimodal palaeocurrents (Poiré et al. 2005). The youngest unit of
the Sierras Bayas Group is the Loma Negra Fm., which is c. 40 m
thick and composed almost exclusively of reddish and black lime-
stone deposited in an open marine ramp to lagoonal environment. It
contains Cloudina, which indicates a late Ediacaran age and tropical
depositional environment (Poiré & Spalletti 2005; Gaucher et al.
2003, 2005; Gómez Peral et al. 2007, Pecoits et al. 2008).
The overlying Cerro Negro Fm. is disconformably separated
from the upper Sierras Bayas Group by a palaeokarst surface
(Barrio et al. 1991). This .150-m-thick unit is composed of
reddish fine-grained sandstone to claystone deposited in a
shallow marine to tidal environment. Poiré & Spalletti (2005) cor-
related the Cerro Negro Fm. with the Sierra del Volcán and Punta
Mogotes formations, which occur in the southeastern part of the
belt, and suggested a Cambrian age for all three units. Gómez
Peral et al. (2007, fig. 2) instead argued that the Cerro Negro
ERAS LITOSTRATIGRAPHIC UNITS SE REGION
PERIODS
NW REGION CENTRAL REGION
ORDOVICIAN
Balcarce Formation
Balcarce Formation
Balcarce Formation
CAMBRIAN
Cerro Negro Cerro Negro
Formation
Formation
Loma Negra Loma Negra
Formation
Palaeohigh ?
Sierras Bayas Group
Fm. is laterally equivalent to the Las Águilas, the Punta Mogotes
and the Sierra del Volcán formations. We propose a distinct strati-
graphic framework (Fig. 54.2) based on a description by Marchese
& di Paola (1975) of the Tandilia stratigraphy in the Punta Magotes
borehole in the southestern part of the belt. These authors recog-
nized eight units within the well. The lowermost, unit 8 (89 m
thick), is composed of metapelite beds that they redefined as the
Punta Mogotes Fm. and correlated with the regional basement.
Units 7– 5 comprise quartz arentite and heterolithic sediments.
Unit 4 is a diamictite (6 m thick), which we suspect to be glacio-
genic (see below). It is followed by more siliciclastic units 2–3,
which may correlate with either the Cerro Negro or Balcarce for-
mations. Unit 1 is assigned to the Balcarce Fm.
Glaciogenic deposits and associated strata
The glaciogenic deposits of the Sierra del Volcán Fm. are extre-
mely thin (10 m thick), and the outcrops are almost entirely
covered by vegetation and debris from the overlying Balcarce
Fm. Exposures occur on private property with restricted access
(Estancia Cerro Del Volcán). The sedimentological description
is based on recent fieldwork and descriptions given in Spalletti
& del Valle (1984). The unit was divided into three intervals: (i)
URUGUAY
Borehole a lower diamictite, (ii) a middle heterolithic and pelitic unit, and
upper
Las Ventanas (iii) an upper diamictite. The lower diamictite is polymictic with
Formation
clasts of the Buenos Aires complex, but no carbonate clasts. The
matrix contains abundant kaolinite, reminiscent of the saprolitized
CerroNegro basement and the matrix of the Balcarce Fm. The middle interval is
Formation Arroyo
(units 2-3)
del Soldado
rich in outsized clasts disrupting the stratification, which exhibits
Group undulatory and faintly developed ripples. The overlying diamictite
is normally graded (Spalletti & del Valle 1984). Evidence of a

Formation
LATE Del Volcán Playa Hermosa glacial influence in deposition of the Sierra del Volcán Fm.
Olavarría
EDIACARAN
Formation Las Águilas
Formation UNIT 4 Formation includes glacial pavements, rainout deposits, faceted clasts and
and Sierras Bayas matrix-supported diamictite beds with immature composition
Cerro Largo Cerro Largo Group
Formation Formation (units 5-7) (Pazos et al. 2008).
Another diamictite occurs at the contact between the Villa
Monica and Cerro Largo formations (Gómez Peral et al. 2007,
Fig. 54.3) in the northwestern sector of the belt. The slumped dia-
Villa Villa mictite contains mainly chert clasts and was interpreted by Poiré &
EDIACARAN
Mónica
Formation
Mónica
Formation
Spalletti (2005) as glacial in origin, although no direct evidence of
glacial influence has been reported. Another diamictite occurs at a
depth of 264 –270 m (unit 4) in the Punta Mogotes borehole
PALAEO-MESO
BUENOS AIRES COMPLEX Punta Mogotes
Formation
ARCHAEAN
COMPLEX
(Marchese & di Paola 1975). This 6-m-thick diamictite contains
PROTEROZOIC (unit 8)
fragments of feldspar and of metamorphosed pelite believed to
be derived from the Punta Mogotes Fm. and is compositionally dis-
Fig. 54.2. Integrated stratigraphy (after Poiré & Spalletti 2005), including the tinct from bracketing sediments. No unequivocal evidence for gla-
Punta Mogotes borehole stratigraphy suggested in this chapter, and the ciation has been observed, but based on the stratigraphic
suggested stratigraphic position for the Sierra del Volcán Fm. Diamictite arguments described below, we tentatively correlate unit 4 with
intervals marked with triangles. the Sierra del Volcán Fm.
 THE TANDILIA SYSTEM, ARGENTINA 567

low-angle unconformity (Spalletti & del Valle 1984). The nature

of the contacts of the diamictite comprising unit 4 in the Punta
Mogotes borehole is unkown.

Chemostratigraphy

Gómez Peral et al. (2007) published reconnaissance chemostrati-

graphic data on the Villa Mónica and Loma Negra formations of
the Sierras Bayas Group. To summarize, the Villa Mónica carbon-
ates are heavily diagenetically altered, with very low Sr concen-
trations and enrichments in Fe and Mn. d13C values define a
narrow range from –0.5 to 2.2‰. The Loma Negra Fm. is moder-
ately altered, with Sr contents of 300 –400 ppm. d13C carbonate
values cluster between 3 and 4‰, whereas organic matter values
lie between – 27 and –28‰.

Palaeolatitude and palaeogeography

The available Neoproterozoic palaeomagnetic data from the Rı́o

Fig. 54.3. Detailed logged section of the Sierra del Volcán Fm. taken from
Pazos et al. (2008). (a) Location of the interval studied. (b) Logged section
originally from Spalletti & del Valle (1984).
Boundary relations with overlying and underlying
non-glacial units
The Sierras Bayas Group is bounded by stratigraphic unconformi-
ties (e.g. Poiré & Spalletti 2005; Gómez Peral et al. 2007). The
non-conformity that separates the basement (Buenos Aires
Complex) from the Sierras Bayas Group represents hundreds of
millions of years during which the basement was intensely chemi-
cally weathered. An unconformity also separates the Villa Mónica
Fm. from the overlying Cerro Largo Fm., but the nature and mag-
nitude of the gap is subject to debate (Pazos et al. 2008).
The Sierra del Volcán Fm. sits non-conformably on the
basement and is separated from the overlying Balcarce Fm. by a
de la Plata (RP) craton were reviewed in Tohver et al. (2006)
and Rapalini (2005, 2006). A preliminary apparent polar wander
track for the interval 600– 500 Ma has been proposed recently
for this craton (Rapalini 2006; Fig. 54.4). The available palaeo-
magnetic database to constrain the palaeogeographical evolution
of the RP craton consists of four palaeomagnetic poles and one
mean geomagnetic pole. The latter means that there is no guarantee
of full compensation of non-dipole components of the Earth
magnetic field.
The oldest available pole was obtained from the Campo Alegre
lavas (CA, D́Agrella Filho & Pacca 1988) in Brazil. These rocks
were accurately dated as 595 + 5 Ma (U – Pb, Citroni et al.
1999). The similar position of the mean geomagnetic pole for
the Playa Hermosa sediments (PH) to CA (Fig. 54.4) suggests a
broadly similar age for both (Pazos et al. 2011).
The other poles in the APWP are nearly identical to those of the
older units of the Sierra de Animas Complex (SA2, Sanchez Bet-
tucci & Rapalini 2002) and Los Barrientos claystone (LB, Rapalini
2006) in the Tandilia System (Table 54.1). Originally, an age of c.
550 Ma was assigned to SA2 based on old K –Ar and Rb –Sr ages.
Oyhantçabal et al. (2007) published an Ar/Ar age of 579 + 2 Ma
on hornblende for the Pan de Azucar syenite, interpreted as a
cooling age of that body. LB is a high-quality pole with no firm
age constraints, although its position being similar to SA2 suggests
an age between 580 and 570 Ma.
Taking into consideration the apparent polar wander path in
Figure 54.4 and Ediacaran to Cambrian poles from other

Table 54.1. Available Neoproterozoic palaeomagnetic poles for the Rio de la Plata

Geographical Formation/lithology Palaeomagnetic pole (PP) Palaeomagnetic results Palaeopoles age Q Ref.
location (lat.) (long.) A95 (dp/dm)

Uruguay Sierra de las Animas 5.98 338.18 (19.68/26.78) AF/Th demag, N ¼ 7 (n ¼ 33), IRM Cambrian (520– 3 1
Complex (SA1) 500 Ma)
Uruguay Sierra de las Animas –16.98 250.98 (15.98/21.58) AF/Th demag, N ¼ 6 (n ¼ 27), IRM, both Ediacaran 4 1
Complex (SA2) polarities (579 Ma)
Tandilla System Sierra de los –15.18 252.68 (10.98/14.28) AF/Th demag, N ¼ 12 (n ¼ 56), both polarities, Ediacaran? 6 2
Barrientos reversal of EMF recorded. Correction for
claystones (LB) remanence anisotropy
Uruguay Playa Hermosa Fm. –43.08 198.48 (8.68/16.08) AF/Th demag, N ¼ 2 (n ¼ 6), IRM, small number Early Ediacaran? 3 1
(PH) of samples
Brazil Campo Alegre lavas –57.08 223.08 9.08 AF/Th demag, N ¼ 6 (n ¼ 46), both polarities. Early Ediacaran 5 3
(CA) (595 Ma)

Q, quality factor of the pole (maximum, 7) according to criteria set by Van der Voo (1990). 1, Sánchez Bettucci & Rapalini (2002); 2, Rapalini (2006); 3, D́Agrella Filho &
Pacca (1988). EMF, Earth magnetic field; IRM, isothermal remanent magnetization; AF, alternating field; Th, thermal.
568 P. J. PAZOS & A. RAPALINI
500
525 SA1
510
535
550
LB
SA2 (579)
CA (595 )
PH
Fig. 54.4. Palaeomagnetic poles and their respective circles (or ovals) of
confidence for the Rio de la Plata (RP) craton (black squares with white ovals),
and other Gondwana blocks (white circles with grey ovals). Poles from
remagnetized units (Rapalini & Sanchez Bettucci 2008) have been omitted. The
hypothetical apparent polar wander track for this craton is shown in dashed
lines. Numbers indicate approximate ages of poles in Ma. Data for RP poles are
presented in Table 54.1. Gondwana poles selected from Trindade et al. (2006).
SA1, Sierra de Animas 1; SA2, Sierra de Animas 2; PH, Playa Hermosa; LB,
Los Barrientos; CA, Campo Alegre poles.
Gondwanan cratonic blocks (Trindade et al. 2006), the RP craton
appears to have lain between 308S and 608S through the Ediacaran
and Early Cambrian.
Geochronology
In the Villa Mónica Fm., Rb –Sr data on diagenetic clay minerals
yielded an age of 795 + 28 Ma on a green shale. A Rb – Sr age of
769 + 12 Ma was obtained on shale deposits of the overlying
Cerro Largo Fm. (Bonhomme & Cingolani 1980), while Cingolani
& Bonhomme (1988) and Kawashita et al. (1999) suggested a
depositional age of c. 730 Ma for the same formation based on
Rb –Sr whole-rock data. However, these ages are all regarded as
too old and, based on the palaeomagnetic considerations described
above, an Ediacaran age for the Sierra Bayas Group is more likely.
Detrital zircon geochronology (Rapela et al. 2007; Gaucher
et al. 2007) has provided no meaningful age constraints on the
Sierras Bayas Group, but has established a maximum Early Ordo-
vician age for the Balcarce Fm. (Rapela et al. 2007).
The body fossil Cloudina riemkeae occurs in the Loma Negra
Fm. (Gaucher et al. 2005; Pecoits et al. 2008) and confirms an
Ediacaran age, at least for the upper Sierras Bayas Group.
Chauria circularis has been reported from Villa Mónica and
Cerro Negro formations and leiosphaeridia-dominated acritarch
assemblages are found from the upper Villa Mónica Fm. to the
middle Cerro Negro Fm. (Gaucher et al. 2005).
Discussion
The diamictite beds of the Sierra del Volcán Fm. are identified as
glacial in origin based on a range of features and interpretations
described in detail in Pazos et al. (2008). The important
observations and interpretations of the Sierra del Volcán Fm. can
be summarized as follows:
† The diamictite at the base and top of the unit (Fig. 54.3) appears
to have been deposited by debris flows, as is particularly
evident in the normal grading of the upper diamictite.
† The clasts in the diamictite beds are derived mainly from the
metamorphic rocks of the Buenos Aires Complex, as well as
unmetamorphosed quartzite clasts and blocks, probably
belonging to the quartzite beds of the Sierras Bayas Group
present in the Villa Mónica (lower quartzite) and Cerro Largo
(upper quartzite) formations.
† The micromorphology of quartz grains is consistent with a
glacial origin (van Staden et al. 2005).
† The middle unit contains dropstones that clearly disrupt
bedding (Spalletti & del Valle 1984) and symmetric ripples
indicating a combination of open water and floating ice.
The diamictite documented at the base of the Cerro Largo Fm. is a
cherty breccia that lacks evidence of glacially influenced sedimen-
tation. The diamictite interval documented in the Punta Mogotes
borehole (unit 4) described in detail by Marchese & di Paola
(1975) is very thin (6 m thick) and differs compositionally with
both underlying and overlying sediments. It contains fresh feldspar
and angular basement clasts floating in a clayish matrix. These fea-
tures indicate a low degree of weathering of clast material, but not
sufficient to assign an unambiguous glacial origin (e.g. Eyles &
Januszczak 2004). The absence of carbonate clasts, despite the
low weathering index, imply that the unit 4 diamictite is older
than the Loma Negra Fm., and thus indisputably Neoproterozoic
in age. The Sierra del Volcán Fm. similarly lacks carbonate
clasts and we suggest that it correlates with the unit 4 diamictite
(Fig. 54.2). This correlation requires an explanation for the
absence of the Loma Negra carbonates in the Punta Mogotes bore-
hole. We interpret this to be the result of a topographic high in the
Punta Mogotes region, which was likely also a source area for
clasts in the Sierra del Volcán Fm.
We thank Dr R. Somoza and Professor V. Ramos for providing bibliography, dis-
cussions and stimulation, although we stress that we accept full responsibility for
the ideas presented here. I want to express my gratitude (P.P.) to Cheryl Turking-
ton for her unconditional support. This is the contribution R-38 of the Instituto de
Estudios Andinos Don Pablo Groeber. This represents a contribution of the IUGS
and UNESCO-funded IGCP (International Geoscience Programme) project #512.
References
Barrio, C. A., Poiré, D. G. & Iñiguez Rodrı́guez, A. M. 1991. El contacto
entre la Formación Loma Negra (Grupo Sierras Bayas) y la Formación
Cerro Negro, un ejemplo de paleokarst, Olavarrı́a, provincia de Buenos
Aires. Revista de la Asociación Geológica Argentina, 46, 69–76.
Bonhomme, M. G. & Cingolani, C. 1980. Mineralogı́a y geocronologı́a
Rb –Sr y K–Ar de fracciones finas de la Formacion ‘La Tinta’,
provincia de Buenos Aires. Revista de la Asociación Geológica
Argentina, 35, 519–538.
Borello, A. 1962. Formación Punta Mogotes (Eopaleozoico Provincia de
Buenos Aires). Comisión Nacional de Investigaciones Cientı́ficas.
Notas 1, 1. La Plata.
Cingolani, C. & Bonhomme, M. G. 1988. Resultados geocronológicos
en niveles pelı́ticos intercalados en las dolomı́as de las Sierras
Bayas (Grupo La Tinta), provincia de Buenos Aires. Actas Segundas
Jornadas Geológicas, Actas I, 2283– 2389.
Cingolani, C. & Dalla Salda, L. 2000. Buenos Aires Cratonic region.
In: Cordani, U., Milani, E., Thomaz Filho, A. & Campos, D. (eds)
Tectonic Evolution of South America. 31st International Geological
Congress, Brazil, Rio Janeiro, 139–147.
Citroni, S. B., Basei, M. A. S., Sato, K. & Siga, O. Jr. 1999. Petrogen-
esis of the Campo Alegre basin magmatism, based on geochemical
and isotopic data. II South American Symposium on Isotope
Geology, Cordoba, Argentina, Actas, 174– 177.
 THE TANDILIA SYSTEM, ARGENTINA
D’Agrella Filho, M. S. & Pacca, I. I. G. 1988. Paleomagnetism of the
Itajaı́, Castro and Bom Jardim Groups from Southern Brazil. Geophy-
sical Journal International, 93, 365–376.
Dalla Salda, L., Bossi, J. & Cingolani, C. 1988. The Rio de la Plata cra-
tonic región of southwestern Gondwana. Episodes, 11, 263–269.
Demoulin, A., Zárate, M. & Rabassa, J. 2005. Long-term landscape
development: a perspective from the southern Buenos Aires ranges
of east central Argentina. Journal of South American Earth Sciences,
19, 193– 204.
Eyles, N. & Januszczak, N. 2004. ‘Zipper-rift’: a tectonic model
for Neoproterozoic glaciations during the break up of Rodinia after
750 Ma. Earth-Science Reviews, 65, 1 – 73.
Gaucher, C., Boggiani, P. C., Sprechmann, P., Sial, A. N. & Fiar-
child, T. 2003. Integrated correlation of the Vendian to Cambrian
Arroyo del Soldado and Corumbã Groups (Uruguay and Brazil) –
palaeogeographic, palaeoclimatic and palaeobiologic implications.
Precambrian Research, 120, 241– 278.
Gaucher, C., Poiré, D. G., Peral, L. & Chiglino, L. 2005. Litoestrati-
grafı́a, Bioestratigrafı́a y correlaciones de las sucesiones sedimentar-
ias del Neoproterozoico –Cambrico del Cratón del Rı́o de la Plata
(Uruguay y Argentina). Latin American Journal of Sedimentology
and Basin Analysis, 12, 145–160.
Gaucher, C., Poiré, D., Finney, S. C., Valencia, V., Blanco, G.,
Pamoukaghlian, K. & Gómez Peral, L. 2007. Zircones detrı́ticos
de secuencias neoproterozoicas de Uruguay y Argentina: inferencias
sobre la evolución paleogeográfica del cartón del Rı́o de la Plata.
V Congreso Uruguayo de Geologı́a, Montevideo, Uruguay.
(CD-ROM, extended paper).
Gómez Peral, L., Poiré, D. G., Strauss, H. & Zimmermann, U. 2007.
Chemostratigraphy and diagenetic constraints on Neoproterozoic
carbonate successions from the Sierras Bayas Group, Tandilia
System, Argentina. Chemical Geology, 237, 109–128.
Hartmann, L. A., Santos, J. O. S., Cingolani, C. A. & MacNaug-
thon, N. J. 2002. Two Paleoproterozoic orogenies in the evolution
of the Tandilia Belt, Buenos Aires, as evidenced by zircon U –Pb
SHRIMP geochronology. Int. Geology Rev., 44, 528– 543.
Iñiguez Rodrı́guez, A. M., Del Valle, A., Poiré, D., Spalletti, L. A.
& Zalba, P. 1989. Cuenca Precámbrica/Palaeozoica inferior de
Tandilia, Provincia de Buenos Aires. In: Chebli, W. & Spalletti,
L. A. (eds) Cuencas Sedimentarias Argentinas. Serie Correlación
Geológica, San Miguel del Tucumán, 6, 245– 263.
Kawashita, K., Varela, R. et al. 1999. Geocronology and chemostrati-
graphy of ‘La Tinta’ Neoproterozoic sedimentary rocks, Buenos
Aires Province, Argentina. Actas II. South American Symposium on
Isotope, 403–407.
Marchese, H. G. & di Paola, E. C. 1975. Miogeosinclinal Tandil.
Revista de la Asociación Geológica Argentina, 30, 161–179.
Massabie, A. C. & Nestiero, O. E. 2005. La estructura del Grupo Sierras
Bayas en el sector norte de las Sierras Septentrionales de Buenos
Aires. Revista de la Asociación Geológica Argentina, 60, 135–146
Oyhantçabal, P., Siegesmund, A., Wemmer, K., Frei, R. & Layere-
tal, P. 2007. Post-collisional transition from calc-alkaline to alkaline
magmatism during transcurrent deformation in the southernmost
Dom Feliciano Belt (Braziliano –Pan-African, Uruguay). Lithos, 98,
141– 159.
Pankhurst, R. J. & Rapela, C. W. (eds) 1998. The Proto-Andean
Margin of Gondwana. Geological Society of London, Special
Publication, 142.
Pankhurst, R. J., Ramos, A. & Linares, E. 2003. Antiquity of the Rı́o de
la Plata craton in Tandilia, southern Buenos Aires Province, Argen-
tina. Journal of South American Earth Sciences, 16, 5 –13.
Pazos, P. J., Sánchez Bettucci, L. & Loureiro, J. 2008. The Neopro-
terozoic glacial record in the Rı́o de la Plata Craton: a critical reap-
praisal. In: Pankhurst, R. J., Trouw, R. A. J., de Brito Neves,
569
B. B. & de Witt, M. J. (eds) West Gondwana Pre-Cenozoic Corre-
lations Across the South Atlantic Region. Geological Society,
London, Special Publications, 294, 343–364.
Pazos, P. J., Rapalini, A., Sánchez Bettucci, L. & Tófalo, R. O. 2011.
The Playa Hermosa Formation, Playa Verde Basin, Uruguay.
In: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The
Geological Record of Neoproterozoic Glaciations. Geological
Society, London, Memoirs, 36, 547– 553.
Pecoits, E., Gingras, M., Aubet, N. & Konhauser, K. 2008. Ediacaran
in Uruguay: palaeoclimatic and palaeobiological implications.
Sedimentology, 55, 689– 721.
Poiré, D. G. & Spalletti, L. A. 2005. La cubierta sedimentaria precám-
brica/paleozoica inferior del Sistema de Tandilia. In: Barrio, R. E.,
Etcheverry, R. O., Caballé, M. F. & Llambı́as, E. J. (eds) Geolo-
gı́a y Recursos Minerales de la provincia de Buenos Aires. Relatorio
del XVI Congreso Geológico Argentino, 51 –68.
Poiré, D. G., Canalicchio, J. & Alonso, G. 2005. Las calizas del
Sistema de Tandilia y su utilización en la industria cementera. In:
Barrio, R. E., Etcheverry, R. O., Caballé, M. F. & Llambı́as, E.
J. (eds) Geologı́a y Recursos Minerales de la provincia de Buenos
Aires. Relatorio del XVI Congreso Geológico Argentino, 387– 396.
Rapalini, A. E. 2005. The accretionary history of Southern South
America from the latest Proterozoic to the Late Paleozoic: some
paleomagnetic constraints. In: Vaughan, A., Leat, P. & Pan-
khurst, R. J. (eds) Terrane Processes at the Margins of Gondwana.
Geological Society, London, Special Publication, 246, 305– 328.
Rapalini, A. E. 2006. New Late Proterozoic paleomagnetic pole for the
Rio de la Plata craton: implications for Gondwana. Precambrian
Research, 147, 223–233.
Rapalini, A. E. & Sanchez Bettucci, L. 2008. Widespread remagneti-
zation of Late Proterozoic sedimentary units of Uruguay and the
apparent polar wander path for the Rio de la Plata craton. Geophysical
Journal International, 174, 55 – 74.
Rapela, C., Dalla Salda, L. & Cingolani, C. 1974. Un intrusivo básico
Ordovı́cico en la Formación La Tinta (Sierra de los Barrientos),
Provincia de Buenos Aires. Revista de la Asociación Geológica
Argentina, 29, 319–331.
Rapela, C. W., Pankhurst, R. J. et al. 2007. The Rio de la Plata craton
and the assembly of SW Gondwana. Earth Science Reviews, 83,
49– 82.
Sánchez Bettucci, L. & Rapalini, A. E. 2002. Paleomagnetism of the
Sierra de Las Animas Complex, Southern Uruguay: Its Implications
in the Assembly of Western Gondwana. Precambrian Research,
118, 243–265.
Spalletti, L. & Del Valle, A. 1984. Las diamictitas del sector oriental
de Tandilia: caracteres sedimentológicos y origen. Revista de la
Asociación Geológica Argentina, 39, 188– 206.
Teruggi, M. E. & Kilmurray, J. 1975. Tandilia. In: Geologı́a de la
Provincia de Buenos Aires. Relatorio del VI Congreso Geológico
Argentino, Buenos Aires, 55 –77.
Tohver, E., D’Agrella Filho, M. S. & Trindade, R. I. F. 2006. Paleo-
magnetic record of Africa and South America for the 1200–500 Ma
interval, and evaluation of Rodinia and Gondwana assemblies.
Precambrian Research, 147, 193– 222.
Trindade, R. I. F., D’Agrella-Filho, M. S., Epof, I. & Brito Neves,
B. B. 2006. Paleomagnetism of Early Cambrian Itabaiana mafic
dikes (NE Brazil) and the final assembly of Gondwana. Earth
Planetary Science Letters, 244, 361– 377.
Van der Voo, R. 1990. The reliability of paleomagnetic data. Tectonophy-
sics, 184, 1– 9.
van Staden, A., Zimmermann, U. & Germs, G. J. 2005. Provenance and
depositional study on tillites from the Volcan Hill, Tandilia System in
east Argentina: preliminary results. XVI Congreso Geológico
Argentino, Actas, I, 239–246.
 Chapter 55

Glacial sediments and associated strata of the Polarisbreen Group,

northeastern Svalbard

GALEN P. HALVERSON1,2
1
School of Earth and Environmental Sciences, The University of Adelaide North Terrace, Adelaide, SA 5005, Australia
2
Present address: Department of Earth and Planetary Sciences, McGill University, 3450 University Street, Montreal,
QC, H3A 2A7, Canada
(e-mail: galen.halverson@mcgill.ca)

Abstract: Northeastern Svalbard hosts exceptionally well-preserved Neoproterozoic sediments. The glaciogenic Petrovbreen Member
and Wilsonbreen Formation (Fm.) of the Polarisbreen Group crop out in a narrow, Caledonian-aged fold-and-thrust belt spanning from
Olav V Land on Spitsbergen in the south to western Nordaustlandet in the north. The older Petrovbreen Member is thin (0– 52 m) and
patchily preserved, comprising mainly poorly stratified, dolomite-matrix diamictite likely deposited in a marine setting. The basal contact
of the Petrovbreen Member erosionally truncates the upper Russøya Member, which preserves a large (12‰) negative C-isotope
anomaly. The Petrovbreen Member is overlain by 200 m of dark, monotonous shales of the MacDonaldryggen Member, followed by
cherty dolomite grainstone and microbiolamintes of the Slangen Member. The upper Slangen Member is an exposure surface in the
southern part of the belt, but in northern Spitsbergen and on Nordaustlandet is transitional into sands of the northward-thickening
Bråvika Member. The Wilsonbreen Fm. is typically 100– 150 m thick and consists mostly of massively to poorly stratified diamictite,
with subordinate sand beds, conglomerate lenses, and carbonates deposited in a terrestrial environment. It is overlain by the colourful
Dracoisen Fm., which records, at its base, a typical post-glacial negative d13C anomaly. There are no direct radiometric age constraints
or reliable palaeomagnetic data from the Polarisbreen Group, but it is widely accepted that northeastern Svalbard was contiguous with
East Greenland during the Neoproterozoic.

Late Neoproterozoic glaciogenic rocks occur in both western and
northeastern Svalbard (Harland et al. 1993). The much better
studied and preserved glaciogenic sediments and bracketing
strata of the upper Hecla Hoek succession in northeastern Svalbard
are the subject of this chapter. These rocks occur in Olav V Land
and Ny Friesland in northeastern Spitsbergen and in Gustav V
Land in northwestern Nordaustlandet (Fig. 55.1). These rocks
are best exposed on nunataks in Olav V Land and southern Ny
Friesland, but occur also over a large area in Nordaustlandet
along the eastern side of Hinlopenstretet, and to a lesser extent,
in coastal exposures in northern Ny Friesland. In Spitsbergen,
two distinct glacial units, named the Petrovbreen Member
(older) and Wilsonbreen Fm. (younger) occur within the middle
of the mixed clastic-carbonate Polarisbreen Group (Fig. 55.2).
Both the Petrovbreen Member and the Wilsonbreen Fm., together
with bracketing strata, are readily identified in Nordaustlandet
(Halverson et al. 2004), where a separate nomenclature has histori-
cally been used (Kulling 1934). Because correlation across Hinlo-
penstretet (Fig. 55.1) is unambiguous (Kulling 1934; Harland et al.
1993; Fairchild & Hambrey 1995), only the better known nomen-
clature from Spitsbergen (Harland 1997) is used here (Fig. 55.2).
The history of geological investigation of Precambrian glacial
deposits was recently summarized in Harland (2007). In short,
the Neoproterozoic sedimentary succession in northeastern Sval-
bard was first studied by Nordenskiöld (1863). A glacial origin
for the Polarisbreen diamictites (specifically the Wilsonbreen
Fm.) was originally proposed by Kulling (1934), who mapped
and named these rocks in Nordaustlandet and northern Ny Friesland
during the Swedish– Norwegian Arctic Expedition of 1931.
Fleming & Edmonds (1941) further investigated coastal exposures
of the Hecla Hoek, including the Wilsonbreen Fm., but it was not
until W. B. Harland and C. B. Wilson worked systematically on
the upper Hecla Hoek succession on both coastal and inland sec-
tions on Spitsbergen (Harland & Wilson 1956; Wilson 1961;
Wilson & Harland 1964) that specific attention was focused on
the Neoproterozoic glacial rocks of Svalbard and their significance
in Earth history (Harland 2007). Since this time, and with their
global importance established (Harland 1964), several detailed
sedimentological studies on the Polarisbreen glaciogenic
sediments have been carried out (e.g. Chumakov 1968; Edwards
1976; Hambrey 1982; Fairchild 1983; Dowdeswell et al. 1985),
with the results culminating in a monograph by Harland et al.
(1993). Carbonates within and bounding the Polarisbreen diamic-
tites have also been the focus of coupled sedimentological –
geochemical investigations (Fairchild & Hambrey 1984; Fairchild
& Spiro 1987; Halverson et al. 2004). More broadly, Neoprotero-
zoic sediments in Svalbard and East Greenland were the subject
of the seminal chemostratigraphic study by Knoll et al. (1986)
that first firmly established the intimate association between nega-
tive C-isotope anomalies and Precambrian glaciation. The upper
Hecla Hoek succession also hosts an unusually rich microfossil
record, which has been studied in detail by A. H. Knoll, N. J. Butter-
field and colleagues (e.g. Knoll 1982, 1984; Butterfield et al. 1988,
1994). The excellent preservation of these sediments and their role
in the ongoing debate over the cause and severity of Neoproterozoic
glaciations overcompensate for the logistical difficulty involved in
studying them. This contribution is intended to update the review
by Hambrey et al. (1981) in the context of more recent research
on the Polarisbreen Group and significant advances in our under-
standing of the Neoproterozoic Earth system since that time.
Structural framework
The Svalbard archipelago, situated on the northwestern corner of
the Barents Shelf, comprises three tectonic terranes that were
amalgamated in the Silurian – Devonian Ny Friesland (Caledonian)
orogeny (Harland & Gayer 1972; Harland et al. 1992; Gee & Page
1994; Lyberis & Manby 1999). The thick Hecla Hoek succession
occurs in the northern part of the Eastern Terrane, which is the
only exposed part of the Barentsia microcontinent (Breivik et al.
2002). The Polarisbreen Group is part of the relatively undeformed
upper Hecla Hoek succession, which includes the Lomfjorden and
Hinlopenstretet supergroups (Fig. 55.2). In Ny Friesland, the basal
Lomfjorden Supergroup is juxtaposed against high-grade meta-
sediments of the Stubendorfbreen Supergroup along the Eolus-
sletta Shear Zone (Fig. 55.1; Lyberis & Manby 1999). The
nature of this contact and the relationship between the lower and

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 571– 579. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.55
572 G. P. HALVERSON
upper Hecla Hoek has been the subject of a long-standing debate,
with one side arguing that the two packages form a single, con-
formable stratigraphic sequence (e.g. Harland 1959; Harland
et al. 1992). Most recent workers, however, agree that this
contact represents a significant hiatus associated with Grenvillian
orogenesis (e.g. Gee et al. 1995; Manby & Lyberis 1995;
Fig. 55.1. Geological sketch map of the
region in northeastern Svalbard where the
Polarisbreen Group and equivalent rocks
crop out. The mapping is from Dallmann
et al. (2002), with modifications to the
Neoproterozoic strata based on mapping by
the author. Letters a to e refer to principal
regions in which the Polarisbreen Group has
been studied: a, Akademikerbreen area;
b, Ditlovtoppen; c, Dracoisen/Oslobreen;
d, Sveanor (south coast of Murchisonfjord),
e, Aldousbreen (north coast of
Wahlenbergfjord). ESZ, Eolusletta Shear
Zone (cf. Lyberis & Manby 1999); LFZ,
Lomfjorden Fault Zone.
Witt-Nilsson et al. 1998), an argument that is supported by
mapping and geochronology in Nordaustlandet where the Lomf-
jord Supergroup rests unconformably on early Neoproterozoic vol-
canics and subvolcanic intrusives (Fig. 55.2), which themselves
intrude and overlie high-grade, late Mesoproterozoic metasedi-
ments (Gee et al. 1995; Johansson et al. 2005). The broad,
 THE POLARISBREEN GROUP, NORTHEASTERN SVALBARD
Fig. 55.2. Stratigraphic nomenclature and correlation of the Neoproterozoic
successions in northeastern Spitsbergen and northwestern Nordaustlandet. Note
that only the nomenclature from Spitsbergen is used in this chapter. Hinlopen.
Spgp., Hinlopenstretet Supergroup; Lom. S., Lomfjorden Supergroup; St. S.,
Stubendorfbreen Supergroup. The ‘upper Hecla Hoek succession’, referred to in
the text and many publications, is the Lomfjorden plus Hinlopenstretet
supergroups.
conformable and aggradational nature of the upper Hecla Hoek
and its continuity with the equivalent units in East Greenland
(Fairchild & Hambrey 1995) imply that the glacial sediments of
the Polarisbreen Group were deposited in a thermally subsiding
basin (Maloof et al. 2006).
The upper Hecla Hoek crops out in a 120-km-long, north – south
belt (Fig. 55.1) that was deformed during the Caledonian orogeny.
The Polarisbreen Group sits in the core of the north-plunging Hin-
lopenstrettet Synclinorium, which spans the length of the belt
(Fig. 55.1). The eastern limb of the synclinorium, exposed in
Nordaustlandet, is a fold-and-thrust belt, dissected by a network
of conjugate NW –SE and SW –NE strike –slip faults (Sokolov
et al. 1968). The western limb of the synclinorium is less
deformed, but is similarly cut by Caledonian-aged reverse and
strike –slip faults. (Lyberis & Manby 1999). In the central part
of the belt, the upper Hecla Hoek is intruded by Silurian granitoids
(Teben’kov et al. 1996). The upper Hecla Hoek was eroded and
draped by Carboniferous sediments during post-orogenic, west–
east extension (Harland 1959), but normal faulting was largely
concentrated along the prominent Lomfjorden Fault Zone (LFZ;
Fig. 55.1). Dolerites related to the Late Jurassic – Early Cretaceous
High Arctic Large Igneous Province (Maher 2001) intrude the
Hecla Hoek within the northern part of the belt in the Hinlopenstre-
tet region (Fig. 55.1). The Late Cretaceous –Early Tertiary West
Spitsbergen Orogen, which heavily deformed the Western
Terrane (Harland 1959), only lightly affected the Eastern
Terrane, but did result in reactivation of major lineaments, such
as the LFZ (Larsen 1987; Lyberis & Manby 1993).
Stratigraphy
The Polarisbreen Group (formerly Series) was originally defined
by Wilson & Harland (1964) in the Dracoisen area in southern
Ny Friesland, near the confluence of the Polarisbreen and Chyde-
niusbreen glaciers (Fig. 55.2). It conformably overlies the thick
stack (2 km) of Akademikerbreen Group carbonates and broadly
marks a transition from predominantly carbonate to mixed
carbonate-clastic sedimentation. The contact between these two
groups is easily identified throughout the outcrop belt, with the
lithologically distinctive, pale yellow ‘Dartboard Dolomite’
(Wilson 1961; Wilson & Harland 1964) of the uppermost Akade-
mikerbreen Group contrasting with the laminated black shale and
bluish-grey to black limestone of the lower Polarisbreen Group
(Halverson et al. 2004). The nature of this contact is variable. In
some places, it is unambiguously transitional. Elsewhere, the
upper Dartboard Dolomite is brecciated and silicified, and contains
tepee-like structures, and the contact with the overlying limestone
573
is sharp (Halverson et al. 2004). The top of the Polarisbreen Group
is bound by the Cambrian Tokammane Fm. (base of the Oslobreen
Group; Fig. 55.2). Although no erosional break at this contact is
evident from mapping or outcrop exposures, a large depositional
hiatus is implied by the biostratigraphy (Knoll & Swett 1987).
The Polarisbreen Group (Figs 55.2 & 55.3) is subdivided into
three formations (Wilson & Harland 1964): Elbobreen, Wilsonb-
reen and Dracoisen. The Elbobreen Fm. is further subdivided
into four regionally mappable members, each of which could be
accorded formation status in its own right: (from bottom to top)
the E1 or Russøya Member, the E2 or Petrovbreen Member, the
E3 or MacDonaldryggen Member, and the E4 or Slangen
Member (Fig. 55.2). A fifth unit that only occurs in the north of
the outcrop belt, the Bråvika member, was tentatively assigned
to the Wilsonbreen Fm. (Halverson et al. 2004), but is clearly tran-
sitional below with Slangen Member, and thus may more appropri-
ately belong to the Elbobreen Fm. The Petrovbreen Member and
Wilsonbreen Fm. comprise the two glaciogenic units, both of
which are distinct from one another sedimentologically and
within the context of their bounding strata (Figs 55.3 & 55.4).
Glaciogenic deposits and associated strata
Russøya Member (E1)
The Russøya Member varies from 75 to 170 m in thickness,
thinning from south to north (Halverson et al. 2004). It is a litho-
logically variable unit, being predominantly carbonate in Nordaus-
tlandet and containing an increasing fraction of sandstone and
siltstone to the south. Despite its heterogeneity, the stratigraphic
geometry of the Russøya Member is consistent, comprising a rela-
tively thick (40– 160 m), basal, shoaling-upward sequence, over-
lain by up to four complete parasequences, 5–20 m thick
(Fig. 55.3). The lowest parasequence consists of a basal dark shale
that grades upward into increasingly carbonate-rich sediments,
including limestone ribbonites and grainstones. Molar tooth struc-
tures are prominent in this sequence, but occur exclusively within
the ribbonite facies (Fig. 55.3).
The upper parasequences typically consist of shale at the base
and sandstone or dolomitic grainstone, stromatolites or microbial
laminites at the top. The uppermost parasequence is capped by a
2– 6-m-thick biostrome of a spiralling, columnar stromatolite
(Kussiella) and brecciated and variably truncated on an outcrop
scale (Halverson et al. 2004). Although erosional truncation
below the Petrovbreen Members is ,2 m on the outcrop scale,
regional correlations, assisted by chemostratigraphy (see below),
suggest as much as 50 m of erosion beneath the Petrovbreen
Member (Halverson et al. 2004).
Petrovbreen Member (E2)
The Petrovbreen Member is on average only about 10 m thick
(Fig. 55.4), but is highly variable (,52 m), its thickness broadly
reciprocating the depth of erosion on the sub-glacial sequence
boundary (Halverson et al. 2004). In Nordaustlandet, the Petrovb-
reen Member is typically very thin and had been recognized only
in northernmost sections (Flood et al. 1969; Hambrey 1982;
Harland et al. 1993), but recent mapping has shown that it can be
identified in most sections (Halverson et al. 2004). The yellow- to
orange-weathering Petrovbreen Member is predominantly com-
posed of poorly stratified and massive diamictite and wackestone
with a yellow- to orange-weathering dolomite matrix. Other
common facies include dolomitic rhythmites, tabular clast conglom-
erates and sandy dolostone (Fairchild & Hambrey 1984). Climbing
ripples occur near the top of the member in some sections. The
clasts in the diamictite range up to 90 cm in diameter, are sub-
angular to angular, and consist mainly of light grey dolomite
574 G. P. HALVERSON
(grainstone and stromatolite) and black chert. Minor sandstone, silt-
stone and rare volcanic clasts are also found (Hambrey 1982;
Harland et al. 1993).
Evidence for direct glacial influence in the Petrovbreen Member
comes from rare striated clasts and the abundance of dropstones in
finely laminated sediments (Hambrey 1982; Harland et al. 1993),
as well as evidence of glacial rock flour in the form of sub-
micrometre-sized dolomite in the matrix of the diamictite
(Fairchild 1983). Fe and 18O-enrichment of this dolomitic matrix
relative to clasts suggests early marine alteration (Fairchild et al.
1989). The absence of features indicative of a terrestrial environ-
ment, coupled with rapid facies changes and the occurrence of
dropstones, suggest glaciomarine (or glaciolacustrine) deposition
near an ice-grounding line (Hambrey 1982; Fairchild &
Hambrey 1984; Harland et al. 1993).
MacDonaldryggen Member (E3)
The contact between the upper Petrovbreen Member and the base of
the MacDonaldryggen Member is everywhere sharp but conform-
able. In Olav V Land, where the MacDonaldryggen Member was
named (Harland et al. 1993), the base consists of a thin (10–40
cm), finely laminated carbonate (variably limestone and dolomite)
with wispy organic-rich laminae (Halverson et al. 2004). Else-
where, the upper Petrovbreen Member is directly overlain by the
same finely laminated, olive green to dark grey mudstone that com-
prises the bulk of the c. 200-m-thick MacDonaldryggen Member.
The fraction of silt increases up-section, with thin, fine sandstone
beds present in some sections. In Nordaustlandet, authigenic
calcite nodules up to 2 cm in diameter, and in places densely clus-
tered within distinct layers, speckle the upper MacDonaldryggen
Member some 20–40 m beneath the contact with the overlying
Slangen Member (Fig. 55.3). These nodules are commonly
Fig. 55.3. Composite stratigraphic column
through the Polarisbreen Group (after
Halverson et al. 2004) and associated
chemostratigraphic and biostratigraphic
data. Sr-isotope data are from Halverson
et al. (2007), C-isotope data (v. VPDB) on
organic matter (d13Corg) from Knoll et al.
(1986) and Kaufman et al. (1997) (as
compiled in the latter), and C-isotope data
on carbonates (d13Ccarb) from Fairchild &
Spiro (1987), Halverson et al. (2004) and
Kaufman et al. (1997). The two principal
acritarch assemblages in the Polarisbreen
Group (right side of figure) are summarized
from Knoll (1982) and Knoll & Swett
(1987).
rosette-shaped and have canted and square prismatic habits (Halver-
son et al. 2004).
The upper MacDonaldryggen shales and siltstones are every-
where transitional into flaggy dolomites that mark the bottom of
the Slangen Member. In southern sections, this transition includes
an increase in carbonate, including limestone ribbons. Mud-cracks
have been reported from this level (Fairchild & Hambrey 1984),
but appear to be rare.
Slangen Member (E4)
The Slangen Member is typically a 20 –30 m regressive parase-
quence of cherty dolostone. The dominant facies in the Slangen
Member is a vuggy, trough and tabular cross-bedded grainstone;
however, fenaestral microbial laminates also occur as thin inter-
beds within the grainstone and as distinct beds up to 5 m thick in
the upper Slangen Member. Ribbons occur mainly in the transition
with the underlying MacDonaldryggen Member, but are also
important in the uppermost Slangen Member in Nordauslandet.
The occurrence of length-slow chalcedony, high Na concen-
trations, and rare anhydrite in the Slangen Member suggest depo-
sition in a restricted environment such as a coastal sabkha
(Fairchild & Hambrey 1984).
The upper contact of the Slangen Member is variable. In most
Spitsbergen sections, it is sharp, silicified and brecciated, with
clasts showing no evidence for reworking (Fairchild & Hambrey
1984). At Ditlovtoppen (Fig. 55.1) the grainstone passes transition-
ally upward into 2.5 m of mud-cracked siltstone beneath the basal
diamictites of the Wilsonbreen Fm. In Nordaustlandet, the grain-
stones and microbial laminites are commonly overlain by silty
dolomite ribbons, which grade upwards through ripple cross-
laminated dolomitic siltstone and fine sandstone and into dolomitic
sandstone of the basal Bråvika Member (Fig. 55.4).
 THE POLARISBREEN GROUP, NORTHEASTERN SVALBARD
The Wilsonbreen Fm.
In most sections, the Wilsonbreen Fm. consists of c. 100–150 m of
green- to red-weathering, massively to poorly bedded diamictite;
however, the diamictite thins dramatically northwards in Nordaus-
tlandet, being reciprocated by a northward-thickening wedge of
quartz sands, informally named the Bråvika member (Halverson
et al. 2004; Fig. 55.4). In the northernmost exposure of the Wil-
sonbreen Fm. (at Langrunesset), the Bråvika member is 250 m
thick and diamictite is absent.
In Spitsbergen, the Wilsonbreen Fm. is subdivided into the
Ormen (W1), Middle Carbonate (W2) and Gropbreen (W3)
members (Harland et al. 1993; Fig. 55.4). The Ormen Member
varies from 20 to 75 m in thickness and consists mainly of
poorly stratified, shaley diamictite, but also includes lenses and
interbeds of sandstone, conglomerate and breccia (Fig. 55.4; Fair-
child & Hambrey 1984). Clast lithology is variable, but dominated
by carbonates derived from the underlying Elbobreen Fm. and
Akademikerbreen Group. Chert, silt, sandstone and basement
clasts, including exotic fragments of gneiss and undeformed pink
granite, are also found. Striated clasts occur throughout the unit
and dropstones, although not common in most sections, are
575
Fig. 55.4. Characteristic stratigraphic
sections of the Wilsonbreen Fm. and
Petrovbreen Member glacials from each of
principal regions in which the glacial
intervals have been studied (locations a to e
are linked to Fig. 55.1; the Petrovbreen
Member does not occur at location e).
Stratigraphic depth is given relative to a 0 m
datum at the top of each glacial unit. Note
the difference in scale between the
Wilsonbreen Fm. and Petrovbreen Member.
The measured sections are compiled from
published columns in Edwards (1976),
Harland et al. (1993), Halverson et al.
(2004) and unpublished data of the author.
The breakdown and naming of members
here follows that of Harland et al. (1993),
but is slightly different from that used in
previous papers, where the names of the
Ormen and Gropbreen Members are
inverted and a fourth Wilsonbreen member
is included (e.g. Hambrey 1982).
abundant at Aldousbreen (Fig. 55.4e; Edwards 1976). Sandstone
wedges occur near or at the top of the Ormen Member in the Aka-
demikerbreen area (Fig. 55.4a) and at Ditlovtoppen (Fig. 55.4b)
(Hambrey 1982; Fairchild & Hambrey 1984).
The Middle Carbonate Member is thin (,25 m) and comprises
mainly dolomitic, medium-grained quartz sandstone. It is,
however, recognized and distinguished by the presence of thin
(,1.5 m) beds, lenses and fragments of primary carbonate. The
nature of the carbonate is variable, being corrugated silty dolomitic
microbial laminites at Ormen (in the Akademikerbreen area),
mixed limestone and dolomite rhythmites and ribbons at Ditlov-
toppen (Fig. 55.4b), and isolated limestone stromatolites within
sandstone at Dracoisen (Fig. 55.4c). Based on their geochemistry,
these carbonates are interpreted to have formed under the influence
of evaporation in periglacial lakes (Fairchild et al. 1989).
The Gropbreen Member is lithologically similar to the Ormen
Member, with dominantly weakly stratified to massive maroon-
weathering diamictite, and subordinate sandstone lenses and
beds. Clast composition is also similar to the Ormen Member,
but with a higher proportion of basement clasts. In many sections,
a medium brown sandstone up to a few metres thick occurs at the
very top of the Gropbreen Member and is continuous with well
576 G. P. HALVERSON
developed sandstone wedges that penetrate up to 2 m into the
underlying diamictite. Whether the sand is present or not, the
contact between the Wilsonbreen Fm. and the overlying Dracoisen
Fm. is always sharp.
In Nordaustlandet, the Wilsonbreen is subdivided into two
members. The Bråvika member overall is a coarsening-upwards
package of quartz sandstone, with irregular dolomite lenses and
intraclasts, and in the northernmost sections, minor limestone
rhythmites and ribbons. The northward-thickening sandstone con-
tains both planar and trough cross-bedding throughout, indicating a
palaeoflow direction to the north (Halverson et al. 2004). The
prevalence of low-angle cross-sets, coupled with pinstripe lami-
nations and a bimodal grain population, suggests an aeolian
origin, although the trough cross-bedding and dolomite precipi-
tation indicate that the sand was in places reworked subaqueously.
In the southern exposures in Nordaustlandet, the Bråvika sand-
stones are transitional with overlying diamictites, but in the
Sveanor (Fig. 55.4d) are also transitional below with the Slangen
Member (Halverson et al. 2004).
The diamictite-dominated unit overlying the Bråvika member in
Nordaustlandet resembles the Ormen and Gropbreen members on
Spitsbergen, only containing a higher percentage of undeformed
igneous clasts. Rhythmites with dropstones and fine interbedded
turbidites are well preserved in the southernmost Nordaustlandet
section at Aldousbreen (Fig. 55.4e), in Wahlenbergfjorden
(Edwards 1976). Unstratified siltstones, interpreted as loessites
(Edwards 1976), also occur in the same section (Fig. 55.4).
The Dracoisen Fm.
The colourful lower part of the Draocoisen Fm. is a superb marker
interval (Wilson & Harland 1964). A yellow-weathering dolomite
bed everywhere overlies the Wilsonbreen Fm., with no evidence
for reworking or depositional hiatus. This dolostone varies in
thickness from as much as 18 m at Ditlovtoppen to less than 3 m
in parts of Nordaustlandet (Halverson et al. 2004) and always com-
prises the transgressive component of a thick (c. 170 m) sequence
(Fig. 55.3). The dolostone is mostly finely laminated dololutite,
with minor low-angle erosional truncations and cross-laminations.
Inversely graded laminae and pockets of peloids are also common.
Large-scale oscillation ripples with amplitudes of up to 40 cm and
wavelengths up to 6 m (Allen & Hoffman 2005) occur in the upper
part of the dolostone in most northern sections.
The top of the dolostone is transitional into marly red siltstone,
which then passes upwards into green then black shales. The
change to black shale, at c. 30–40 m above the base of the Dracoi-
sen Fm., marks the maximum flooding surface of the sequence.
Black shales with common carbonate concretions persist for over
100 m and are transitional into mud-cracked, variegated, finely
bedded siltstone and shale that constitutes much of the upper
part of the formation. The interval of mud-cracked siltstone is
interrupted by 9 m of microbial-laminated dolomite, internally
deformed by cauliflower chert.
Chemostratigraphy
In one of the earliest systematic chemostratigraphic studies of a
Neoproterozoic succession, Knoll et al. (1986) produced a
coupled d13Ccarb – d13Corg record through the Veteranen, Akademi-
kerbreen and Polarisbreen groups in both Spitsbergen and Nor-
daustlandet (as well as East Greenland). Although by modern
standards the resolution of this record is low, with only a
handful of data points from the Polarisbreen Group, this paper
established the concurrence of glaciations and negative
C-isotope anomalies, as well as the prevalence of high d13C
values during the much of the Neoproterozoic. Subsequent
C-isotope stratigraphy on the Polarisbreen Group was published
by Fairchild & Spiro (1987), Kaufman et al. (1997) and Halverson
et al. (2004, 2005). Separate studies have focused on the isotopic
composition of dolomites and limestones within the Wilsonbreen
Fm. as a means of reconstructing the palaeoenvironment of
syn-glacial carbonate precipitation (Fairchid & Spiro 1987,
1990; Fairchild et al. 1989).
A compilation of C-isotope data from these various publications
is plotted alongside a composite stratigraphic column in
Figure 55.3. The Polarisbreen Group d13Ccarb and d13Corg records
broadly mirror one another (Fig. 55.3), with an average isotopic
difference of c. 30‰, which is typical for the Neoproterozoic
(Hayes et al. 1999). The most striking feature of the C-isotope
record is a 12‰ negative anomaly just below the Petrovbreen
Member. This anomaly is reproduced in multiple sections and
essentially spans the upper two parasequences of the Russøya
Member (Halverson et al. 2004). d13C values are still negative in
carbonates within the lowermost MacDonaldyrggen Fm., but
quickly rise to positive values, where they remain into the Wil-
sonbreen Fm. Intraglacial carbonates show a range in d18O and
d13C values between –11 and þ11‰ and þ1 and 5‰, respectively
(Fairchild & Spiro 1987; Halverson et al. 2004). Fairchild & Spiro
(1987, 1990) argued that the Wilsonbreen carbonates were precipi-
tated by evaporation in a periglacial lake based, in part, on these
unusually high d18O compositions. More recently, Bao et al.
(2009) reported D17O (triple oxygen isotopes) compositions in
carbonate-associated sulphate (CAS) within these carbonates of
as low as –1.6‰, which are the lowest values ever reported
from natural specimens. Bao et al. (2009) interpreted these anom-
alous values to record either extraordinarily high pCO2 or unu-
sually sluggish O2 cycling during the glaciation.
A second negative C-isotope anomaly is recorded in the basal
Dracoisen dolostone and reproduced in multiple sections, where
d13C values gradually decline from –3 to – 5‰ (Halverson et al.
2004). d13C values then spike to .10‰ in evaporitic dolomites
in the upper Dracoisen Fm. (Fig. 55.3).
Four Sr-isotope data from the Polarisbreen Group were first
published by Derry et al. (1989), and a few new analyses from
the original sample set were added in Jacobsen & Kaufman
(1999). Additional 87Sr/86Sr data from the Polarisbreen were pub-
lished in Halverson et al. (2005, 2007). The highest-quality data
from all these studies consistently show 87Sr/86Sr ¼ 0.7067–
0.7068 in the Russøya Member (Fig. 55.3). The absence of unal-
tered marine limestone above the Russøya Member precludes
application of Sr-isotope chemostratigraphy to the remainder of
the Polarisbreen Group.
Palaeolatitude and palaeogeography
Barentsia is widely assumed to have been contiguous with East
Greenland during the Neoproterozoic based on the remarkable
stratigraphic similarity between the two regions (Harland &
Gayer 1972; Knoll et al. 1986; Fairchild & Hambrey 1995).
However, the precise location of Eastern Svalbard relative to the
East Greenland Caledonides (Johansson et al. 2005) and the geo-
metry and origin of the East Greenland –eastern Svalbard platform
within the broader framework of the connection and eventual
rifting between Laurentia, Baltica and Amazonia are controversial.
In the conventional configuration, western Baltica is the conjugate
margin to eastern Greenland (and Barentsia), virtually identical to
the Caledonian fit (e.g. Weil et al. 1998). In a different reconstruc-
tion, Hartz & Torsvik (2002) placed the East Greenland –eastern
Svalbard platform adjacent to the southern peri-Urals of Baltica
and proposed that it developed as a sinistral pull-apart basin
between Baltica and Amazonia. This model has been challenged
by more recent palaeomagnetic data from Baltica (Cawood &
Pisarevsky 2006), and in a modification of earlier west-Baltica –
eastern Greenland fits, Pisarevsky et al. (2008) proposed that
the west Norwegian margin was conjugate to southeastern
Greenland. This model implies that the eastern Svalbard and the
 THE POLARISBREEN GROUP, NORTHEASTERN SVALBARD
Neoproteroozic successions in northern Norway were part of a
long passive margin isolated from the other coeval sedimentary
basins to the south.
Direct palaeomagnetic constraints for Svalbard’s Neoproterozoic
palaeography are limited to a new suite of palaeomagnetic data
from the Akademikerbreen Group that show a stable, pre-
Caledonian remanent magnetization indicating that the EGES
platform resided in tropical latitudes (I  158) during the mid-
Neoproterozoic (Maloof et al. 2006). No reliable palaeomagnetic
results have been obtained from the Polarisbreen Group, and Sval-
bard’s complicated and uncertain tectonic history since this time
precludes any extrapolation of palaeolatitude for the glacial depos-
its from elsewhere on the Laurentia craton (Evans 2000).
Geochronological constraints
No direct radiometric ages have been obtained on the upper Hecla
Hoek. U –Pb ages on detrital zircons in the Planetfjella Group in
Spitsbergen and presumed primary zircons from volcanics and
subvolcanic intrusives beneath the Veteranen Group in Nordaus-
tlandet indicate a maximum age for the base of the Lomfjorden
Supergroup of c. 946 Ma (Johansson et al. 2000, 2005;
Fig. 55.2). Similar ages have been obtained from igneous clasts
in the Wilsonbreen Fm. in Nordaustlandet (Johansson et al. 2000).
The top of the Polarisbreen Group, which is in disconformable
contact with the overlying Cambrian Tokamanne Fm. (Knoll &
Swett 1987), is inferred to be ,542 Ma (the age of the
Precambrian –Cambrian boundary; Amthor et al. 2003). If the
Dracoisen Fm. is correlative with the Maieberg Fm. in northern
Namibia and the Doushantuo Fm. in South China (Hoffmann
et al. 2004; Condon et al. 2005), as implied by the chemostrati-
graphy and biostratigraphy of the Polarisbreen Group, then the
Wilsonbreen glaciation ended at 635 Ma (Halverson et al. 2005).
Biostratigraphy
In contrast to the rich and diverse fossil assemblage from the
underlying Akademikerbreen Group (Knoll 1982; Butterfield
et al. 1994, and references therein), the biostratigraphic record of
the Polarisbreen Group is most notable for its depauperate micro-
fossil assemblage, despite the generally mild thermal history of the
rocks (Knoll & Swett 1987). In the MacDonaldryggen Member
and in shales within the Wilsonbreen Fm., the acritarch Bavlinella
faveolata dominates the microfossil assemblage (Knoll 1982;
Knoll & Swett 1987). B. faveolata also occurs in the Dracoisen
Fm. and is locally abundant, but overall is subordinate to a low
diversity assemblage of small, smooth-walled leiosphere acri-
tarchs (Fig. 55.3). The diverse assemblage of the large, acantho-
morphic acritarchs that characterize middle– upper Ediacaran
successions in Australia (Grey et al. 2003) and elsewhere (Knoll
2000) is absent in the Polarisbreen Group (Knoll & Swett 1987).
Coupled with the lack of Ediacaran fauna, this lacuna strongly
suggests that the contact between the Polarisbreen Group and the
overlying Oslobreen Group represents a significant depositional
hiatus (Knoll & Swett 1987) and that at least the last 35 myr of
the Ediacaran Period are missing in Svalbard.
Discussion
The Neoproterozoic succession in Svalbard has played a promi-
nent role in questions and controversies regarding Precambrian
glaciations and environmental change since Harland (1964) first
proposed a pan-global infra-Cambrian glaciation. Although a
few researchers maintained scepticism over the glacial origin of
diamictites in the Polarisbreen Group (Krasil’shchov 1973; Scher-
merhorn 1974), the detailed sedimentological work of Edwards,
Hambrey, Fairchild and others (e.g. Edwards 1976; Hambrey
577
1982; Fairchild & Hambrey 1984; Dowdeswell et al. 1985) effec-
tively put to rest any doubt. Evidence for a glacial environment,
specifically in the Wilsonbreen Fm., is found in the abundance
of striated and faceted clasts, dropstones, the extrabasinal origin
of some stones, and sandstone wedges.
The Petrovbreen Member diamictite and associated lithofacies
are typically interpreted to have been deposited in relatively
deep water, most likely in a marine environment (Fairchild &
Hambrey 1984; Harland et al. 1993). This interpretation is
consistent with the lack of evidence for any obvious change in
relative sea level at the contact between the Petrovbreen and
MacDonaldryggen Members (Halverson et al. 2004). In contrast,
the heterogeneous Wilsonbreen Fm., which includes diamictites
interpreted as lodgement tills (Dowdeswell et al. 1985), direct
evidence for subaerial exposure (sandstone wedges), and lacus-
trine carbonates is inferred to have been deposited in a pre-
dominantly terrestrial setting (Fairchild & Hambrey 1984;
Dowdeswell et al. 1985; Halverson et al. 2004). Dowdeswell
et al. (1985) argued that the Wilsonbreen diamictites were depos-
ited beneath ice sheets or ice caps (rather than valley glaciers)
based on the preponderance of exotic clasts and a lack of
angular supraglacial debris.
A great deal of effort has been expended on attempting to corre-
late the Polarisbreen diamictites with other Neoproterozoic glacial
units in the North Atlantic (e.g. Harland & Gayer 1972; Hambrey
1983; Nystuen 1985; Fairchild & Hambrey 1995) and globally
(e.g. Kaufman et al. 1997; Kennedy et al. 1998; Halverson et al.
2005). Based on biostratigraphic and chemostratigraphic data, it
has generally been argued that the Polarisbreen Group is late
‘Vendian’ in age, meaning essentially that it post-dates the earliest
Neoproterozoic glaciations. Unfortunately, the initial optimism that
the chronostratigraphy and correlations would be firmed up by geo-
chronological and palaeomagnetic data (Hambrey 1983) has not
been realized, and no undisputed correlation scheme exists for the
North Atlantic region, let alone the whole of the globe.
In a twist on the conventional Vendian interpretation of the age of
the Polarisbreen Group diamictites, Halverson et al. (2004) argued
that both the Petrovbreen Member and Wilsonbreen Fm. belonged
to a single glacial episode, ending at 635 Ma (the approximate age
of the end of the Ghaub glaciation in Namibia and Nantuo glaciation
in South China; Hoffmann et al. 2004; Condon et al. 2005). This
model was based on (i) the similar stratigraphic context and magni-
tude of the upper Russøya negative d13C anomaly (Fig. 55.3) and
the so-called Trezona anomaly, which precedes the Ghaub glacia-
tion in Namibia (Halverson et al. 2002), coupled with lack of evi-
dence for an analogous anomaly preceding any of the older
glaciations; (ii) the occurrence of glendonites in the upper MacDo-
naldryggen Member; and (iii) the virtually identical geochemistry,
sedimentology and stratigraphy of the Dracoisen and post-Ghaub
Maieberg (and other) cap-carbonate sequences. In light of new
Sr-isotope data from the level of the negative anomaly in the
Russøya Member (87Sr/86Sr ¼ 0.7068 v. 0.7072 in the Trezona
anomaly; Halverson et al. 2007) and evidence that the lowermost
of three distinct glacial units in the Dalradian Supergroup is
preceded by a negative C-isotope anomaly of similar magnitude
to that in the upper Russøya Member (McCay et al. 2006), it now
seems more likely that the Petrovbreen Member diamictite must
represent a separate, older glaciation. Although it remains likely
that the Wilsonbreen Fm. is c. 635 Ma, the ongoing ambiguity in
Cryogenian–Ediacaran age constraints precludes any unequivocal
assignment of age to either of the Polarisbreen glaciations.
The more conventional two-glaciation interpretation also raises
the question of the climatic significance of the glendonite nodules
in the upper MacDonaldryggen Member. Glendonites have also
been reported in possibly equivalent-aged strata in the Windermere
Supergroup in the Mackenzie Mountains (James et al. 2005).
Insofar as the MacDonaldryggen pseudomorphs originated as
ikaite, which forms at near-freezing temperatures in alkaline-
charged waters (Buchart et al. 1997), then their occurrence in
578 G. P. HALVERSON
continental shelf sediments with no other evidence of glaciation,
and just below sabkha deposits, poses yet another Neoproterozoic
climatic conundrum. Thus, it seems certain that the Polarisbreen
Group will continue to play a central role in the ongoing effort
to understand Neoproterozoic climate.
This chapter stems in part from the author’s PhD and subsequent research in
Svalbard, funded by NSF and NASA. Original mapping, measured sections,
and other contributions presented here are the results of collaborative fieldwork
with A. C. Maloof, P. F. Hoffman, E. W. Domack, M. T. Hurtgen, J. Eigenbrode,
and many field assistants. M. Hambrey and I. Fairchild provided constructive
reviews of the manuscript. The contribution forms TRaX record #169. This rep-
resents a contribution of the IUGS and UNESCO-funded IGCP (International
Geoscience Programme) project #512.
References
Allen, P. A. & Hoffman, P. F. 2005. Extreme winds and waves in the
aftermath of a Neoproterozoic glaciation. Nature, 433, 123– 127.
Amthor, J. E., Grotzinger, J. P., Schröder, S., Bowring, S. A., Rame-
zani, J., Martin, M. W. & Matter, A. 2003. Extinction of Cloudina
and Namacalathus at the Precambrian – Cambrian boundary in Oman.
Geology, 31, 431–434.
Bao, H., Fairchild, I. J., Wynn, P. M. & Spötl, C. 2009. Stretching the
envelope of past surface environments: Neoproterzoic glacial lakes
from Svalbard. Science, 323, 119–122.
Breivik, A. J., Mjelde, R., Grogan, P., Shimamura, H., Murai, Y.,
Nishimura, Y. & Kuwana, A. 2002. A possible Caledonide arm
through the Barents Sea imaged by OBS data. Tectonophysics, 355,
67 –97.
Buchart, B., Seaman, P. et al. 1997. Submarine columns of ikaite tufa.
Nature, 390, 129–130.
Butterfield, N. J., Knoll, A. H. & Swett, K. 1988. Exceptional pres-
ervation of fossils in an Upper Proterozoic shale. Nature, 334,
424– 427.
Butterfield, N. J., Knoll, A. H. & Swett, K. 1994. Paleobiology of the
Neoproterozoic Svanbergfjellet Formation, Spitsbergen. Fossils and
Strata, 34, 1– 81.
Cawood, P. A. & Pisarevsky, S. A. 2006. Was Baltica right-way-up or
upside-down in the Neoproterozoic? Journal of the Geological
Society, London, 163, 753–759.
Chumakov, N. 1968. On the character of the Late Precambrian glaciation
of Spitsbergen (translated title). Doklady An SSSR, Seriya Geologi-
cheskaya, 180, 1446– 1449.
Condon, D., Zhu, M., Bowring, S., Jin, Y., Wang, W. & Yang, A. 2005.
From the Marinoan glaciation to the oldest bilaterians: U– Pb ages
from the Doushantou Formation, China. Science, 308, 95– 98.
Derry, L. A., Keto, L. S., Jacobsen, S. B., Knoll, A. H. & Swett, K.
1989. Strontium isotopic variations in Upper Proterozoic carbonates
from Svalbard and East Greenland. Geochimica et Cosmochimica
Acta, 53, 2331– 2339.
Dallmann, W. K., Ohta, Y., Elvevold, S. & Blomeier, D. 2002.
Bedrock map of Svalbard and Jan Mayen (1:750,000 scale geological
map). Norsk Polarinstitutt Temekart, 33.
Dowdeswell, J. A., Hambrey, M. J. & Wu, R. 1985. A Comparison of
clast fabric and shape in Late Precambrian and Modern glaciogenic
sediments. Journal of Sedimentary Petrology, 55, 691– 704.
Edwards, M. B. 1976. Sedimentology of Late Precambrian Sveanor
and Kapp Sparre Formations at Aldousbreen, Wahlenbergfjorden,
Nordaustlandet. Norsk Polarinstitutt Årbok, 1974, 51 –61.
Edwards, M. B. 1978. Late Precambrian glacial loessites from north
Norway and Svalbard. Journal of Sedimentary Petrology, 49, 85 – 91.
Evans, D. A. D. 2000. Stratigraphic, geochronological and paleomagnetic
constraints upon the Neoproterozoic climatic paradoxes. American
Journal of Science, 300, 347– 443.
Fairchild, I. J. 1983. Effects of glacial transport and neomorphism on
Precambrian dolomite crystal sizes. Nature, 304, 714– 716.
Fairchild, I. J. & Hambrey, M. J. 1984. The Vendian succession of
northeastern Spitsbergen: petrogenesis of a dolomite– tillite associ-
ation. Precambrian Research, 26, 111–167.
Fairchild, I. J. & Hambrey, M. J. 1995. Vendian basin evolution in East
Greenland and NE Svalbard. Precambrian Research, 73, 217– 233.
Fairchild, I. J. & Spiro, B. 1987. Petrological and isotopic implications
of some contrasting Late Precambrian carbonates, NE Spitsbergen.
Sedimentology, 34, 973– 989.
Fairchild, I. J. & Spiro, B. 1990. Carbonate minerals in glacial sedi-
ments: geochemical clues to palaeoenvironment. In: Dowdeswell,
J. A. & Scourse, J. D. (eds) Glacimarine Environments: Processes
and Sediments. Geological Society, London, Special Publications,
53, 201– 216.
Fairchild, I. J., Hambrey, M. J., Spiro, B. & Jefferson, T. H. 1989. Late
Proterozoic glacial carbonates in northeast Spitsbergen: new insights
into the carbonate –tillite association. Geological Magazine, 126,
469– 490.
Fleming, W. L. S. & Edmonds, J. M. 1941. Hecla Hoek rocks of New
Friesland (Spitsbergen). Geological Magazine, 78, 405– 428.
Flood, B., Gee, D. G., Hjelle, A., Siggerud, T. & Winsnes, T. 1969.
The geology of Nordaustlandet, northern and central parts. Norsk
Polarinstitutt Skrifter, 146, 1– 139 (þ1:250,000 map).
Gee, D. G. & Page, L. M. 1994. Caledonian terrane assembly on Svalbard:
new evidence from 40Ar/39Ar dating in Ny Friesland. American
Journal of Science, 294, 1166–1186.
Gee, D. G., Johansson, Å. et al. 1995. Grenvillian basement and a major
unconformity within the Caledonides of Nordaustlandet, Svalbard.
Precambrian Research, 70, 215– 234.
Grey, K., Walter, M. R. & Calver, C. R. 2003. Neoproterozoic biotic
diversification: Snowball Earth or aftermath of the Acraman
impact. Geology, 31, 459– 462.
Halverson, G. P., Hoffman, P. F., Schrag, D. P. & Kaufman, A. J.
2002. A major perturbation of the carbon cycle before the Ghaub
glaciation (Neoproterozoic) in Namibia: prelude to snowball
Earth? Geochemistry, Geophysics, Geosystems, 3, doi: 10.1029/
2001GC000244.
Halverson, G. P., Hoffman, P. F., Schrag, D. P., Maloof, A. C. & Rice,
A. H. 2005. Towards a Neoproterozoic composite carbon-isotope
record. Geological Society of America Bulletin, 117, 1181–1207.
Halverson, G. P., Maloof, A. C. & Hoffman, P. F. 2004. The Marinoan
glaciation (Neoproterozoic) in Svalbard. Basin Research, 16,
297– 324.
Halverson, G. P., Dudas, F. O., Maloof, A. C. & Bowring, S. A. 2007.
Evolution of the 87Sr/86Sr composition of Neoproterozoic
seawater. Palaeogeography, Palaeoclimatology, Palaeoecology,
256, 103– 129.
Hambrey, M. J. 1982. Late Precambrian diamictites of northeastern Sval-
bard. Geological Magazine, 119, 527– 551.
Hambrey, M. J. 1983. Correlation of late Proterozoic tillites in the North
Atlantic region and Europe. Geological Magazine, 120, 290– 320.
Hambrey, M. J., Harland, W. B. & Waddams, P. 1981. Late Precam-
brian tillites of Svalbard. In: Hambrey, M. J. & Harland, W. B.
(eds) Earth’s Pre-Pleistocene Glacial Record. Cambridge University
Press, 592–600.
Harland, W. B. 1959. The Caledonian sequence in Ny Friesland, Spits-
bergen. Quarterly Journal of the Geological Society, 114, 307– 342.
Harland, W. B. 1964. Critical evidence for a great infra-Cambrian
glaciation. Geologische Rundschau, 54, 45 –61.
Harland, W. B. 1997. The Geology of Svalbard. Geological Society of
London Memoir, 17, London, 521.
Harland, W. B. 2007. The Ny Friesland Orogen, Spitsbergen. Geological
Magazine, 144, 633– 642.
Harland, W. B. & Gayer, R. A. 1972. The Arctic Caledonides and
earlier oceans. Geological Magazine, 109, 289–314.
Harland, W. B. & Wilson, C. B. 1956. The Hecla Hoek succession in Ny
Friesland, Spitsbergen. Geological Magazine, 93, 265– 286.
Harland, W. B., Scott, R. A., Auckland, K. A. & Snape, I. 1992.
The Ny Friesland Orogen, Spitsbergen. Geological Magazine, 129,
679– 708.
Harland, W. B., Hambrey, M. J. & Waddams, P. 1993. Vendian
Geology of Svalbard. Norsk Polarinstitutt Skrifter (Oslo), 193, 150.
Hartz, E. & Torsvik, T. 2002. Baltica upside down: a new plate tectonic
model for Rodinia and the Iapetus Ocean. Geology, 30, 225– 258.
Hayes, J. M., Strauss, H. & Kaufman, A. J. 1999. The abundance of C
in marine organic carbon and isotopic fractionation in the global
 THE POLARISBREEN GROUP, NORTHEASTERN SVALBARD
biogeochemical cycle of carbon during the past 800 Ma. Chemical
Geology, 161, 103– 125.
Hoffmann, K. H., Condon, D. J., Bowring, S. A. & Crowley, J. L.
2004. A U– Pb zircon date from the Neoproterozoic Ghaub
Formation, Namibia: constraints on Marinoan glaciation. Geology,
32, 817–820.
Jacobsen, S. B. & Kaufman, A. J. 1999. The Sr, C and O isotopic evol-
ution of Neoproterozoic seawater. Chemical Geology, 161, 37 –57.
James, N. P., Narbonne, G. M., Dalrymple, R. W. & Kyser, C. 2005.
Glendonites in Neoproterozoic low-latitude, interglacial sedimentary
rocks, northwest Canada: insights on the Cryogenian ocean and Pre-
cambrian cold-water carbonates. Geology, 33, 9 –12.
Johansson, Å., Larianov, A. N., Tebenkov, A. M., Gee, D. G., White-
house, M. J. & Vestin, J. 2000. Grenvillian magmatism of western
and central Nordaustlandet, northeastern Svalbard. Transactions of
the Royal Society of Edinburgh, 90, 221– 234.
Johansson, A., Gee, D. G., Larionov, A. N., Ohta, Y. & Tebenkov,
A. M. 2005. Grenvillian and Caledonian evolution of eastern Sval-
bard – a tale of two orogenies. Terra Nova, 17, 317– 325.
Kaufman, A. J., Knoll, A. H. & Narbonne, G. M. 1997. Isotopes, ice
ages and terminal Proterozoic earth history. Proceedings of the
National Academy of Science (USA), 94, 6600–6605.
Kennedy, M. J., Runnegar, B., Prave, A. R., Hoffmann, K.-H. &
Arthur, M. 1998. Two or four Neoproterozoic glaciations?
Geology, 26, 1059 – 1063.
Knoll, A. H. 1982. Microfossil based biostratigraphy of the Precambrian
Hecla Hoek sequence of Nordaustlandet, Svalbard. Geological
Magazine, 199, 269– 279.
Knoll, A. H. 1984. Microbiotas of the late Precambrian Hunnberg
Formation Nordaustlandet, Svalbard. Journal of Palaeontology,
58, 131– 162.
Knoll, A. H. 2000. Learning to tell Neoproterozoic time. Precambrian
Research, 100, 3– 20.
Knoll, A. H. & Swett, K. 1987. Micropaleontology across the Precam-
brian – Cambrian boundary in Spitsbergen. Journal of Palaeontology,
61, 898– 926.
Knoll, A. H., Hayes, J. M., Kaufman, A. J., Swett, K. & Lambert, I. B.
1986. Secular variation in carbon isotope ratios from Upper Protero-
zoic successions of Svalbard and east Greenland. Nature, 321,
832– 837.
Krasil’shchikov, A. A. 1973. The stratigraphy and paleotectonics of the
Precambrian – Early Paleozoic of Spitsbergen (translated title). Trudy
Arkticheskogo Nauchno-Issledovatel’skogo Instituta, 172, 1 –120.
Kulling, O. 1934. The Hecla Hoek Formation round Hinlopenstredet.
Geografiska Annaler, 14, 161–253.
Larsen, V. B. 1987. A synthesis of tectonically related stratigraphy in the
North Atlantic – Arctic region from Aalenian to Cenomanian time.
Norsk Geologisk Tidsskrift, 67, 323– 338.
Lyberis, N. & Manby, G. 1993. The origin of the West Spitsbergen Fold
Belt from geological constraints and plate kinematics: implications
for the Arctic. Tectonophysics, 224, 371–391.
579
Lyberis, N. & Manby, G. 1999. Continental collision and lateral escape
deformation in the lower and upper crust: an example from Caledo-
nide Svalbard. Tectonics, 18, 40 – 63.
Maher, H. D. Jr. 2001. Manifestations of the Cretaceous High Arctic
Large Igneous Province in Svalbard. Journal of Geology, 109,
91– 104.
Maloof, A. C., Halverson, G. P., Kirschvink, J. L., Schrag, D. P.,
Weiss, B. P. & Hoffman, P. F. 2006. Combined paleomagnetic, iso-
topic and stratigraphic evidence for true polar wander from the Neo-
proterozoic Akademikerbreen Group, Svalbard, Norway. Geological
Society of America Bulletin, 118, 1099– 2014.
Manby, G. & Lyberis, N. 1995. Discussion on the Ny Friesland Orogen,
Spitsbergen. Geological Magazine, 132, 351– 356.
McCay, G. A., Prave, A. R., Alsop, G. I. & Fallick, A. E. 2006. Glacial
trinity: Neoproterozoic Earth history within the British –Irish Caledo-
nides. Geology, 34, 909– 912.
Nordenskiöld, A. E. 1863. Geografisk och geognostisk beskrifning over
noröstra delarna af Spetbergen och Hinlopen Strait [Geographic and
geognostic descriptions of the northeast part of Spitsbergen and
Hinlopen Straight]. Kungliga Svenska Vetenskapsakademiens Han-
dlingar, 4. Stockholm.
Nystuen, J. P. 1985. Facies and preservation of glaciogenic sequences
from the Varanger ice age in Scandinavia and other parts of the
North Atlantic region. Palaeogeography, Palaeoclimatology,
Palaeoecology, 51, 209–229.
Pisarevsky, S. A., Murphy, J. B., Cawood, P. A. & Collins, A. S. 2008.
Late Neoproterozoic and Early Cambrian palaeogeography: models
and problems. In: Pankhurst, R. J., Trouw, R. A. J., de Brito
Neves, B. B. & de Wit, M. (eds) West Gondwana: Pre-Cenozoic
Correlations Across the South Atlantic Region. Geological Society,
London, Special Publications, 294, 9 –31.
Schermerhorn, L. J. G. 1974. Late Precambrian mixtites: glacial and/or
non-glacial. American Journal of Science, 274, 673–824.
Sokolov, V. N., Krasil’chov, A. A. & Livshits, YU. YA. 1968. The main
features of the tectonic structure of Spitsbergen. Geological Maga-
zine, 105, 95– 115.
Teben’kov, A. M., Ohta, Y., Balashov, J. A. & Sirotkin, A. N. 1996.
Newtontoppen granitoid rocks; their geology, chemistry and Rb –Sr
age. Polar Research, 15, 67– 80.
Weil, A. B., Van der Voo, R., Mac Niocaill, C. & Meert, J. G. 1998.
The Proterozoic supercontinent Rodinia: paleomagnetically-derived
reconstructions for 1100– 800 Ma. Earth and Planetary Science
Letters, 154, 13– 24.
Wilson, C. B. 1961. The upper Middle Hecla Hoek rocks of Ny Friesland,
Spitsbergen. Geological Magazine, 98, 89 – 116.
Wilson, C. B. & Harland, W. B. 1964. The Polarisbreen Series and other
evidences of late Pre-Cambrian ice ages in Spitsbergen. Geological
Magazine, 101, 198– 219.
Witt-Nilsson, P., Gee, D. G. & Hellman, F. J. 1998. Tectonostrati-
graphy of the Caledonian Atomfjella Antiform of the northern Ny
Friesland, Svalbard. Norsk Geologiske Tidsskrift, 78, 67 –80.
 Chapter 56

Neoproterozoic (Cryogenian –Ediacaran) deposits in East and North-East Greenland

SVEND STOUGE1*, JØRGEN LØYE CHRISTIANSEN2, DAVID A. T. HARPER1, MICHAEL HOUMARK-NIELSEN3,

KASPER KRISTIANSEN4, CONALL MACNIOCAILL5 & BJØRN BUCHARDT-WESTERGÅRD6
1
Geological Museum, University of Copenhagen, DK-1350 Copenhagen K, Denmark
2
University College of Zealand, DK-4300 Holbaek, Denmark
3
GeoGenetics Centre, Natural History Museum, University of Copenhagen, DK-1350 Copenhagen K, Denmark,
4
Maersk Oil and Gas, Denmark
5
Department of Earth Sciences, Oxford University, UK
6
Department of Geography and Geology, University of Copenhagen, DK-1350 Copenhagen K, Denmark
*Corresponding author (e-mail: svends@snm.ku.dk)

Abstract: The Neoproterozoic succession of East and North-East Greenland (over 14 000 m thick) includes the Eleonore Bay Super-
group (?Tonian– Cryogenian) and the Tillite Group (Cryogenian–Ediacaran). The upper units of the Eleonore Bay Supergroup
consist of shallow to deeper-water carbonates, succeeded by siliciclastic fine-grained sediments (Bedgroup 19) that characterize the
top unit of the supergroup.
The Tillite Group includes two diamictite-bearing units (Ulvesø and Storeelv formations) of glaciogenic origin and two upper,
upwards-shallowing strata (Canyon and Spiral Creek formations) that were deposited during semiarid conditions and concluded the
Neoproterozoic depositional cycle. Diamictite is preserved on the craton and compares with the Storeelv Formation (Fm.) of the
Tillite Group.
Detailed investigations of the diamictite-bearing units (i.e. Ulvesø and Storeelv formations) demonstrate that the lower of the two for-
mations is mainly of marine origin, whereas the upper one has both marine and terrestrial origins. Chemostratigraphic data include ana-
lyses on total carbon (TC), total organic carbon (TOC), total sulphur (TS) and d13C. The data set for d13C shows a substantial and abrupt
shift towards negative values of 10%, from below Bedgroup 19. Low-diversity acritarch assemblages (Cryogenian) are recorded from
the Andrée Land and Tillite groups; a thin cherty dolostone unit present above the Storeelv Fm. suggests that the diamictite units are of
late Cryogenian age and the upper part of the Tillite Group is Ediacaran.
Bedgroup 19 disconformably overlies older carbonates and the unit is a prelude to the succeeding (upper Cryogenian– lower
Ediacaran) diamictite sediments of the Tillite Group. A disconformity separates the Tillite Group from the overlying Lower
Palaeozoic sediments. Both disconformities are, according to palaeomagnetic data, related to rift–drift episodes that occurred during
the late Neoproterozoic. Alternatively, the isotope data suggest that the diamictites were deposited during the late Cryogenian glaciation
and the older disconformity may be interpreted as a significant gap developed by the lowering of sea level during an early
Cryogenian glaciation.

The Neoproterozoic sediments in East and North-East Greenland
are well preserved and, despite their remote location, the area is
well known and has served as key area for the study of Neoproter-
ozoic successions in the Arctic (Figs 56.1 & 56.2). It should be
noted that this region was previously referred to as East Greenland
in the literature but is now referred to as East and North-East
Greenland according to new administrative boundaries.
Two diamictite bearing units (Ulvesø and Storeelv formations,
formerly referred to as lower and upper tillite respectively) are
thought to record glacial conditions in the region during this
time period. Other diamictite units occur in para-autochthonous
windows and are broadly correlated with the Storeelv Fm. The
Ulvesø and Storeelv formations are part of the Tillite Group
(Haller 1971; Henriksen & Higgins 1976; Hambrey & Spencer
1987), which Kulling (1929) initially referred to as the Tillite
Series. Even though the name does not follow modern guidelines
for stratigraphic terminology, it has been used continuously in
the literature since 1929, and it is retained here following the
recommendation of Hambrey & Spencer (1987, p. 6). The Tillite
Group is exposed in a north – south-trending belt extending
through the central fjord region from Canning Land in the south
(728250 N) to Payer Land in the north (748200 N). The exposure is
excellent in many places, especially along the coastlines of
the fjords.
The Ulvesø Fm. was formally defined by Hambrey & Spencer
(1987) with its type locality in Kløftelv, c. 200 m north of
Ulvesø on Ella Ø (728520 N, 258050 W) and with a coastal section
beginning in Tømmerbugt and extending into Bastion Bugt,
serving as the reference section (Figs 56.3 & 56.4). It formed
part of the Cape Oswald Fm. of Poulsen (1930), and Schaub
(1950) placed the unit in his Tømmerbugt Group. The Storeelv
Fm. takes its name from Storeelv on Ella Ø (728510 N, 258070 W)
(Fig. 56.1).
Systematic research of the upper Neoproterozoic sediments
began in the mid-1920s and continued in the 1950s, when several
expeditions visited the area under the leadership of Lauge Koch
(Koch 1929, 1930; Koch & Haller 1971; for a summary see
Haller 1971 and Henriksen & Higgins 2008). At that time, ship
and aircraft support had become widely available, which made it
possible to visit these remote parts of East and North-East
Greenland. Stratigraphical, sedimentological and facies studies
from the mid-1970s onwards were undertaken by the Geological
Survey of Greenland as part of its general mapping programme
(Henriksen & Higgins 1976; Bengård 1991; Henriksen 1999,
2003; Henriksen & Higgins 2008). Geologists from Harvard and
Cambridge University participated in expeditions to the region
in the 1970s and 1980s, which resulted in several new stratigraphi-
cal and sedimentological findings (Hambrey 1983, 1989; Hambrey
& Spencer 1987; Moncrieff & Hambrey 1988, 1990; Hambrey et al.
1989; Herrington & Fairchild 1989; Manby & Hambrey 1989;
Swett & Knoll 1989) and descriptions of microfossils assemblages
(Knoll et al. 1986; Green et al. 1987, 1988, 1989). The present

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 581– 592. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.56
582 S. STOUGE ET AL.
lithostratigraphic scheme for the upper Neoproterozoic deposits in
North-East Greenland (Hambrey & Spencer 1987) appeared at
around the same time, when Knoll et al. (1986) published their
geochemistry-driven data for the broadly correlative successions
on Spitsbergen. More recent research on stratigraphy, sedimentol-
ogy, micropalaeontology, palaeomagnetism and geochemistry
started in the late 1990s (Frederiksen et al. 1999; Frederiksen
2000) and continues now into the 21st century (Stouge et al.
2001; Sønderholm et al. 2008).
Fig. 56.1. Location map showing positions
of exposures, and the structure and geology
of East and North-East Greenland.
(Modified from Frederiksen 2000 and
Higgins et al. 2004). Used with permission
from the National Geological Survey of
Denmark and Greenland (GEUS) #.
Structural framework
Most of the Neoproterozoic deposits of East and North-East
Greenland accumulated in the Eleonore Bay basin that existed
for c. 350 Ma. This sedimentary basin was nearly 500 km long,
NE –SW trending and 200– 400 km wide, and a pile of c. 14–
15 km of sediments accumulated (Sønderholm et al. 2008).
Eleonore Bay Basin sediments, preserved today in the Franz
Joseph Allochton of the Caledonian orogen were deposited

Era Period Supergroup Group
Palaeozoic Cambrian (not named) Kong Oscar Fjord
Ediacarian
Spiral Creek
Tillite
Neoproterozoic
Eleonore Bay
Group
Cryogenian
Bedgroup 20
Bedgroup 19
Andrée Land
Bedgroup 18
c. 200– 400 km away, farther to the east (Higgins & Leslie 2000,
2008) than the present day. The Neoproterozoic succession in
the basin is thought to record a rift –drift transition associated
with the fragmentation of Rodinia (Herrington & Fairchild 1989;
Stouge et al. 2001; Eyles & Januszczak 2004; Nystuen et al.
2008; Sønderholm et al. 2008), although volcanic deposits typi-
cally associated with rifted margins have not been reported from
the succession (Soper 1994; Higgins & Leslie 2008; Sønderholm
et al. 2008).
This Neoproterozoic sedimentary pile is incorporated in the
Caledonian mountain belts of eastern Greenland, extending from
708 to 828N and flanking the eastern edge of the Greenland ice sheet
(Henriksen & Higgins 1976, 2008; Henriksen 1999; Higgins et al.
2004; Gee et al. 2008; Higgins & Leslie 2008). The outcrop of Cale-
donian bedrock in eastern Greenland (Fig. 56.1) is c. 300 km wide
in the south, at 708N, and narrows northwards to c. 100 km, striking
obliquely offshore into the northern Greenland continental shelf. It
forms the western part of the Caledonian Orogen dominated by
thrust sheets, which were emplaced west-northwestwards onto
the Laurentian Craton with its Lower Palaeozoic cover.
The allochthons are composed of two major thrust sheets, the
Niggli Spids thrust sheet and the Hagar Bjerg thrust sheet (Higgins
et al. 2001, 2004; Higgins & Leslie 2008) (Fig. 56.1). The lower
Niggli Spids thrust sheet comprises Archaean and Palaeoproterozoic
basement gneisses overlain by upper Mesoproterozoic to lower
Neoproterozoic metasediments or the Krummedal supracrustal
sequence, which is several kilometres thick. The Hagar Bjerg
thrust sheet overlies the Niggle Spids thrust sheet; it includes the
Krummedal supracrustal sequence intruded by c. 950–920 Ma gran-
ites (Kalsbeek et al. 2001, 2008). These are overlain by the
c. 18.5-km-thick Neoproterozoic siliciclastic and carbonate succes-
sions (Eleonore Bay Supergroup), siliciclastic sediments with dia-
mictite (Tillite Group), and Cambrian–Middle Ordovician
siliciclastic and platform carbonate formations (Kong Oscar Fjord
Group). This sedimentary package comprises the upper part of the
Hagar Bjerg thrust sheet and is recognized as the Franz Joseph
allochthon (Higgins et al. 2004; Higgins & Leslie 2008).
All the transported rocks in eastern Greenland were derived
from the Laurentian margin, including both Archaean and Palaeo-
proterozoic crystalline basement and the younger, shallow-marine
sedimentary cover. WNW-directed thrust transport of the alloch-
thons has been estimated to be at least 200 km (Higgins & Leslie
2000, 2008), along with sinistral strike displacements; hence,
prior to the Caledonian Orogeny, the platform margin of Laurentia
was nearly twice as wide as it is today.
NEOPROTEROZOIC NE GREENLAND 583
Formation Sequence Ma Events
Kløftelv Sauk Iapetus passive margin
542
Hiatus Uplift
580
Canyon III
635
Storeelv
Maximal separation Fig. 56.2. Stratigraphical nomenclature
SQB used in this chapter and summary of the
Arena
series of events that occurred in the
II
Eleonore Bay basin during the Cryogenian
Ulvesø
Drift and Ediacaran. The succession comprises
SQB
three stratigraphic sequences, and the hiatus
Final carbonate deposition
between the Tillite Group and the Lower
I
Foundering of Palaeozoic Kløftelv Fm. is extensive. No
Rifting
carbonate platform radiometric ages are available in this
SQB
succession; ages are inferred based on
Carbonate platform interpretation of the stratigraphic and
biostratigraphic record. See text
for discussion.
The upper part of the Andrée Land Group is of special interest in
this context, and together with the Tillite Group crop out over an
area of c. 200 km from north to south and c. 50 km in an east –
west direction (Fig. 56.1). The outcrop area is bounded to the
west by the older sediments of the Eleonore Bay Supergroup and
to the east by the Lower Palaeozoic outcrops or by major north –
south-trending faults. The sediments are gently folded into syn-
forms and antiforms within the East and North-East Greenland
Caledonides and are only mildly affected by Caledonian meta-
morphism; the succession is locally jointed and faulted by
younger late Palaeozoic (Carboniferous) orogenic events.
Additional occurrences of upper Neoproterozoic sediments are
preserved in the Charcot Land, Gåseland and Målebjerg windows
in the lower allochthon and are para-autochthonous deposits
(Fig. 56.1; Wenk 1961; Phillips et al. 1973; Montcrieff 1989;
Smith & Robertson 1999; Higgins et al. 2001, 2004; Sønderholm
et al. 2008). The deposits are overlain by the Hagar Bjerg and the
Niggli Spids thrust sheets (Higgins et al. 2004, 2008; Fig. 56.1).
These sediments accumulated at least 200 km to the west of the
Eleonore Bay basin but are thought to be broadly correlative
with the Tillite Group (Higgins & Leslie 2000, 2008).
Stratigraphy
The allochthonous upper Neoproterozoic sediments in East and
North-East Greenland are collectively referred to as the Eleonore
Bay Supergroup (Andrée Land Group) and the overlying Tillite
Group (Haller 1971; Sønderholm & Tirsgaard 1993; Sønderholm
et al. 2008) (Fig. 56.2). Associated para-autochtonous deposits
are broadly correlated with the Storeelv Fm. of the Tillite Group
in the Eleonore Bay Basin (Montcrieff 1989).
Andrée Land Group
The top unit of the Eleonore Bay succession is the Andrée Land
Group (Teichert 1933; Tirsgaard 1993, 1996; Sønderholm & Tirs-
gaard 1993; Tirsgaard & Sønderholm 1997) (Fig. 56.2), which
consists primarily of shallow to relatively deeper-water marine
carbonates succeeded by marine deep-water siltstone-shale,
cherty rhythmites and carbonate. It totals c. 1500 m in thickness
and was recognized previously as the ‘Die Kalk-Dolomit Serie’
(i.e. ‘Limestone-Dolomite Series’; Teichert 1933), which was sub-
divided into a number of ‘bed groups’ (Teichert 1933; Schaub
584 S. STOUGE ET AL.
1950; Katz 1952; Eha 1953; Fränkl 1953; Haller 1953, 1971;
Cowie & Adams 1957; Sommer 1957). These informal bed
groups have since been widely used as standard reference nomen-
clature in the literature for the Eleonore Bay Supergroup (Haller
1971; Hambrey & Spencer 1987; Herrington & Fairchild 1989;
Swett & Knoll 1989; Fairchild & Hambrey 1995).
The Andrée Land Group was formally established by Sønder-
holm & Tirsgaard (1993), who also introduced informal litho-units
(AL1 –7), which should serve as replacements for the bedgroups.
However, the older stratigraphic terminology summarized by
Haller (1971) is still applicable and is used here as stratigraphical
framework (Fig. 56.2).
The upper Andrée Land Group, named Bedgroup 18 (¼ AL5
of Sønderholm & Tirsgaard 1993), is composed of marine,
Fig. 56.3. Stratigraphy of the uppermost
Andrée Land Group (Bedgroups 19 and 20)
and Tillite Group. The sedimentary logs
from Andrée Land and Ella Ø represent the
northern and southern development of the
succession respectively. AS, sandstone
(aeolian) unit forming at the top of
Bedgroup 20; CD, cap dolomite following
above the diamictite of the Storeelv Fm.
shallow-water carbonates that accumulated on a ramp situated
along a former low-latitude passive continental margin (Frederik-
sen et al. 1999; Frederiksen 2000; Stouge et al. 2001; Mac Niocaill
et al. 2004, 2008; Sønderholm et al. 2008). The uppermost strata of
the Andrée Land Group are referred to as Bedgroup 19 and Bed-
group 20 (¼ AL6 and AL7 of Sønderholm & Tirsgaard 1993).
These are up to c. 400 m thick and consist of fine-grained clastic
sediments and carbonate.
Tillite Group
The upper part of the Neoproterozoic succession comprises the
siliciclastic deposits and minor carbonates of the Tillite Group
NEOPROTEROZOIC NE GREENLAND 585

Fig. 56.4. Detailed stratigraphic logs of the

Ulvesø and Storeelv formations. The Ulvesø
Formation is from the type section at Ulvesø
in Kløftelv on Ella Ø and the reference
section from Tømmerbugt til Bastion Bugt,
Ella Ø. Stratigraphic log of the Storeelv
Formation is from the coast section in
Bastion Bugt, Ella Ø.
586 S. STOUGE ET AL.
(c. 900 m), representing terrestrial and marine depositional settings
in five different formations (Figs 56.2 & 56.3). The marine depos-
its accumulated at some distance from the continental margin in a
deep-water, oceanic setting. The diamictite facies of the Ulvesø
and Storeelv formations are separated by mudstone and sandstone
facies of the Arena Fm. and succeeded, first by deep-water sedi-
ments and then by shallowing-upwards successions, where the
uppermost unit concludes with evaporitic environments (Canyon
and Spiral Creek formations; Fig. 56.2).
A regional synthesis of the depositional environments and
stratigraphical nomenclature was reviewed by Haller (1971) and
revisions were published by Cowie & Adams (1957), Hambrey
& Spencer (1987), Fairchild & Hambrey (1995), Sønderholm &
Tirsgaard (1993) and Sønderholm et al. (2008). The switch from
deposition of carbonate to deep-water, mostly fine-clastic, turbi-
dites on the slope in the upper part of the Andrée Land Group her-
alded the deposition of the Cryogenian –Ediacaran diamictite units
in the region.
The Eleonore Bay Supergroup disconformably overlies the
Krummedal supracrustal succession (Higgins et al. 2004, 2008;
Nystuen et al. 2008). Granites, which are between 950 and
920 Ma old (Kalsbeek et al. 2001, 2008), intrude the Krummedal
supracrustal sequence, and the c. 900 Ma date for the granites
may represent a maximum age for the initial start of deposition
in the Eleonore Bay Basin. The late Neoproterozoic succession
is unconformably overlain by the quartzitic sandstone of the Kløf-
telv Fm. (Lower Cambrian) of the Kong Oscar Fjord Group in the
central fjord zone (Fig. 56.2).
Associated para-autochthonous strata
Diamictite beds have been recorded from Charcot Land (Mon-
tcrieff 1989), Gåseland (Wenk 1961; Phillips et al. 1973) and Mål-
ebjerg windows (Smith & Robertson 1999) (Fig. 56.1). The
diamictite deposits are preserved in pockets on the surface of
the basement and below the higher thrusts. The Støvfanget Fm.
(Montcrieff 1989) is named from the succession in the Gåseland
window. The strata from the Charcot Land and Målebjerg
windows were named the Tillite Nunatak Fm. (Montcrieff 1989;
Smith & Robertson 1999). The sediments are considered to be
lateral equivalents, so the name Støvfanget Fm. is used here for
these deposits.
Glaciogenic deposits and associated strata
Bedgroup 19, Andrée Land Group
Bedgroup 19 is characterized by deep-water black, green, grey and
red shales, interbedded with chert, dolostone, minor limestone,
interbedded limestone and shale and carbonate breccias and con-
glomerates (Fig. 56.2). Bedgroup 19 is overlain either by Bedgroup
20 or the Ulvesø Fm. of the Tillite Group (Sønderholm et al. 2008).
The lateral distribution of facies is heterogeneous and the succes-
sion shows a northern and a southern development (Fig. 56.3). Bed
group 19 is here subdivided into two informal units: Unit A and
Unit B. Unit A is well developed throughout the basin, whereas
Unit B is only found in the southern portion of the basin.
Unit A is the stratigraphically oldest strata of Bedgroup 19 of the
upper Andrée Land Group (Sønderholm et al. 2008). It is 200 m or
more thick, and is characterized by mixed chert and fine-grained
to very fine-grained siliclastic sediments associated with minor
carbonate. Lithologies include argillite (shale); siltstone; bedded
light grey, dense thin-bedded chert, developed as rhythmites;
and siliceous argillite interbedded with variable amounts of thin-
bedded and very fine-grained grey, green to violet chert. Yellow
to yellow-brown weathering is characteristic of the finely lami-
nated grey to green to violet chert.
In detail, Unit A begins with quartzitic, fine-grained sandstone
overlying the black limestone of Bedgroup 18. The quartzitic sand-
stone is succeeded by bedded chert and rhythmically bedded grey
chert. The chert beds are 1–5 cm thick and have millimetre-scale
partings of grey-green shale. This unit is succeeded by rhythmi-
cally bedded chert with partings consisting of 0.1 –0.5 mm thin
green-grey shale as well as interbeds of grey-red-brown shale over-
lain by finely laminated chert and grey shale. The chert is charac-
terized by light yellow to brown weathering surfaces. Besides
occasional ripples, sedimentary structures are rare to absent in
the unit.
The basal quartzitic unit of Unit A is sparsely exposed, so it has
not been reported previously. It is well exposed in the region from
Andrée Land and east of Eleonore Bay and towards the north
(Figs 56.1 & 56.3). On Ella Ø, Schaub (1950) and Hambrey &
Spencer (1987) described Unit A in detail and divided it into
several beds, but Unit A has also an extensive lateral distribution.
It is recorded across the whole basin and with a nearly uniform
development. Many of the individual beds can be traced over
long distances; one prominent chert/carbonate horizon with
shale partings, 7 to 10 m thick, can easily be traced laterally
from Lyell Land to Ella Ø to Ole Rømer Land and farther to the
north (i.e. for more than 200 km); it can thus be used as a
marker bed across the whole basin.
Unit B is composed mainly of grey and black relatively
organic-rich shale that is interbedded with subordinate parted to
ribbon limestone and limestone breccia. The ribbon and parted
limestone forms the base of the unit and is very fine-grained,
dark grey to black, finely laminated and with limestone beds
varying in thickness from 4 to 40 cm. Several horizons of the
ribbon and parted limestone are slumped and contorted. Above
the interval with parted to ribbon limestone and breccia beds
follows a unit up to 100 m thick, dark grey to black shale.
Unit B is recorded only in the southern extension of the basin
(Sønderholm et al. 2008).
The limestone breccias of Unit B have a chaotic fabric with a
matrix composed of grey to black, dirty green to dull brown,
purple-to-yellow weathering calcareous mud and lime mudstone.
Purple-to-yellow weathering carbonate and elongated plates of
low lithological variety are the most common clasts. They are com-
posed of grey, blue-grey, to dark grey limestone and dolomite
plates. The lime breccias are mostly thin but well exposed. On
Ella Ø, one breccia horizon up to 10 m thick serves as a local
lithological marker band in this region (Hambrey & Spencer
1987, fig. 6; Herrington & Fairchild 1989).
Bedgroup 20, Andrée Land Group
Bedgroup 20 (AL7 as per Sønderholm & Tirsgård 1993) is mainly
a carbonate unit (Katz 1952; Eha 1953; Fränkl 1953; Haller
1953; Hambrey & Spencer 1987; Herrington & Fairchild 1989;
Sønderström et al. 2008), which conformably overlies Unit A of
Bedgroup 19. Bedgroup 20 is typically composed of (i) silty,
planar bedded ribbon limestone at the base, (ii) lime mudstone
and bedded limestone in the main part of the unit (Fig. 56.3) and
(iii) planar lamellites associated with small domal stromatolites
and minor nodular cherts. Oolitic-pisolitic limestone, up to 40 m
thick, prevails in the higher part of the unit on Ole Rømers Land
(Swett & Knoll 1989; Frederiksen 2000).
Bedgroup 20 varies in thickness from 240 m in Ole Rømer Land
and decreases towards the south where 2 m are recorded on the
southern part of Ymer Ø. The unit extends northwards as far as
Hudson Land. Farther to the south and towards the southern exten-
sion of the Eleonore Bay Basin, the carbonates of Bedgroup
20 are absent.
The uppermost part of Bedgroup 20 is composed of yellow,
medium-grained sandstone with well-developed and large cross-
beds (Fig. 56.3). This sandstone subunit (c. 40 m) is shown in
 NEOPROTEROZOIC NE GREENLAND
Fig. 56.3 as ‘aeolian sandstone’ (AS). This subunit displays the
same geographical distribution as Bedgroup 20 (Sønderholm
et al. 2008).
Ulvesø Fm., Tillite Group
At Ella Ø, the Ulvesø Fm. comprises a generally coarsening
upwards succession composed of diamictite, associated with
minor laminite and shale, massive sandstone and conglomerate
(Figs 56.3 & 56.4). The crudely bedded diamictite is composed
predominantly of carbonate clasts derived mainly from the older
Andrée Land Group and sediments of Bedgroup 19 in the southern
part of the basin, and from carbonate units of the older Andrée
Land Group and Bedgroup 20 to the north of Ymer Ø. Beds are
normally 30–50 cm thick, and clasts are occasionally up to 1 m
in size. The matrix is grey-green and composed of shale, silt and
sand, which weather characteristically as dull yellow to brown.
The formation varies laterally in thickness from nearly 10 m
in the north to a maximum thickness of c. 320 m on northern
Scoresby Land.
The erosive-based diamictite is rich in striated outsized clasts
up to a metre in size and slump folds and mega breccias
occur frequently throughout the succession. Towards the top, dia-
mictite interfingers with erosive-based sandstone and clast-
supported conglomerate, beds of shale and brecciated limestone
derived from the breakdown of the underlying carbonate-shale
platform. Together with the lowermost part of the sandstones of
the overlying Arena Fm., the diamictite of the Ulvesø Fm.
appears strongly deformed, occurring as metre-sized ball and
pillow structures.
Arena Fm., Tillite Group
The Arena Fm. is composed of grey, dark-grey to green shale,
siltstone and sandstone (Figs 56.2 & 56.3). The formation was
formally defined by Hambrey & Spencer (1987), with its type
locality at Arenaen on Gunnar Andersson Land, Ymer Ø
(738200 N, 248500 W). The formation was previously included in
the Tillite Series (Kulling 1929; Schaub 1955; Katz 1954, 1961)
and the Cape Oswald Fm. of Poulsen (1930). Later it was named
the Inter Tillite Member (Schaub 1950) or Inter-Tillite (Haller
1971; Henriksen & Higgins 1976; Higgins 1981).
The formation varies from c. 100 m in thickness in Strindberg
Land and Ole Rømer Land, through 220– 360 m in thickness in
Albert Heim Bjerge (Hudson Land). The formation begins with
shale and silt beds followed by thick-bedded green to grey sand-
stone. The shale and siltstone are the dominant lithologies, with
a minor increase in the amount of sandstone occurring towards
the top of the Arena Fm. The recessive Arena Fm. separates the
diamictite units and is not easy to access in inland outcrops, and
the beds of the unit are commonly covered.
On Ella Ø, the lowest part of the unit is composed of yellow
sandstone, which is similar to the sandstone on Scoresby
Land. The latter includes thin diamictite horizons (Hambrey &
Spencer 1987). Sedimentary structures in the Arena Fm. comprise
cross-lamination, wave ripples, graded bedding, load-casts and
slump structures.
Storeelv Fm., Tillite Group
The Storeelv Fm. ranges in thickness from 220 m on Ella Ø
(Fig. 56.3) and Arenaen to 60 m in Strindberg Land in the north
and Canning Land in the south. On Ella Ø, the formation conform-
ably rests on a striated glacial boulder pavement of outsized
far travelled erratics. Crudely bedded reddish diamictite with
clasts of local and extra-basinal origin, including igneous and
metamorphic rocks of the Archaean basement, occasionally
587
alternate with breccias of limestone and sandstone. More com-
monly, they interfinger with cross-bedded sandstone and laminated
shale with outsized clasts (Figs 56.3 & 56.4). Conglomerate-filled
mega-channels truncate the diamictite, mainly in the middle
part of the formation. Towards the top, diamictite units once
more dominate the succession. However, here it interfingers
with minor beds of laminated shale with dropstones and
bedded sandstone.
Canyon Fm., Tillite Group
The Canyon Fm. was first named the Canyon Series by Schaub
(1955) and Fränkl (1953). Later, Haller (1971) modified it to the
Canyon Fm. The name is derived from Tillite Canyon Fm.
(Poulsen 1930), but the name Tillite was dropped by subsequent
authors.
In the northern part of the basin, this formation includes a cherty
dolostone horizon at the base. The cherty carbonate unit, up to
10 m thick, is situated at the base of the formation and extends
laterally from Andrée Land (Tillite Kløft and west of Kap
Weber) and northwards to Albert Heim Bjerge (Cowie & Adams
1957). The unit is composed of light grey, dense and finely lami-
nated chert or siliceous dolostone. Above follows a rhythmically
bedded shale and chert. The shale is commonly very red, and the
chert beds are pale grey to nearly white, so the shale/chert unit
is easy to recognize from a distance. This cherty carbonate
unit is not present on Ella Ø, and red shale follows above the
Storeelv diamictite. The characteristic, rhythmically bedded, red
shale and grey finely laminated chert appear c. 10 m above the
Storeelv Fm.
The subsequent and main part of the succession of the Canyon
Fm. is composed of red, grey green and black shale. Towards
the top and on Ella Ø, the Canyon Fm. becomes increasingly
sandy and the sandstone and shale display well-developed,
upwards coarsening sequences.
On Ella Ø, the top of the formation is developed as c. 50 m
yellow weathering dolostone with small stromatolitic reefs and
algal laminations. This facies of the Canyon Fm. is only seen on
Ella Ø. Elsewhere, black shale with prominent yellow-weathering
surfaces marks the top of the formation.
Spiral Creek Fm., Tillite Group
Spiral Creek Fm. is the topmost unit of the Neoproterozoic
succession in East and North-East Greenland (Fig. 56.2). The for-
mation is named after Spiral Creek, a locality on the northern shore
of Andrée Land (Poulsen 1930, p. 307), where the unit is well dis-
played. It is also known from Lyell Land, Ella Ø and Ole Rømer
Land. Spiral Creek Fm. is up to 45 m thick and is composed of silt-
stone, sandstone and stromatolitic dolostone. The colours of the
sediments vary from yellow, grey to green and maroon. The for-
mation is divided into several beds or horizons (Poulsen 1930;
Schaub 1950; Cowie & Adams 1957). Halite pseudomorphs,
mudcracks, ripples and intraformational breccia are frequent in
the beds. The sediments probably deposited in a playa-like
setting and under semiarid conditions (Hambrey & Spencer
1987; Fairchild & Hambrey 1995).
Støvfanget Fm., para-autochthonous deposits
The sediments of the Støvfanget Fm. include diamictite, sandstone
and laminated mudstone with dropstone (Smith et al. 2004;
Sønderholm et al. 2008). The composition of the clasts is hetero-
geneous and derived from local sources; granitic blocks are
common. The blocks vary in size and are up to 2 m, rarely 6 m,
in size. The succession varies from a couple of tens of metres to
a maximum of c. 200 m in thickness.
588 S. STOUGE ET AL.
Boundary relations with overlying and underlying
non-glacial units
Boundary relations within the Neoproterozoic succession
The lower boundary of Bedgroup 19 is not well exposed in the
southern part of its extent of outcrop. The boundary relationships,
however, are well displayed in Andrée Land and Strindberg Land,
where the boundary is a disconformity. The carbonate top surface
of Bedgroup 18 is dissolved and has karst-like features (Frederik-
sen 2000; Sønderholm et al. 2008). This carbonate surface is
overlain by a thin, quartzitic and medium-grained sandstone.
The quartzitic bedded sandstone is succeeded by bedded fine-
grained to grey chert deposits with thin shale partings. The lower
boundary of Bedgroup 20 is conformable, with a gradual transition
from dark grey to black shale, through interbedded silty laminated
limestone to bedded limestone.
The base of the overlying Tillite Group is usually sharp in the
deep part of the basin, where black shale (Unit B) of Bedgroup
19 is overlain directly by the Ulvesø Formation diamictite
(Hambrey & Spencer 1987). The boundary towards the north is
different and the top carbonate strata of Bedgroup 20 are separated
from the diamictite of the Ulvesø Fm. by a yellow (aeolian), quart-
zitic sandstone. This sandstone is conformably overlain by the
Ulvesø Fm. of the Tillite Group. On Ymer Ø, Bedgroup 20 is in
faulted contact with the overlying Ulvesø Fm. and boundary
relationships are obscure (Hambrey & Spencer 1987; Moncrieff
& Hambrey 1988).
The upper boundary of the Ulvesø Fm. is in most places
gradual and is marked by a transition from dark grey diamictite,
conglomerate or sandstone to the lower sandstone of the Arena
Fm. On Ella Ø the base of the overlying Arena Fm. is sharp and
strongly deformed with the underlying Ulvesø into metre-sized
load-casts.
The upper Neoproterozoic para-authochthonous sediments rest
directly on basement rocks and are tectonically constrained
upwards by thrust sheets; however, the diamictite in the Målebjerg
window is overlain disconformably by Lower Cambrian quartzitic
sandstone referred to as the Slottet Fm. (Smith et al. 2004).
Boundary relations with the Lower Palaeozoic succession
The Lower Cambrian Kløftelv Fm. of the Kong Oscar Fjord Group
(Lower Cambrian to Middle Ordovician) unconformably overlies
the Tillite Group across the whole Eleonore Bay basin and the
boundary is a low-angle angular unconformity (Henriksen &
Higgins 1976; Hambrey 1989; Hambrey et al. 1989; Stouge
et al. 2001, 2002; Sønderholm et al. 2008). Thus the boundary
relationships from the north to south are variably developed.
In Andrée Land and on Ella Ø, the upper Neoproterozoic eva-
poritic sediments of Spiral Creek Fm. are overlain unconfomably
by the Lower Cambrian Kløftelv Fm., and the Lower Palaeozoic
succession begins with a basal conglomerate (Stouge et al. 2001,
2002). Away from Andrée Land and Ella Ø, the Spiral Creek
Fm. and the upper stromatolitic limestone facies of the Canyon
Fm. are absent and black shales of Canyon Fm. are directly over-
lain by a basal conglomerate followed by the arenitic sandstone
of the Lower Cambrian Klöftelv Fm. The hiatus separating the
Neoproterozoic from the Palaeozoic strata is proposed to have
lasted for about 35 Ma (see Sønderholm et al. 2008 for further
discussion).
Chemostratigraphy
Early studies of carbonate and organic carbon geochemistry were
carried out by Knoll et al. (1986). Samples from Bedgroup 7
through 20 (Eleonore Bay Supergroup) as well as from the
Canyon and Spiral Creek formations (Tillite Group) revealed nega-
tive d13Corg values between –20 and – 35‰PDB. The d13Ccarb
values were predominantly positive throughout the Eleonore Bay
Supergroup with a sharp decrease to negative values ( –4 to
–9‰PBD) in the uppermost Bedgroup 19/20 and negative
values (c. 0 to –5‰PDB) in the Canyon and Spiral Creek for-
mations. Both geochemical data sets followed very similar profiles
suggesting a primary signature and an unusual secular variation in
the Neoproterozoic carbon cycle (Knoll et al. 1986).
Total carbon (TC), total organic carbon (TOC) and total sulphur
(TS) have been recorded through an interval ranging from the
upper part of the Andrée Land Group and up to and including
the top of Tillite Group; the results will be presented elsewhere.
Stable C-isotopes from the upper part of the Andrée Land Group
have also been analysed and will be presented elsewhere (see
Halverson et al. 2005; Sønderholm et al. 2008 for further details).
Other characteristics
Organic constructed organisms (acritarchs) in the succession
represent the plankton of the Andrée Land –Tillite Group basin.
The microflora assemblage is of low diversity and consists of
both small and large smooth leosperoid types associated with
only few acanthomorphs (Vidal 1976, 1979, 1985). The biostrati-
graphic value of the acritarch assemblage is at present limited, but
in general the microflora from Andrée Land Group is assigned to
the Cryogenian (Vidal 1976, 1979; Green et al. 1987, 1988,
1989). Other microfossils reported from the succession are vase-
shaped protists (Green et al. 1988), similar to those reported in
other Neoproterozoic successions (Porter & Knoll 2000; Porter
et al. 2003).
Palaeolatitude and palaeogeography
Palaeogeographical reconstructions of East and North-East Green-
land have long been tied to the palaeogeographical evolution of the
Neoproterozoic succession in northeastern Svalbard based on
detailed lithostratigraphic similarity between these two basins
(Harland & Gayer 1972; Fairchild & Hambrey 1995; Halverson
et al. 2004; Sønderholm et al. 2008; Halverson 2011).
The Hekla Hoek succession in northeastern Spitsbergen is
similar to the Eleonore Bay Supergroup of northeastern Greenland
and shows development from siliciclastic to carbonate deposition
followed by glaciogenic deposits. This similarity is also strength-
ened by comparable and substantial thicknesses of the lithological
units in the two regions (Harland & Geyer 1972; Hambrey 1989;
Fairchild & Hambrey 1995; Harland 1997).
The chemostratigraphy shows similarities (Knoll et al. 1986)
suggesting close affinities between the two areas. Halverson
et al. (2005) identified the distinctive pre-Marinoan negative
d13C anomaly or the ‘Trezona anomaly’ beneath the lowest dia-
mictite in northeastern Svalbard. The same anomaly has been
recognized from within the upper unit of the Eleonore Bay Super-
group in East and North-East Greenland (Sønderholm 2008; Kris-
tiansen unpublished data), which also provides a strong argument
for the similarity of the two regions. Palynology studies in both
East and North-East Greenland and Svalbard (Vidal 1985; Green
et al. 1989; Swett & Knoll 1989) also indicate close relationships
between the two areas and as a whole the two regions are con-
sidered to be deposited within the same basin or series of ensialic
basins, with East and North-East Greenland located near eastern
Svalbard (Sønderholm et al. 2008).
Recent and ongoing palaeomagnetic studies have shown that
East and North-East Greenland was situated at equatorial latitudes
at about 38S during the Cryogenian, with palaeomagnetic data
from 23 sites (105 specimens) towards the top of the Eleonore
Bay Supergroup yielding a mean palaeomagnetic inclination
 NEOPROTEROZOIC NE GREENLAND
of 58 (k ¼ 12; a95 ¼ 98) (Mac Niocaill et al. 2004, 2008; Kilner
2005). These data are constrained by a positive fold test (Caledo-
nian folding) and a positive conglomerate test on clasts in the
overlying lower Tillite. At that time the carbonate sedimentary
succession of the Andrée Land Group was deposited in a platform
setting on a ramp. The region formed the eastern part of the north-
eastern margin of the Rodinia palaeocontinent. A change in the
latitudinal position began at the base of Bedgroup 19 as the conti-
nental margin region started to move southwards. During depo-
sition of the Storeelv Fm., the area was situated at c. 668
southern latitude, with sparse palaeomagnetic data from six sites
(nine specimens) in the Tillite Group yielding a mean palaeomag-
netic inclination of 788 (k ¼ 10; a95 ¼ 228) (Mac Niocaill et al.
2004, 2008; Kilner 2005), although this awaits confirmation with
further analyses in progress. The movement towards higher,
southern latitudes is associated with the shift from stable carbonate
accumulation of Andrée Land Group to the siliciclastic deposition
of Bedgroup 19 and the start of foundering of the platform. The
presence of Bedgroup 20 signifies that carbonate accumulation –
for a short while – returned to the region, but finally ceased in
the late Cryogenian.
Geochronological constraints
Radiometric ages have not been obtained from the Andrée Land
and Tillite groups. A Cryogenian age is confirmed based on the
biostratigraphic record (see previous section).
Discussion
Depositional setting
The stratigraphy of the upper Andrée Land Group has been con-
sidered to conform to a normal superposition succession, where
Bedgroup 20 overlies Bedgroup 19. However, Eha (1953) pro-
posed that the carbonate sediments of Bedgroup 19 (¼ base of
Unit B of this chapter) represented the lateral equivalents of Bed-
group 20. In this review, the upper part of Bedgroup 19, that is, the
sediments of Unit B and Bedgroup 20 (¼ AL7), are considered
coeval (Fig. 56.3), as this interpretation is supported by the sedi-
mentology, intragroup stratigraphy and the overall stratigraphical
succession developed during recent fieldwork. Unit B is con-
sidered the fine-grained and deep-water equivalent of Bedgroup
20, dominated by parted to ribbon limestone, debris-flow deposits,
lime-breccias containing intrabasinal clasts and distal turbiditic
shale at the basin margin and organic-rich shale in the basin.
Bedgroup 19 (Eha 1953; Fränkl 1953; Haller 1971) of the upper-
most Andrée Land Group significantly provides the first record of
fine-grained to very fine-grained siliciclastic sediment and chert
rhythmites (i.e. Unit A), which followed disconformably above
the shallow-water stromatolite-bearing limestone and slightly
deeper-water carbonate rocks. Limestone (Bedgroup 20) and
shale and ‘lime-breccias’ (i.e. Bedgroup 19, Unit B) underlie the
first diamictite unit (i.e. Ulvesø Fm.) of the Tillite Group. This suc-
cession is interpreted as recording deepening within the basin and
foundering of a carbonate platform that had developed on a stable
craton and had been in existence for a long period of time (i.e.
during deposition of the underlying Ymer Ø Group and most of
the Andrée Land Group). In addition, the rapid facies change
from carbonate accumulation to siliciclastic deposition shows
that rift-related subsidence was initiated in the Cryogenian, creat-
ing north –south elongated basins with isolated and carbonate-
capped platforms and deep-water troughs. Bedgroup 20 carbonates
probably developed a low-relief bank profile with mobile pisolite
shoals at the margin, and shallow sub-tidal carbonate and peritidal
deposits in the interior. The final carbonate deposition in the region
(i.e. before the occurrence of diamictite and represented by
589
Bedgroup 20) was due to a marked regression and perhaps also
due to a change in palaeo-climate.
The diamictites of the Tillite Group have long been considered
glaciogenic deposits (Kulling 1929; Poulsen 1930 and subsequent
authors). Hambrey & Spencer (1987) and Moncrieff & Hambrey
(1988, 1990) provided detailed models for glacial depositional
environments. Halverson et al. (2004, 2005) placed the Tillite
Group deposits within the scope of ‘Snowball Earth’.
The Ulvesø and Storeelv formations are indeed dominated by
diamictite, with megabreccias and sculptured clasts alternating
with minor quantities of shale, sandstone and conglomerate. A ter-
restrial origin can most probably be applied to the lowermost and
middle parts of the Storeelv Fm.; the facies successions here indi-
cate a highly diverse depositional system. Glaciers grounded on
the shelf or in fjords delivered till and glacio-fluvial deposits inter-
bedded with flow diamictites and lacustrine and aeolian facies.
Other parts of the Storeelv and most of the Ulvesø Fm. were
most probably deposited below the grounding line in sub-aqueous
environments. Consequently, the Tillite Group contains a limited
amount of tillite, and although most components in the succession
originated from glacial debris, most diamictites finally settled in
marine settings, especially those of the Ulvesø Fm. A marine
origin is supported by the occurrence of algal life forms throughout
the Tillite Group, except where sediments show signs of oxidation
(Stouge & Piasecki unpublished data). The diamictite units were
principally deposited from suspensions of glacial debris released
beneath floating ice shelves and by ice-rafting in a glaciomarine
environment. These deposits suffered slumping and reworking
by sediment gravity flows in a near shelf break environment.
Only parts of the Storeelv Fm. indicate that glaciers occasionally
reached grounding line or more rarely that terrestrial ice sheets
deposited tillite and glaciofluvial deposits. It is not clear how
much conglomerate and sandstone that originated in on-shore
fluvial and aeolian environments was actually reworked to even-
tually settle in submarine canyons and in current- and wave-
dominated shallow water.
Sequence stratigraphy
The upper part of the Andrée Land Group and the Tillite Group
represent three depositional sequences that mostly accumulated
as deeper- to deep-water sediments in oceanic settings (Fig. 56.2;
Sønderholm et al. 2008). The oldest sequence is represented by
Bedgroups 19 and 20. The lower sequence boundary is an erosional
or a dissolution surface. Siliciclastic silt, shale and chert were
deposited in the transgressive system tract during the progradation
of the coastline outside the study area. The maximum flooding
surface can be traced across the whole basin, and the following
highstand deposition created the regressive platform of Bedgroup
20 and the distal deposits of Unit B. The regressive top of
Bedgroup 20 is composed of shallow-water silt and sandstone
deposited in tidal flat environment and aeolian sandstones.
The second sequence comprises the Ulvesø and Arena for-
mations (Fairchild & Hambrey 1995: Sønderholm et al. 2008),
which are developed exclusively in clastic facies. The third
sequence comprises the transgressive Storeelv Fm. – including
the thin carbonate unit – and the Canyon Fm., which becomes
regressive upwards, permitting shallow-water carbonates to
accumulate. The overlying Spiral Creek Fm. represents the ulti-
mate part of the highstand, and the regressive top of the Tillite
Group accumulated in shoreface and semi-arid, evaporitic lagoo-
nal environments (Fairchild & Herrington 1989).
Timing of glaciation
Possibly two widespread or even global glaciations took place
during the late Neoproterozoic (Hambrey & Harland 1995; Knoll
590 S. STOUGE ET AL.
2000; Hoffman & Schrag 2002). It is, however, uncertain if the two
diamictite units in East and North-East Greenland represent one or
two separate glaciations and in what respect they correlate with
other Neoproterozoic phases of glaciations (e.g. Kennedy et al.
1988; Kaufman et al. 1997; Brasier & Shields 2000; Hoffmann
& Schrag 2002; Robb et al. 2004).
Halverson et al. (2004, 2005) suggested that the two diamictite
units on Svalbard represent one glaciation. This idea is based on
the d13C-isotope curve (Knoll et al. 1986; Halverson et al. 2004,
2005) and the presence of a distinctive d13C anomaly (¼‘Trezona
anomaly’) found beneath the lowest tillite unit in Svalbard. In
addition, the presence of the thin carbonate unit, which should
be equal to a ‘cap dolomite’ (Fairchild & Hambrey 1995), is
typical of the late Cryogenian glaciation. The diamictite units in
Svalbard were therefore related genetically to the late Cryogenian
glaciation (Marinoan) by Halverson et al. (2004, 2005).
Based on their stratigraphic similarity (mentioned previously),
the two diamictite units in East and North-East Greenland may
also contain tillite deposits of the Marinoan (late Cryogenian) gla-
ciation. The trend of the d13C-isotope curve (Knoll et al. 1986;
Kristiansen et al. unpublished data) from the upper part of the
Andrée land Group carbonates, combined with the presence of a
possible cap dolomite above the Storeelv Formation in the northern
part of the basin, suggests that the two diamictite units in East and
North-East Greenland could be related to one glaciation.
A problem with this interpretation arises, however, because
any trace of an older Cryogenian glaciation often found in other
Neoproterozoic successions is apparently missing in the Greenland
succession. Although no diamictite-bearing units occur below the
Ulvesø Fm., the extensive disconformity developed at the bound-
ary between Bedgroup 18 and Bedgroup 19 may be the response to
a global or eustatic lowering of sea level causing subaerial
exposure and subsequent dissolution of the shallow marine sea
bed in the region. The subsequent deepening (see ‘Sequence stra-
tigraphy’ section) could thus be a result of the rapid melting of the
ice sheet producing fine- to very fine-grained sedimentation.
Palaeogeographical setting
The transition from the upper Andrée Land Group to Bedgroup 19
and further to the Tillite Group has also been interpreted as depo-
sition in an early rift basin related to the splitting of Rodinia and
before the opening phase of the Iapetus Ocean (Fig. 56.2; Herring-
ton & Fairchild 1989; Stouge et al. 2001; Eyles & Januszczak
2004; Nystuen et al. 2008; Sønderholm et al. 2008). According
to the interpretation of Eyles & Januszczak (2004), the upper
part of the Andrée Land Group and the main part of the Tillite
Group accumulated during active rifting, and deep-water sedi-
ments accumulated in oceanic settings bordered by glaciated con-
tinents. Upwards, the uppermost part of the Andrée Land Group
and the top of the Tillite Group became shallower and the regres-
sive top of the Andrée Land Group is marked by a prominent
dissolution surface, which is overlain by marine sandstone.
The palaeomagnetic signal supports rifting and the development
of an extension-related basin and the rapid drift of the region from
low to higher southern latitudes (Mac Niocaill et al. 2004, 2008). A
Cryogenian age for the onset of rifting as potentially recorded in
the East and North-East Greenland succession is earlier than that
initially proposed for the region by previous authors (i.e. during
the Ediacaran; e.g. Soper 1994; Higgins & Leslie 2000). An
additional and more significant problem with the rift-basin
model is that no evidence or structures such as rift-related dolerite
dykes or volcanic or pyroclastic deposits have been observed in the
region, which might suggest a rifting stage (Soper 1994; Higgins &
Leslie 2008; Sønderholm et al. 2008). One explanation could
be that the eastern Laurentian margin in East and North-East
Greenland was, in fact, located in a cratonward position relative
to the rift zone.
The authors thank the reviewers M. J. Hambrey and A. H. Knoll for their valuable
comments and suggestions, from which the manuscript greatly benefited. Field-
work in Greenland was funded both by the Carlsberg Foundation and the
Danish Research Council. The Geological Survey of Denmark and Greenland
(GEUS) gave permission to reproduce elements in Figure 56.1. This represents
a contribution of the IUGS- and UNESCO-funded IGCP (International
Geoscience Programme) Project #512.
References
Bengaard, H.-J. 1991. Upper Proterozoic (Eleonore Bay Supergroup) to
Devonian central fjord zone, East Greenland, geological map, 1:
250000. Copenhagen, Geological Survey of Greenland.
Brasier, M. D. & Shields, G. 2000. Neoproterozoic chemostratigraphy
and correlation of the Port Askaig glaciation, Dalradian Supergroup
of Scotland. Journal of the Geological Society, London, 157,
909– 914.
Cowie, J. W. & Adams, P. J. 1957. The geology of the Cambro-
Ordovician rocks of central Greenland. Part 1: Stratigraphy and struc-
ture. Meddelelser om Grønland, 153, 193.
Eha, S. 1953. The pre-Devonian sediments on Ymers Ø, Suess Land,
and Ella Ø (East Greenland) and their tectonics. Meddelelser om
Grønland, 111, 105.
Eyles, N. & Januszczak, N. 2004. ‘Zipper-rift’: a tectonic model for
Neoproterozoic glaciations. Earth Science Reviews, 65, 1 –73.
Fairchild, I. J. & Herrington, P. M. 1989. A tempestite-stromatolite-
evaporite association (Late Vendian, East Greenland): a shoreface-
lagoon model. Precambrian Research, 43, 101– 127.
Fairchild, I. J. & Hambrey, M. J. 1995. Vendian basin evolution in
East Greenland and NE Svalbard. Precambrian Research, 73,
217– 223.
Fränkl, E. 1953. Geologische Untersuchungen in Ost-Andrees Land
(NE-Grønland). Meddelelser om Grønland, 113, 160.
Frederiksen, K. S. 2000. Evolution of a Late Proterozoic carbonate ramp
Ymer Ø and Andrée Land Groups, Eleonore Bay Supergroup, East
Greenland: response to relative sea-level rise. Polarforschung, 68,
125– 130.
Frederiksen, K. S., Craig, L. E. & Skipper, C. B 1999. New observations
of the stratigraphy and sedimentology of the Upper Proterozoic
Andrée Land Group, East Greenland: supporting evidence for a
drowned carbonate ramp. Danmarks og Grønlands Geologiske
Undersøgelse Rapport, 1999/19, 145– 158.
Gee, D. G., Fossen, H., Henriksen, N. & Higgins, A. K. 2008.
From the Early Paleozoic platforms of Baltica and Laurentia
to the Caldonide Orogen of Scandinavia and Greenland. Episodes,
31, 1 – 8.
Green, J. W., Knoll, A. H., Golubic, S. & Swett, K. 1987. Paleobio-
logy of distinctive benthic microfossils from the Upper Proterozoic
limestone-dolomite ‘Series’, central East Greenland. American
Journal of Botany, 74, 928– 940.
Green, J. W., Knoll, A. H. & Swett, K. 1988. Microfossils from oolites
and pisolites of the Upper Proterozoic Eleonore Bay Group, central
East Greenland. Journal of Paleontology, 62, 835– 852.
Green, J. W., Knoll, A. H. & Swett, K. 1989. Microfossils from silici-
fied stromatolitic carbonates of the Upper Proterozoic Limestone-
Dolomite ‘Series’, central East Greenland. Geological Magazine,
126, 567– 585.
Haller, J. 1953. Geologie und Petrographie von West-Andrées Land
und Ost-Frænkels Land (NE-Grönland). Meddelelser om Grønland,
113, 196.
Haller, J. 1971. Geology of the East Greenland Caledonides. Inter-
science Publishers, New York.
Halverson, G. P. 2011. Glacial sediments and associated strata of the
Polarisbreen Group, northeastern Svalbard. In: Arnaud, E.,
Halverson, G. P. & Shields-Zhou, G. (eds) The Geological
Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 571–580.
Halverson, G. P., Hoffman, P. F., Schrag, D. P., Maloof, A. C. &
Rice, A. H. N. 2005. Toward a Neoproterozoic composite carbon-
isotope record. Geological Society of America, Bulletin, 117,
1181– 1207.
 NEOPROTEROZOIC NE GREENLAND
Halverson, G. P., Maloof, A. C. & Hoffman, P. F. 2004. The Marinoan
glaciation (Neoproterozoic) in northeast Svalbard. Basin Research,
16, 297– 324.
Hambrey, M. J. 1983. Correlation of late Proterozoic tillites in the
North Atlantic region and Europe. Geological Magazine, 120,
290– 320.
Hambrey, M. J. 1989. The Late Proterozoic sedimentary record of East
Greenland: its place in understanding the evolution of the Caledonide
Orogen. In: Gayer, R. A. (ed.) The Caledonide Geology of Scandina-
via. Graham & Trotman, London, 352, 257– 262.
Hambrey, M. J. & Spencer, A. M. 1987. Late Precambrian glaciation
of central East Greenland: Meddelelser om Grønland, Geoscience,
19, 50.
Hambrey, M. J. & Harland, W. B. 1995. The Late Proterozoic glacial
era. Palaeogeography, Palaeoclimatology, Palaeoecology, 51,
255– 272.
Hambrey, M. J., Peel, J. S. & Smith, M. P. 1989. Upper Proterozoic and
Lower Palaeozoic in northern East Greenland. Rapport Grønlands
Geologiske Undersøgelse, 145, 103– 108.
Harland, W. B. 1997. The Geology of Svalbard. Geological Society of
London, Memoirs, 17, 521.
Harland, W. B. & Gayer, R. A. 1972. The Arctic Caledonides and
earlier oceans. Geological Magazine, 109, 289–314.
Henriksen, N. 1999. Conclusion of the 1 : 500000 mapping project in the
Caledonian fold belt in North-East Greenland. Geology of Greenland
Survey Bulletin 183, 10– 22.
Henriksen, N. 2003. Caledonian orogen East Greenland 708– 828N, litho-
tectonic map 1: 1 000000. Geological Survey of Denmark and Green-
land, Copenhagen.
Henriksen, N. & Higgins, A. K. 1976. East Greenland Caledonian fold
belt. In: Escher, A. & Watt, W. S. (eds) Geology of Greenland.
Geological Survey of Greenland, Copenhagen, 182–246.
Henriksen, N. & Higgins, A. K. 2008. Geological research and
mapping in the Caledonian orogen in East Greenland, 708N–828N.
1 – 28. In: Higgins, A. K., Gilotti, J. & Smith, P. M. (eds) The
Greenland Caledonides: Evolution of the Northeast Margin of Laur-
entia. Memoir Geological Society of America, 202, 368.
Herrington, P. M. & Fairchild, I. J. 1989. Carbonate shelf and slope
facies evolution prior to Vendian glaciation, central East Greenland.
In: Gayer, R. A. (ed.) The Caledonide Geology of Scandinavia.
Graham & Trotman, London, 263–273.
Higgins, A. K. 1981. The Late Precambrian Tillite Group of the Kong
Oscars Fjord and Kejser Franz Josefs Fjord region of East Greenland.
In: Hambrey, M. J. & Harland, W. B. (eds) Earth’s Pre-Pleistocene
Glacial Record. Cambridge University Press, Cambridge, 778– 781.
Higgins, A. K. & Leslie, A. G. 2000. Restoring thrusting in the East
Greenland Caledonides. Geology, 28, 1019– 1022.
Higgins, A. K. & Leslie, A. G. 2008. Architecture and evolution of the
East Greenland Caledonides – an introduction. In: Higgins, A. K.,
Gilotti, J. A. & Smith, M. P. (eds) The Greenland Caledonides:
Evolution of the Northeast Margin of Laurentia. Geological Society
of America, Memoir, 202, 369, 29– 53.
Higgins, A. K, Leslie, A. G. & Smith, M. P. 2001. Neoproterozoic –
Lower Palaeozoic stratigraphical relationships in the marginal
thin-skinned thrust belt of the East Greenland Caledonides: compari-
sons with the foreland of Scotland. Geological Magazine, 138,
143– 160.
Higgins, A. K., Elvevold, S. et al. 2004. The foreland-propagating
thrust sheet architecture of the East Greenland Caledonides 728 –
758N. Journal of the Geological Society (London), 161, 1009– 1026.
Higgins, A. K., Gilotti, J. A. & Smith, M. P. (eds) 2008. The Greenland
Caledonides: Evolution of the Northeast Margin of Laurentia. Geo-
logical Society of America, Memoir, 202, 369.
Hoffman, P. F. & Schrag, D. P. 2002. The snowball Earth hypothesis:
testing the limits of global change. Terra Nova, 14, 129– 155.
Kalsbeek, F., Thrane, K., Nutman, A. P. & Jepsen, H. F. 2000. Late
Mesoproterozoic to early Neoproterozoic history of the East Green-
land Caledonides: evidence for Grenvillian orogenesis? Journal of
the Geological Society (London), 157, 1215– 1225.
Kalsbeek, F., Jepsen, H. F. & Jones, K. A. 2001. Geochemistry and
petrogenesis of S-type granites in the East Greenland Caledonides.
Lithos, 57, 91 – 109, doi: 10.1016/S0024–4937(01)00038-X.
591
Kalsbeek, F., Higgins, A. K. & Jepsen, H. F. 2008. Granites and granites
in the East Greenland Caledonides. 227– 249. In: Higgins, A. K.,
Gilotti, J. A. & Smith, M. P. (eds) The Greenland Caledonides:
Evolution of the Northeast Margin of Laurentia. Memoir, Geological
Society of America, 202, 369.
Katz, H. R. 1952. Zur Geologie von Strindbergs Land (NE-Grönland).
Meddelelser om Grønland, 111, 150.
Katz, H. R. 1954. Einige Benmerkungen zur Lithologie und Stratigraphie
der Tillitprofile in Gebiet des Kejser Franz Josephs Fjord, Ostgrön-
land. Meddelelser om Grønland, 72, 63.
Katz, H. R. 1961. Late Precambrian to Cambrian stratigraphy in East
Greenland. In: Raasch, G. O. (ed.) Geology of the Arctic. Vol. 1.
University of Toronto Press, Toronto, 299– 328.
Kaufman, A. J., Knoll, A. H. & Narbonne, G. M. 1997. Isotopes, ice
ages, and terminal Proterozoic earth history. Proceedings of the
National Academy of Science, 94, 6600– 6605.
Kennedy, M. J., Runnegar, B., Prave, A. R., Hoffman, K. H. &
Arthur, M. 1998. Two or four Neoproterozoic glaciations?
Geology, 26, 1059– 1063.
Kilner, B. R. 2005. Low latitude Neoproterozoic glaciation: palaeomag-
netic investigations of the Precambrian sequences of Oman and East
Greenland. Unpublished D.Phil thesis, University of Oxford.
Knoll, A. H. 2000. Learning to tell Neoproterozoic time. Precambrian
Research, 100, 3– 20.
Knoll, A. H., Hayes, J. M., Kaufman, A. J., Swett, K. & Lambert, I. B.
1986. Secular variation in carbon isotope ratios from Upper Proterozoic
successions of Svalbard and East Greenland. Nature, 321, 832–838.
Koch, L. 1929. The geology of East Greenland. Meddelelser om Grøn-
land, 73, II(I), 204.
Koch, L. 1930. Die technische Entwicklung Grönlands. Geologische
Rundschau, 21, 345– 347.
Koch, J. & Haller, J. 1971. Geological map of East Greenland 72 –76
N. Lat. (1 : 250.000). Meddelelser om Grønland, 183, 26, 13 maps.
Kulling, O. 1929. Stratigraphic studies of the geology of North-east
Greenland. Meddelelser om Grønland, 74, 317–346.
Mac NioCaill, C., Stouge, S., Harper, D. A. T., Christiansen, J.,
Kilner, B., Johnson, A. & Watts, C. 2004. Preliminary paleomag-
netic results from the late Neoproterozoic of eastern Greenland: A
low-latitude Sturtian glaciation? EOS Transactions of the American
Geophysical Union, 85, 165.
Mac Niocaill, C., Kilner, B., Stouge, S., Knudsen, M. F., Harper,
D. A. T. & Christiansen, J. L. 2008. The Neoproterozoic drift
history of Laurentia: a critical evaluation and new palaeomagnetic
data from Northern and Eastern Greenland. EOS Transactions of
the American Geophysical Union, 89, S514– 05.
Manby, G. M. & Hambrey, M. J. 1989. The structural setting of the Late
Proterozoic tillites of East Greenland. In: Gayer, R. A. (ed.) The
Caledonide Geology of Scandinavia. Graham & Trotman, London,
352, 299–312.
Moncrieff, A. C. M. 1989. The Tillite Group and related rocks of East
Greenland: implications for Late Proterozoic palaeogeography. In:
Gayer, R. A. (ed.) The Caledonian Geology of Scandinavia.
Graham and Trotman, London, 352, 285–297.
Moncrieff, A. C. M. & Hambrey, M. J. 1988. Late Precambrian
glacially-related grooved and striated surfaces in the Tillite Group
of Central East Greenland. Palaeogeography, Palaeoclimatology,
Palaeoecology, 65, 183–200.
Moncrieff, A. C. M. & Hambrey, M. J. 1990. Marginal-marine glacial
sedimentation in the late Precambrian succession of East Greenland.
In: Dowdeswell, J. A. & Scourse, J. D. (eds) Glacimarine Environ-
ments: Processes and Sediments. Geological Society, London,
Special Publication, 53, 397– 410.
Nystuen, J. P., Andresen, A., Kumpulainen, R. A. & Siedlecka, A.
2008. Neoproterozoic basin evolution in Fennoscandia, East Green-
land, and Svalbard. Episodes, 31, 35 – 43.
Phillips, W. E. A, Stillman, C. J., Friderichsen, J. D. & Jemelin, L.
1973. Preliminary results of mapping in the western gneiss and
schist zone around Vestfjord and inner Gåsefjord, south-west Scor-
esby Sund. Rapport Grønlands Geologiske Undersøgelse, 58, 17– 32.
Porter, S. M. & Knoll, A. H. 2000. Testate amoebae in the Neoproter-
ozoic era: evidence from vase-shaped microfossils in the Chuar
Group, Grand Canyon. Paleobiology, 26, 360–385.
592 S. STOUGE ET AL.
Porter, S. M., Meisterfeld, R. & Knoll, A. H. 2003. Vase-shaped
microfossils from the Neoproterozoic Chuar Group, Grand Canyon:
a classification guided by modern testate amoebae. Journal of
Paleontology, 77, 409–429.
Poulsen, C. 1930. Contributions to the stratigraphy of the Cambro – Ordo-
vician of East Greenland. Meddelelser om Grønland, 74, 297–316.
Robb, L. J., Knoll, A. H., Plumb, K. A., Shields, G. A., Strauss, H. &
Veizer, J. 2004. The Precambrian: the Archaean and Proterozoic
eons. In: Gradstein, F. M., Ogg, J. G. & Smith, A. G. (eds) A Geo-
logic Time Scale 2004. Cambridge University Press, Cambridge,
129– 140.
Schaub, H. P. 1950. On the pre-Cambrian sedimentation in NE-
Greenland. Meddelelser om Grønland, 114, 50.
Schaub, H. P. 1955. Tectonics and morphology of Kap Oswald
(NE-Greenland). Meddelelser om Grønland, 103, 1– 33.
Smith, M. P. & Robertson, S. 1999. Vendian – Lower Palaeozoic strati-
graphy of the parautochthon in the Målebjerg and Eleonore Sø
windows, East Greenland Caledonides. Danmarks og Grønlands
Geologiske Undersøgelse Rapport, 1999/19, 169– 182.
Smith, M. P., Rasmussen, J. A., Robertson, S., Higgins, A. K & Leslie,
A. G. 2004. Lower Palaeozoic stratigraphy of the East Greenland
Caledonides. Geological Survey of Denmark and Greenland Bulletin,
6, 5 –28.
Sommer, M. 1957. Geologie von Lyells Land (NE-Grönland). Meddelel-
ser om Grønland, 155, 157.
Soper, N. J. 1994. Neoproterozoic sedimentation on the NE margin of
Laurentia and the opening of Iapetus. Geological Magazine, 131,
291– 299.
Stouge, S., Boyce, W. D., Christiansen, J. L., Harper, D. A. T. &
Knight, I. 2001. Vendian – Lower Ordovician stratigraphy of Ella
Ø, North-East Greenland: new investigations. Geology of Greenland
Survey Bulletin, 189, 107–114.
Stouge, S., Boyce, W. D., Christiansen, J. L., Harper, D. A. T. &
Knight, I. 2002. Lower –Middle Ordovician stratigraphy of North-
East Greenland. Geology of Greenland Survey Bulletin, 191,
117– 125.
Sønderholm, M. & Tirsgaard, H. 1993. Lithostratigraphic framework
of the Upper Proterozoic Eleonore Bay Supergroup of East and
North-East Greenland. Bulletin Grønlands Geologiske Undersøgelse,
167, 38
Sønderholm, M., Frederiksen, K. S., Smith, M. P. & Tirsgaard, H.
2008. Neoproterozoic sedimentary basins with glacigenic deposits
of the East Greenland Caledonides. In: Higgins, A. K., Gilotti, J.
A. & Smith, M. P. (eds) The Greenland Caledonides: Evolution of
the Northeast Margin of Laurentia. Memoir, Geological Society of
America, 202, 99– 136.
Swett, K. & Knoll, A. H. 1989. Marine pisolites from Upper Proterozoic
carbonates of East Greenland and Spitsbergen. Sedimentology, 36,
75 – 93.
Teichert, C. 1933. Untersuchungen zum Bau des kaledonischen Gebirges
in Ostgrönland. Meddelelser om Grønland, 95, 121.
Tirsgaard, H. 1993. The architecture of Precambrian high energy tidal
channels: an example from the Lyell Land Group (Eleonore Bay
Supergroup), East Greenland. Sedimentary Geology, 88, 137–152.
Tirsgaard, H. 1996. Cyclic sedimentation of carbonate and siliciclastic
deposits on a late Precambrian ramp: the Elisabeth Bjerg Formation
(Eleonore Bay Supergroup), East Greenland. Journal of Sedimentary
Research, 66B, 699– 712.
Tirsgaard, H. & Sønderholm, M. 1997. Lithostratigraphy, sedimentary
evolution and sequence stratigraphy of the Upper Proterozoic Lyell
Land Group (Eleonore Bay Supergroup) of East and North-East
Greenland. Geology of Greenland Survey Bulletin, 178, 60.
Vidal, G. 1976. Late Precambrian acritarchs from the Eleonore Bay
Group and Tillite Group in East Greenland. Grønlands Geologiske
Undersøgelse Rapport, 78, 19.
Vidal, G. 1979. Acritarchs from the Upper Proterozoic and Lower
Cambrian of East Greenland. Grønlands Geologiske Undersøgelse
Rapport, 134, 40.
Vidal, G. 1985. Biostratigraphic correlation of the Upper Proterozoic and
Lower Cambrian of the Fennoscandian Shield and the Caledonides of
East Greenland and Svalbard. In: Gee, D. G. & Sturt, B. A. (eds)
The Caledonide Orogen – Scandinavia and Related Areas. John
Wiley and Sons Ltd., London, 331– 338.
Wenk, E. 1961. On the crystalline basement and the basal part of the pre-
Cambrian Eleonore Bay Group in the southwestern part of Scoresby
Sund. Meddelelser om Grønland, 168, 54.
 Chapter 57

Glaciogenic rocks of the Neoproterozoic Smalfjord and Mortensnes formations,

Vestertana Group, E. Finnmark, Norway

A. HUGH N. RICE1 *, MARC B. EDWARDS2, TOR A. HANSEN3, EMMANUELLE ARNAUD4 & GALEN P. HALVERSON5,6
1
Vienna University, Structural Processes Group, Department of Geodynamics and Sedimentology, Althanstrasse 14,
1090 Vienna, Austria, Europe
2
5430 Dumfries Drive, Houston, Texas, USA
3
Talisman Energy Norge AS, Verven 4, PO Box 649, Stavanger, Norway
4
Department of Land Resource Science, University of Guelph, Guelph, Ontario, N1G 2W1, Canada
5
Department of Geology & Geophysics, School of Earth & Environmental Sciences, University of Adelaide,
North Terrace, Adelaide, SA 5005, Australia
6
Present address: Department of Earth and Planetary Sciences, McGill University, 3450 University Street, Montreal,
Quebec, H3A 2A7, Canada
*Corresponding author (e-mail: alexander.hugh.rice@univie.ac.at)

Abstract: The Vestertana Group in East Finnmark, North Norway, contains two Neoproterozoic glaciogenic sequences, the Smalfjord
and Mortensnes formations, preserved on the northern edge of Baltica. The former comprises up to 420 m of aeolian, fluvioglacial and
glaciomarine sediments and terrestrial diamictite. The latter consists of up to 50 m of predominantly diamictite. The Smalfjord Formation
(Fm.) is underlain by dolostones (Grasdalen Fm., Tanafjorden Group), only locally preserved due to the sub-Smalfjord Fm. unconfor-
mity, which cuts down-section through a c. 2.5 km dominantly clastic sequence to rest on Baltic Shield gneisses. The two glaciogenic
successions are separated by c. 350 m of mostly clastic sediments (Nyborg Fm.), with thin dolostones at the base and towards the top. The
latter are generally absent due to the sub-Mortensnes Fm. unconformity, which also cuts down southwards through the Nyborg and
Smalfjord formations to the Baltic Shield. No robust isotopic age constraints are available for the succession. d13C data, together
with cap dolostone characteristics, offer paradigmic correlations with other areas (Smalfjord ; Marinoan; Mortensnes ; Gaskiers).
A limited Ediacaran fauna, including Aspidella, give only broad age constraints. Palaeomagnetic data are ambiguous; some suggest
Baltica lay at equatorial (158S) to mid-latitudes (508S) for the period 750– 550 Ma, respectively, while other interpretations place it
at either 308N or S at c. 550 Ma.

This chapter is concerned with the glaciogenic sediments of the
Smalfjord and Mortensnes formations (Lower and Upper Tillites,
respectively, in older literature), lying at the base of the Vestertana
Group, within a c. 4.7-km-thick, predominantly clastic sequence of
Tonian/Cryogenian to Tremadocian age. The former glaciogenic
unit only occurs in three areas, all in E Finnmark; Laksefjordvidda,
Tanafjord and Varangerfjord (Fig. 57.1). In contrast, the latter unit
has tentatively been correlated with many other diamictite-bearing
units in the Scandinavian Caledonides (Kumpulainen 2011; Kum-
pulianen & Greiling 2011; Nystuen & Lamminen 2011; Stodt et al.
2011). Owing to significant facies variations, no single outcrop/
area can be taken as a type locality/region.
The rocks have been studied by many authors (see references in
Føyn 1937; Bjørlykke 1967; Føyn & Siedlecki 1980; Laajoki
2002), but detailed sedimentological studies remain few
(Reading & Walker 1966; Edwards 1972, 1975, 1979, 1984;
Hansen 1992; Arnaud & Eyles 2002; Baarli et al. 2006; Levine
et al. 2006; Arnaud 2008). An overview of d13C data from associ-
ated dolostones was given by Halverson et al. (2005).
The Smalfjord Fm. is significant in having provided some of the
first evidence for a glacial event specifically attributed to a pre-
Pleistocene age, at Oaibaččanjar’ga (formerly Bigganjar’ga;
Reusch 1891). Interpretation of this outcrop has remained a bone
of contention since its discovery; recent results indicate that
intense high-temperature brittle deformation occurred during
rapid glacial movement (Bestmann et al. 2006).
The two glaciogenic sequences became an informal ‘type
sequence’ for late Precambrian glaciations in the North Atlantic
region (e.g. Hambrey 1983), known as the Varang(er)ian
glaciation (Varangeristiden; Kulling 1951), later developed into
the Varanger Epoch (cf Harland et al. 1989). Use of the term
Varang(er)ian is here discouraged for reasons given below.
Structural framework
During the Neoproterozoic in northern Scandinavia, the NNE –
SSW-trending western Baltica continental shelf, now preserved
in the Scandinavian Caledonides, was joined to the WNW –ESE-
trending Timan Basin in NE Baltica (Siedlecka 1985; Gayer &
Rice 1989; Siedlecka et al. 2004). Both of these margins are
thought to have experienced extensional rift tectonics from
c. 1000 to c. 630 Ma, associated with the break-up of Rodinia, fol-
lowed by relatively quiescent passive margin conditions during the
deposition of the two glaciogenic units (Røe 2003; Siedlecka et al.
2004). The Timan Basin has been divided into northerly basinal
and southerly shelf regions (Siedlecka et al. 1995), with the
WNW – ESE-trending dextral strike – slip Trollfjorden – Koma-
gelva Fault (cf Rice et al. 1989) forming the boundary between
these two regions (Fig. 57.1). The Smalfjord Fm. was deposited
(or has only been preserved in; but see Kumpulainen 2011) the
eastern part of the Gaissa Basin, which palaeogeographically
formed a linking area between the Timanian shelf region and the
western Baltica continental shelf. In contrast, the Mortensnes
Fm. and its presumed correlatives were deposited and preserved
over much of western Baltica.
The rocks were shortened predominantly during Silurian-
Devonian Caledonian orogenesis (Scandian event). Deformation
was essentially in-sequence, with c. 50% shortening in the external
imbricate zone (Gaissa Thrust Belt; imbricated Gaissa Basin

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 593– 602. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.57
594 A. H. N. RICE ET AL.

Fig. 57.1. Map of NE Scandinavian Caledonides showing the main localities and distribution of glaciogenic lithologies. A, Alduskai’di; An, Andabakoai’vi;
C, Čikkojåkka; D, Digermul; G, Gaessenjar’ga; Gr, Grasdal; Gu, Gulgofjord; H, Handelsneset; L, Lappaluokoai’vi; Lp, Leirpollen; Lv, Laksefjordvidda; M, Mortensnes;
Mi, Miel’kejåkka; Mn, Manjunnas; N, Nyborg; Ne, Nesseby; Nj, Njukčagai’sa; O, Oaibaččannjar’ga; R, Ruos’soai’vi; S, Selesnjar’ga; Sk, Skjåholm; Sm, Smalfjord;
St, Stappugied’di; U, Uccaskai’di; V, Vestertana; Vb, Vaððasbak’te; Vr, Vieranjar’ga.

sediments) west of Laksefjordvidda (Townsend et al. 1986). East
of Laksefjordvidda, shortening decreased to c. 15% (Chapman
et al. 1985) and dies out very gradually across Varangerhalvøya.
In the Digermul –Tanafjord region, large-scale folds developed,
sometimes with a back-thrusting vergence (Reading 1965;
Chapman et al. 1985; Siedlecka 1987). These likely formed by but-
tressing, as the sole thrust attempted to footwall-shortcut through
NNE –SSW-trending extensional structures.
Although the strain decreases to the east, most pelitic rocks have a
cleavage, locally with crenulations of sedimentary laminae in
flexural-slip fold hinges, while diamictites may carry a spaced
(approximately centimetre scale) anastamosing cleavage. Pressure-
solution was common in carbonate rocks, including dolomite-rich
diamictites. However, in sandier lithologies, finite strains are gener-
ally low, with sedimentary structures well preserved.
Stratigraphy
The glaciogenic sequences in E. Finnmark lie at the base of the
Vestertana Group, within the c. 4.7-km-thick, predominantly
shallow marine and siliciclastic, Tonian (inferred) to Tremadocian
East Finnmark Supergroup (Fig. 57.2; Føyn 1937; Johnson et al.
1978; Siedlecka et al. 2004). The sub-Smalfjord Fm. unconformity
at the base of the Vestertana Group cuts down-section to the south
until it lies on the Baltic Shield around Varangerfjord (Fig. 57.1).
Between the Tana River west of Varangerfjord and Andabakoaivi
(Fig. 57.1), the Vestertana Group forms the most southerly Caledo-
nian Neoproterozoic rocks, although the basement –cover contact
is not exposed.
The Smalfjord Fm. has been preserved in two major palaeoval-
leys (Varanger and Krokvatn palaeovalleys; Bjørlykke 1967; Føyn
& Siedlecki 1980) and in the intervening Tanafjord area. The
maximum preserved thickness occurs in the Krokvatn palaeoval-
ley (c. 420 m), but facies variations preclude formation-wide
member names. The Mortensnes Fm. (and equivalents; see Kum-
pulainen 2011; Kumpulainen & Greiling 2011; Nystuen & Lammi-
nen 2011; Stodt et al. 2011), which is more widespread across the
orogen, often lying unconformably on autochthonous or allochtho-
nous basement, is ,50 m thick. The two glaciogenic units are sep-
arated by c. 350 m of clastic and minor carbonate sediments of
the Nyborg Fm. (Edwards 1984). The uppermost sediments of
the Tanafjorden Group that underlie the Smalfjord Fm. are the
mixed carbonate and clastic sediments of the Grasdalen and Fad-
nuvag’gi formations. The Mortensnes Fm. is overlain by mud-
stones and sandstones of the Stappogiedde Fm. (Lillevannet
Member).
The Precambrian –Cambrian boundary lies near the base of the
overlying Breidvika Fm. (Føyn & Glaessner 1979; Farmer et al.
1992; Crimes & McIlroy 1999).
Glaciogenic deposits and associated strata
Grasdalen Fm., Tanafjord
The Grasdalen Fm. (230 m), which only crops out in the Grasdal-
Gulgofjord area, very close to the top of the Tanafjorden Group,
was divided into three major parts by Siedlecka & Siedlecki
(1971) and Johnson et al. (1978): a lower mixed sand/silt-marl-
dolostone unit (111 m), a massive dolostone (58 m) and an inter-
bedded dolostone-clastic sequence (61 m). In detail, however,
the stratigraphy is considerably more complex (Rice, unpublished
data; Fig. 57.2).
 NEOPROTEROZOIC SMALFJORD AND MORTENSNES FORMATIONS 595

Fig. 57.2. Regional stratigraphic profiles showing position of glaciogenic units (modified from Johnson et al. 1978; Føyn & Siedlecki 1980; Edwards 1984; Rice &
Townsend 1996; Halverson et al. 2005). The Lille Molvika Fm. and Ekkerøya group (formerly known as the Ekkerøya Fm., within the Vadsø Group) were informally
defined based on regional correlations by Rice & Townsend (1996). UE, LE, position of uppermost and lowermost Ediacaran fauna; Pl, position of Platysolenites
antiquissimus (Eichw.). Note different vertical scales for columns (a) and (c) compared to (b).

Fadnuvag’gi Fm., Tanafjord deposited within an essentially north –south-oriented palaeovalley

Although the Grasdalen Fm. is usually taken to continue up to the
base of the Smalfjord Fm. (Siedlecka & Siedlecki 1971), the inter-
bedded dolostone-clastic sequence is overlain by c. 20 m of a dis-
tinctly more fissile, darker, pyritiferous and finely interbanded
shale silt/sandstone without carbonate interlayers, informally
termed the Fadnuvaggi Fm. (Halverson et al. 2005).
Smalfjord Fm., Krokvatn Palaeovalley, Laksefjordvidda
Føyn & Siedlecki (1980) documented three diamictite units with
intervening sandstones (Lower, Middle and Upper Krokvatn
Diamictites, Lower and Upper Krokvatn Sandstones; Fig. 57.2)
cut into the Tanafjorden Group on Laksefjordvidda (Fig. 57.1).
Figure 57.3a shows a semi-schematic profile across the northern
part of the palaeovalley (Uccaskai’di-Vaððasbak’te); further
south (Čikkojokka; Fig 57.1), the sequence comprises only the
middle diamictite and upper sandstone (Føyn & Siedlecki 1980).
The Tanafjorden Group dips to the NW compared to the
420-m-thick palaeovalley infill, with 600 m of tilted stratigraphy
eroded.
The Lower and Middle Krovatn Diamictites are both ,25 m
thick, with a grey to reddish colour and a sandy matrix. The
former includes a 2-m-thick sandstone, whereas the latter has inter-
layered sandstones in the basal part and has faint banding and rare
dropstones (Føyn & Siedlecki 1980). Clasts, sometimes faceted,
are predominantly from the Tanafjorden Group, with rare
596
Fig. 57.3. (a) Schematic profile from Uccaskai’di to Vaððasbak’te through the glacial successions in the Krokvatn Palaeovalley (modified from Føyn & Siedlecki
1980). LKD, MKD, UKD, Lower, Middle and Upper Krokvatn Diamictites; (b) Schematic profile from Varangerfjord to Gulgofjord (modified from Edwards & Føyn
1981). Nyborg facies: N2, fan channel; N3, N4, fine and coarse submarine fans, respectively; N5, tidal distributary; N6, bay/lagoon; N7, offshore barrier. lsm, usm, lower
and upper sub-members of the Lillevannet Member.
kilometre-sized dolostone blocks at the base (Edwards et al. 1973)
and some basement material. The upper diamictite (,100 m thick)
is more variable, with dolomitic and sandy matrices and a buff or
grey colour. These are interbedded with sandstones and siltstones
with a few lonestones. On Uccaskai’di, the Upper Krokvatn Dia-
mictite has abundant dolostone clasts, possibly reflecting the
nearby presence of Tanafjorden Group dolostones (Fig. 57.1;
Føyn & Siedlecki 1980). In contrast, in the northeastern part of
the valley, the Nyborg Fm. rests unconformably on the Lower
Krokvatn Diamictite and Lower Sandstone, the higher units
having been removed either during deposition of the third diamic-
tite elsewhere, or afterwards.
The massive, homogeneous very fine-grained sandstones and
very coarse-grained siltstones are essentially devoid of sedimen-
tary structures throughout the succession; ripples occur at only
one locality. Beds are from 1 m to several metres thick, without
sorting of the angular grains, although larger grains (.0.2 mm)
are rounded. Hand-sized basement lonestones occur at two
localities. Directly above the Middle Krokvatn Diamictite, the
sediments are red (Føyn & Siedlecki 1980).
Smalfjord Fm., Varanger Palaeovalley, Varangerfjord
The Smalfjord Fm. in the Varangerfjord area (Bjørlykke 1967;
Edwards 1975, 1984; Arnaud & Eyles 2002; Arnaud 2008;
Laajoki 2001, 2002; Baarli et al. 2006) has been correlated
with the lower to middle part of the Krokvatn succession (Føyn
& Siedlecki 1980). The following descriptions are based on
A. H. N. RICE ET AL.
outcrops at Vieranjar’ga, Nesseby, Handelsneset, Skjåholm, Oai-
baččanjar’ga and Selesnajar’ga (Edwards 1975, 1984; Arnaud &
Eyles 2002; Baarli et al. 2006; Arnaud 2008).
Two diamictite and four sandstone and conglomerate (S1– S4)
facies have been identified in this area. The first type of diamictite
is a massive, matrix-supported, pink diamictite with a medium-
to coarse-grained matrix with only a few granitic and sandstone
clasts. These are up to 20 cm across, sub-rounded to sub-angular,
and without striae or faceting. This unit, in beds up to 1 m thick,
can be traced over several hundred metres. The second diamictite,
which is also mostly matrix-supported and massive, with a dis-
tinctly grey colour in outcrop, consists of a muddy, medium
grained sandstone matrix and abundant granite and sandstone
clasts, ,50 cm across, some with faceting. Bjørlykke (1967)
also noted striated clasts in grey diamictite.
The S1 facies comprises well-stratified pebble conglomerates,
occasionally imbricated, sandstones showing cross-bedding and
parallel lamination, with fining-upwards trends between these
lithologies and rare mud-drapes. S2 is typified by large-scale
inclined bedding, 5–10 m thick and continuous for up to several
kilometres. These include steeply dipping poorly sorted pebble
conglomerates to well-sorted gently dipping parallel-laminated
sandstones. S3 comprises medium-bedded parallel-sided to slightly
lenticular sandstones, usually internally massive, but at times
parallel-laminated or rippled. Mudstone partings are thin or absent
but soft-sediment deformation, with convolute bedding, recum-
bent folds and faulting occur. Steep-sided isolated channels filled
with poorly sorted conglomerate are also present. S4 comprises
interbeddded sandstones and mudstones, the former in thin to
 NEOPROTEROZOIC SMALFJORD AND MORTENSNES FORMATIONS
medium, laterally continuous beds. These display many turbiditic
features (erosive surfaces, sole marks, grading, convolute bedding).
A 3-m-thick, crudely stratified breccia within S4 contains angular
sandstone clasts ,40 cm across, sometimes graded, in a coarse-
grained sandstone matrix.
The stratigraphic and lateral distribution of these facies vary
from place to place. Laterally, lithofacies can often be traced for
several tens of metres, although abrupt facies changes are common
(cf Edwards 1984). Complex soft-sediment deformation in sand-
stone and conglomerate facies is particularly notable at the base
of the Smalfjord Fm. at Handelsneset (Arnaud 2008). Overall, the
diamictite, breccia and S4 facies tend to occur near the base of
the sequence, whereas the S1 facies occurs near or at the top.
Smalfjord Fm., Tanafjord area
The Smalfjord Fm. in this region (Reading & Walker 1966;
Edwards 1984; Hansen 1992; Arnaud & Eyles 2002) has been cor-
related with the Upper Krokvatn Diamictite (Laksefjordvidda
region; Føyn & Siedlecki 1980; Edwards 1984). The best outcrops
in this area are those exposed at Gaessenjar’ga and Luovtat
(Fig. 57.1).
Based on outcrops in the Vestertana to Smalfjord area, with
additional important outcrops between Njukčagai’sa and east of
the Tana River, Edwards (1984) recognized a fivefold (A –E)
repetitive cycle of erosion surface, diamictite and laminated mud-
stone, occasionally with intervening sandstones. Some units can be
traced over hundreds of square kilometres. Diamictite units are from
2–40 m thick, structureless or stratified, and have erosive bases,
at both outcrop and regional scales. Diamictites are all
matrix-supported, with clasts up to 1 m in diameter (most are
,35 cm) in a siltstone to sandstone matrix. Material from underly-
ing beds was often incorporated into the overlying diamictite,
forming inclusions of diamictite, conglomerate, sandstone or mud-
stone, folds and faults, load-casts, boudinage, shear bands and
flame structures. Some diamictite units have gradational contacts
(Hansen 1992; Arnaud & Eyles 2002). Stratified sandstone/con-
glomerate bodies are present within the diamictite; these deposits
resemble the facies in the Varangerfjord area (see above). In some
cases, the basal diamictite represents mixing of far-travelled and
locally derived material. The dominant clast and matrix source
areas determine the diamictite lithology; buff and brown colours
indicate a dolostone source (cycles B, C and E). Cycle A is purple,
with a hematitic matrix, and was derived from ferruginous sand-
stones of the Dakkovarre Fm. (Fig. 57.2). Diamictites are also
often distinguished by variable clast abundance. Some diamictite
units exhibit normal grading with a sandier matrix at the base and
muddier matrix at the top (Hansen 1992; Arnaud & Eyles 2002).
The Smalfjord Fm. in the Tanafjord area has a much higher
proportion of diamictite and is laterally much more consistent
compared to outcrops and lithofacies distributions in the Varan-
gerfjord area; in the Gaessenjar’ ga area, over 85% of the sections
consist of diamictite units and these tend to be traceable over
several kilometres (Edwards 1984; Arnaud & Eyles 2002). One
exception is that of a 1–5-m-thick purple diamictite that grades
into a laminated mudstone with outsized clasts within several
hundred metres of lateral exposure at Gaessenjar’ga (Arnaud &
Eyles 2002).
Sandstone beds are massive, graded (normal or reverse), rippled
and cross-bedded (trough and planar). Laminated mudstones, 0.3–
10 m thick, vary in colour, lamination prominence, whether this is
random or rhythmic, and size, abundance and composition of lone-
stones. The latter are predominantly dolostones, ,1–30 cm in size
and are either dispersed throughout the lithology or lie in discrete
layers. Some laminated mudstone units are interbedded with sand-
stone or diamictite.
In the Gaessenjar’ga area, Hansen (1992) and Arnaud & Eyles
(2002) recognized essentially the same lithologies and units as
597
Edwards (1984), but ordered them differently, without any
depositional cyclicity.
In the Grasdal area, Edwards (1979) described a 3 –6-m-thick
dark grey siltstone with an unusually high silt and low clay
content; this may also occur on the southern side of Leirpollen
(Fig. 57.1). Clasts consist of angular quartz and feldspar grains
of very fine sand to coarse silt size, forming c. 77% of the rock,
with a statistically non-random preferred north –south to NE–
SW orientation. This unit has sharp contacts with the over- and
underlying units (the latter is a very dark diamictite with abundant
dolomitic clasts) and an internal stratification defined by isolated,
typically dolomitic, clasts (,7.5 mm sized) in discrete layers
c. 30 cm apart, together with very faint grey colour variations.
Nyborg Fm., E. Finnmark (0– 350 m)
The Smalfjord and Mortensnes formations are separated by the
Nyborg Fm. The lowest part of this unit (Member A) comprises
lateral and vertical gradations between massive buff-weathering
dolomicrites with sheet cracks and pseudo-tepees, through
massive to thinly bedded dolomicrites to interbeddded red sand-
stones/shales and dolomicrites, locally reworked as edgewise
breccias, to dolomite-cemented orange-red sandstones, as well as
white-weathering green-grey sandstones and red and dark grey
shales (Edwards 1984; Reading & Walker 1966; Rice &
Hofmann 2001). Such sequences are usually only few tens of
metres thick (Reading & Walker 1966; Edwards 1984) although
the succession is c. 160 m thick at Alduskaidi (Fig. 57.1; Føyn &
Siedlecki 1980). Generally, successions become more clastic
upwards. On Ruos’soai’vi, west of Varangerfjord, 10-mm sulphate
crystal fans (now pseudomorphed by quartz) grew on the base-
ment, preserved within a 12-mm-thick biotite extraclast recrystal-
lized dolostone overlain by white-weathering sandstones. Similar
white-weathering sandstones, likely cannibalized from the Smal-
fjord Fm., also directly overlie very irregular basement surfaces on
Ruos’soai’vi. White-weathering green-grey sandstones (18.5 cm
thick) also unconformably overlie c. 14 m of massive sheet-
cracked dolomicrite in the Miel’kejåkka area, SW of Tanafjord
(Fig. 57.1). Members B –D of the Nyborg Fm. comprise, respect-
ively, interbedded reddish-purple shales/sandstones (c. 200 m),
interbedded grey-green shales/sandstones (c. 150–200 m) and
purple sandstone/grey-green shales (c. 70 m; Edwards 1984).
Member E (c. 25 m thick), exposed only in the Gulgofjord-Grasdal
area (Fig. 57.1), consists of white to grey sandstones with two thin
(c. 1 m thick) buff-weathering siliclastic-extraclast dolomicrites at
the base (Edwards 1984).
Mortensnes Fm., E. Finnmark (,50 m)
Edwards (1984) recognized three members in this formation, the
lowest (,30 m) being a northwards-thinning wedge of predomi-
nantly grey-green to purple massive diamictite (depending on
the substrate colour) of highly variable clast/matrix composition,
with basement clasts and minor intrabasinal clasts from the under-
lying substrate. This dies out around the southern end of Vester-
tana. Brecciation of the substrate has been found below some
diamictites, with substrate blocks up to 20 m long and 1 m thick
incorporated into the overlying diamictite. Lenses and bands of
extra-basinal dark diamictite occur within more mixed lithologies.
Basic clasts may show facets and striations (Banks et al. 1971),
while clast sizes decrease overall to the north (Edwards 1984).
The middle member has a gradational contact to the underlying
rocks and is distinguished from it by its buff-brown weathering,
dolomitic composition of matrix and clasts with subordinate
chert. This member rapidly increases in thickness from ,4 m in
the south to .10 m, approximately coincident with the southern
end of Vestertana, giving two sub-members. The thin sub-member
598 A. H. N. RICE ET AL.
comprises 2– 4 m of stratified diamictite with primary and soft-
sediment deformation structures. The thick sub-member comprises
five lithofacies; a blanket of massive purple to grey-green defor-
mation diamictite, a zone of large tabular blocks (of the Nyborg
Fm., diamictite and white sandstone), a relatively rare stratified dolo-
mitic diamictite, a prominent buff-brown diamictite with a sandy
matrix, which thins from 20 to 8 m from south to north, and, at the
top, a bedded buff brown diamictite. The unconformably overlying
upper member (,40 m) consists of dark grey massive diamictite
overlain north of Stappugied’di and Leirpollen by a 40-cm-thick
dolomitic matrix diamictite with small dolostone and occasional
large basement clasts. A 20–30-cm-thick polymict conglomerate
overlies the formation over a large area (Edwards 1984).
Lillevannet Member, Stappogiedde Fm. (40 – 110 m)
The Mortensnes Fm. is overlain by the Lillevannet Member of the
Stappogiedde Fm. (Reading & Walker 1966), which thins from
south to north, approximately coincident with the southern end
of Vestertana (Edwards 1984). The lower sub-member (3 –55 m)
comprises grey, parallel laminated mudstones, silty to sandy in
the north, grading upwards into a siltstone with some ripple cross-
lamination and with fine- to medium-grained lenticular sandstones
(Edwards 1984). The upper sub-member is a complex assemblage
of sandstone and shale facies, including coarse arkosic sandstones,
poorly cross-bedded and granule conglomerates, medium-grained
sub-arkosic sandstone, relatively well sorted and rounded sand-
stones, thin to medium bedded fine to very fine lenticular, erosive-
based sandstones, dark grey, brown weathering rippled and finely
laminated micaceous silty-sandy mudstone, sometimes in
coarsening-upwards cycles and finely parallel laminated grey
mudstones (Edwards 1984).
Boundary relations with overlying and underlying
non-glacial units
Southwards, the Smalfjord Fm. cuts down through up to 2.5 km of
the Tanafjorden, Ekkerøya and Vadsø Groups, to lie on the base-
ment around Varangerfjord. Minor outliers of Neoproterozoic
sediments overlie the basement, and minor inliers of basement
occur within the Smalfjord Fm., together preserving an irregular
unconformity, locally palaeo-frost-shattered (Bjørlykke 1967;
Siedlecka 1990; Rice & Hofmann 2000; Rice et al. 2001; Laajokki
2001, 2002; Edwards 1984, 1975). The base of the Smalfjord Fm.
is characterized by rare east –west to WNW –ESE oriented sub-
glacial striations; the best developed occur at Oaibaččanjar’ga
(Bigganjar’ga) in Varangerfjord (Reusch 1891; Strahan 1897),
although others have been reported (Bjørlykke 1967; Rice &
Hoffman 2000; Laajoki 2002). Rice & Hofmann (2000) and Best-
mann et al. (2006) documented a thin glacially formed breccia
within and as ridges beside the striations at Oaibaččanjar’ga.
Regionally, the upper contact of the Smalfjord Fm. with the
overlying Nyborg Fm. is also an unconformity, because the latter
rests on different parts of the former in different areas; specifi-
cally, the upper part of the Smalfjord Fm. in the Vaððasbak’te
area of the Krokvatn Palaeovalley and in the Varanger palaeoval-
ley are absent (Fig. 57.1; Føyn & Siedlecki 1980; Edwards 1984).
In the Ruos’soai’vi-Lap’paluokoai’vi area (Fig. 57.1), the Nyborg
Fm. rests directly on the basement (Siedlecka 1990; Rice unpub-
lished data).
The base of the Mortensnes Fm. is an invariably planar surface
at outcrop-scale that cuts down-section towards the south at a very
low angle. The contact often exhibits brecciation or homogeniz-
ation of the underlying sediments (deformation diamictite;
Edwards 1984). At its upper contact, the Mortensnes Fm. is over-
lain with a sharp or rapid transition by the Lillevannet Member
(Fig. 57.2; Reading & Walker 1966; Edwards 1984).
Chemostratigraphy
d13C data from E. Finnmark were briefly reviewed by Halverson
et al. (2005); although more data are now available the results
are essentially the same as those previously reported. All data
are standardized to VPDB; see Halverson et al. (2005) for
analytical methods.
The Grasdalen Fm. has variable values, passing from –2 to
–3‰ at the base, gradually rising to þ6‰ and then falling very
rapidly to – 3‰; the two stratigraphically highest samples record
values of þ3 to þ4‰.
Ten profiles through the dolostone (or interlayered dolostones–
shales/siltstones/sandstones) sequences lying directly above the
Smalfjord Fm. have been analysed over a wide geographical
area and include the c. 160-m-thick dolostone at Alduskaidi
(Fig. 57.1; Føyn & Siedlecki 1980). Although few profiles
expose both the bottom and top of the dolomite unaffected by
either tectonic or penecontemporaneous erosional processes, all
samples show d13C values between –1.0 and – 6.0‰ (VPDB).
Most have essentially constant d13C values or show either a
slight increase or decrease in concentration. Only the 14-m-thick
profile at Miel’kejåkka (Fig. 57.1) shows a major variation,
passing from – 2.55‰ at the base to –5.92‰ at the erosive
upper contact. These data show a poor concave upwards distri-
bution, comparable to the uppermost part of the slope-bank d13C
profile documented by Hoffman et al. (2007).
Thin dolostones in Member E of the Nyborg Fm., lying c. 20 m
below the sub-Mortensnes unconformity in the Gulgofjord area
(Edwards 1984; Figs 57.1 & 57.2) gave d13C values between
–7.6 and – 9.9‰ (VPDB). Significantly, the thin carbonate-matrix
diamictite forming the uppermost part of the Mortensnes Fm. on
the NW side of Tanafjord (Edwards 1984) also gave extreme nega-
tive values (down to – 10.44‰ VPDB), taken to reflect erosion of
the underlying dolostones of Member E in the Nyborg Fm. (Rice &
Halverson, unpublished data).
Palaeolatitude and palaeogeography
Palaeomagnetic analyses on the Nyborg Fm. have yielded palaeo-
latitudes for this area of 338S to 418S (Torsvik et al. 1995),
although no robust geochronological constraint is associated
with this palaeolatitude. Samples were subjected to thermal
demagnetization and some also to alternating field demagnetiza-
tion. High-temperature components from two field areas (55
samples) were identified with a positive fold test at 95% signifi-
cance level (k ¼ 4.9 in situ and 14.6 100% unfolded; a95 ¼
33.8 in situ and 18.1 100% unfolded, Torsvik et al. 1995).
The palaeogeography of Baltica as a whole has been constrained
from other sites in Baltica, although well-dated palaeopoles remain
few (see reviews in Bingen et al. 2005, Cocks & Torsvik 2005). At
c. 750 Ma, a southwards facing Finnmark-Kola region at c. 158S
formed the southern margin of Baltica (Hartz & Torsvik 2002).
This region lay adjacent to a roughly east –west-trending rift
(Timanian margin) that to the west linked with the north –south
trending rift/spreading axis developing between Baltica and Laur-
entia (Greenland). At 616 Ma, Baltica is thought to have been at
polar latitudes (758S) based on the well-dated Egersund dolerite
dykes in southern Norway (Bingen et al. 2005). By c. 550 Ma,
E. Finnmark was lying at c. 508S (Cocks & Torsvik 2005). In con-
trast, Cawood & Pisarevsky (2006) and Pisarevsky et al. (2008)
place Baltica in a more equatorial position (c. 308) although
whether it lay in the northern or southern hemisphere is unclear.
Other characteristics
No economic deposits have been reported in these rocks. Acri-
tarchs have been documented from the Vadsø, Ekkerøya,
 NEOPROTEROZOIC SMALFJORD AND MORTENSNES FORMATIONS
Tanafjorden and Vestertana Groups, although only reworked acri-
tarchs and vase-shaped fossils have so far been found in the Smal-
fjord and Mortensnes formations (Vidal 1981; Vidal & Siedlecka
1983). A limited Ediacaran fauna (Farmer et al. 1992; Crimes &
McIlroy 1999), dominated by discoidal forms (cf Gehling et al.
2000) and including Aspidella (Narbonne, pers. comm. 2008),
has been documented, the oldest forms occurring in the Innerelv
Member (Fig 57.2).
Geochronological constraints
No robust and high-resolution geochronological constraints are
available. Rb –Sr dating of shales associated with the glaciogenic
strata provides broad constraints for the deposition of the Smal-
fjord and Mortensnes formations (630 – 560 Ma; Pringle 1973;
Gorohkov et al. 2001).
Discussion
Føyn & Siedlecki (1980) interpreted the Lower Krokvatn Diamic-
tite as a massive indurated ground moraine and the Middle and
Upper Krokvatn diamictites as partly ground moraine and partly
deposited under water. Aeolian or fluvial processes deposited
the intervening sandstones into quiet lakes or marine basins in
the palaeovalley, while the red sediments above the middle diamic-
tite are likely loessites. Essentially, Føyn & Siedlecki (1980) inter-
preted the sequence as alternating glacial stadial and interstadial
events. However, Eyles (1993) argued that this model was pre-
sumptive, with the interpretation that the rocks are glacial in
origin being based on their stratigraphic position rather than any
unequivocal diagnostic glacial criteria, and thus the alternations
were simply presumed to represent cyclical glacial advances and
retreats, without substantive proof.
Bjørlykke (1967) and Edwards (1984) interpreted the Varan-
gerfjord succession as an infill of a glacially scoured valley, the
ice moving towards the NW (Laajoki 2003; Baarli et al. 2006).
Irregular glacier retreat left ice-cored moraines and basement
highs that were submerged by rising water levels (Levine et al.
2006). Bathymetric lows were filled first by sediment gravity
flows and overlain subsequently by rapidly prograding deltas and
sandur plains. Later diamictites may represent glacial advances
and/or sediment slumping.
At Oaibaččanjar’ga, Edwards (1975) suggested that the diamic-
tite overlying the striations formed as a meltout diamictite from
sediment-laden dead-ice, with the margins slumping and being
eroded as the ice-entrained sediment lost cohesion. In contrast,
Schermerhorn (1974), Jensen & Wulff-Pedersen (1996), Crowell
(1999) and Arnaud & Eyles (2002) argued that the diamictite
formed during slumping, possibly of glaciogenic deposits, with
some suggestions that the striations also formed during slumping
as the overriding diamictite was deposited; some of these authors
did not realize that the pavement is part of a major, deep regional
angular unconformity. The very thin breccia preserved in a number
of striations and as ridges beside striations on the pavement has
been ascribed to glacio-tectonic brittle deformation processes, pre-
served by the instantaneous recrystallization of the highly strained
comminuted material, associated with flash-heating of the sub-
strate, possibly up to c. 1700 8C, during rapid glacial movement.
This was likely associated with glacial earthquakes (Rice &
Hofmann 2000; Bestmann et al. 2006). The comminuted material
represents proto-rock-flour.
In the Tanafjord area, Edwards (1984) interpreted the Smalfjord
Fm. (equivalent to the Upper Krokvatn Diamictite of Føyn &
Siedlecki 1980; Fig. 57.2) as a cyclic deposit of glacial advances
and retreats, with the former yielding a basal massive lodgement
tillite, locally with sandstones deposited in situ by sub-glacial
meltwater at the ice margin. At the base of cycles, ice movement
599
resulted in the formation of deformation tillites, comprised solely
of deformed substrate sediments, while mixing of local and far-
travelled material formed banded tillites. Erosional bases of the
cycles are recorded at the outcrop-scale by deformation structures
and mixing with subjacent material and on a regional scale by the
absence of stratigraphy below diamictites. During retreat, poorly
sorted sand and silt with pebbles were deposited at the ice margin
by tractional underflows and gravity flows. In interglacial periods,
finer-grained sediments accumulated, occasionally with dropstones,
attesting to a glaciomarine environment. The dark silt-dominated
deposit at Grasdal was interpreted to be a loessite (Edwards 1979).
Hansen (1992), while agreeing that most of the diamictites
were of glacial origin, argued for a shelf depositional environment
dominated by undermelt- and flow-tillites, some heavily deformed
by an advancing glacier. Rapid vertical and lateral thickness vari-
ations and terminations of both tillites and associated facies were
interpreted to have been mostly associated with changes in glacial
movement directions and local bathymetrical conditions rather
than glacial erosion and lodgement. Glaciogenic mass-movement
tillites (flow-tillite) laterally grading into sandstones, and at other
stratigraphic levels grading into rhythmites, indicates transport
directions and the relative location of the basin/continent.
In contrast, Arnaud & Eyles (2002) suggested that the diamic-
tites in both the Gaessenjar’ga and the southern and western
shores of Varangerfjord accumulated from subaqueous gravity
flows from the basin margin, with the latter forming part of a
debris apron (see also Crowell 1999; Schermerhorn 1974; Jensen
& Wulff-Pedersen 1996). Thus, although the slumped material
may have been of glacial origin, seen in rare faceted and striated
clasts and lonestones, the diamictites were not directly deposited
by ice and so cannot be considered to be tillites. Instead, Arnaud
& Eyles (2002) suggested that they record deposition of unstable
sediments on the edge of an active extensional basin (but see
Edwards 2004; Arnaud & Eyles 2004), in which icebergs or
sea-ice contributed dropstones. Ice proximal settings were inferred
only for the eastern shore of Skjåholmen, where sediment gravity
flow diamictite is closely associated with well stratified glacioflu-
vial sandstones and conglomerates and at Handelsneset, where
complex deformation observed in conglomerates and sandstones
suggests active deformation by overriding ice (Arnaud 2008).
The proposed source areas, and thus sediment transport (ice or
otherwise) directions for the Smalfjord Fm., also vary. Føyn & Sie-
dlecki (1980) argue that the Middle and probably also the Lower
Krokvatn Diamictite were derived from the south. Arnaud &
Eyles (2002) also proposed a southerly source area for the probable
equivalent rocks at Vieranjarg’a in the Varangerfjord area.
Edwards (1984) inferred a westward ice flow along the Varanger
palaeovalley, with two dominant striation directions documented:
NW –SE and east – west (Bjørlykke 1967; Edwards 1975; Jensen &
Pedersen 1996; Rice & Hofmann 2000; Laajoki 2002). Edwards
(1984) also proposed that of the five advance/retreat cycles ident-
ified in the Tanafjord area, the first was southerly derived, the sub-
sequent three northerly derived and the last of unknown origin.
Hansen (1992), in contrast, suggested that the lower part of the suc-
cession was NE-derived and the latter part was SE-derived. These
differences are not simply due to the different order of the units
proposed by the two authors.
Member A of the Nyborg Fm. represents post-glacial transgres-
sion. Although Edwards (1984) suggested that the dolostones
formed around topographically high areas, correlation of the carbon-
ates with Marinoan cap dolostones (see below) suggests that depo-
sition was not directly related to palaeodepth (cf Hoffman et al.
2007). Variations in the clastic input certainly controlled which
lithology formed. Subsequently, Members B–D reflect a rapid dee-
pening of the basin (Fig. 57.3), followed by a more gradual shallow-
ing, with Member E representing a barrier lagoon facies, with
intermittent carbonate deposition (Edwards 1984).
The Mortensnes Fm. represents two advances and retreats of ice
(Edwards 1984); the first was initially derived both from the south
600 A. H. N. RICE ET AL.
(lower member) and later from the north (middle member).
The source area of the second advance (upper member), which
had a major change in provenance, is not known, although the
extreme negative d13C values of the dolomitic diamictite at
the top of the formation are thought to indicate derivation from
the underlying Member E of the Nyborg Fm. For both cycles, lod-
gement tillite was followed by floating ice, giving finer-grained
sedimentation and dropstones. The polymict conglomerate
draping the formation has been interpreted as a lag-conglomerate,
formed during local isostatic uplift, after glacial retreat (Edwards
1984). The succession passed essentially conformably upwards
into the post-glacial Lillevannet Member, representing fluvio-
deltaic and subsequent marine deposition.
Halverson et al. (2005) gave probable correlations of the glacio-
genic units in E. Finnmark with those documented elsewhere,
based on chemostratigraphy and carbonate lithologies. Dolostones
at the base of the Nyborg Fm. have negative d13C values (–1 to
–5.9‰), and, where massive, have sheet cracks, pseudo-tepees
and, at one locality, sulphate crystal fans. This, in combination
with their buff-weathering colour strongly suggests that this is a
Marinoan-type cap dolostone (cf Kennedy et al. 1998). d13C data
from the 230-m-thick Grasdalen Fm. in the Gulgofjord area
(Fig. 57.1) show positive values (6‰), falling rapidly to negative
values and then recovering to positive values below the
c. 20-m-thick Fadnuvag’gi Fm., which lies directly under the Smalf-
jord Fm. in the Grasdal area (Fig. 57.1). Halverson et al. (2005)
equated this d13C pattern with the Trezona anomaly, but since a
broadly similar anomaly occurs under the Port Askaig Fm., which
is no longer considered to be a Marinoan equivalent (McCay et al.
2006), this is not a robust constraint. Nevertheless, the character-
istics of the cap dolostone secure the Smalfjord Fm. as an equivalent
of the 635 Ma (Hoffmann et al. 2004) Marinoan glaciogenic
succession.
Thin dolostones in the upper part of the Nyborg Fm. (Member
E), with d13C values of –7.6 to –9.9‰, are likely correlatives
of the Wonoka anomaly, which elsewhere has d13C values as
low as –12.81‰ (VPDB: Le Guerroué 2006); this implies that
the almost immediately overlying Mortensnes Fm. is broadly a
correlative of the 580 Ma Gaskiers diamictite (cf Halverson
et al. 2005).
Kulling (1951) introduced the term Varangeristiden in a discus-
sion of diamictites in northern Sweden. Although those diamictites
are now correlated solely with the Mortensnes Fm. (cf. Stodt et al.
2011), the term originally included all the Neoproterozoic glacio-
genic rocks in E. Finnmark and implied one glacial period.
However, as the Smalfjord and Mortensnes formations are now cor-
related with two distinct glacial events (Halverson et al. 2005), one
of which was worldwide and the other only localized, the concept of
a Varangeristiden (Varangian/Varangerian ice age) is no longer
valid. Further, taking a 12 Ma duration for the Marinoan glaciation
(Bodiselitsch et al. 2005) and 1 Ma for the Gaskiers (Bowring et al.
2003) implies a total time span of 647–579 Ma for the two glacial
events, of which only 13 Ma (18%) was actually spent under ice;
scarcely enough to justify the term ‘ice age’. For these reasons,
use of the term Varang(er)ian ice age is here discouraged.
No evidence of an older, Sturtian equivalent, glaciation has been
recorded in this area. The Lille Molvika Fm. (Ekkerøya Group;
Fig. 57.2) is bounded by two regional unconformities (Rice 1994;
Siedlecka 1995), either of which may ‘hide’ a pre-Marinoan
glacial event. These unconformities are of post-800 Ma age, based
on correlations in the Manjunnas area (Fig. 57.1), where the
Ekkerøya Group overlies the Båtsfjord Fm. of the Barents Sea
Group (Rice 1994). d13C data from the 1400–1600-m-thick Båts-
fjord Fm. in the North Varanger Region (Fig. 57.1) are consistently
negative (Rice & Halverson, unpublished data), suggesting a corre-
lation with the c. 800 Ma old Bitter Springs anomaly (cf Halverson
et al. 2005).
In the North Varanger Region (Fig. 57.1), the Barents Sea (max.
9 km thick) and overlying Løkvikfjell (c. 5.7 km) Groups are
separated by a major angular unconformity (Johnson et al. 1978;
Siedlecki & Levell 1978); no glacial deposits occur in these
sequences. Biostratigraphic data (Vidal & Siedlecka 1983) are
too imprecise to estimate the age of the youngest rocks in the
Barents Sea Group (Moczydlowska-Vidal, pers. comm. 2007),
although it is younger than 800 Ma (see above). Mafic dykes
cutting the unconformity and basal part of the Løkvikfjell Group
were intruded at c. 577 + 14 Ma (Rice et al. 2004). Thus evi-
dence for glacial conditions recorded by the Smalfjord and Mor-
tensnes formations may not have been preserved in this basin.
A. H. N. R. thanks A. and J. Pettersen, G. Stensvold, P. Sørflaten, Ø. Hauge
and the extended Larsen family (Vestertana) for hospitality and boating assistance
during fieldwork in E. Finnmark, and C. and R. Hofmann and M. Ebner
for assistance in the field. M. B. E. thanks A. and J. Pettersen for their
hospitality; H. Reading, S. Føyn, S. Siedlecki, A. Siedlecka, S.-L. Røe,
T. Spencer, K. Bjørlykke and G. Boulton for scientific assistance in the field
and/or discussions; the New York State Higher Education Assistance Program
and family for financial and moral support. Other acknowledgements are given
in Edwards (1984). E.A. thanks the Natural Sciences and Engineering Research
Council of Canada for funding, and C. Trotter and S. Aspden for assistance in
the field. This represents a contribution of the IUGS- and UNESCO-funded
IGCP (International Geoscience Programme) Project #512.
References
Arnaud, E. 2008. Deformation in the Neoproterozoic Smalfjord For-
mation, Northern Norway: an indicator of glacial depositional con-
ditions? Sedimentology, 55, 335– 356.
Arnaud, E. & Eyles, C. H. 2002. Glacial influence on Neoproterozoic
sedimentation: the Smalfjord Formation, northern Norway. Sedimen-
tology, 49, 765– 788.
Arnaud, E. & Eyles, C. H. 2004. Glacial influence on Neoproterozoic
sedimentation: the Smalfjord Formation, northern Norway — reply.
Sedimentology, 51, 1423– 1430.
Baarli, B. G., Levine, R. & Johnson, M. E. 2006. The Late Neoproter-
ozoic Smalfjord Formation of the Varanger peninsula in northern
Norway: a shallow fjord deposit. Norwegian Journal of Geology,
86, 133– 150.
Banks, N. L., Edwards, M. B., Geddes, W. P., Hobday, D. K. &
Reading, H. G. 1971. Late Precambrian and Cambro-Ordovician
sedimentation in East Finnmark. Norges geologiske Undersøkelse,
269, 197– 236.
Bestmann, M., Rice, A. H. N., Langenhorst, F., Grasemann, B. &
Heidelbach, F. 2006. Subglacial bedrock welding associated with
glacial earthquakes. Journal of the Geological Society, 163, 417– 420.
Bingen, B., Griffin, W. L., Torsvik, T. H. & Saeed, A. 2005. Timing of
Late Neoproterozoic glaciation on Baltica constrained by detrital
zircon geochronology in the Hedmark Group, south-west Norway.
Terra Nova, 17, 250–258.
Bjørlykke, K. O. 1967. The Eocambrian ‘Reusch Moraine’ at Biggan-
jargga and the geology around Varangerfjord, Northern Norway.
Norges geologiske Undersøkelse, 251, 18 –44.
Bodiselitsch, B., Koeberl, C., Master, S. & Reimold, W. U. 2005.
Estimating duration and intensity of Neoproterozoic snowball glacia-
tions from Ir anomalies. Science, 308, 239–242.
Bowring, S., Myrow, P., Landing, E., Ramezani, J. & Grotzinger, J.
2003. Geochronological constraints on terminal Neoproterozoic
events and the rise of metazoans. Geophysical Research Abstracts,
5, 13219.
Cawood, P. A. & Pisarevsk, S. A. 2006. Was Baltica right-way-up or
upside-down in the Neoprotereozoic? Journal of the Geological
Society, London, 163, 753–759.
Chapman, T. J., Gayer, R. A. & Williams, G. D. 1985. Structural cross-
sections through the Finnmark Caledonides and timing of the Finn-
markian event. In: Gee, D. G. & Sturt, B. A. (eds) The Caledonide
Orogen — Scandinavia and Related Areas. J. Wiley & Sons, Chi-
chester, 593–610.
Cocks, L. R. M. & Torsvik, T. H. 2005. Baltica from the late Precambrian
to mid-Palaeozoic times: the gain and loss of a terranes identity. Earth
Science Reviews, 72, 39 –66.
 NEOPROTEROZOIC SMALFJORD AND MORTENSNES FORMATIONS
Crimes, T. P. & McIlroy, D. 1999. A biota of Ediacaran aspect from the
Lower Cambrian strata on the Digermul Peninsula, Arctic Norway.
Geological Magazine, 136, 633– 642.
Crowell, J. C. 1999. Pre-Mesozoic Ice-ages. Their bearing on Under-
standing the Climate System. Geological Society of America
Memoir, 192, 106.
Edwards, M. B. 1972. Glacial, interglacial and postglacial sedimentation
in a late Precambrian shelf environment. DPhil, Oxford University.
Available at http://www.marcedwards.com/publications_nnorway.
htm.
Edwards, M. B. 1975. Glacial retreat sedimentation in the Smalfjord
Formation, Late Precambrian, North Norway. Sedimentology, 22,
75 – 94.
Edwards, M. B. 1979. Late Precambrian glacial loessites from North
Norway and Svalbard. Journal of Sedimentary Petrology, 49, 85– 92.
Edwards, M. B. 1984. Sedimentology of the Upper Proterozoic glacial
record, Vestertana Group, Finnmark, North Norway. Norges geolo-
giske Undersøkelse Bulletin, 394, 1– 76.
Edwards, M. B. 2004. Glacial influence on Neoproterozoic sedimen-
tation: the Smalfjord Formation, northern Norway — discusssion.
Sedimentology, 51, 1409– 1417.
Edwards, M. B. & Føyn, S. 1981. Late Precambrian tillites in Finnmark,
North Norway. In: Hambrey, M. J. & Harland, W. B. (eds) Earth’s
Pre-Pleistocene Glacial Record. Cambridge University Press, Cam-
bridge, 606– 610.
Edwards, M. B., Baylis, P., Gibling, M., Goffe, W., Potter, M. &
Suthren, R. J. 1973. Stratigraphy of the ‘Older Sandstone Series’
(Tanafjord Group) and Vestertana Group North of Stallogaissa, Lak-
sefjord District, Finnmark. Norges geologiske Undersøkelse, 294,
25 – 41.
Eyles, N. 1993. Earth’s glacial record and its tectonic setting. Earth
Science Reviews, 35, 1 –248.
Farmer, J., Vidal, G., Moczydłowska, A., Strauss, H., Ahlberg, P. &
Siedlecka, A. 1992. Ediacaran fossils from the Innerelv Member
(late Proterozoic) of the Tanafjorden area, northeastern Finnmark.
Geological Magazine, 129, 181– 195.
Føyn, S. 1937. The Eo-Cambrian series of the Tana District, northern
Norway. Norsk Geologisk Tidsskrift, 17, 65– 164.
Føyn, S. & Glaessner, M. F. 1979. Platysolenites, other animal fossils,
and the Precambrian – Cambrian transition in Norway. Norsk Geolo-
gisk Tiddskrift, 59, 25 – 46.
Føyn, S. & Siedlecki, S. 1980. Glacial Stadials and Interstadials in
the Late Precambrian Smalfjord Tillite on Laksefjordvidda,
Finnmark, North Norway. Norges geologiske Undersøkelse, 358,
31 – 45.
Gayer, R. A. & Rice, A. H. N. 1989. Palaeogeographic reconstruction of
the pre- to syn-Iapetus rifting sediments in the Caledonides of
Finnmark, N. Norway. In: Gayer, R. A. (ed.) The Caledonide
Geology of Scandinavia. Graham & Trotman, London, 127– 139.
Gehling, J. G., Narbonne, G. M. & Anderson, M. M. 2000. The first
named Ediacaran body fossil, Aspidella terranovic. Palaeontology,
43, 427– 456.
Gorokhov, I. M., Siedlecka, A., Roberts, D., Melnikov, N. N. &
Turchenko, T. L. 2001. Rb –Sr dating of diagenetic illite in Neopro-
terozoic shales, Varanger Peninsula, northern Norway. Geological
Magazine, 138, 541– 562.
Halverson, G. P., Hoffman, P. F., Schrag, D. P., Maloof, A. C. &
Rice, A. H. N. 2005. Towards a Neoproterozoic composite carbon
isotope record. Geological Society of America Bulletin, 117,
1181– 1207.
Hambrey, M. J. 1983. Correlation of Late Proterozoic tillites in
the North Atlantic region and Europe. Geological Magazine, 120,
209–232.
Hansen, T. A. 1992. Sedimentologiske og Stratigrafiske Undersøkelser av
den Sen-Prekambriske Smalfjord formasjonen, Øst-Finnmark. Cand.
Real thesis, Universitet i Troms.
Harland, W. B., Armstrong, R. L., Cox, A. V., Craig, L. E., Smith,
A. G. & Smith, D. G. 1989. A Geological Time Scale 1989.
Cambridge University Press.
Hartz, E. H. & Torsvik, T. H. 2002. Baltica upside down: a new plate
tectonic model for Rodinia and the Iapetus Ocean. Geology, 30,
255– 258.
601
Hoffman, P. F., Halverson, G. P., Domack, E. W., Husson, J. M.,
Higgins, J. A. & Schrag, D. P. 2007. Are basal Ediacaran (635
Ma) post-glacial ‘cap dolostones’ diachronous. Earth and Planetary
Science Letters, 258, 114–131.
Hoffmann, K.-H., Condon, D. J., Bowring, S. A. & Crowley, J. L.
2004. U –Pb zircon date from the Neoproterozoic Ghaub
Formation, Namibia: constraints on Marinoan Glaciation. Geology,
32, 817– 820.
Jensen, P. A. & Wulff-Pedersen, E. 1996. Glacial or non-glacial origin
for the Bigganjarga tillite, Finnmark, northern Norway. Geological
Magazine, 133, 137– 145.
Johnson, H. D., Levell, B. K. & Siedlecki, S. 1978. Late Precambrian
sedimentary rocks in East Finnmark, North Norway and their
relationship to the Trollfjord-Komagelva Fault. Journal of the
Geological Society, London, 135, 517– 533.
Kennedy, M. J., Runnegar, B., Prave, A. R., Hoffman, K. H. &
Arthur, M. A. 1998. Two or four Neoproterozoic glaciations?
Geology, 26, 1059– 1063.
Kulling, O. 1951. Spår av Varangeristiden i Norbotten. Sveriges geolo-
giska Undersökning Årsbok 43 (1949), C 503, 1 –45.
Kumpulainen, R. A. 2011. The Neoproterozoic Lillfjället Formation,
southern Swedish Caledonides. In: Arnaud, E., Halverson, G. P.
& Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic
Glaciations. Geological Society, London, Memoirs, 36, 629– 634.
Kumpulainen, R. A. & Greiling, R. O. 2011. Evidence for late Neopro-
terozoic glaciation in the central Scandinavian Caledonides. In:
Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geo-
logical Record of Neoproterozoic Glaciations. Geological Society,
London, Memoirs, 36, 623–628.
Laajoki, K. 2001. Additional observations on the late Proterozoic Varan-
gerfjorden unconformity, Finnmark, northern Norway. Bulletin of the
Geological Survey of Finland, 73, 17 –34.
Laajoki, K. 2002. New evidence of glacial abrasion of the Late Protero-
zoic unconformity around Varangerfjorden, northern Norway.
Special Publication of the International Association of Sedimentolo-
gists, 33, 405– 436.
Laajoki, K. 2003. The Larajæg’gi outcrop — a large combined Neopro-
terozoic/Pleistocene roche moutonnée at Karlebotn, Finnmark,
North Norway. Norwegian Journal of Geology, 84, 107– 115.
Le Guerroué, E., Allen, P. A. & Cozzi, A. 2006. Chemostratigraphic
and sedimentological framework of the largest negative carbon isoto-
pic excursion in Earth history: the Neoproterozoic Shuram Formation
(Nafun Group, Oman). Precambrian Research, 146, 68 –92.
Levine, R., Baarli, G. & Johnson, M. E. 2006. Glacial and rocky-shore
dynamics of the northern Karlebotn monadnocks: late Neoprotero-
zoic of northern Norway. Canadian Journal of Earth Sciences, 43,
1215–1228.
McCay, G. A., Prave, A. R., Alsop, G. I. & Fallick, A. E. 2006. Glacial
trinity: Neoproterozoic Earth history within the British –Irish Caledo-
nides. Geology, 34, 909– 912.
Nystuen, J. P. & Lamminen, J. T. 2011 Neoproterozoic glaciation of
South Norway: from continental interior to rift and pericratonic
basins in western Baltica. In: Arnaud, E., Halverson, G. P. &
Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic
Glaciations. Geological Society, London, Memoirs, 36, 613–622.
Pisarevsky, S. A., Murphy, J. B., Cawood, P. A. & Collins, A. S. 2008.
Late Neoproterozoic and Early Cambrian palaeogeography: models
and problems. In: Pankhurst, R. J., Trouw, R. A. J., de Brito
Neves, B. B. & de Wit, M. J. (eds) West Gondwana: Pre-Cenozoic
Correlations Across the South Atlantic Region. Geological Society,
London, Special Publication, 294, 9–31.
Pringle, I. R. 1973. Rb –Sr age determinations on shales associated with
the Varanger ice age. Geological Magazine, 109, 465– 472.
Reading, H. G. 1965. Eocambrian and Lower Palaeozoic geology of the
Digermul Peninsula, Tanafjord, Finnmark. Norges geologiske Under-
søkelse, 234, 167– 191.
Reading, H. G. & Walker, R. G. 1966. Sedimentation of Eocambrian
tillites and associated sediments in Finnmark, North Norway. Palaeo-
geography, Palaeoclimatology, Palaeoecology, 2, 177– 212.
Reusch, H. 1891. Skuringsmærker og morængrus eftervist i Finnmarken
fra en periode meget ældre end “istiden”. Norges geologiske Under-
søkelse, 1, 78 –85 (97–100 English summary).
602 A. H. N. RICE ET AL.
Rice, A. H. N. 1994. Stratigraphic overlap of the late Proterozoic Vadsø
and Barents Sea Groups and correlation across the Trollfjorden–
Komagelva Fault, Finnmark, North Norway. Norsk Geologisk Tids-
skrift, 74, 48 –57.
Rice, A. H. N. & Townsend, C. 1996. Correlation of the late Precambrian
Ekkerøya Formation (Vadsø Group; E. Finnmark) and the Brennelv-
fjord Interbedded member (Porsangerfjord Group; W. Finnmark),
N. Norwegian Caledonides. Norsk Geologisk Tidsskrift, 76, 55– 61.
Rice, A. H. N. & Hofmann, Ch.-Ch. 2000. Evidence for a glacial origin
of the Neoproterozoic III striations at Oaibaččanjar’ga, Finnmark,
northern Norway. Geological Magazine, 137, 355– 366.
Rice, A. H. N. & Hofmann, Ch.-Ch. 2001. The transition from Neo-
proterozoic glacial to interglacial sedimentation near Hammarnes,
East Finnmark, North Norway. Norwegian Journal of Geology, 81,
257– 262.
Rice, A. H. N., Gayer, R. A., Robinson, D. & Bevins, R. E. 1989. Strike-
slip restoration of the Barents Sea Caledonides Terrane, Finnmark,
North Norway. Tectonics, 8, 247– 264.
Rice, A. H. N., Hofmann, Ch.-Ch. & Pettersen, A. 2001. A new sedi-
mentary succession from the southern margin of the Neoproterozoic
Gaissa Basin, south Varangerfjord, North Norway; the Lattanjaŕga
unit. Norsk Geologisk Tidsskrift, 81, 41 –48.
Rice, A. H. N., Ntaflos, T., Gayer, R. A. & Beckinsale, R. D. 2004.
Metadolerite geochronology and dolerite geochemistry from East
Finnmark, Northern Scandinavian Caledonides. Geological Maga-
zine, 141, 301– 318.
Røe, S.-L. 2003. Neoproterozoic peripheral-basin deposits in eastern
Finnmark, N. Norway: stratigraphic revision and palaeotectonic
implications. Norwegian Journal of Geology, 83, 259– 274.
Schermerhorn, L. J. G. 1974. Late Precambrian mixtites: glacial or
non-glacial. American Journal of Science, 274, 673–824.
Siedlecka, A. 1985. Development of the Upper Proterozoic sedimentary
basins of the Varanger Peninsula, East Finnmark. Geological Survey
of Finland Bulletin, 331, 175–185.
Siedlecka, A. 1987. Trollfjorden berggrunnskart 2336 3, 1:50,000,
foreløpig utgave. Norges geologiske Undersøkelse.
Siedlecka, A. 1990. Varangerbotn berggrunnskart 2335 3, 1:50,000, fore-
løpig utgave. Norges geologisk Undersøkelse.
Siedlecka, A. 1995. Neoproterozoic sedimentation on the Rybachi and
Sredni Peninsulas and Kildin Island, NW Kola, Russia. Norges
geologiske Undersøkelse Bulletin, 427, 52– 55.
Siedlecka, A. & Siedlecki, S. 1971. Late Precambrian sedimentary
rocks of the Tanafjord-Varangerfjord region of Varanger Peninsula,
Northern Norway. Norges geologiske Undersøkelse, 269, 246–294.
Siedlecka, A., Lyubtsov, V. V. & Negrutsa, V. Z. 1995. Correlations
between Upper Proterozoic successions in the Tanafjorden-
Varangerfjorden Region of Varanger Peninsula, northern Norway,
and on Sredni Peninsula and Kildin Island in the northern coastal
area of Kola Peninsula in Russia. Norges geologiske Undersøkelse
Special Publication, 7, 217– 232.
Siedlecka, A., Roberts, D., Nystuen, J. P. & Olovyanishnikov, V. G.
2004. Northeastern and northwestern margins of Baltica in Neopro-
terozoic time: evidence from the Timanian and Caledonian
Orogens. In: Gee, D. G. & Pease, V. (eds) The Neoproterozoic Tima-
nide Orogen of Eastern Baltic. Geological Society, London,
Memoirs, 30, 169–190.
Siedlecki, S. & Levell, B. K. 1978. Lithostratigraphy of the Late
Precambrian Løkvikfjell Group on Varanger Peninsula, East
Finnmark, North Norway. Norges geologiske Undersøkelse, 343,
73 – 85.
Stodt, F., Rice, A. H. N., Björklund, L., Bax, G. & Pharaoh, T. C.
2011. Evidence of late Neoproterozoic glaciation in the Caledonides
of NW Scandinavia. In: Arnaud, E., Halverson, G. P. & Shields-
Zhou, G. (eds) The Geological Record of Neoproterozoic Glacia-
tions. Geological Society, London, Memoirs, 36, 603–611.
Strahan, A. 1897. On Glacial Phenomena of Palaeozoic age in the
Varanger Fjord. Quarterly Journal of the Geological Society, 53,
137– 146.
Torsvik, T., Lohmann, K. C. & Sturt, B. A. 1995. Vendian glaciations
and their relation to the dispersal of Rodinia: palaeomagnetic con-
straints. Geology, 23, 727– 730.
Townsend, C., Roberts, D., Rice, A. H. N. & Gayer, R. A. 1986. The
Gaissa Nappe, Finnmark, North Norway: an example of a deeply
eroded external imbricate zone within the Scandinavian Caledonides.
Journal of Structural Geology, 8, 431–440.
Vidal, G. 1981. Micropalaeontology and biostratigraphy of the Upper
Proterozoic and Lower Cambrian sequences in East Finnmark, north-
ern Norway. Norges geologiske Undersøkelse, 362, 1 –53.
Vidal, G. & Siedlecka, A. 1983. Planktonic, acid-resistant microfossils
from the Upper Proterozoic strata of the Barents Sea region of Varan-
ger Peninsula, East Finnmark, Northern Norway. Norges geologiske
Undersøkelse, 382, 45– 79.
 Chapter 58

Evidence of late Neoproterozoic glaciation in the Caledonides of NW Scandinavia

F. STODT1, A. H. N. RICE2*, L. BJÖRKLUND3, G. BAX4, G. P. HALVERSON5,6 & T. C. PHARAOH7

1
Savignystrasse 9, 35037 Marburg, Germany
2
Vienna University, Structural Processes Group, Department of Geodynamics and Sedimentology, Althanstrasse 14,
1090 Vienna, Austria, Europe
3
Göteborg University, Earth Sciences Centre, Box 460, SE-405 30 Göteborg, Sweden
4
Department of Earth Sciences & Centre for Image Analysis, Uppsala University, SE-752 36, Uppsala, Sweden
5
Department of Geology & Geophysics, School of Earth & Environmental Sciences, University of Adelaide, North Terrace,
Adelaide, SA 5005, Australia
6
Present address: Department of Earth and Planetary Sciences, McGill University, 3450 University Street, Montreal,
Quebec, H3A 2A7, Canada
7
Geophysics & Marine Geosciences, British Geological Survey, Keyworth, Nottinghamshire, NG12 5GG, UK
*Corresponding author (e-mail: alexander.hugh.rice@univie.ac.at)

Abstract: The northwestern part of the Scandinavian Caledonides, formed by SE- to ESE-directed thrusting through the Neoproterozoic
W. Baltica continental shelf, contains numerous small and often isolated outcrops of diamictite and associated strata. No precise bio-
stratigraphic or isotopic data are available to constrain the age of these sediments, but, on the basis of their stratigraphic position,
most are correlated with the Mortensnes Formation (Fm.) in E. Finnmark and also presumed to be of glaciogenic origin. The Mortensnes
Fm. has been correlated with the 580 Ma Gaskiers glacial event on the basis of d13C isotope studies. Structurally, the deposits occur in the
Autochthon (below the Torneträsk Fm.), within an external imbricate zone (Lower Allochthon), within cover successions lying uncon-
formably on allochthonous basement (Window Allochthon) palaeogeographically derived from below or outboard of the Lower Alloch-
thon and, more rarely, within the Middle Allochthon, derived from outboard of the Window Allochthon. Evidence for a glaciogenic
origin is typically poor or lacking. Only in the Komagfjord Antiformal Stack (Window Allochthon), where an up to 40-m-thick succes-
sion of three fining upwards cycles has been mapped, are the deposits comparable in thickness and complexity to the Mortensnes Fm.
Other sequences are sometimes ,1 m thick and unconformably overlain by post-‘glacial’ deposits. The Vakkejokk Breccia, a submarine
slump in the Torneträsk area of the Autochthon closely underlies the correlative Precambrian –Cambrian lithostratigraphic boundary in
E. Finnmark but overlies the first appearance of the boundary marker fossil Treptichnus pedum. Although sometimes interpreted as peri-
glacial, this seems unlikely in view of the 30–508 palaeolatitude during deposition. Calcite nodules (,1 cm size) in the Vakkejokk
Breccia have previously been interpreted as glendonite, but the microstructure and palaeolatitude makes this unlikely; they are likely
a replacement of gypsum. Diamictites of uncertain origin have also been found in the Ediacaran Lower Siltstone Member of the Torne-
träsk Fm. and unconformably under the ?Lower Cambrian Lomvatn Fm. in the Komagfjord Antiformal Stack.

This chapter primarily covers (meta-)sediments correlated with the
Mortensnes Fm., the younger of the two glaciogenic units of
E. Finnmark, N. Norway (cf Rice et al. 2011). Such deposits
typically occur as small and isolated outcrops within both the
Autochthon and nappes of the Scandinavian Caledonides, often
resting unconformably on basement rocks (Kumpulainen 2011;
Kumpulainen & Greiling 2011; Nystuen & Lamminen 2011)
After restoration of the Caledonian nappes, these diamictite out-
crops indicate that glacial deposits covered an area of some
140 000 km2 in N. Scandinavia (including E. Finnmark).
Studies in Norrbotten, Sweden, in the southernmost part of the
area covered here (Fig. 58.1), led Kulling (1951) to introduce the
term Varangeristiden (Varang(er)ian glaciation) to encompass
both the glacial events described here and the earlier Smalfjord
Fm. (Marinoan). However, for reasons summarized below, this
term should no longer be used.
Structural framework
Restoration of the Scandinavian Caledonides, using balanced
cross-sections, demonstrates that the NNE –SSW-trending
(present orientation) continental shelf of W. Baltica comprised
an outboard basin lying adjacent to the continental edge and an
intermittently developed, somewhat shallower inboard basin
(Fig. 58.2; Gayer et al. 1987; Gayer & Greiling 1989; Rice
1998, 2001, 2006). In NE Scandinavia, this continental shelf
linked with the WNW –ESE-trending Timanian Basin (Siedlecka
1985; Gayer & Rice 1989; Siedlecka et al. 2004; cf Rice et al.
2011) that formed after the accretion of several minor terranes to
Baltica in the middle Neoproterozoic (cf Cocks & Torsvik
2005). The Timanian Basin has been divided into northerly deep
water and southerly shelf parts (Siedlecka et al. 1995, 2004), sep-
arated by the Trollfjorden – Komagelva Fault (cf Rice et al. 1989).
In Scandinavia, the shelf part comprises the Gaissa Basin, which,
in the west, was also an inboard basin of the NNE –SSW-trending
W. Baltica shelf.
Predominantly in-sequence ESE- to SE-directed Caledonian
shortening occurred mostly in Silurian – Devonian times. This
passed through the outboard basin (Middle Allochthon), then a
palaeogeographic basement-high (Window Allochthon, Rice
2001) and then the inboard basin, forming an external imbricate
zone (external Lower Allochthon), finally emplacing the nappe
pile onto the Autochthon. In some areas, the internal part of the
Lower Allochthon restores to above the Window Allochthon and
was imbricated earlier (Fig. 58.2). Both basins have essentially
the same lithostratigraphy; in most areas, the basement-high was
finally drowned in late Ediacaran to mid-Cambrian –Tremadocian
times, essentially contemporary with the mainland to the SE. In the
Tysfjord-Akkajaure area, both the Middle and Lower Allochthons

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 603– 611. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.58
604 F. STODT ET AL.

Fig. 58.1. Map of north Scandinavian Caledonides showing the distribution of glaciogenic lithologies and non-glaciogenic diamictites outside the E. Finnmark
(Laksefjordvidda-Tanafjord-Mortensnes) area. A, Altenes Window; An, Andabakoaivi; Ak, Akkajaure; AK, Alta-Kvænangen Window; Au, Autajaure Window;
B, Bulljovagge; G, Gæv’dnjajav’ri; H, Halkavarre; J, Jerta; K, Komagfjord Window; Ka, Karnjelajåkka; Ko, Kuokkel Window; Lv, Laksefjordvidda; Mn, Mauken
Window; P, Porsavatn; Po, Porsangerfjord; Pa, Paittasjärvi; Ri, Ritsemjaure Window; RS, Rombak-Sjangeli Window; Ru, Ruoddojokka; Si, Sitojaure; Sj, Stora
Sjöfallet; Sk, Skoganvarre; Sv, Sarvapakte; T, Torneträsk; Ty, Tysfjord; Vk, Vakkejåkka (now called Sarvájohka); Vu, Vuojtasrika.

are dominated by basement rocks, with only a thin cover succes-
sion (Björklund 1985, 1989), indicating that very restricted basin
development occurred in this region.
In the area covered here, much of the Middle Allochthon com-
prises pre-Caledonian basement (Björklund 1985, 1989; Kirkland
et al. 2006). Window Allochthon rocks are exposed in the cores
of large domal/periclinal structures, with the major units being
the Komagfjord Antiformal Stack (Komagfjord, Altenes & Alta-
Kvaenangen Windows combined), the Rombak-Sjangeli, and
adjacent Kuokkel Windows, and to the south, the relatively
small Autajaure and Ritsemjaure Windows (Fig. 58.1). Although
a basal thrust is typically not exposed, their allochthonous status
has been inferred from structural and metamorphic criteria
(Gayer et al. 1987; Anderson 1989; Bax 1989; Björklund 1989;
Rice 2001).
The Gaissa Thrust Belt (derived from the inboard, Gaissa
Basin), with 50% shortening in the west (Townsend et al. 1986,
1989), is the main development of the Lower Allochthon. On

Fig. 58.2. Schematic (V = H) WNW– ESE-oriented palinspastic cross-section showing the development of inboard and outboard basins in pre-diamictite times,
separated by a basement high. The basins and basement high were then buried under glacial and post-glacial deposits. The main tectonic units are numbered in the order of
their imbrication, with schematic main and minor thrusts. Localities along the top are given in the caption to Figure 58.1. Note that in many areas only one basin formed
and in some areas neither basin formed.
 THE CALEDONIDES OF NW SCANDINAVIA 605

Laksefjordvidda and to the east (Fig. 58.1), shortening decreases to Stratigraphy

c. 15% (Chapman et al. 1985). SW of Ruoddojokka, most of the
Lower Allochthon has been eroded away (cf. Hossack & Cooper
1986; Anderson 1989), although several relicts have been pre-
served (Fig. 58.1; Jerta Nappe, Rautas Complex, Lower Thrust
Complex – see below).
The deposits documented here are associated with condensed
sequences correlated directly or indirectly with the late Precam-
brian sediments exposed in E. Finnmark (predominantly the
Vestertana Group; cf. Rice et al. 2011; Fig. 58.3).

Fig. 58.3. Regional stratigraphic profiles. (a) Alta-Kvaenangen, S. Komagfjord and N. Komagfjord refer to where each part of the log-section was measured. In the
S. Komagfjord zone, Borras Group, Rafsbotn Fm. and Slettfjell Fm. are the three lithostratigraphic equivalents from the three tectonic windows (from south to north) in
the Komagfjord Antiformal Stack. The section shown is of the Slettfjell Fm. (Pharaoh 1985). (b) Section from Torneträsk (Thelander 1982; modified after Bax 1984 and
Stodt 1987). (c) E. Finnmark and W. Finnmark refer to the successions in the eastern (Johnson et al. 1978) and western (Williams 1976; Rice & Townsend 1996) parts of
the Gaissa Basin. Pl, Platysolenties antiquissimus; T, Treptichnus pedum; UE, LE, upper and lower limit of reported Ediacaran fauna; C, calcite nodules.
606 F. STODT ET AL.
West of Andabakoaivi (Fig. 58.1), the Caledonian Autochthon
comprises the Dividal Group, correlated with the post-Mortensnes
Fm. part of the Vestertana Group and overlying Digermul Group
in E. Finnmark (Fig. 58.3; Kulling 1964; Føyn 1967; Vogt 1967;
Thelander 1982). The constituent Torneträsk (c. 100– 260 m) and
overlying Alum Shale (c. 80 m) formations are locally separated
from the underlying pre-Caledonian basement (often a regolith;
Bax 1984; Stodt 1987) by small patches of diamictite (Fig. 58.1);
typically these are unnamed. The locally developed, but impor-
tant Vakkejokk Breccia (cf. Kulling 1964; Thelander 1982; Stodt
1987) cuts across the lower part of the Torneträsk Fm.
(Fig. 58.3). Although Kulling (1964) defined a Middle Sandstone
Fm. (; unit C of Vogt 1967), Thelander (1982), who downgraded
the rank of all the units from formations to members, merged this
into the Lower Siltstone Member (Fig. 58.3). (Note that the Vak-
kejokk Breccia type locality, Vakkejåkka, or Orddajohka on
some maps, is now called Sarvájohka; but this is not the same as
the Sarvajåkka, now Sarvágorsa, described by Thelander, 1982).
In the western part of the Gaissa Thrust Belt (external Lower
Allochthon; imbricated Gaissa Basin; Fig. 58.2; Townsend et al.
1986, 1989), the Tanafjord Group is thinner than in the type
area, although correlatives of the Grasdal and Lille Molvika for-
mations (Stabbursdal plus Porsanger formations and Brennelvfjord
Fm., respectively; Fig. 58.3) are thicker (cf. Rice & Townsend
1996). The poorly studied Airoaivi Group is thinner and lithologi-
cally different to the Vadsø Group, its possible chronostratigraphic
equivalent (Townsend et al. 1989; Rice & Townsend 1996). No
diamictite has been preserved from this part of the Gaissa Basin.
In the Jerta Fm. (Jerta Nappe; Lower Allochthon), Skjerlie &
Tan (1961; Fig. 58.1) recorded a predominantly clastic sequence
with a basal diamictite. This whole sequence, estimated at
c. 500 m thick, is likely a correlative of the Vestertana Group,
although the presence of pyritiferous black shales suggest it may
be partially correlated with the Digermul Group/Alum Shale
Fm. (Reading 1965; Thelander 1982). The successions in both
the Rautas Complex and the Lower Thrust Complex (near Torne-
träsk and Stora Sjöfallet, respectively; Fig. 58.1) have been
reliably correlated with the tectonically underlying autochthonous
Dividal Group (Björklund 1985; Bax 1989).
In the Komagfjord Antiformal Stack (Window Allochthon), the
cover sequence youngs northwards, with ,185 m of Bossekop
Group (; Tanafjord Group; Fig. 58.3; Føyn 1964, 1985) lying
unconformably on the basement in the Alta-Kvaenangen
Window (Fig. 58.1). The Bossekop Group is unconformably over-
lain by the Borras Group, correlated with the post-Nyborg Fm. part
of the Vestertana Group, including c. 10 m of basal diamictite
(Fig. 58.3; Føyn 1964, 1985). In the Altenes and southern part of
the Komagfjord Window, patchy outcrops of diamictite up to
40 m thick rest directly on the basement and are overlain by a
,150-m-thick sequence again directly comparable to the lower
part of the Torneträsk Fm. (Rafsbotn & Slettfjell formations;
Roberts & Fareth 1974; Pharaoh 1985) whereas in the northern
part of the Komagfjord Window a small patch of diamictite
(here termed the Porsavatn Diamictite Bed) is overlain by
,160 m of the Lomvatn Fm. (Pharoah 1985), broadly correlated
in this chapter with the upper parts of the Torneträsk Fm.
(Fig. 58.3).
Other successions within the area, from the Lower, Window and
Middle Allochthons, that overlie diamictites have been directly
correlated with the Dividal Group and have not been given local
stratigraphic names.
Glaciogenic and associated deposits
Autochthon
Near Skoganvarre, Siedlecka (1987; Fig. 58.1) recorded three out-
crops of polymict conglomerate within a 1km distance, comprising
1–50 cm clasts of amphibolite, gneiss and quartzite, lying below
the distinctive basal conglomerate of the Torneträsk Fm. To the
SW of Ruoddojokka, Holmsen (1956) noted a very variable stra-
tigraphy under the Torneträsk Fm.; in some places only occasional
lonestones are present at the base of the sequence, essentially lying
on the basement, whereas elsewhere 2–4 m of grey diamictite,
with 1 m clasts, is overlain by 0.5– 1m of carbonate-bearing
sandstone. At Bulljovagge, c. 2 m of red brown and grey diamictite
overlie quartzitic sandstone, but the contact with the basement
is not exposed (Holmsen 1957; Mathiesen in Skjerlie & Tan
1960). In these outcrops, the diamictite is often dominated
by clasts derived from the immediately underlying rocks. In the
Paittasjärvi – Sitojaure area and further south (Fig. 58.1), Kautsky
(1949) described undeformed chloritic diamictites (Sito diamic-
tite; Stromberg 1981) up to 4 m thick, with a dark matrix. The
unsorted, angular to poorly rounded blocks are up to cubic
metres in size and were derived from the underlying basement,
which shows no signs of pre-diamictite weathering. The matrix
is siliciclastic, comprising small quartz and felspar fragments; no
carbonate is present. Near Sitojaure, the diamictite is overlain by
.30 m of green mudstones with frequent blue quartzites and
these are probably overlain by the Dividal Group, which directly
overlies diamictites in nearby areas (Kulling 1951). Kulling
(1951) also recorded banded silts and clays in the Sitojaure area.
Other thin diamictites in this region were documented by
Kulling (1951; Fig. 58.1), some with faceted clasts, although it is
often unclear whether the outcrops documented belong to the
Autochthon or overlying Lower Allochthon (cf Strömberg 1981).
The Vakkejokk Breccia lies in the upper part of the Lower Silt-
stone Member of the Torneträsk Fm. on the NE side of Torneträsk
(Kulling 1964; Thelander 1982; Stodt 1987; Figs 58.1 & 58.3),
thinning from 3– 10 m (authors disagree on the thickness) in the
Vakkejåkka (Sarvágorsa) area to c. 0.5 m some 7.5 km to the
ESE. This locally cuts down through the Lower Sandstone
Member to the basement (Stodt 1987). Where the breccia is thin,
it may be overlain by sandstones of the uppermost part of the
Lower Siltstone Member (Thelander 1982). Stodt (1987) docu-
mented four breccia types.
† Type A (recorded only at Gaev’dnjajav’ri; Fig. 58.1): conglom-
erate with ,4-cm-sized, well-rounded dark-grey clay-silt clasts
and some quartz fragments, forming the base of a 12-cm-thick
BoumaABCD sequence lacking a structureless sandstone and
dewatering structures. This lies with an erosional contact on
the underlying thin-bedded sandstones of the Lower Siltsone
Member, c. 1 m below the base of the Middle Siltstone Member.
† Type B: angular to rounded granitic, vein-quartz, silt and shale
matrix-supported clasts up to 15 cm in size in sandstone beds up
to 30 cm thick. Platy clasts are bedding parallel. In some cases,
the beds show normal grading. Matrix sandstone grains are
rounded to well-rounded and predominantly quartz. The car-
bonate cement may have been derived from a carbonate-mud
matrix.
† Type C: normally .90% granitic clasts with minor elongate,
randomly oriented sedimentary clasts, often folded. The
former are predominantly clast-supported, equidimensional to
slightly elongate, mostly angular but sometimes well-rounded.
Intraclasts are platy and show abundant fold structures. The
matrix comprises grey, green and red silt/sandstone, often
internally brecciated. Two such flows have been recognized,
except at Vakkejåkka, where five flows occur. South of Sarva-
pakte (Sarvabakti; Fig. 58.1) one flow contains up to
1-cm-sized calcite nodules interpreted as glendonite by Stodt
(1987; but see ‘Discussion’). The flows usually lie conformably
or slightly erosively on the Lower Siltstone Member.
† Type D: granitic blocks 2– 100 m in size, lying either within
and above siltstones that are often folded and disturbed or on
type C breccias, along sharp boundaries, exposed for c. 4 km
east of Vakkejåkka. SE of Vakkejåkka, 2– 3 m clasts lie directly
 THE CALEDONIDES OF NW SCANDINAVIA
on the basement. Some breccia clasts, derived from earlier
flows, are also present (breccia-in-breccia structure). Granitic
blocks and clasts are basement-derived.
Lower Allochthon
No outcrops of diamictites have been recorded within the Gaissa
Thrust Belt in the Porsangerfjord area; exposures from areas
further east are reviewed in Rice et al. (2011). In the Jerta Fm.,
Skerlie & Tan (1960; Fig. 58.1) described a diamictite with
5–30 cm angular to rounded, unsorted clasts of light grey, yellow-
weathering dolomite, as well as greenstones and quartzites. The
matrix is dark to brownish-grey, with angular dolomite and
smaller poorly rounded quartz grains. Some finer, stratified and
possibly graded beds are also present. No thicknesses were
given, but photographs indicate a c. 2 m or greater thickness.
To the north of Sitojaure, in the Stora Sjöfället area, Kulling
(1948, 1951, 1982) reported a close association of thinly bedded
calcareous shales and diamictite or diamictite-like deposits with
lithologies similar to the diamictites in the Ritsemjaure Window
(see below).
Window Allochthon
In the Alta-Kvaenangen window, the most southerly part of the
Komagfjord Antiformal Stack, 185 m of quartzites and shales of
inferred shallow marine origin (Bossekop Group) are unconform-
ably overlain by c. 10 m of red-brown diamictite at the base of the
Borras Group, passing up gradually to conglomerates correlated
with the base of the Torneträsk Fm. and thence to sandstones
and shales (Føyn 1964, 1985; Fig. 58.3). Further north, in the
Altenes Window, Roberts & Fareth (1974) very locally found
c. 2 m of reddish-brown diamictite with clasts (,30 cm size;
30% of the rock) of mixed cover and basement lithologies, many
derived from the immediate substrate, lying on the basement.
Nearby, the Rafsbotn Fm., comprising a basal well-rounded
pebble-sized quartz conglomerate, c. 1 m thick, and then by
green and red mudstones, slates and siltstones, directly overlies
the basement. The Rafsbotn Fm. is equivalent to the basal part
of the Torneträsk Fm.
Along the southern rim of the Komagfjord Window (northern
part of the Komagfjord Antiformal Stack), outcrops of the essen-
tially undeformed Nyvoll Tillite Member (Slettfjell Fm.) are pre-
served as lenses up to 1km long and 44 m thick (although this
particularly thick outcrop thins to nothing in 300 m along strike;
Pharaoh 1980, 1985). Pharaoh (1980) described four facies, with
up to three fining-upwards cycles:
† Facies A: poorly sorted, unstratified polymict diamictites with
.20% clasts in a reddish-brown matrix, locally clast supported
at the base, where the clast content may rise to 40%. Most clasts
are ,5 cm across, but range up to 60 cm in size and vary from
angular to sub-rounded, and are sometimes facetted. A large
percentage was locally derived, although gneiss, granite and
quartzite clasts might have been externally derived. This
facies is up to 5 m thick and always forms the base of the
sequence, grading up into Facies B, although it reappears
higher in the sequence.
† Facies B: poorly sorted polymict diamictite similar to facies A,
but with significantly fewer (,20%) and smaller clasts
(c. 10 cm max.), locally with a carbonate matrix/cement
towards the top. Thickness varies from 1 to over 20 m.
† Facies C: poorly sorted, poorly stratified diamictite, gradational
from facies B by a continued decline in clast size (,1cm) and
content (,5%). ‘Bedding’ is defined by thin laminae of quart
grains in a reddish brown muddy matrix. This facies is gener-
ally thinly developed, although a thickness of 23 m occurs in
one section.
607
† Facies D: intercalated sandstones and mudstone in bands 1–
5 cm thick, usually reddish brown, although the sandstones
may be paler and greyish-green. Sandstones are frequently
graded, with microconglomeratic erosive bases, ripples, load
casts and intraformational mud-flake breccias. Impact pits are
present, as is evidence of soft-sediment folding and both
thrust and normal soft-sediment faulting.
Bedding in the diamictite is parallel to that in the overlying basal
conglomerate of the Vargelv Member (Fig. 58.3), indicating that
diamictite thickness variations reflect a syn-sedimentary uneven
basement surface; a palaeorelief, up to 3 m high, has been pre-
served at outcrops and an overall pre-diamictite topography of
40 m has been mapped. Pharaoh (1980) suggested that the
Nyvoll Member was deposited in a NW – SE-trending palaeovalley
c. 1km wide and up to 40 m deep, accounting for the sequence
having a thickness atypical outside the E. Finnmark area (cf Rice
et al. 2011). The northern margin of the valley is poorly defined
due to scarce exposure. Clasts in the diamictites comprise felds-
pathic metasandstones, jasper, vein quartz, metagabbro, serpenti-
nite and trondjemite, all locally derived; sometimes they can be
linked to nearby palaeo-topographic highs. These sediments
are overlain by a quartz conglomerate, with little topographic
relief at the contact, and then by red and green siltstones/shales,
comparable with the Borras Group/Rafsbotn Fm. to the south
(Fig. 58.3).
In the northern part of the Komagfjord Window, a
,0.5-m-thick, very poorly sorted matrix-supported diamictite
(Pharaoh 1985), here termed the Porsavatn Diamictite Bed, com-
prising 0.5 –3 cm angular clasts of vein quartz and quartzite in a
grey muddy matrix, underlies the 180-m-thick Lomvatn Fm. at
one area (Figs 58.1 & 58.3). Elsewhere, the 0.6–2-m-thick basal
conglomerate of this formation, consisting of well-rounded vein
quartz and quartzite pebbles, rests unconformably on the base-
ment and is overlain by sandstones and shales (Pharaoh 1985),
probably correlatives of the upper part of the Torneträsk Fm.
(Fig. 58.3).
The Gearbeljávri Fm., a thin sedimentary veneer that crops out
around the rim of the Rombak-Sjangeli and Kuokkel Windows and
on minor tectonic klippen within them, has also been correlated
with the basal part of the Torneträsk Fm. (Fig. 58.1, Brown &
Wells 1966; Tull et al. 1985; Bax 1989, 2001). In places, a basal
conglomerate lies at the basement –cover contact, but elsewhere
the contact grades from unweathered basement, through regolith
into quartzites. At Vuojtasrika, Brown & Wells (1966) documen-
ted 0.5-m-deep, diamictite-filled fissures in the basement, but
Bax (unpublished data) reinterpreted these as fault breccias.
Along the northern margin of the Rombak-Sjangeli Window,
Tull et al. (1985) described grey to white arkosic cross-bedded
sandstones, interlayered with conglomerates and pelitic schists,
locally overlain by diamictites with clasts of granitic, ?dioritic
and pelitic lithologies up to 25 cm in diameter. Coarsely graded
sequences are common. Current directions indicate flow to the
south to SSW, although the basement –cover contact is a palaeo-
peneplain.
At one outcrop in the SE part of the Kuokkel Window, a
1-m-thick diamictite is underlain by quartzites correlated with
Lower Sandstone Member (Torneträsk Fm.). This diamictite, which
lies c. 2 m above the basement –cover unconformity, although the
contact is not exposed directly here (Bax 1984), comprises angular,
unsorted, carbonate and sandstone clasts (no basement clasts) up to
20 cm in size, surrounded by a fine-grained matrix. No clast sorting
or preferred orientation has been observed. Upwards in the section,
above the overlying cross-bedded sandstones, ripple marks
are common in sandy layers interbedded with brown shales
(Bax 1984).
In the Ritsemjaure Window, within the Middle Allochthon
(Akkajaure Nappe Complex), vertical fissure fillings trending
0108 occur in mesoperthitic granite. The fissures have highly
608 F. STODT ET AL.
irregular margins and widths of up to 5 m, but are of unknown
depths (Fig. 58.1; Björklund 1989). The fissures are filled with
an unsorted sedimentary breccia, consisting of a dark grey, silty
matrix, chaotically mixed with all sizes (,1m) of angular clasts
from the host granite and very subordinate shale and dolomite
clasts. The same type of breccia overlies the basement granite,
but with smaller (,10 cm) clasts. Microscopic clasts reflect the
mesoperthitic feldspars of the subjacent granite. The breccias
grade upwards into dark conglomerates with up to pebble-sized
clasts and a weak clast size stratification and thence into dark
shales with ,5 cm lonestones. Local shearing within the diamic-
tite sequence makes thickness estimates uncertain, but the diamic-
tite lying on the granite is 5 –8 m thick and the overlying
conglomeratic to shaly part is c. 5 m thick in the least disturbed
areas. Overlying this, along a sharp contact, is a ,3-m-thick, typi-
cally blue-grey orthoquartzite with a basal quartz conglomerate
including porphyry clasts, correlated with the base of the Torne-
träsk Fm. (Björklund 1989).
Further west, on the NE side of the Tysfjord Culmination,
several local pockets of strongly sheared ,0.5-m-thick schistose
unsorted diamictites with granite clasts in a dark grey matrix have
been mapped with irregular contacts on the basement (Fig. 58.1;
Björklund 1989). These are overlain by c. 1-m-thick white to
grey quartzites and thence by overthrust rocks of the Lower
Allochthon.
Middle Allochthon
In the eastern part of the lowest imbricate of the Akkajaure Nappe
Complex (in Karnjelajåkka; Fig. 58.1), a 1–2-m-thick diamictite
with centimetre-sized angular mesoperthitic granite clasts in a
carbonate-bearing matrix rests on an uneven granitic basement
surface. The diamictite is sharply overlain by a c. 5-m-thick
conglomerate with sub-rounded to sub-angular quartz and less
common mesoperthite clasts fining upwards to gravelly quartzite,
followed by grey quartz phyllite, all correlated with the base of the
Torneträsk Fm. Westwards within this imbricate, a transition from
mesoscopic clast-rich diamictites to strongly weathered calcite-
dolomite mica schists occurs, overlain by quartz mylonites. Clast
size diminishes westwards. The schists have microscopic sub-
angular clasts of mesoperthite, quartz and albite similar to those
in the diamictite; chemically, the calcite-dolomite mica schists and
the diamictite matrix are very similar (Björklund 1989). On the
higher, more westerly derived thrust sheets, micaceous marbles
are interlayered with these schists.
Boundary relations with overlying and underlying
non-glacial units
The sediments (mostly diamictites) generally lie unconformably
on autochthonous or allochthonous Baltic Shield-derived base-
ment rocks and are unconformably overlain by the distinctive
quartz-rich basal conglomerate of the Torneträsk Fm. or
(deformed) correlatives. In some cases, a few metres of finer-
grained sediments lie between the diamictites and the basal con-
glomerate. The succession in the Komagfjord Antiformal Stack
differs in that in the south, around the Alta-Kvaengen Window,
the diamictites lie on a condensed sequence correlated with the
mid- to upper part of the Tanafjord Group (Fig. 58.3; Føyn 1964,
1985), whilst further north they lie on allochthonous basement
(Roberts & Fareth 1974; Pharoah 1985). The upper contact in
the Komagfjord Antiformal Stack is also atypical, in that it is
gradational with the Borras Group and Slettfjell formations,
correlatives of the Lillevatn Fm. (¼ base Torneträsk Fm.)
in E. Finnmark (Føyn 1964; Pharaoh 1980, 1985). The upper
contact of the Porsavatn Diamictite Bed is erosively overlain by
the Lomvatn Fm. (Fig. 58.3).
In contrast, the diamictite in the Kuokkel Window lies within
a sandstone/quartzite succession correlated with the Lower
Sandstone Member of the Torneträsk Fm., above the basal con-
glomerate. Rather similarly, the Vakkejokk Breccia rests with
a marked erosive unconformity on the underlying sediments of
the Torneträsk Fm., cutting down-section to, very locally, the
basement, and is conformably overlain by younger sediments
(Stodt 1987).
Chemostratigraphy
No stable isotope data are available due to the general lack of car-
bonates in the sequence. d13C values from calcite nodules in the
Vakkejokk Breccia range between –4.03 and 0.79‰ (VPDB;
mean, – 1.57‰, +1.38, N ¼ 11; crystals were drilled out and
analysed using the technique of Halverson et al. 2005).
Palaeolatitude and palaeogeography
Well-dated palaeopoles that can be used to constrain the Ediacaran
palaeogeogeography of Baltica are scarce (cf Torsvik et al. 1996;
Bingen et al. 2005; Cocks & Torsvik 2005). The southern rim of an
inverted Baltica lay at c. 158S at c. 750 Ma (Hartz & Torsvik 2002).
At 616 Ma, data from mafic dykes in southern Scandinvia indicate
that Baltica lay at 758S although by c. 550 Ma, the area covered
here lay at c. 45–508S (Cocks & Torsvik 2005). In contrast,
Cawood & Pisarevsky (2006) and Pisarevsky et al. (2008) place
Baltica in a more equatorial position (c. 308), although whether
it lay in the northern or southern hemisphere is unclear.
Geochronological constraints
No robust isotopic age constraints are available from this area.
Correlation of the sub-Dividal Group sediments with the
Mortensnes Fm. gives a broad age constraint of c. 580 Ma, by cor-
relation with the Gaskiers glaciation (Bowring et al. 2003).
Although it is likely that the Gaskiers event was diachronous,
the close association of the Mortensnes Fm. with extremely nega-
tive d13C values in Finnmark, correlated with the Wonoka anomaly
(Halverson et al. 2005; Rice et al. 2011), probably makes the effect
of diachroneity on the timing of glaciation relatively small.
The Vakkejokk Breccia is closely underlain (within a few
metres) by the fossil-rich Kullingia Beds, within which Treptich-
nus pedum has been recorded (Stodt 1987). Although Treptichnus
pedum was established as the index trace fossil for the base of the
Cambrian (Brasier et al. 1994), it has subsequently been found in
the youngest Ediacaran successions (Gehling et al. 2001). Thus the
Vakkejokk Breccia is constrained to very latest Precambrian to
early Cambrian times. This boundary lies somewhat below the
lithostratigraphic correlation with the proposed Precambrian –
Cambrian boundary from E. Finnmark (Fig. 58.3).
Discussion
In East Finnmark, the glaciogenic Smalfjord Fm. has a typical
Marinoan-type cap dolostone, forming the base of the Nyborg
Fm. (cf Halverson et al. 2005; Rice et al. 2011). Thin dolostones
in the upper part of the Nyborg Fm. (Member E; Edwards 1984;
Fig. 58.3) have d13C values of –7.6 and –9.9‰ (VPDB); these
have been correlated with the extreme negative d13C values
recorded in the Wonoka anomaly in other parts of the world
(down to – 12‰; cf. Halverson et al. 2005, Le Guerroué et al.
2006) to suggest that the almost immediately overlying Mort-
nesnes Fm. is broadly a correlative of the 580 Ma Gaskiers diamic-
tite (cf Halverson et al. 2005). Correlation of the sub-Tornetäsk
 THE CALEDONIDES OF NW SCANDINAVIA
Fm. diamictites and other ‘glaciogenic’ sediments in the present
area with the Mortensnes Fm. is based on the robust lithostrati-
graphic correlation of the Stappogiedde Fm. in the Vestertana
area with the Dividal Group in the Andabakaoaivi-Halkavarre
area (Føyn 1967) and from there to regions to the SW (Føyn
1964; Vogt 1967; Thelander 1982, and references cited above).
The limited palaeontological data (Platysolenites antiquissimus
Eichwald, Treptichnus pedum) support these correlations
(Kulling 1964; Hamar 1967; Føyn & Glaessner 1979; Stodt
1987; Crimes & McIlroy 1999; Fig. 58.3).
The outcrops documented here are often small and rather unin-
spiring. Some have little or no direct (reported) evidence of glacial
activity and it is their stratigraphic position, underlying the distinc-
tive quartz-conglomerate forming the base of the Dividal Group,
that has been used to infer a glacial (direct or very indirect)
origin. Exposures such as the polymict conglomerates near Sko-
ganvarre (Siedlecka 1987) are clearly of this type. Similarly, the
finer-grained, stratified deposits associated with diamictites in
the Jerta Fm. (Skerlie & Tan 1960) were inferred to be fluvioglacial
solely based on the presence of diamictites; however, these are also
an unreliable indicator of glaciogenic facies.
Criteria used for suggesting a glacial origin include faceted
clasts, impact pits (reflecting dropstones), soft-sediment folding
and faulting, taken to reflect glacial shear stresses, and the diamic-
titic composition and texture of the rocks. In the Komagfjord
Window, Pharoah (1980) recognized small roches moutonées
forming part of the 3-m-high palaeorelief within the 40-m-deep
palaeovalley. The presence of regular ‘varved’ sequences associ-
ated with diamictite has also been cited as evidence of glacial
activity (Kulling 1951). Although most of these criteria are no
longer regarded as being unequivocally diagnostic of glacial
activity, their co-occurrence, together with their stratigraphic
position, forms a reasonably compelling argument for a glacial
(sensu lato) origin; whether the diamictites are tillites or slightly
reworked material is unknown.
Within the lowest imbricate of the Akkajaure Nappe Complex
there is some evidence for a transition from coarse diamictites in
the ESE to finer-grained diamictites (now schists) in the WNW.
Both lithologies have comparable angular mesoperthitic clasts,
with clast sizes decreasing to the west, and similar geochemistries.
This has been inferred to reflect a transition from a proximal glacial
to a more distal, possibly glaciomarine, depositional environment
(Björklund 1989).
Kulling (1951) suggested that the Vakkejokk Breccia is a tillite,
but later studies indicated a mass-flow origin (cf Stodt 1987). A
peri-glacial environment has been proposed (Strömberg 1981;
Thelander 1982) and this seemed to be confirmed by the discovery
of glendonite in the breccia (Stodt 1987). Similarly, the breccia
clasts found in the Type D breccia (breccia-in-breccia structure)
were interpreted to indicate that the material was frozen during
transportation. Stodt (1987) also reported up to 20-cm-diameter
lonestones in the Lower Siltstone Member, under the Vakkejokk
Breccia, although no dropstones were seen. However, detailed
examination of the ‘glendonite’ does not show any evidence for
the 30% volume loss associated with the breakdown of ikaite.
Instead, each radial ‘arm’ comprises a single, slightly distorted
calcite crystal, with curved (strained) twin planes and radially
varying (undulose) extinction orientations. Essentially, the internal
microstructure seems too regular to be a replacement of ikaite and
is instead thought to represent replacement of an initial gypsum
nodule (Peckmann, pers. comm. 2011). The nodules are too irregu-
lar, partly due to bedding parallel pressure solution, to make a diag-
nostic determination of the crystal form. This non-glacial
reinterpretation is consistent with the palaeomagnetic data and
the lack of widespread glacial deposits in places nearer the poles
at that time.
The type A conglomerate in the Vakkejokk Breccia is likely a
rapidly accumulated mid-fan deposit. In the overlying breccias,
the parallelism of platy clasts indicates laminar flow and the
609
internal brecciation is indicative of rigid plugs, both typical of
debris flows (Stodt 1987). Stodt (1987) proposed that the breccia
formed due to the uplift and collapse of a palaeotopographic
high; based on the asymmentry of soft-sediment folds within the
breccia, this lay to the west, with east-directed slumping.
Similar, but slightly younger normal faulting has been documented
at the Caledonian front SW of Akkajaure (Hansen 1989).
The diamictites at the northern margin of the Rombak-Sjangeli
Window and in the SE corner of the Kuokkel Window also lie
within, rather than under, sediments correlated with the lower
part of the Torneträsk Fm. (Bax 1984; Tull et al. 1985). Although
it cannot be wholly discounted, it seems unlikely that these are
tillites. A more likely origin is that they are debris-flows, possibly
reworked nearby sub-Torneträsk Fm. tills or the commonly
reported regolith in the area (Stodt 1987; Bax 1989).
The most northerly part of the Komagfjord Antiformal Stack, a
palaeo-basement high, lay well above the Mortensnes post-glacial
sea level, with drowning occurring some time after it occurred in
more southerly parts of the high; glacial sediments deposited in
this area would thus have been strongly affected by erosion, with
a low preservation potential. The Porsavatn Diamictite Bed,
lying under a sharp erosional contact with the basal conglomerates
of the Lomvatn Fm. is the only potential relict of such deposits
found (Fig. 58.3) and, if of glaciogenic origin (sensu lato) is
most likely reworked (debris-flow), rather than primary.
Kulling (1951) introduced the term Varangeristiden (Varan-
g(er)ian ice age) in a discussion of diamictites in northern
Sweden. Although these deposits are now correlated solely with
the Mortensnes Fm., the term originally encompassed all the Neo-
proterozoic glaciogenic rocks of the Smalfjord and Mortensnes
formations in E. Finnmark (and elsewhere in Scandinavia). The
Smalfjord and Mortensnes formations have now been correlated
with the Marinoan and Gaskiers glacial events based on, respect-
ively, the development of a typical Marinoan cap dolostone and
a close association with extreme negative d13C values taken to
reflect the Wonoka anomaly (Halverson et al. 2005; Rice et al.
2011). One glacial event was worldwide in scope, while the
other was much more localized. Taking a 12 Ma duration of the
Marinoan glaciation (Bodiselitsch et al. 2005) and 1 Ma for the
Gaskiers (Bowring et al. 2003) implies a total time span of
647– 579 Ma for the two glacial events, of which only 13 Ma
(18%) were actually spent under ice, scarcely enough to justify
the term ‘ice age’. Further, Harland et al. (1989) used the term
for a Precambrian epoch. In view of the very confused implications
given by the term, we discourage the use of the term Varang(er)ian
in any sense.
In summary, the many isolated, but regionally persistent sub-
Dividal Group diamictites in northern Scandinavia, occurring in
the Autochthon and at all tectonic levels of the Caledonian
nappes derived from Baltica, testify to a mid-Ediacaran glaciation
broadly equated with the 580 Ma Gaskiers event (which was
probably diachronous). Lithological, structural and mineralogi-
cal characteristics suggest terrestrial as well as glaciomarine
depositional environments. These deposits covered an area of
140 000 km2 in the region considered. Applying the same struc-
tural model as used here to the whole orogen (cf Gayer & Greiling
1989; Rice 2006) indicates that patches of Mortensnes Fm. equiva-
lents (Kumpulainen 2011; Kumpulainen & Greiling 2011; Nystuen
& Lamminen 2011) cover a restored area of c. 780 000 km2.
A. H. N. R. thanks A. and J. Pettersen and U. and P.-Ø. Gjøvik for many years of
hospitality during fieldwork in Finnmark, and C. and R. Hofmann for assistance in
the field. T.C.P.’s contribution appears with the permission of the Director, British
Geological Survey (Natural Environment Research Council). D. Roberts and R.
Kumpulainen are thanked for providing information. The latter and G. Shields
are thanked for their reviews. G. Narbonne is thanked for confirming the identi-
fication of Treptichnus pedum in the Torneträsk Fm. This represents a con-
tribution of the IUGS- and UNESCO-funded IGCP (International Geoscience
Programme) Project #512.
610 F. STODT ET AL.
References
Anderson, M. W. 1989. Basement-cover evolution during Caledonian
Orogenesis, Troms, N. Norway. In: Gayer, R. A. (ed.) The Caledo-
nide Geology of Scandinavia. Graham & Trotman, London,
101– 110.
Bax, G. 1984. Geologie des Tornehamngebietes am Westufer des Torne-
träsk, Schwedisch-Lappland. Diploma thesis, University of Marburg.
Bax, G. 1989. Caledonian structural evolution and tectonostratigraphy in
the Rombak-Sjangeli Window and its covering sequences, northern
Scandinavian Caledonides. Norges geologiske Undersøkelse Bulle-
tin, 415, 87 –104.
Bax, G. 2001. Application of remote sensing techniques for the geologi-
cal mapping of tectonic klippen in the northern Scandinavian Caledo-
nides. In: Buchroithner, M. F. (ed.) A Decade of Trans-European
Remote Sensing Cooperation. Balkema, Rotterdam, 245– 249.
Bingen, B., Griffin, W. L., Torsvik, T. H. & Saeed, A. 2005. Timing of
Late Neoproterozoic glaciation on Baltica constrained by detrital
zircon geochronology in the Hedmark Group, SE Norway. Terra
Nova, 17, 250– 258.
Björklund, L. 1985. The Middle and Lower Allochthons in the
Akkajaure-Tysfjord area, northern Scandinavian Caledonides. In:
Gee, D. G. & Sturt, B. A. (eds) The Caledonide Orogen —
Scandinavia and Related Areas. J. Wiley and Sons, Chichester,
515– 528.
Björklund, L. 1989. Geology of the Akkajaure-Tysfjord-Lofoten tra-
verse, N. Scandinavian Caledonides. PhD thesis, Göteborg Univer-
sity Publ. A 59, ISSN 0348-2367.
Bodiselitsch, B., Koeberl, C., Master, S. & Reimold, W. U. 2005.
Estimating duration and intensity of Neoproterozoic snowball glacia-
tions from Ir anomalies. Science, 308, 239– 242.
Bowring, S., Myrow, P., Landing, E., Ramezani, J. & Grotzinger, J.
2003. Geochronological constraints on terminal Neoproterozoic
events and the rise of metazoans. Geophysical Research Abstracts,
5, 13219.
Brasier, M., Cowie, J. & Taylor, M. 1994. Decision on the Precam-
brian –Cambrian boundary stratotype. Episodes, 17, 95 –100.
Brown, B. R. & Wells, M. L. 1966. A contribution to the geology of the
Vassijaure-Sjangel area of Swedish Lapland. Geologiska Förenin-
gens i Stockholm Förhandlingar, 87, 527– 547.
Cawood, P. A. & Pisarevsky, S. A. 2006. Was Baltica right-way-up or
upside-down in the Neoprotereozoic? Journal of the Geological
Society, London, 163, 753–759.
Chapman, T. J., Gayer, R. A. & Williams, G. D. 1985. Structural cross-
sections through the Finnmark Caledonides and timing of the
Finnmarkian event. In: Gee, D. G. & Sturt, B. A. (eds) The Caledo-
nide Orogen — Scandinavia and Related Areas. J. Wiley and Sons,
Chichester, 593–610.
Cocks, L. R. M. & Torsvik, T. H. 2005. Baltica from the late Precambrian
to mid-Palaeozoic times: the gain and loss of a terranes identity. Earth
Science Reviews, 72, 39 –66.
Crimes, T. P. & McIlroy, D. 1999. A biota of Ediacaran aspect from the
Lower Cambrian strata on the Digermul Peninsula, Arctic Norway.
Geological Magazine, 136, 633–642.
Edwards, M. B. 1984. Sedimentology of the Upper Proterozoic glacial
record, Vestertana Group, Finnmark, North Norway. Norges geolo-
giske Undersøkelse Bulletin, 394, 1– 76.
Føyn, S. 1964. Den tillitførende formasjonsgruppe i Alta — en jevnføring
med Øst-Finnmark og med indre Finnmark. Norges geologiske
Undersøkelse, 228, 139– 150.
Føyn, S. 1967. Dividal-gruppen (HYOLITHUS-sonen) i Finnmarke
og dens forhold til de eokambrisk-kambriske formasjoner. Norges
geologiske Undersøkelse, 249, 1– 84.
Føyn, S. 1985. The Late Precambrian in northern Scandinavia. In: Gee,
D. G. & Sturt, B. A. (eds) The Caledonide Orogen — Scandinavia
and related areas. J. Wiley and Sons, Chichester, 233– 245.
Føyn, S. & Glaessner, M. F. 1979. Platysolenites, other animal fossils,
and the Precambrian –Cambrian transition in Norway. Norsk Geolo-
gisk Tidsskrift, 59, 25– 46.
Gayer, R. A. & Greiling, R. O. 1989. Caledonian nappe geometry in
north-central Sweden and basin evolution on the Baltoscandian
margin. Geological Magazine, 126, 499– 513.
Gayer, R. A. & Rice, A. H. N. 1989. Palaeogeographic reconstruction
of the pre- to syn-Iapetus rifting sediments in the Caledonides
of Finnmark, N. Norway. In: Gayer, R. A. (ed.) The Caledonide
Geology of Scandinavia. Graham & Trotman, London, 127–139.
Gayer, R. A., Rice, A. H. N., Roberts, D., Townsend, C. & Welbon, A.
1987. Restoration of the Caledonian Baltoscandian margin from
balanced cross-sections: the problem of excess continental crust.
Transactions of the Royal Society of Edinburgh: Earth Sciences,
78, 197– 217.
Gehling, J. G., Jensen, S., Droser, M. L., Myrow, P. M. & Narbonne,
G. M. 2001. Burrrowing below the basal Cambrian GSSP, Fortune
Head, Newfoundland. Geological Magazine, 138, 213– 218.
Halverson, G. P., Hoffman, P. F., Schrag, D. P., Maloof, A. C. & Rice,
A. H. N. 2005. Towards a Neoproterozoic composite carbon isotope
record. Geological Society of America Bulletin, 117, 1181–1207.
Hamar, G. 1967. Platysolenites antiquissimus Eich. (Vermes) from the
Lower Cambrian of northern Norway. Norges geologiske Undersø-
kelse, 249, 87 – 95.
Hansen, L. 1989. Age relationships between normal and thrust faults near
the Caledonian front at the Vietas hydropower station, northern
Sweden. In: Gayer, R. A. (ed.) The Caledonide Geology of Scandi-
navia. Graham & Trotman, London, 91 – 100.
Harland, W. B., Armstrong, R. L., Cox, A. V., Craig, L. E., Smith,
A. G. & Smith, D. G. 1989. A Geological Time Scale 1989.
Cambridge University Press, Cambridge.
Hartz, E. H. & Torsvik, T. H. 2002. Baltica upside down: a new plate
tectonic model for Rodinia and the Iapetus Ocean. Geology, 30,
255– 258.
Holmsen, P. 1956. Hyolithus-sonens basale lag i Vest-Finnmark. Norges
geologiske Undersøkelse, 195, 65 –72.
Holmsen, P. 1957. De eokambriske lag under hyolithussonen mellen
Čarajavvre og Časkias, Vestfinnmark. Norges geologiske Undersø-
kelse, 200, 47 – 50.
Hossack, J. R. & Cooper, M. A. 1986. Collision Tectonics in the
Scandinavian Caledonides. Geological Society of London, Special
Publications, 19, 287–304.
Johnson, H. D., Levell, B. K. & Siedlecki, S. 1978. Late Precambrian
sedimentary rocks in East Finnmark, North Norway and their
relationship to the Trollfjord–Komagelva Fault. Journal of the
Geological Society, London, 135, 517–533.
Kautsky, G. 1949. Eokambrische Tillitvorkommen in Norbotten, Schwe-
den. Geologiska Föreningens i Stockholm Förhandlingar, 71,
595– 603.
Kirkland, C. L., Daly, S. J. & Whitehouse, M. J. 2006. Provenance and
Terrane Evolution of the Kalak Nappe Complex, Norwegian Caledo-
nides: Implications for Neoproterozoic Palaeogeography and Tec-
tonics. Journal of Geology, 115, 21– 41.
Kulling, O. 1948. Om berggrunden i Sareks randområden. Geologiska
Föreningens i Stockholm Förhandlingar, 70, 661–672.
Kulling, O. 1951. Spår av Varangeristiden i Norbotten. Sveriges geolo-
giska Undersökning Årsbok 43 (1949), C 503, 1 –45.
Kulling, O. 1964. Översikt over norra Norrbottensfjällens kaledonberg-
grund (with English summary). Sveriges geologiska Undersökning,
Ba 19, 166.
Kulling, O. 1982. Översikt över södra Norrbottenfjällens Kaledonberg-
grund. Sveriges Geologiska Undersökning, Ba 26, 295.
Kumpulainen, R. A. 2011 The Neoproterozoic Lillfjället Formation,
southern Swedish Caledonides. In: Arnaud, E., Halverson, G. P.
& Shields-Zhou, G. (eds) The Geological Record of Neoprote-
rozoic Glaciations. Geological Society, London, Memoirs, 36,
629– 634.
Kumpulainen, R. A. & Greiling, R. O. 2011. Evidence for late Neopro-
terozoic glaciation in the central Scandinavian Caledonides. In:
Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geo-
logical Record of Neoproterozoic Glaciations. Geological Society,
London, Memoirs, 36, 623–628.
Le Guerroué, E., Allen, P. A. & Cozzi, A. 2006. Chemostratigraphic
and sedimentological framework of the largest negative carbon isoto-
pic excursion in Earth history: the Neoproterozoic Shuram Formation
(Nafun Group, Oman). Precambrian Research, 146, 68– 92.
Nystuen, J. P. & Lamminen, J. T. 2011. Neoproterozoic glaciation of
South Norway: from continental interior to rift and pericratonic
 THE CALEDONIDES OF NW SCANDINAVIA
basins in western Baltica. In: Arnaud, E., Halverson, G. P. &
Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic
Glaciations. Geological Society, London, Memoirs, 36, 613– 622.
Pharaoh, T. C. 1980. The geological history of the Komagfjord Tectonic
Window, Finnmark, Northern Norway. PhD thesis, University of
Dundee.
Pharaoh, T. C. 1985. The stratigraphy and sedimentology of autoch-
thonous metasediments in the Repparfjord-Komagfjord Tectonic
Window, west Finnmark. In: Gee, D. G. & Sturt, B. A. (eds) The
Caledonide Orogen — Scandinavia and Related Areas. J. Wiley
and Sons, Chichester, 347– 357.
Pisarevsky, S. A., Murphy, J. B., Cawood, P. A. & Collins, A. S. 2008.
Late Neoproterozoic and Early Cambrian palaeogeography: models
and problems. In: Pankhurst, R. J., Trouw, R. A. J., De Brito
Neves, B. B. & De Wit, M. J. (eds) West Canadian: Pro-Cenozoic
Correlations Across the South Atlantic Region. Geological Society,
London, Special Publication, 294, 9 – 31.
Reading, H. G. 1965. Eocambrian and Lower Palaeozoic geology of the
Digermul Peninsula, Tanafjord, Finnmark. Norges geologiske Under-
søkelsei, 234, 167– 191.
Rice, A. H. N. 1998. Stretching lineations and structural evolution of the
Kalak Nappe Complex (Middle Allochthon) in the Repparfjord-
Fægfjord area, Finnmark, N. Norway. Norsk Geologisk Tidsskrift,
78, 277– 289.
Rice, A. H. N. 2001. Field evidence for thrusting of the basement rocks
coring tectonic windows in the Scandinavian Caledonides; an
insight from the Kunes Nappe, Finnmark, Norway. Norsk Geologisk
Tidsskrift, 81, 321– 328.
Rice, A. H. N. 2006. Quantifying the exhumation of UHP-rocks in the
Western Gneiss Region, S. W. Norway: a branch-line — balanced
cross-section model. Austrian Journal of Earth Sciences, 98, 2 –21.
Rice, A. H. N. & Townsend, C. 1996. Correlation of the late Precambrian
Ekkerøya Formation (Vadsø Group; E. Finnmark) and the Brennelvf-
jord Interbedded member (Porsangerfjord Group; W. Finnmark,
N. Norwegian Caledonides. Norsk Geologisk Tidsskrift, 76, 55 –61.
Rice, A. H. N., Edwards, M. B., Hansen, T. A., Arnaud, E. & Halver-
son, G. P. 2011. Glaciogenic rocks of the Smalfjord and Mortensnes
Formations, Vestertana Group, E. Finnmark, Norway. In: Arnaud,
E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological
Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 593–602.
Rice, A. H. N., Gayer, R. A., Robinson, D. & Bevins, R. E. 1989. Strike-
slip restoration of the Barents Sea Caledonides Terrane, Finnmark,
North Norway. Tectonics, 8, 247– 264.
Roberts, D. & Fareth, E. 1974. Correlations of autochthonous stratigra-
phical sequences in the Alta-Reppparfjord Region, West Finnmark.
Norsk Geologisk Tidsskrift, 54, 123– 129.
Siedlecka, A. 1985. Development of the Upper Proterozoic sedimentary
basins of the Varanger Peninsula, East Finnmark. Geological Survey
of Finland Bulletin, 331, 175–185.
611
Siedlecka, A. 1987. Skoganvarre berggrunnskart 2034 4, 1:50,000.
Foreløpig utgave. Norges geologiske Undersøkelse.
Siedlecka, A., Lyubtsov, V. V. & Negrutsa, V. Z. 1995. Correlations
between Upper Proterozoic successions in the Tanafjorden-
Varangerfjorden Region of Varanger Peninsula, northern Norway,
and on Sredni Peninsula and Kildin Island in the northern coastal
area of Kola Peninsula in Russia. Norges geologiske Undersøkelse,
Special Publicationi, 7, 217– 232.
Siedlecka, A., Roberts, D., Nystuen, J. P. & Olovyanishnikov, V. G.
2004. Northeastern and northwestern margins of Baltica in Neopro-
terozoic time: evidence from the Timanian and Caledonian Orogens.
In: Gee, D. G. & Pease, V. (eds) The Neoproterozoic Timanide
Orogen of Eastern Baltic. Geological Society, London, Memoirs,
30, 169– 190.
Skjerlie, F. J. & Tan, T. H. 1960. The geology of the Caledonides of
the Reisa Valley area, Troms-Finnmark, Northern Norway. Norges
geologiske Undersøkelse, 213, 175–196.
Stodt, F. 1987. Sedimentologie, Spurenfossilien und Weichkörper-
Metazoan der Dividal Gruppe (Wendium-Unterkambrium) im
Torneträskgebiet/Nordschweden. PhD thesis, Phillips-University
Marburg-Lahn.
Stromberg, A. G. B. 1981. The Late Precambrian Sito tillite and the Vak-
kejokk breccia in the northern Swedish Caledonides. In: Hambrey,
M. J. & Harland, W. B. (eds) Earth’s Pre-Pleistocene Glacial
Record. Cambridge University Press, Cambridge, 611– 614.
Thelander, T. 1982. The Torneträsk Formation of the Dividal Group,
northern Swedish Caledonides. Sveriges geologiska Undersökning,
C 789, 1– 41.
Torsvik, T. H., Smethurst, M. A. et al. 1996. Continental break-up and
collision in the Neoproterozoic and Palaeozoic — a tale of Baltica and
Laurentia. Earth Science Reviews, 40, 229– 258.
Townsend, C., Roberts, D., Rice, A. H. N. & Gayer, R. A. 1986. The
Gaissa Nappe, Finnmark, North Norway: an example of a deeply
eroded external imbricate zone within the Scandinavian Caledonides.
Journal of Structural Geology, 8, 431– 440.
Townsend, C., Rice, A. H. N. & Mackay, A. 1989. The structure and stra-
tigraphy of the southwestern portion of the Gaissa Thrust Belt and
adjacent Kalak Nappe Complex, Finnmark, N Norway. In: Gayer,
R. A. (ed.) The Caledonide Geology of Scandinavia. Graham &
Trotman, London, 111–126.
Tull, J. F., Bartley, J. M., Hodges, K. V., Andresen, A., Stelten-
pohl, M. G. & White, J. M. 1985. The Caledonides in the
Ofoten region (68 –698N), north Norway: key aspects of tectonic
evolution. In: Gee, D. G. & Sturt, B. A. (eds) The Caledonide
Orogen — Scandinavia and Related Areas. J. Wiley and Sons,
Chichester, 553– 568.
Vogt, T. 1967. Fjellkjedestudier in den ostlige del av Troms. Norge
geologiske Undersøkelse, 248, 1 –59.
Williams, D. M. 1976. A revised stratigraphy of the Gaissa Nappe,
Finnmark. Norges geologiske Undersøkelse, 324, 63 –78.
 Chapter 59

Neoproterozoic glaciation of South Norway: from continental interior to rift

and pericratonic basins in western Baltica

JOHAN P. NYSTUEN* & JARKKO T. LAMMINEN

Department of Geosciences, University of Oslo, P.O. Box 1047 Blindern, NO-0316 Oslo, Norway
*Corresponding author (e-mail: j.p.nystuen@geo.uio.no)

Abstract: Neoproterozoic glacial deposits of South Norway comprise the Moelv and Koppang formations. The former occurs on
Baltican crystalline basement in autochthonous position at the Caledonian erosional nappe front, on basement windows and basement
thrust sheets in the Caledonian nappe region, and in thick sedimentary successions in the allochthonous Hedmark and Valdres rift
basins. The Koppang Formation (Fm.) occurs on top of platform carbonates in the allochthonous pericratonic Engerdalen Basin. The
glacial deposits are dominated by diamictite interpreted as basal till from warm-based grounded ice, whereas stratified successions of
diamictite beds, sandstone and laminated siltstone with outsized stones represent local ice-margin deposits and/or subglacially infilled
water bodies, and the final glaciomarine stage. Palinspastic reconstruction of Caledonian nappe complexes carrying the glacial for-
mations indicates that the glacial deposits were deposited over a wide area by a large western Baltoscandian ice sheet, probably
during the Gaskiers (c. 580 Ma) glacial event (or events), but the age of the glaciation in South Norway needs to be better constrained.

Neoproterozoic glacial deposits in South Norway have been
recognized since Holtedahl (1922) interpreted the ‘Moelv con-
glomerate’ by Lake Mjøsa as an ‘Eo-Cambrian’ tillite. The
‘Moelv tillite’, formally termed the Moelv Fm. (Bjørlykke et al.
1967), has since been recorded in a series of outcrops in autochtho-
nous position and in the Caledonian Osen-Røa Nappe Complex
and the Valdres Nappe Complex in the Sparagmite region of
central and eastern South Norway (the term sparagmite, from
Greek ‘sparagma’ meaning fragment, was introduced by Esmark
(1829) for the arkosic Neoproterozoic sandstones that dominate
the region). The Koppang Fm., a diamictite unit that has been cor-
related with the Moelv Fm., occurs in the Kvitvola Nappe Complex
of the northern part of the Sparagmite region (Bjørlykke &
Nystuen 1981) (Figs 59.1 & 59.2). The outcrops of both formations
are scattered and generally small, from a few to several hundreds of
square metres. The best exposures are present in road and railway
sections and along very steep cliff sides. The largest outcrop of the
Moelv Fm. measures c. 3 km in length along the western side of the
mountain Fonnåsfjellet (Fig. 59.2). In its type area, the Moelv Fm.
has recently been exposed in a series of new road sections and
bedrock surfaces in a residential area within the town of Moelv
(Nystuen 2008).
The Koppang Fm. (Törnebohm 1896, p. 31), being poorly
exposed in the Koppang type area, is best exposed in its hypostra-
totype in the Engerdalen area further to the east (Nystuen 1980).
The Moelv and Koppang formations were recorded during
regional mapping in the 1960s–1980s (e.g. Loesche & Nickelsen
1968; Nickelsen 1974; Siedlecka et al. 1987). Sedimentological
and stratigraphic studies of the Moelv Fm. were also performed
during this period (Bjørlykke 1966, 1969, 1974; Englund 1966,
1973a; Løberg 1970; Bjørlykke et al. 1976; Nystuen 1976a, b,
1985; Nystuen & Sæther 1979; Nystuen & Ilebekk 1981; Sæther
& Nystuen 1981; Siedlecka & Ilebekk 1982). The glaciogenic for-
mations were shown to have been deposited in four major principal
depositional settings: directly upon crystalline basement and in
three palaeobasins derived from the western marginal zone of Neo-
proterozoic Baltica, represented by the Moelv Fm. in the Hedmark
and Valdres basins and the Koppang Fm. in the Engerdalen Basin
(Nystuen 1985; Siedlecka et al. 2004). These palaeobasins,
together with similar allochthonous basins with Neoproterozoic
glacial formations in Sweden (Kumpulainen 2011; Kumpulainen
& Greiling 2011) are referred to the western Baltoscandian
basins (Kumpulainen & Nystuen 1985).
In the present review of the Neoproterozoic glacial deposits in
South Norway, main emphasis will be given to facies, stratigraphic
and structural position and implication for depositional extent,
palaeogeography, character of glaciation, correlation and age.
Geochronological and geochemical data given without any refer-
ences correspond to the authors’ own yet unpublished results
from provenance studies.
Structural framework
Structural setting of the Moelv Fm.
The Moelv Fm. rests with depositional contact on crystalline base-
ment rocks in three types of structural settings: (i) in autochthonous
position at the erosional nappe front at the eastern side of lake
Storsjøen, (ii) in thrust sheets within the Osen-Røa Nappe
Complex east of lake Storsjøen and along the river Mistra at the
northern end of lake Storsjøen and in the Valdres Nappe
Complex at Ormtjernskampen and (iii) in the Snødøla, Atnsjøen,
Øversjødalen and Tufsingdalen basement windows in the northern
part of the Sparagmite region (Siedlecka et al. 2004) (Figs 59.2 &
59.3). Basement underlying the diamictite in the Valdres Nappe
Complex consists of gabbroic rocks, whereas in the other settings
the crystalline basement rocks are granitic with minor bodies of
gabbro and dolerite (Siedlecka et al. 1987).
In the window structures, Precambrian basement rocks and
a succession of late Neoproterozoic –Ordovician rocks up to
c. 130 m in thickness occur beneath the Osen-Røa and Kvitvola
nappe complexes; the Moelv Fm. is the oldest unit in this cover
succession. The granitic rocks of the structural windows yield
U –Pb zircon ages of c. 1650 Ma, whereas similar rocks at the
nappe front have been dated to 1680 Ma. The granitic rocks
belong to the younger intrusives of the Transscandinavian
Igneous Belt (TIB) in central southern Fennoscandia (e.g. Heim
et al. 1996; Högdahl et al. 2004; Andersen et al. 2009). The
window structures in the northern part of the Sparagmite region
reveal the same lithology and stratigraphy as the autochthon at
the nappe front. For this reason the window structures have been
interpreted as domes of the underlying Baltica basement and its
sedimentary cover (Nystuen 1981). Morley (1986) suggested the
window structures to be allochthonous. This is supported by
regional studies of similar basement windows in the Scandinavian

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 613– 622. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.59
614 J. P. NYSTUEN & J. T. LAMMINEN

Fig. 59.1. Location of outcrop area (framed area) for the Neoprotereozoic glacial Moelv and Koppang formations in South Norway, with outcrops shown in the map
of Figure 59.2. Major geological crustal domains within the pre-glacial Fennoscandian basement are shown (modified from Bingen et al. 2008). The Caledonian
crustal segment Western Gneiss Region (WGR) comprises the northwesternmost coastal part of South Norway. The Cambrian– Devonian Caledonian domain and small
areas with Phanerozoic cover rocks in the Fennoscandian shield are not shown.

Caledonides (e.g. Gayer et al. 1987; Gayer & Greiling 1989; Gee
et al. 2010). In accordance with Morley (1986) and Rice (2005),
the Snødøla, Atnsjøen, Spekedalen, Øversjødalen, Tufsingdalen
and other window structures are here considered allochthonous,
in a tectonostratigraphic position beneath the Osen-Røa Nappe
Complex. The windows may thus represent structural elements
(horsts) of large duplex structures in nappe complexes of the
Lower and Middle Allochthon, with the Caledonian basal thrust
as the lower bounding surface, a thrust not being exposed in the
northern part of the Sparagmite region.

Fig. 59.2. Simplified geological map of the

Sparagmite region in South Norway,
showing outcrops of the Neoproterozoic
glacial Moelv and Koppang formations.
The Western Gneiss Region (WGR)
mentioned in the text is located NW of the
area covered by the map (Figs 59.1 & 59.4).
Map modified from Nystuen (1987).
 NEOPROTEROZOIC GLACIATION OF SOUTH NORWAY 615

Fig. 59.3. Representative examples of facies and stratigraphic context of the glacial Moelv and Koppang formations. The inset map covers the same area as the map of
Figure 59.2, in which localities are also shown. The Moelv Fm. in the Hedmark Group is in an autochthonous position in the Osen-Røa Nappe Complex and tectonic
windows: 1, Slemdalen (Sl); 2, Osdalen (Os); 3, Sjølisand (Sj); 4, Andrå (An); 5, Rena; 6, Moelv; 7, Åstdalen (Ås); 8, Kjølsjøberget (Kj); 9, Øvre Rendal (ØR);
10, Atnsjøen window (At); 11, Famphøgdene (Fa); 12, Prestkampen (Pr). Koppang Fm. in the Engerdalen Group of the Kvitvola Nappe Complex: 13, Hylleråsen (Hy);
14, Sjoa (Sjo). Equivalent to the Moelv Fm. in the Valdres Group of the Valdres Nappe Complex: 15, Ormtjernskampen East (OrE); 16, Ormtjernskampen West (OrW);
17, Mellene (Mel).

The second major structural setting of the Moelv Fm. is within
the thick rift basin successions of the Hedmark and Valdres basins
of the Osen-Røa and Valdres nappe complexes, respectively
(Fig. 59.2). The rift character of these allochthonous palaeobasins
is reflected by at least 5000 –6000-m-thick successions with large
lateral facies variations, from marginal alluvial or deep-marine
conglomerate fans to central basinal facies of either deep-marine
sandstone and shale or alluvial plain sandstone, and in the
Hedmark Basin also by tholeiitic basalt flows. The Moelv Fm.
occurs both in marginal and distal positions of the basin succes-
sions (Siedlecka et al. 2004).
In most settings where the Moelv Fm. rests on crystalline base-
ment, the primary features are well preserved with little distortion
of clasts and matrix of the diamictite, except close to overlying
thrust boundaries. In the Hedmark and Valdres basin successions,
deformation and metamorphism generally increase from south to
north. In high-strain zones along thrusts and shear zones, clasts
in the diamictite can be flattened and elongated parallel to foliation
in the matrix. The metamorphism is of lower greenschist facies.
Non-deformed diamictite is, however, very common in most
parts of the outcrop area of the Moelv Fm.
basin, the Engerdalen Basin (Siedlecka et al. 2004). In this
basin, sandstone, carbonate and diamictite formations have a
wide lateral extent. The Kvitvola Nappe Complex is located
above the Osen-Røa Nappe Complex and beneath the Upper
Allochthonous Trondheim Nappe Complex (Fig. 59.2).
Compared to the Moelv Fm., the Koppang Fm. is generally more
deformed. The formation is located very close to the thrust boundary
of the Kvitvola Nappe Complex. Together with the underlying
carbonate-shale formation, the diamictite of the Koppang Fm. has
acted as a detachment horizon during the Caledonian thrust move-
ments. In some areas, stretching and flattening of clasts makes the
identification of clast lithologies very difficult; in some thrust
zones the diamictite lithology is camouflaged by intense shear defor-
mation. The metamorphism is of greenschist facies with chlorite and
pale mica formed in the matrix. The diamictite of the Koppang Fm.
is best preserved in the eastern part of the Kvitvola Nappe Complex
in the Engerdalen district (Nystuen 1980) (Figs 59.2 & 59.3).
Stratigraphy and sedimentology of the Moelv Fm.

Moelv Fm. deposited directly on Precambrian basement

Structural setting of the Koppang Fm.
The Moelv Fm. is developed as brown to grey diamictite in all
The Koppang Fm., occurring in the Kvitvola Nappe Complex, is localities where the formation rests directly on granitic basement
interpreted to have been deposited in a wide pericratonic shelf rocks, in autochthonous position, in basement thrust sheets or in
616 J. P. NYSTUEN & J. T. LAMMINEN
structural windows (Figs 59.2 & 59.3). In these settings, the dia-
mictite unit is up to c. 20 m in thickness, structureless, unsorted
and with clasts up to about 1 m in diameter. The matrix is
mud-rich sandstone. The clasts are dominated by different types
of granitic rocks and rhyolitic porphyries; grey and red quartzite
and sandstone, dolerite and vein quartz are also common
(Nystuen & Sæther 1979; Nystuen & Ilebekk 1981). On the north-
ern side of the Atnsjøen window, the 20-m-thick diamictite is
dominated by granitic clasts of the same type as in the basement
below; however, gabbro, which is also a prominent basement
rock in this window structure, is not recorded among the clast
population (Siedlecka & Ilebekk 1982). The diamictite resting
on gabbro in Ormtjernskampen in the Valdres Nappe Complex
(Ormtjernskampen West, Fig. 59.2) is dominated by gabbro
clasts, but granite and quartzite clasts also occur in this
2-m-thick diamictite that is succeeded by laminated siltstone and
quartz arenite formations correlated with the Ekre and Vangsås
formations, respectively, in the Hedmark Group (Nickelsen
1974; Nickelsen et al. 1985).
A conglomerate rests on gabbro basement in the Feforkampen
outlier of the Valdres Nappe Complex (Fig. 59.2). According to
Englund (1973a), the gabbroic basement rock is overlain by a
thin quartz- and quartzite pebble conglomerate, followed by 5–
8-m-thick diamictite with clasts up to 1 m across of quartz diorite,
granite and anorthosite. The present authors visited the Feforkam-
pen locality during the summer of 2008 and found this unit to be a
matrix- to clast-supported conglomerate dominated by quartzite
boulders of up to c. 1.5 m.
The basement –diamictite boundary in outcrops east of Lake
Storsjøen and in the windows in the northern part of the Sparag-
mite region is a smooth unconformity with relief of up to some
few metres. Parallel striations with direction from ESE to WNW
are present on the ‘polished’ granite surface below the diamictite
in the hill at Andrå east of Lake Storsjøen. The diamictite is pre-
served as erosional remnants, in thickness varying from some
few centimetres (Tufsingdalen) to 5–20 m (Atnsjøen, Spekeda-
len and Øversjødalen) (Figs 59.2 & 59.3). At some localities,
diamictite is overlain by lower Cambrian quartz arenite (Vangsås
Fm.) and greenish grey shale, followed by Middle Cambrian
black alum shale.
The uniform lithology of the diamictite, the polished and striated
crystalline substratum and the wide regional extent of the
Fig. 59.4. Palaeogeographical
reconstruction of South Norway at the time
of the Neoproterozoic glaciation, showing
the Western Baltoscandian Ice Sheet and the
Hedmark and Valdres rift basins and the
pericratonic Engerdalen Basin, together
with the present outcrop area of the Moelv
and Koppang glacial formations in the
Sparagmite region. Note the reference
locality of autochthonous diamictite of the
Moelv Fm. at the Baltica basement east of
Lake Storsjøen. The TIB domain and the
Gothian and Sveconorwegian domain are
explained in terms of geological age in
Figure 59.1. The Western Gneiss Region
(WGR) also discussed in the text comprises
the present northwestern coastal regions of
South Norway. The Engerdalen Basin was
located west of the pre-Caledonian position
of WGR.
 NEOPROTEROZOIC GLACIATION OF SOUTH NORWAY
diamictite unit are features favouring an origin as basal till of a
continental ice sheet. The clast assemblage is dominated by grani-
tic and porphyritic rock types common for the TIB and associated
sandstones, quartzites and dolerites to the east and NE of the
present outcrop region for the Moelv diamictites. In the Valdres
Nappe Complex the grounded ice sheet has picked up local
bedrock types. The clast-supported conglomerate resting on top
of the gabbro in Feforkampen is lithologically quite different
from the other diamictites and a glacial origin is here
considered uncertain.
Moelv Fm. in the Hedmark Basin
The Moelv Fm. rests with depositional contact within the Hedmark
Group on (1) coarse-grained fan-delta sandstones of the Ring Fm.,
(2) carbonate platform and shale beds of the Biri Fm., (3) fluvial
sandstones of the Rendalen Fm., (4) shallow-marine sandstones
of the Atna Fm. and (5) basalt flows of the Svarttjørnkampen
Fm. (Figs 59.2 & 59.3). Except for some deep marine sub-basins,
most of the Hedmark Basin was filled up before the onset of the
glaciation, in the east with at least 4000-m-thick alluvial conglom-
erates and fluvial sandstones (Rendalen Fm.) and in the west by
more than 4000 m of marine turbidites and black shale (Brøttum
Fm.), submarine conglomerate fans (Biskopåsen Fm.) and coarse-
grained fan-delta accumulations (Ring Fm.). Local basalt flows of
the Svarttjørnkampen Fm. on top of the Rendalen Fm. and within
and on top of the Atna Fm. in the northern and eastern part of the
basin are interpreted as being related to fissure eruptions (Nystuen
1982, 1987; Furnes et al. 1983). Basalt clasts in marine canyon
infill conglomerates of the Biskopåsen Formation located at the
top of the Brøttum Fm. in the western part of the basin indicate
that local volcanic eruptions may have taken place repeatedly at
several places in the rift basin (Siedlecka et al. 2004). The
pre-glacial basin infill culminated with the formation of carbonate
platforms, and black mud in adjacent deep sub-basins (Biri Fm.).
Acritarchs in the upper part of the Brøttum Fm. and in the Biri
Fm. indicate a late Riphean to early Vendian age (Cryogenian to
Ediacaran) (Vidal & Nystuen 1990).
The lower boundary of the Moelv Fm. is a distinct erosional
unconformity where the formation’s basal lithology is diamictite.
A sand-filled wedge in underlying fluvial Rendalen sandstone is
interpreted as a fossil ice-wedge (Nystuen 1976b). At localities
where the formation rests on thick marine shales of the Biri Fm.,
the lower boundary appears gradual from shale to laminated grey
silt-rich mudstone with scattered outsized stones of the Moelv Fm.
The Moelv Fm. usually varies in thickness from 1 to 2 m up to
150 m. The laminated facies with outsized stones overlying the
Biri shale is thinnest. The most frequently encountered thickness
is in the order of 10 –15 m, and the dominant lithology is brownish,
greenish or grey diamictite. In many localities, the diamictite
passes upwards into laminated mudstone with outsized stones, a
facies interval that may be up to c. 2 m in thickness. An upwards
decrease in clast content marks the transition of this facies into
the overlying Ekre Fm., a laminated green or red siltstone
without any outsized stones.
The thickness increases to 100–150 m in Fonnåsfjell, Slemda-
len and Brennåsen (Figs 59.2 & 59.3). In the eastern hillside of
the mountain Fonnåsfjell, the formation consists of structureless
to faintly stratified light grey diamictite with clasts up to c. 1 m
in diameter. In Slemdalen and Brennåsen, the c. 150-m-thick
Moelv Fm. consists of structureless and poorly stratified brown
diamictite beds, stratified sandstone beds with scattered pebbles
and boulders, and laminated siltstone with outsized stones up to
c. 1 m in diameter.
The clast (larger than 2 cm) content of the diamictite facies
ranges from c. 7.5 to 30% (Nystuen 1976a). The extrabasinal
clast suite in eastern and northern outcrops is very similar to that
of the Moelv diamictite located directly on crystalline basement
617
rocks. Recent radiometric datings and isotopic studies (U – Pb
and Lu –Hf in zircons) of the igneous clast material have confirmed
that the TIB, or its equivalents in terms of age and petrogenetic
history, contributed significantly to the clastic material in the
eastern Hedmark Basin with ages ranging from 1770 Ma to 1010
Ma, most being c. 1670 Ma. Only one clast with an age of
c. 1540 Ma has been found, which means that clasts of Late
Gothian (c. 1590 –1520 Ma; Åhäll & Connelly 2008) age are
rare in the Moelv Fm. In the western part of the Hedmark Basin,
the clast assemblage also includes fine-grained quartzite, grey
and pale granite, gneisses and amphibolite (Løberg 1970;
Englund 1973a). Many Sveconorwegian (Grenvillian) c.960 Ma
granitoid clasts occur in the western and southwestern parts of
the basin (Lamminen et al. 2009). Clasts corresponding to
typical rock types dominating in the Gothian-Sveconorwegian
basement adjacent to the southernmost outcrops of the Moelv
Fm., such as coarse-grained red granodiorites, various types of
gneisses, amphibolites and gabbros (Nordgulen 1999), are not
present in the diamictite. Most clasts of crystalline rock types are
angular to sub-rounded, whereas sandstone, quartzite and vein
quartz clasts of pebble size are skewed towards the sub-rounded
to well-rounded range (Nystuen 1976a).
The lithology of formations in the Hedmark Group directly
underlying the Moelv Fm. is reflected in the composition of the
diamictite close to the lower boundary. The matrix is enriched in
coarse-grained sand particles, granules and pebbles of quartz and
feldspar when resting upon the coarse-clastic Rendalen and Ring
formations. The basal zone of the diamictite carries small clasts
of limestone and basalt at sites where Biri limestone and basalt
are the underlying stratigraphic units, respectively. Limestone
clasts are also abundant at localities where limestone is not directly
underlying the diamictite, as at the type locality in Moelv. This
may be due to the presence of a carbonate platform of the Biri
Fm. that was located along the western margin of the Hedmark
Basin at the time when the diamictite was formed (Løberg 1970;
Bjørlykke et al. 1976). Clasts of siltstone and sandstone show
glacial striations (Bjørlykke 1974; Nystuen & Sæther 1979).
The diamictite deposits of the Moelv Fm. are unsorted to poorly
sorted and display grain size curves very similar to those of Pleis-
tocene diamictons. Fabric studies in the eastern part of the
Hedmark Basin reveal an east – west-oriented peak of preferred
long-axis orientation that has been inferred to represent an original
depositional fabric formed by subglacial westward transport of
debris (Nystuen 1976a).
The structureless diamictite is interpreted as tillite sensu
stricto, formed as basal till beneath a grounded ice sheet, as also
suggested for the similar type of diamictite in settings directly
on basement. The siliciclastic beds in the Rendalen and Ring for-
mations, directly underlying the Moelv Fm., may have been poorly
consolidated at the time of glaciation, thus supplying clastic grains
to the till matrix.
Successions consisting of structureless and faintly stratified dia-
mictite beds, stratified sandstone and laminated siltstone with out-
sized stones have been interpreted as ice-margin deposits, formed
at sites where water depth was close to the buoyancy depth of the
ice sheet (Nystuen 1976a). Another possible depositional environ-
ment for these localized stratified successions may have been small
subglacial water bodies. Laminated siltstone with outsized stones
within successions of mixed lithology at the top towards the over-
lying Ekre Fm. is thought to represent glacial mud with ice-
dropped debris formed in an iceberg zone. This facies probably
formed during the final stage of glaciation, maybe during eustatic
sea-level rise.
Moelv Fm. in the Valdres Basin
In the Valdres Basin, the best outcrops of diamictite correlated
with the Moelv Fm. are in the Ormtjernskampen area (Figs 59.2 &
618 J. P. NYSTUEN & J. T. LAMMINEN
59.3). In addition to the diamictite resting on gabbroic basement
rock, described above from the outcrop in the western part of the
Ormtjernskampen area, a phyllitic diamictite unit also occurs in
the eastern part of the area on top of the 800-m-thick Ormtjerns-
kampen Fm. of alluvial conglomerate and below a sandstone
unit correlated with the Vangsås Fm. in the Hedmark Group (Nick-
elsen 1974; Nickelsen et al. 1985) (Ormtjernskampen East,
Fig. 59.2). This diamictite contains the same clast lithologies as
the diamictite resting on the gabbroic crystalline basement in the
outcrop Ormtjernskampen West (Fig. 59.2).
At Mellene (Figs 59.2 & 59.3), a 0.5–3-m-thick diamictite, cor-
related with the Moelv Fm., occurs on top of the 3000-m-thick
Olefjell Fm. of arkosic sandstone. As with the Moelv Fm. in the
Hedmark Basin, the diamictite is succeeded by siltstone and sand-
stone, possible correlatives to the Ekre and Vangsås formations,
and Cambro-Ordovician strata (Loeschke & Nickelsen 1968).
The diamictite unit recorded within the rift basin succession
in the Valdres Basin reveals similar lithologies and stratigraphic
characteristics to those of the Moelv Fm. in the Hedmark Basin,
which supports the interpretation of this diamictite unit as a
glacial deposit. The stratigraphic position of the Moelv Fm. in
the Valdres Basin reflects the geological setting of glaciation in
a continental rift basin with crystalline basement forming the
basin margins, together with coarse-clastic marginal alluvial
fans, and very thick fluvial accumulations in the central part, all
settings in which glacial diamictite was deposited. Thus, in the
Ormtjernskampen area, the glacial diamictite was deposited on
top of an alluvial fan succession and on crystalline basement,
and these two sites were likely separated by a fault. In the
Mellene area, the glacial formation was deposited in a more
central position of the rift basin.
Stratigraphy and sedimentology of the Koppang Fm.
The Koppang Fm. occurs in the lower part of the Engerdalen
Group in the Kvitvola Nappe Complex. The best outcrops of the
succession are in the Engerdalen area (Fig. 59.2).
The formation is a 5–15-m-thick diamictite and rests with
primary depositional contact on a carbonate unit, the Hylleråsen
Fm. (Fig. 59.3). The Hylleråsen Fm. is 15–30 m thick and consists
predominantly of dolomite with thin siltstone laminae, but also con-
tains magnesite with chert nodules (Nystuen 1969), black shale and
laminated fine-grained sandstone, locally with dolomite pseudo-
morphs after gypsum crystals (Nystuen 1980). A unit of shallow-
marine sandstone locally occurs below the carbonate formation.
The diamictite in the Koppang Fm. is structureless, except close
to its upper boundary at some localities, where it may be weakly
laminated. Its lower boundary is erosional, and the diamictite is
enriched with dolomite clasts (up to 90% in the basal zone), with
the largest clast having a diameter of 80 cm. Extrabasinal clasts
are sub-angular to sub-rounded red and grey granite, with boulders
up to 1 m in diameter, rhyolitic porphyry, red sandstone, light grey
quartzite and vein quartz pebbles. The clast assemblage resembles
that of the Moelv Fm. in the eastern part of the Hedmark Basin. The
diamictite is overlain by medium-grained, well-sorted sandstone
inferred to be of shallow-marine origin (Nystuen 1980). The
upper contact towards the overlying sandstone formation is
lithologically sharp and often tectonically deformed.
The carbonate–diamictite–sandstone succession can be traced
along the sole of the Kvitvola Nappe Complex for c.140 km from
west to east (Figs 59.2 & 59.3). The Hylleråsen Fm. at its base is
commonly cut by the sole thrust of the nappe complex and has
obviously acted as a detachment horizon during emplacement of
the Kvitvola Nappe Complex. The carbonate–diamictite–sandstone
tripartite succession is frequently strongly tectonized and deformed.
A glacial origin is inferred for the Koppang Fm. from its diamic-
tite lithology and clast content, erosional lower boundary with
clasts derived from underlying formation, its laterally wide
extent and setting within a pericratonic basin (Nystuen 1980;
Kumpulainen & Nystuen 1985; Siedlecka et al. 2004). The
facies association in the underlying Hylleråsen Fm. implies that
this carbonate unit was formed during warm-water to tropical con-
ditions in a carbonate platform setting including evaporitic lagoo-
nal sub-environments favourable for magnesite formation, anoxic
sub-basins and sabkhas.
Chemostratigraphy
No data are available from the glacial units themselves. The
distribution of major and some trace elements through the
Hedmark Group and the overlying Cambro –Silurian succession
was presented by Englund (1973b) and Bjørlykke & Englund
(1979). The stratigraphic resolution for the interval containing
the Moelv Fm. is too low for chemostratigraphic correlation.
Tucker (1983), in his study of the Biri Fm., published stable
isotope analyses with an average d18O of –11.4‰ and d13C of
þ1.4‰ relative to PDB, indicating that the limestone was
formed primarily from aragonite in seawater of normal compo-
sition. The data do not give any stratigraphic trend that can be
correlated by isotope curves obtained from other Neoproterozoic
carbonate successions.
Palaeolatitude
No data are available.
Palaeogeography
Pre-Caledonian position of the Hedmark, Valdres and
Engerdalen basins
The Hedmark and Valdres rift basins and the pericratonic
Engerdalen Basin were formed at the western Baltoscandian
margin (present coordinates) (Siedlecka et al. 2004; Nystuen
et al. 2008). Their pre-Caledonian structural setting and relative
position to the Baltoscandian craton is crucial for the interpretation
of the extent and character of the Neoproterozoic glaciation of
western Baltica. Palaeogeography of the glaciated and deposi-
tional area during the Neoproterozoic glaciation in South
Norway can first be inferred after palinspastic restoration of the
allochthonous successions in the Osen-Røa, Kvitvola and
Valdres nappe complexes.
Oftedahl (1943) first quantified the shortening of the southern-
most part of the Osen-Røa Nappe Complex, represented by the
Cambro –Silurian decollement succession now preserved in the
Oslo Graben, and concluded that the succession in the northern
part of the lake Mjøsa area had been dislocated c. 150 km to the
SSE during the Caledonian orogeny. Nystuen (1981) showed
that the Osen-Røa Nappe Complex, riding on top of the window
structures in northern part of the Sparagmite region, had to be
restored to a position NW of the windows, giving a minimum
thrust distance of Osen-Røa Nappe Complex rocks at the erosional
nappe front within the Mjøsa-Trysil sector (Fig. 59.2) of c. 140 km.
Morley (1986) restored hanging- and footwall cut-offs within
the Osen-Røa Nappe Complex and obtained a minimum thrust dis-
tance of 130 km of rocks in the nappe complex at the northern part
of Lake Mjøsa, about the same shortening as quantified by Ofte-
dahl (1943). Morley (1986) restored the window structures to a
position NW of the restored position of the Hedmark Basin,
outside the present coastline of mid-Norway. This restoration
model was further developed by Rice (2005), who included the
window structures in a hypothetical ‘Jostedal platform’ of
Baltica basement northwestward of the Hedmark Basin.
However, the thrust position of the Osen-Røa Nappe Complex
 NEOPROTEROZOIC GLACIATION OF SOUTH NORWAY
above the structural windows and their sedimentary cover succes-
sion shows that the windows must be restored to a position craton-
ward of the Hedmark Basin, and not the opposite as proposed in the
models of Morley (1986) and Rice (2005).
Hossack et al. (1985) restored the package of thrust sheets in the
Valdres area to the NW by c. 315 km, west and NW of the original
position of the Hedmark Basin. The Kvitvola Nappe Complex with
the Engerdalen Basin, located on top of the Osen-Røa Nappe
Complex, has been displaced an even greater distance, probably
in the order of 400 –600 km, from the NW to the SE (Siedlecka
et al. 2004). The Caledonian nappe pile in South Norway is
located on top of the Western Gneiss Region of the southern Scan-
dinavian Caledonides (Fig. 59.1). Consequently, the nappe com-
plexes in discussion were likely derived from positions west of
the Western Gneiss Region. In the gneiss region, Precambrian
basement rocks with overlying and infolded Neoproterozoic to
lower Palaeozoic sedimentary cover successions and Caledonian
thrust sheets are strongly deformed and metamorphosed. Rice
(2005) restored the rocks of the Western Gneiss Region to the
NW as one single crustal block close to the position of the
Valdres Basin and west of the Hedmark Basin. However, the het-
erogeneous structural architecture of the Western Gneiss Region
indicates that the displacement history of this region and its
cover of sedimentary successions and thrust sheets are very com-
plicated (Tucker et al. 2004), and accordingly difficult to
restore palinspastically.
It is beyond the scope of this paper to discuss various palinspas-
tic models of the southern Scandinavian Caledonides. However,
irrespective of models, it must be concluded that during Neoproter-
ozoic times when the Moelv and Koppang formations were depos-
ited, a rather well-denudated Baltica craton extended several
hundred kilometres further west of the outcrop east of lake Stors-
jøen where the Moelv diamictite is located autochthonously on
striated basement rocks, a reference locality of crucial importance
for any reconstruction of the palaeogeography of Neoproterozoic
glaciation in South Norway. The Hedmark and Valdres rift
basins and the Engerdalen shelf basin were located in outboard
positions of the denuded Baltica.
Geochronological constraints
No radiometric age data are available to unequivocally constrain
the ages of the Moelv and Koppang formations. Rankama (1973)
reported a Rb – Sr whole-rock age of 612 + 18 Ma on an argilla-
ceous sample from the Ekre Fm., overlying the Moelv Fm. in the
western part of the Hedmark Basin. Bingen et al. (2005) identified
a 677 + 15 to 620 + 14 Ma old detrital zircon U – Pb age popu-
lation in the Rendalen Fm. at Hanestad in the eastern part of the
Hedmark Basin (Fig. 59.2). This is a very uncommon age range
within Fennoscandian crystalline rocks, and the source for these
zircon grains remains enigmatic. Within the range of this age popu-
lation, a granite in the allochthonous Seve Nappe in northern
Sweden gave a U –Pb titanite age of 637 + 3 Ma; the granite
was considered to be related to terrane accretion in the western-
most part of Baltica during the Timanian orogeny (Rehnström
et al. 2002). Recent detrital zircon U –Pb analyses from the upper-
most parts of the Brøttum Fm. have yielded a wide range of Neo-
proterozoic ages. The same c. 620 Ma population also shows up in
the Brøttum Fm. Because the Brøttum Fm. underlies the Moelv
Fm., this confirms the findings of Bingen et al. (2005) that the
Moelv Fm. appears to be younger than c. 620 Ma.
Hannah et al. (2007) reported a c. 560 Ma Re –Os age for a black
shale in the Biri Fm., beneath the Moelv Fm. in the Mjøsa area.
However, the same authors also reported a c. 300 Ma age for the
same formation at another locality further NE. The latter age
was explained by thermal disturbance in the Re – Os isotopic
system due to the Permian Oslo rift. These authors later investi-
gated scatter in their Re –Os data and related it to oxidation
619
processes (Hannah et al. 2008). Because the Re –Os system is sus-
ceptible to disturbances that are difficult to detect, unlike the U – Pb
system in zircon, the present authors think that a robust age for the
Moelv Fm. still remains to be established (see below).
Discussion
The western Baltoscandian ice sheet
According to the tectonostratigraphic setting of the Moelv and
Koppang glacial formations and the palinspastic restoration of
the Caledonian nappe complexes carrying the Neoproterozoic
basins (see above), the area covered with glacial ice must have
had a width of at least 400–600 km from SE to NW, measured
from the autochthonous reference outcrop of the Moelv Fm. at
Storsjøen to the restored position of the Engerdalen Basin
at the Baltoscandian margin (Figs 59.2 –59.5). This western
Baltoscandian ice sheet also covered carbonate platforms of the
Risbäck rift basin and the pericratonic Tossåsfjället Basin along
the Baltoscandian margin (Kumpulainen 2011; Kumpulainen &
Greiling 2011).
Clast lithologies of the Moelv and Engerdalen formations indi-
cate that the glacial ice streams were sourced from areas located
within the TIB and the Sveconorwegian domain in the central,
and the western to south-western part of Fennoscandia, respect-
ively. The Hedmark and Valdres rift basins probably formed
along zones of weakness between these crustal blocks (Lamminen
et al. 2009). Uplifted shoulders of the rift basins may have given
rise to locally glaciated mountain ridges (Figs 59.4 & 59.5).
Ice streams that originated in highland areas must have coalesced
into extensive ice sheets that moved across lowland regions, now
represented by denuded Baltica basement and the carbonate
platform of the Engerdalen Basin, both types of substratum
being covered by glaciogenic diamictite. Striations on the granitic
surface beneath diamictite east of lake Storsjøen demonstrate the
warm-based character of the glacier ice at the time. Lithoclasts
and the textural and mineralogical composition of basal tills are
generally dominated by the local substratum. This is also revealed
by the concentration of carbonate clasts in the lower part of the
Koppang Fm. where the diamictite rests directly on dolomite,
and by basalt clasts in diamictite of the Moelv Fm. at sites where
this formation has been deposited upon basalt of the Svarttjørn-
kampen Fm. (Figs 59.2 & 59.3). The absence of Gothian gneisses,
which are typical in the autochthonous basement, among clasts in
the Moelv Fm. at Mjøsa is thus in accordance with the allochtho-
nous position of the glacial formation in the area.
The sedimentological character of the carbonate units under-
lying the Moelv and Koppang formations, the Biri and Hylleråsen
formations, respectively, indicates dominating warm and arid
climatic conditions before the onset of glaciation. The presence
of fossil ice wedges in alluvial sediments beneath the Moelv Fm.
shows that periglacial climatic conditions prevailed in the area in
front of the advancing ice sheet.
Correlation and age
The Moelv and Koppang formations are generally correlated with
the Mortensnes Fm. in Finnmark, the younger of the two Neopro-
terozoic glacial units traditionally referred to the ‘Varangerian
glacial period’ in northern Norway, and to the Långmarkberg
and Lillfjället glacial formations in Sweden (e.g. Bjørlykke &
Nystuen 1981; Kumpulainen & Nystuen 1985; Siedlecka et al.
2004; Nystuen et al. 2008). Halverson et al. (2005) correlated
the Mortensnes Fm. with the Gaskiers glaciation and the older
glacial Smalfjord Fm. with the Marinoan glaciation, from d13C
profiles obtained from carbonate deposits associated with
the two formations. The Marinoan glacial units are dated to
620 J. P. NYSTUEN & J. T. LAMMINEN
Fig. 59.5. Simplified and composite section across the western Baltoscandian craton and the Hedmark and Valdres rift basins and the pericratonic Engerdalen Basin,
showing the Western Baltoscandian Ice Sheet from the cratonic interior in the SE to the Baltoscandian Margin in the NW, with local areas of glaciation along mountain
ridgers along the rift basins. Note the reference locality of autochthonous diamictite of the Moelv Fm. at the Baltica basement east of Lake Storsjøen.
c. 635 Ma (Hoffmann et al. 2004) and the Gaskiers and Squantum
glacial deposits of Avalonia to c. 580 Ma (Myrow & Kaufman
1999; Thompson & Bowring 2000; Bowring et al. 2003; Thomp-
son et al. 2007). The Marinoan glacial episode may have lasted
at least 3 million, and most likely 12 million, years (Bodiselitsch
et al. 2005). Duration and variation in age for the Gaskiers
glacial event or events are also very uncertain.
On the basis of d13C values in cap carbonates of the youngest of
two diamictite units in Svalbard, Halverson et al. (2004) correlated
the Wilsonbreen Fm. with the Marinoan glacial event and the
inferred global extent of this glaciation, in accordance with
the Snowball Earth hypothesis (Hoffman & Schrag 2002). In the
British-Irish Caledonides, three stratigraphic levels of diamictite
have been correlated as equivalents to the c. 700 Ma Sturtian,
c. 635 Ma Marinoan and c. 580 Ma Gaskiers glacials, respectively
(McCay et al. 2006).
Bingen et al. (2005) suggested that the Moelv Fm. corresponds
to the Gaskiers glacial event, constrained by the 620 + 14 Ma det-
rital zircon age obtained from the Rendalen Fm. at Hanestad. The
Lillfjället Fm. in the Tossåsfjället Basin, Sweden, an alleged cor-
relative to the Moelv Fm. (Nystuen et al. 2008), is penetrated by
the Ottfjället dolerite dyke swarm (Kumpulainen 1980, 2011).
This dyke swarm is generally considered to be the result of pre-
Iapetus rifting at the western Baltoscandian craton, contempora-
neous with the intrusion of the 616 + 3 Ma Egersund dolerite
dyke swarm (Bingen et al. 1998) in southwestern Norway (e.g.
Siedlecka et al. 2004; Nystuen et al. 2008). There are presently
several data and arguments favouring a mid-Ediacaran (possible
Gaskiers equivalent) age for Neoproterozoic glaciation in South
Norway and Sweden, represented by glacial diamictite formations.
The glacial formations are located in the uppermost part of
Neoproterozoic successions, with the Moelv Fm. above a strati-
graphic level in the uppermost part of the Brøttum Fm. that con-
tains c. 620 Ma zircon grains, and are overlain by Ediacaran
and/or Lower Cambrian formations. The glacial deposits also
lack cap carbonates, typical of most Gaskiers glacial deposits
according to Halverson et al. (2005).
If all glacial diamictite units in central western Sweden and
in South Norway are Gaskiers equivalents, then deposits of the
allegedly global end-Cryogenian or Marinoan glacial period
appear not to be represented within these western Baltoscandian
basin successions, at least not as glacial diamictites. In this
connection, it should be emphasized that the various glacial for-
mations in South Norway, Finnmark of North Norway, and
Sweden are lithostratigraphically, not chronostratigraphically, cor-
related. A Gaskiers age is not presently unequivocally constrained
for the formations, neither by radiometric age determinations,
nor by d13C studies on carbonate units beneath the glacial units.
If the Ottfjället dolerite dykes were shown to have about the
same age as the Egersund dyke swarm (c. 616 Ma), this would
put the age of the glacial Lillfjället Fm., and maybe also the
Moelv Fm., to be older than Gaskiers. It should also be emphasized
that the detrital zircon U –Pb age of c. 620 Ma is problematic in the
sense that no likely source of these zircons is known, and this age
appears to belong to a zircon U – Pb age population with a wide
time range (Bingen et al. 2005). The meaning of this is not known.
Improved constraining of the depositional age for Neoprotero-
zoic glacial formations in central and southern Scandinavia may
be obtained by radiometric age determination of the Svarttjørn-
kampen basalt directly underlying the Moelv Fm. and the Ottfjället
dolerites cutting the Lillfjället Fm. Such studies are in progress. In
addition, d13C studies of carbonate formations underlying the
glacial formations (cf. Siedlecka et al. 2004; Nystuen et al.
2008) should be performed.
Conclusions
The Neoproterozoic Moelv and Koppang glacial formations were
mainly formed as basal till from a warm-based, grounded, western
Baltoscandian ice sheet that covered large parts of the northwes-
tern Baltica, and the sedimentary Hedmark, Engerdalen and
Valdres basins. Accumulation areas of the glacial ice were in the
zone of the TIB in the east and NE, and highland areas dominated
by rocks with metamorphic and igneous ages corresponding to the
Gothian and the Sveconorwegian (Grenvillian) orogeneses in the
SE and SW. Local ice-marginal and/or subglacial waterbody suc-
cessions were formed in deeper parts of the rift basins. Laminated
mudstone with ice-dropped stones originated towards the end of
the glaciation due to a rise in sea level and the change from
grounded to floating ice sheet margins and icebergs. The glaciation
may correspond to the c. 580 Ma Gaskiers glacial event of
 NEOPROTEROZOIC GLACIATION OF SOUTH NORWAY
Avalonia, but this interpretation still needs to be better constrained
by radiometric age determinations of igneous rocks associated
with the glacial successions and d13C studies on carbonate for-
mations underlying the glacial formations.
We thank R. Bäckmark and H. B. Totland for making drawings (Figs 59.2, 59.3
& 59.5). The manuscript has benefited from the criticism of an earlier version
by H. Rice and by corrections of the English text by A. Read. We also thank
editor G. Shields for corrections and suggestions for improvements to this contri-
bution. This represents a contribution of the IUGS- and UNESCO-funded IGCP
(International Geoscience Programme) Project #512.
References
˚ häll, K.-I. & Connelly, J. N. 2008. Long-term convergence along
A Gee, D. G., Juhlin, C., Pascal, C. & Robinson, P. 2010. Collision
SW Fennoscandia: 330 m.y. of Proterozoic crustal growth. Pre-
cambrian Research, 163, 402–421.
Andersen, T., Andersson, U. B., Graham, S., A ˚ berg, G. & Simonsen,
S. 2009. Granitic magmatism by melting juvenile continental crust:
new constraints on the source of Palaeoproterozoic granitoids in
Fennoscandia from Hf isotopes in zircon. Journal of Geological
Society, London, 166, 1 –15.
Bodiselitsch, B., Koeberl, C., Master, S. & Reimold, W. U. 2005.
Estimating duration and intensity of Neoproterozoic snowball glacia-
tions from Ir anomalies. Science, 308, 239–242.
Bingen, B., Demaiffe, D. & Van Breemen, O. 1998. The 616 Ma old
Egersund basalt dike swarm, SW Norway, in the context of late-
Neoproterozoic opening of the Iapetus Ocean. Norsk geologisk tids-
skrift, 79, 69 – 86.
Bingen, B., Griffin, W. L., Torsvik, T. H. & Saeed, A. 2005. Timing of
Late Neoproterozoic glaciation on Baltica constrained by detrital
zircon in the Hedmark Group, south-east Norway. Terra Nova, 17,
250– 258.
Bingen, B., Andersson, J., Söderlund, U. & Möller, C. 2008. The
Mesoproterozoic in the Nordic countries. Episode, 31, 29 –34.
Bjørlykke, K. 1966. Studies on the Latest Precambrian and Eocambrian
rocks in Norway. 1. Sedimentary petrology of the Sparagmites of
the Rena district, S. Norway. Norges geologiske undersøkelse, 238,
5 – 53.
Bjørlykke, K. 1969. Geologien i sentrale deler av Österdalen. Norsk geo-
logisk tidsskrift, 49, 313– 318.
Bjørlykke, K. 1974. Glacial striations on clast from the Moelv Tillite
of the Late Precambrian of Southern Norway. American Journal of
Science, 274, 443– 448.
Bjørlykke, K. & Englund, J. O. 1979. Geochemical response to upper
Precambrian rift basin sedimentation and lower Palaeozoic epiconti-
nental sedimentation in South Norway. Chemical Geology, 27,
271– 295.
Bjørlykke, K. & Nystuen, J. P. 1981. Late Precambrian tillites of South
Norway. In: Hambrey, M. J. & Harland, W. B. (eds) Earth’s
Pre-Pleistocene Glacial Record. Cambridge University Press,
Cambridge, 624–628.
Bjørlykke, K., Englund, J.-O. & Kirkhusmo, L. 1967. Latest
Precambrian and Eocambrian stratigraphy of Norway. Norges geo-
logiske undersøkelse, 251, 5– 17.
Bjørlykke, K., Elvsborg, A. & Høy, T. 1976. Late Precambrian sedi-
mentation in the central sparagmite basin of South Norway. Norsk
geologisk tidsskrift, 56, 233–290.
Bowring, S. A., Myrow, P. M., Landing, E., Ramezani, J. & Grotzin-
ger, J. 2003. Geochronological constraints on terminal Neoprotero-
zoic events and the rise of metazoans. Geophysical Research
Abstract, 5, 13219.
Englund, J.-O. 1966. Sparagmittgruppens bergarter ved Fåvang,
Gudbrandsdalen-En sedimentologisk og tektonisk undersøkelse.
Norges geologiske undersøkelse, 238, 55 – 103.
Englund, J.-O. 1973a. Stratigraphy and structure of the Ringebu-Vinstra
district, Gudbrandsdalen; with a short analysis of the western part of
the sparagmite region in Southern Norway. Norges geologiske
undersøkelse, 293, 1 –58.
Englund, J.-O. 1973b. Geochemistry and mineralogy of pelitic
rocks from the Hedmark Group and the Cambro-Ordovician
621
sequence, Southern Norway. Norges geologiske undersøkelse, 286,
1– 60.
Esmark, J. 1829. Reise fra Christiania til Trondhjem. Christiania,
Oslo, 81.
Furnes, H., Nystuen, J. P., Brunfelt, A. O. & Solheim, S. 1983.
Geochemistry of Upper Riphean-Vendian basalts associated with
the ‘sparagmites’ of southern Norway. Geological Magazine, 120,
349– 361.
Gayer, R. A. & Greiling, R. O. 1989. Caledonian nappe geometry in
north-central Sweden and basin evolution of the Baltoscandian
margin. Geological Magazine, 126, 499–513.
Gayer, R. A., Rice, A. H. N., Roberts, D., Townsend, C. & Welbon, A.
1987. Restoration of the Caledonian Baltoscandian margin from
balanced cross-sections: the problem of excess continental crust.
Transactions of the Royal Society of Edinburgh, 78, 197– 217.
orogeny in the Scandinavian Caledonides (COSC). Geologiska Före-
ningens i Stockholm Förhandlingar, 132, 29– 44.
Halverson, G. P., Maloof, A. C. & Hoffman, P. F. 2004. The Marinoan
glaciation (Neoproterozoic) in northeast Svalbard. Basin Research,
16, 297– 324.
Halverson, G. P., Hoffman, P. F., Schrag, D. P., Maloof, A. C. &
Rice, A. H. N. 2005. Toward a Neoproterozoic composite carbon-
isotope record. Geological Society of America Bulletin, 117,
1181–1207.
Hannah, J. L., Yang, G., Bingen, B., Stein, H. J. & Zimmerman, A.
2007.  560 Ma and 300 Ma Re –Os ages constrain Neoprotero-
zoic glaciation and record Variscan hydrocarbon migration on exten-
sion of Oslo rift. Goldschmidt Conference Abstracts 2007, A378.
Hannah, J. L., Yang, G., Xu, G., Zimmerman, A., Stein, H. J. & Egenh-
off, S. O. 2008. Re –Os Isotopic disturbances at unconformities:
challenges and opportunities. American Geophysical Union, Fall
Meeting 2008, abstract #PP31C-1519.
Heim, M., Skiöld, T. & Wolff, F. C. 1996. Geology, geochemistry and
age of the ‘Tricolor’ granite and some other Proterozoic (TIB)
granitoids at Trysil, southeast Trysil. Norsk geologisk tidsskrift, 76,
45– 54.
Hoffman, P. F. & Schrag, D. P. 2002. The snowball Earth hypothesis:
testing the limits of global change. Terra Nova, 14, 129– 155.
Hoffman, K.-H, Conodon, D. J., Bowring, S. A. & Crowley, J. L.
2004. U– Pb zircon date from the Neoproterozoic Ghaub Formation,
Namibia: constraints on Marinoan glaciation. Geology, 32, 817– 820.
Holtedahl, O. 1922. A tillite-like conglomerate in the ‘Eo-Cambrian’
Sparagmite of Southern Norway. American Journal of Science, 4,
165– 173.
Hossack, J. R., Garton, M. R. & Nickelsen, R. P. 1985. The geological
section from the foreland up to the Jotun thrust sheet in the Valdres
area, south Norway. In: Gee, D. G. & Sturt, B. A. (eds) The Cale-
donide Orogen – Scandinavia and Related Areas. John Wiley &
Sons, Chichester, 443–456.
Högdahl, K., Andersson, U. B. & Eklund, O. (eds) 2004. The Trans-
scandinavian Igneous Belt (TIB) in Sweden; A Review of its Character
and Evolution. Geological Survey of Finland, Special Paper, 37, 123.
Kumpulainen, R. 1980. Upper Proterozoic stratigraphy and depositional
environments of the Tossåsfjället Group, Särv Nappe, southern
Swedish Caledonides. Geologiska Föreningen i Stockholm Förhand-
lingar, 102, 531–550.
Kumpulainen, R. A. 2011. The Neoproterozoic Lillfjället Formation,
southern Swedish Caledonides. In: Arnaud, E., Halverson, G. P.
& Shields-Zhou, G. (eds) The Geological Record of Neopro-
terozoic Glaciations. Geological Society, London, Memoirs, 36,
629– 634.
Kumpulainen, R. A. & Greiling, R. O. 2011. Evidence for Neoproter-
ozoic glaciation in central Scandinavian Caledonides. In: Arnaud,
E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological
Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 623–628.
Kumpulainen, R. & Nystuen, J. P. 1985. Late Proterozoic basin evol-
ution and sedimentation in the westernmost part of Baltoscandia.
In: Gee, D. G. & Sturt, B. A. (eds) The Caledonide Orogen –
Scandinavia and Related Areas. John Wiley & Sons, Chichester,
213– 232.
622 J. P. NYSTUEN & J. T. LAMMINEN
Lamminen, J., Nystuen, J. P. & Andersen, T. 2009. U –Pb ages and
Lu –Hf isotopes of granitoid clasts and basement rocks of the
Hedmark Group: implications for structural setting of the Hedmark
Basin at the Baltoscandian margin. NGF Vinterkonferansen 2009,
Bergen 13 – 15 January 2009. Abstracts and Proceedings of the Geo-
logical Society of Norway, 60.
Loeschke, J. & Nickelsen, R. P. 1968. On the age and tectonic position
of the Valdres Sparagmite in Slidre (Southern Norway). Neues Jahr-
buch für Geologie und Paläontologie, 131, 337– 367.
Løberg, B. E. 1970. Investigations at the south-western border of the
sparagmite basin (Gausdal Vestfjell and Fåberg Vestfjell), Southern
Norway. Norges geologiske undersøkelse, 266, 160– 205.
McCay, G. A., Prave, A. R., Alsop, G. I. & Fallick, A. E. 2006. Glacial
trinity: Neoproterozoic Earth history within the British– Irish Caledo-
nides. Geology, 34, 909–912.
Morley, C. K. 1986. The Caledonian thrust front and palinspastic restor-
ations in the southern Norwegian Caledonides. Journal of Structural
Geology, 8, 753– 765.
Myrow, P. M. & Kaufman, A. J. 1999. A newly discovered cap carbonate
above Varanger-age glacial deposits in Newfoundland, Canada.
Journal of Sedimentary Research, 69, 784– 793.
Nickelsen, P. R. 1974. Geology of the Røssjøkollan – Dokkvatn area,
Oppland. Norges geologiske undersøkelse, 314, 53– 100.
Nickelsen, P. R., Garton, M. & Hossack, J. R. 1985. Late Precambrian
to Ordovician sedimentology and stratigraphic correlation of the
Valdres and Synnfjell thrust sheets in the Valdres area, central
Norway. In: Gee, D. G. & Sturt, B. A. (eds) The Caledonide
Orogen – Scandinavia and Related Areas. John Wiley & Sons,
Chichester, 369–378.
Nordgulen, Ø. 1999. Geologisk kart over Norge, berggrunnskart
(bedrock map) HAMAR, M 1:250 000. Norges geologiske
undersøkelse.
Nystuen, J. P. 1969. On the paragenesis of chert and carbonate minerals
in chert-bearing magnesitic dolomite from the Kvitvola nappe,
Southern Norway. Norges geologiske undersøkelse, 258, 66 –78.
Nystuen, J. P. 1976a. Facies and sedimentation of the Late Precambrian
Moelv Tillite in the eastern part of the Sparagmite Region, southern
Norway. Norges geologiske undersøkelse, 329, 1 – 70.
Nystuen, J. P. 1976b. Late Precambrian Moelv Tillite deposited on a
discontinuity surface associated with a fossil ice wedge, Rendalen,
southern Norway. Norsk geologisk tidsskrift, 56, 29 –56.
Nystuen, J. P. 1980. Stratigraphy of the Upper Proterozoic Engerdalen
Group, Kvitvola Nappe, southeastern Scandinavian Caledonides.
Geologiska Föreningen i Stockholm Förhandlingar, 102, 551– 560.
Nystuen, J. P. 1981. The Late Precambrian ‘Sparagmites’ of southern
Norway: a major Caledonian allochthon – The Osen-Røa Nappe
Complex. American Journal of Science, 281, 69– 94.
Nystuen, J. P. 1982. Late Proterozoic basin evolution on the Baltoscan-
dian Craton: the Hedmark Group, southern Norway. Norges geolo-
giske undersøkelse, 375, 1 –74.
Nystuen, J. P. 1985. Facies and preservation of glaciogenic sequences
from the Varanger Ice Age in Scandinavia and other parts of the
North Atlantic region. Palaeogeography, Palaeoclimatology, Palaeo-
ecology, 51, 209–229.
Nystuen, J. P. 1987. Synthesis of the tectonic and sedimentological
evolution of the late Proterozoic-early Cambrian Hedmark Basin,
the Caledonian Thrust Belt, southern Norway. Norsk geologisk tids-
skrift, 67, 395– 418.
Nystuen, J. P. 2008. Neoproterozoic Moelv Tillite and the Hedmark
Basin, Mjøsa Area, South Norway. 33 IGC Excursion No 101, 33rd
International Geological Congress, Oslo, Norway, available at
http://www.33igc.org/coco/LayoutPage.aspx, 59.
Nystuen, J. P. & Sæther, T. 1979. Clast studies in the Late Precambrian
Moelv Tillite and Osdal Conglomerate, Sparagmite Region, south
Norway. Norsk geologisk tidsskrift, 59, 239– 251.
Nystuen, J. P. & Ilebekk, S. 1981. Stratigraphy and Caledonian struc-
tures in the area between the Atnsjøen and Spekedalen windows,
Sparagmite Region, southern Norway. Norsk geologisk tidsskrift,
61, 17 – 24.
Nystuen, J. P., Andresen, A., Kumpulainen, R. & Siedlecka, A. 2008.
Neoproterozoic basin evolution in Fennoscandia, East Greenland
and Svalbard. Episodes, 31, 35– 43.
Oftedahl, C. 1943. Overskyvninger i den norske fjellkjede. Naturen
(Oslo), 5, 143–150.
Rankama, K. 1973. The Late Precambrian glaciation, with particular
reference to the Southern Hemisphere. Journal of Proceedings of
the Royal Society of New South Wales, 106, 89– 97.
Rehnström, E. F., Corfu, F. & Torsvik, T. H. 2002. Evidence of a Late
Precambrian (637 Ma) deformational event in the Caledonides of
Northern Sweden. Journal of Geology, 110, 591– 601.
Rice, A. H. N. 2005. Quantifying the exhumation of UHP-rocks in the
Western Gneiss Region, S.W. Norway: a branch-line – balanced
cross section model. Austrian Journal of Earth Sciences, 98, 2– 21.
Sæther, T. & Nystuen, J. P. 1981. Tectonic framework, stratigraphy,
sedimentation and volcanism of the late Precambrian Hedmark
Group, Østerdalen, south Norway. Norsk geologisk tidsskrift, 61,
193– 211.
Siedlecka, A. & Ilebekk, S. 1982. Forekomster av tillitt på nordsiden av
Atnsjøvinduet, Syd-Norge. Norges geologiske undersøkelse, 373,
33 – 37.
Siedlecka, A., Nystuen, J. P., Englund, J.-O. & Hossack, J. 1987.
LILLEHAMMER – berggrunnskart M. 1:250 000. Norges geolo-
giske undersøkelse.
Siedlecka, A., Roberts, D., Nystuen, J. P. & Olovyanishnikov, V. G.
2004. Northeastern and northwestern margins of Baltica in Neopro-
terozoic time: evidence from the Timanian and Caledonian
Orogens. In: Gee, D. G. & Pease, V. (eds) The Neoproterozoic Tima-
nide Orogen of Eastern Baltica. Geological Society, London,
Memoirs, 30, 169–190.
Thompson, M. D. & Bowring, S. A. 2000. Age of the Squantum ‘tillite’,
Boston basin, Massachusetts: U–Pb zircon constraints on terminal
Neoproterozoic glaciation. American Journal of Science, 300,
630– 655.
Thompson, M. D., Grunow, A. M. & Ramezani, J. 2007. Late Neopro-
terozoic paleogeography of the Southeastern New England Avalon
Zone: insights from U –Pb geochronology and paleomagnetism.
Geological Society of America Bulletin, 119, 681–696.
Tucker, M. E. 1983. Sedimentation of organic-rich limestones in the
late Precambrian of southern Norway. Precambrian Research, 22,
295– 315.
Tucker, R. D., Robinson, P. et al. 2004. Thrusting and extension in the
Scandian hinterland, Norway: new U– Pb ages and tectonostrati-
graphic evidence. American Journal of Science, 304, 477– 532.
Törnebohm, A. E.1896. Grunddragen af det centrala Skandinaviens
bergbyggnad. Kongl. svenska vetenskaps-akademiens handlingar,
28, 1 – 210.
Vidal, G. & Nystuen, J. P. 1990. Micropaleontology, depositional
environment, and biostratigraphy of the Upper Proterozoic
Hedmark Group, southern Norway. American Journal of Science,
290-A, 170–211.
 Chapter 60

Evidence for late Neoproterozoic glaciation in the central Scandinavian Caledonides

RISTO A. KUMPULAINEN1* & REINHARD O. GREILING2*

1
Department of Geological Sciences, Stockholm University, SE-106 91 Stockholm, Sweden
2
Institut für Angewandte Geowissenschaften, Abteilung Strukturgeologie und Tektonophysik,
Karlsruher Institut für Technologie (KIT), Hertzstraße 16, D-76187 Karlsruhe, FR Germany
*Corresponding authors (e-mail: risto.kumpulainen@geo.su.se; er8@agk.uka.de)

Abstract: The Långmarkberg Formation (Fm). rests on either the Neoproterozoic rift-related Risbäck Group or Palaeoproterozoic base-
ment in central Scandinavia. It is always succeeded in stratigraphy by the marine Gärdsjön Fm. The Långmarkberg Fm. is a thin, gen-
erally less than 50-m-thick, discontinuous, but persistent unit, primarily in the Lower Allochthon; subordinately also in the Autochthon,
and Middle Allochthon of the central Scandinavian Caledonides. The formation is composed of a lower diamictite and an upper lami-
nated, lonestone-bearing mudstone.
Criteria supporting the glaciogenic origin of the formation are (i) a large regional extension of the formation (.5000 km2) and (ii) its
fixed position in the regional stratigraphy, (iii) the ‘tillite-like’ appearance of the lower part of the formation, (iv) the lonestone-bearing
mudstone in the upper part, (v) the presence of striated clasts and (vi) striated pavement. It is interpreted as having formed during one
single glacial–deglacial cycle.
Information on chemostratigraphy and realistic geochronology from this region is missing. The Långmarkberg Fm. has commonly
been correlated with the Moelv Fm. in southern Norway and the Mortensnes Fm. in northernmost Norway. The Lower Allochthon in
this region also contains an older, potentially glacially influenced diamictite unit at the base of the Risbäck Group, which is also
briefly discussed.

The Långmarkberg Fm. composed of diamictite and mudstone is
part of the Jämtland Supergroup (Gee 1975), which comprises
the Neoproterozoic to Lower Palaeozoic sedimentary succession
of the Caledonian Lower Allochthon (Fig. 60.1a) in central Scan-
dinavia. The formation occurs in small outcrops, which typically
may be up to a few hundred square metres in size and extend up
to a few hundred metres along strike. The formation is a thin, dis-
continuous, but regionally persistent unit resting on either the
Risbäck Group or the Palaeoproterozoic basement and is always
overlain by the Gärdsjön Fm. (Fig. 60.2). The preserved thickness
of the unit varies from less than a metre commonly to less than
20 m, but ranges locally to more than 50 m. The majority of the
Långmarkberg Fm. outcrops occur within the imbricated Lower
Allochthon and are indicated (by stars) in Figure 60.1. Palinspastic
restoration of the imbricated thrust units to the original pre-
tectonic (pre-Caledonian) basin(s) suggests that the minimum
depositional area exceeded 5000 km2 (Kumpulainen & Nystuen
1985; Gayer et al. 1987; Gayer & Greiling 1989, Siedlecka et al.
2004; Nystuen et al. 2008). Only a few outcops of the formation
are known between the main study area (Fig. 60.1b) and Laisvall
in the north, where a diamictite unit of uncertain origin is described
from the Autochthon (Ackerselet Fm., Willdén 1980; Greiling
et al. 1999a). The unit reappears again in the Akkajaure area
further north (Stodt et al. 2011). Towards the south, only a few out-
crops of the formation are known in the Lower Allochthon
(Fig. 60.1a), before reaching southeastern Norway (Nystuen &
Lamminen 2011). The Middle Allochthon (Gayer & Greiling
1989; Greiling & Zachrisson 1999a, b) hosts sections of the Lång-
markberg Fm. NE of Stalon (Fig. 60.1b) and south of Östersund
(Fig. 60.1a).
This chapter focuses primarily on the Långmarkberg Fm.
between Ströms Vattudal and Dikanäs (Fig. 60.1b). However, dia-
mictite units of uncertain origin, which occur locally at the base of
the Risbäck Group, are also discussed.
Historically, a ‘tillite-like sedimentary breccia’ was described by
Asklund & Thorslund (1934) in Sjoutälven (old spelling, Sjougdäl-
ven; Fig. 60.1b) of the Lower Allochthon of west-central Sweden.
Later, Asklund (1938) named this unit ‘Tillitabteilung/Tillit Fm.’
and correlated it with the Moelv Conglomerate of Holtedahl (1922)
in southern Norway. Subsequently, Kulling (1942) described about
20 diamictite sections in this region. Although unable to produce a
proper description of the section at Långmarkberget (Fig. 60.1b;
648320 N/158210 E), Kulling (1942) selected it to be the type
section and named the unit the Långmarkberg Fm. Recent
mapping in the area reveals that this type section is complicated
tectonically (Zachrisson 1997d ). Work in the late 1960s to
mid-1990s revealed more outcrops, which amount today to more
than 60. Some of the other sections are much better exposed
than Långmarkberget. For this reason the unit was renamed the
Dabbsjön Fm. (Gee et al. 1974) after a new type section at Dabbs-
jön (Fig. 60.1b; 648410 N/158180 E). However, encouraged by the
‘rule of priority’, Gee et al. (1978) revived the name Långmark-
berg Fm. and the sections at Dabbsjön remained reference sections.
The work up to about the mid-1970s was summarized by Thelan-
der (1981). Some additional sedimentological information derives
from geological mapping of the areas north of Bijelite and Malgo-
maj (Greiling & Zachrisson 1999a, b; Greiling et al. 1999b).
Structural framework
The Scandinavian Caledonides are a typical thrust-and-fold belt
and are subdivided into the Caledonian Autochthon, Lower,
Middle, Upper and Uppermost Allochthons (e.g. Gee et al.
1985). Only the Autochthon and Lower and Middle Allochthons
are relevant for the discussion in this chapter.
In the Autochthon of central Scandinavia, the sedimentary suc-
cession of sandstone and shale of Cambrian age was deposited on
the passive Baltica margin (Willdén 1980; Greiling et al. 1999a).
The succession is subhorizontal and undeformed. The alteration in
the Autochthon is diagenetic with peak temperatures below 200 8C
(Warr et al. 1996).
The Lower Allochthon is composed of the Neoproterozoic to
Lower Palaeozoic sedimentary succession of the Jämtland Super-
group (Gee 1975) and slices of the Palaeoproterozoic basement.
The Jämtland Supergroup represents (Fig. 60.2) a gradual develop-
ment from an initial continental rift setting (Risbäck Group) to
passive, Iapetus marginal setting (Tåsjön Group) along the

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 623– 628. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.60
624 R. A. KUMPULAINEN & R. O. GREILING
western (present coordinates) Baltica margin (Kumpulainen &
Nystuen 1985; Gayer & Greiling 1989). Subsequent Caledonian
orogenic shortening formed hinterland dipping duplexes and
imbricate fans (Gee et al. 1978; Gayer & Greiling 1989; Zachris-
son & Greiling 1993a, b; Febbroni 1997; Greiling & Zachrisson
1999a, b; Greiling et al. 1999b, c). In the interior parts of the
fold belt, antiformal stacks of the Lower Allochthon are exposed
in tectonic windows, for example Børgefjellet, Bångonåive and
Nasafjäll (Fig. 60.1), (Greiling 1988; Greiling et al. 1993). The
sedimentary sequence is faulted and sheared, with one or two clea-
vages in shaly rocks. Metamorphism is of sub-greenschist grade
(anchimetamorphism) in the marginal fold belt (Warr et al.
1996) and up to biotite grade in the tectonic windows (Greiling
et al. 1993; our data). Primary sedimentary structures and textures,
as well as fossil assemblages, are widely preserved.
Similar to the Lower Allochthon, the Middle Allochthon
is formed of imbricate units with crystalline basement of
Palaeo- and Mesoproterozoic age and metamorphosed units of
Fig. 60.1. (a) Simplified geological map of
the central Scandinavian Caledonides. (b)
Simplified geological map of the Ströms
Vattudal–Dikanäs areas, modified from
Kumpulainen & Thelander (1978), Gee
et al. (1985), Zachrisson (1997a, b, c, d) and
the sources referred to in the text.
conglomerate, sandstone and shale of probable Neoproterozoic
age. The deformation is polyphase, including a pervasive cleavage
in the pelitic rocks (Greiling 1985; Greiling & Zachrisson 1999a,
b). Primary sedimentary features are preserved only in competent,
psammitic rocks.
Stratigraphy
The Neoproterozoic to Lower Palaeozoic sedimentary succession,
the Jämtland Supergroup (Gee 1975), is subdivided into the
Risbäck, Sjoutälven, Tåsjön and Änge Groups (Fig. 60.2).
The Neoproterozoic Risbäck Group (Gee et al. 1974) is domi-
nated by alluvial siliciclastic rocks. It may reach a thickness of
c. 1.5 km, but wedges out laterally to nothing. In its lower part,
the Cryogenian to Cambrian Sjoutälven Group is composed of
the glaciogenic Långmarkberg Fm. (Kulling 1942; Thelander
1981), which is stratigraphically always succeeded by the
 CENTRAL SCANDINAVIAN CALEDONIDES
Gärdsjön Fm. (Gee et al. 1974). The overlying Tåsjön and Änge
groups are not discussed here.
Sedimentary facies types in the rift-related Risbäck Group
suggest a general change from a proximal alluvial facies, domi-
nated by arkosic sandstone with minor (braided-stream-type)
conglomerate, mudstone + occasional beds of poorly sorted
matrix-supported conglomerate in the east to a marine facies of
sandstone and shale including a dolomite unit, the Kalvberget
Fm., in the west and NW (Kumpulainen & Nystuen 1985;
Nystuen et al. 2008). In its type area the dolomite is microcrystal-
line, but contains scattered euhedral dolomite crystals (,1 mm). In
other areas it is strongly recrystallized, and a network of quartz
veins occurs also in Kalvberget.
The Långmarkberg Fm. forms a thin, discontinuous, but
regionally persistent unit and rests on the Risbäck Group.
Outside the Risbäck basin, the Långmarkberg Fm. was deposited
on top of the slightly undulating Palaeoproterozoic basement
(Fig. 60.2).
The Långmarkberg Fm. is, in all outcrops if not disturbed by
fault surfaces, succeeded by the Gärdsjön Fm. formed during
several transgression –regression cycles in a shallow-marine
setting. Regressions produced quartz-arenite units, whereas trans-
gressions deposited mudstone-dominated units (e.g. Kumpulainen
& Nystuen 1985; Greiling et al. 1999a).
The Gärdsjön Fm. rocks are distinctly different, in lithology and
facies, from those in the Risbäck Group, so mistaking these strati-
graphic units in an outcrop is unlikely, also outside the Ströms
Vattudal –Dikanäs area (Fig. 60.1b), despite the fact that this sedi-
mentary succession is tectonically imbricated.
Glaciogenic deposits and associated strata
The Långmarkberg Fm.
The description of the Långmarkberg Fm. in this chapter is essen-
tially a summary of the general characteristics of the formation
based on Thelander (1981), Kulling (1942) and, north of Bijelite-
Malgomaj, on Greiling & Zachrisson (1999a, b). On the basis of
colour and clast content, Kulling (1942) subdivided the Långmark-
berg Fm. into a lower diamictite and an upper diamictite. This subdi-
vision has not been confirmed by later studies (Thelander 1981). The
stratigraphy and facies in the formation vary greatly from one section
to another (Fig. 60.3). However, generally (for exception, see
Fig. 60.3, section 5), the lower part of the formation is dominated
by coarse-clastic, matrix-supported diamictite beds, whereas the
625
Fig. 60.2. Jämtland Supergroup
stratigraphy (Gee 1975) with outlines of the
geological development of the area.
upper part is mudstone-dominated. The boundary from the lower
unit to the upper unit may be gradational or sharp (Fig. 60.3).
The thickness of the formation varies from less than a metre gen-
erally to c. 20 m (Fig. 60.3), but thicknesses exceeding 50 m have
been encountered at Bergvattnet (Fig. 60.1b) and Jillesnåle
(Fig. 60.1a); there are no systematic changes of the thickness in
any particular direction.
The coarse-clastic lower part of the formation may be composed
of up to three diamictite units that are poorly sorted, structureless
and matrix-supported, although changes in the grain size of the
matrix may produce an indistinct stratification. The poorly sorted
matrix in the diamictite varies from clayey to coarse sandy. The
proportion of matrix and framework clasts varies greatly. Stratified
sandstone- or mudstone beds with lonestones up to c. 30 cm in
diameter occur between diamictite beds.
The size of the clasts in the diamictite varies from pebble to
boulder, with the largest boulder measuring 2 m in diameter.
Arkose and other feldspar-bearing sandstone, quartzite, mudstone
Fig. 60.3. Sections through the Långmarkberg Fm. in the Dabbsjön –
Långvattnet areas. The locations are shown in Figure 60.1b: 1– 3, on the
northeastern shore of Dabbsjön (648410 N/158180 E), redrawn from Thelander
(1981); 4, south of Blaikliden, south of Marsån (65820 400 N/158470 700 E); 5, SE
Grytsjö, south of the Grytsjö–Blaikliden road (65800 2100 N/158360 4600 E); 6, east
of Grytsjö, road section at north side of road Grytsjö– Blaikliden (65800 3400 N/
158370 1000 E); 7, western shore of Långvattnet (658100 1100 N/168310 5300 E).
626 R. A. KUMPULAINEN & R. O. GREILING
and dolomite are the dominant clast types of sedimentary origin;
gneiss clasts of sedimentary origin also occur. The majority of the
crystalline rocks are represented by various types of granite, felsic
porphyry and syenite. With few exceptions, the clast content in
the lower part of the diamictite sections reflects the composition
of the substratum. The proportion of crystalline clasts increases
up-section and they always dominate in the lonestone-bearing, lami-
nated upper mudstone unit. The proportion of crystalline clasts in the
formation varies from 10 to 100%, whereas the proportion of sedi-
mentary clasts varies from 0 to 90 %. Sub-angular to sub-rounded
clasts dominate, but shapes range from angular to well-rounded.
Some stones are faceted and a few display striations (Kulling 1942).
Laminated mudstone with or without normal grading is a common
component in these diamictite sections, particularly in the upper part
of the formation. Occasionally, the mudstone unit contains lenses of
muddy diamictite and some stratified gravel beds. Thin sandy beds
are particularly common in the basal part of a mudstone unit. Lone-
stones range from sand-grade to boulders in size. The preserved
thickness of this facies type is generally in the order of 5–10 m.
Vindelälven River
Diamictite outcrops in the Lower Allochthon north of the Vindeläl-
ven River at Jillesnåle rest on the Risbäck Group (Grambow 2001,
unpublished diploma thesis, Heidelberg University). It is similar in
composition to the Långmarkberg Fm. diamictite, although its
exact stratigraphic relationship remains uncertain. Basal, lami-
nated mudstone, c. 5 m thick, with lonestones is overlain by
more than 50 m of coarse, matrix-supported diamictite. This dia-
mictite contains boulders of meta-arkose (26%), syenite (26%),
rhyolite (19%), quartzite (18%), granite (10%) and meta-pelite
(1%). Ten percent of the boulders are angular, 75% sub-angular
and 15% rounded. No striated stones are reported.
Bångonåive window
Diamictite of uncertain origin and stratigraphic position is exposed
in the Lower Allochthon of the Bångonåive window (Fig. 60.1a)
NE of Tärnaby (648430 4700 N/158240 1400 E) at the base of the Oltok-
ken Fm. (Stephens 1977) interpreted as an equivalent to the Lång-
markberg and Gärdsjön formations combined (Greiling et al.
1993). The diamictite forms a lensoidal body, with a maximum
thickness of c. 1 m, only a few metres stratigraphically above
the crystalline substratum. The exposed lateral extent is in the
order of 100 m. The diamictite is composed of a fine sandy
matrix and enclosed angular clasts of quartzite, feldspathic quart-
zite and subordinate felsic crystalline rocks. The clast size varies
from a few centimetres to decimetres.
Børgefjellet window
Close to the eastern margin of the Børgefjellet window (Fig. 60.1a),
a diamictite unit of uncertain stratigraphic position overlies the
Proterozoic basement at the northern slope of the Sipmekfjäll hill
(658430 4700 N/158240 1400 E; Greiling 1988). The unit, c. 1 m thick,
occurs in a tectonic slice and extends laterally for a few hundred
metres. The diamictite is characterized by a quartz-rich, shaly
matrix, which is intensely cleaved and contains angular fragments
of white vein quartz and quartzite up to c. 10 cm in diameter. The dia-
mictite is overlain by white, coarse feldspathic quartzite, interpreted
as an equivalent of the Gärdsjön Fm. (Gayer & Greiling 1989).
The Risbäck Group and diamictites older than the Risbäck Group
In the Risbäck area, the type area of the Risbäck Group, a succes-
sion of maroon to grey, arkose and feldspathic sandstone
dominates the group, whereas units of greenish grey to dark grey
mudstone are subordinate. Dolomites of the Kalvberget Fm. up
to c. 100 m thick are a discontinuous but significant rock unit in
the upper part of the group.
A unit of diamictite in the Flakaträsk and Vojmsjön areas
(Fig. 60.1), described by Kulling (1942) as ‘red conglomerate’,
underlies the Risbäck Group composed of arkose and feldspathic
quartzite (Febbroni 1997; Greiling & Zachrisson 1999b; Greiling
et al. 1999c). At Vojmsjön, a similar matrix-supported diamictite
contains boulders of c. 5– 50 cm in diameter (Grambow 2001,
unpublished diploma thesis, Heidelberg University). The boulders
are composed of granitoid gneiss (44%), rhyolite (36%), vein
quartz (10%), meta-arkose and meta-pelite (5%), epidote frag-
ments (4%) and amphibolite (1%). Most of these boulders are
angular or sub-angular; no striated boulders have been reported.
The Gärdsjön Fm.
The Gärdsjön Fm. always succeeds the Långmarkberg Fm. The
Gärdsjön Fm. has been divided (Gee et al. 1974) into ten informal
subunits, which may be named M1 to M10; five are sandstone
members, representing sea-level lowstands, and the remaining five
are mudstone-dominated members, representing sea-level high-
stands. Tectonic deformation disturbs the preservation of sedimen-
tary structures, but in many sections they are common, and readily
discernible in both rock types. The Gärdsjön Fm. is a common com-
ponent in the Lower Allochthon and certain units, particularly a
maroon mudstone member in the middle of the formation, and
may be traced long distances along most of the thrust belt (e.g. The-
lander 1982; Greiling et al. 1999a). Owing to the onlapping charac-
ter of the Gärdsjön Fm., only some of the upper members are present
in the Autochthon (Fig. 60.1b), such as the Storuman–Laisvall area,
where they are referred to as the Såvvovare Fm. (Willdén 1980;
Eliasson et al. 2003; Greiling et al. 1999b).
Boundary relations with overlying and underlying
non-glacial units
The Långmarkberg Fm.
The Långmarkberg Fm. rests on three different rock types: (i) Pro-
terozoic crystalline rocks, (ii) arkose and feldspathic sandstone of
the Risbäck Group and (iii) dolomite of the Kalvberget Fm. The
lower boundary of the Långmarkberg Fm. at a regional scale con-
forms to bedding in the underlying Risbäck Group. At a local scale,
grooves and linear depressions and disconformities are observed
(Thelander 1981; Greiling & Kumpulainen 2004). Långmarkber-
get (Fig. 60.1b) exposes a low-angle disconformity at the top of
the Risbäck Group. In a minor outcrop SE of Långmarkberget,
bedding in the Risbäck Group is cut at a high angle by the for-
mation boundary, which Thelander (1981) and Gee et al. (1990)
suggested was a major unconformity. However, soft-sediment
deformation occurs frequently in this part of the Risbäck Group
and it is possible that the erosional surface only incidentally trun-
cated an overturned soft-sediment fold in the Risbäck Group.
Asklund (1960) reported a striated pavement below the Långmark-
berg Fm., but later that interpretation was questioned by Crowell
(1964). However, still later, Thelander (1981) reported such struc-
tures in this region.
In most cases and independent of the substratum, the boundary
is sharp. A close correlation of clast lithologies in the basal parts of
the diamictites and their substratum additionally indicates that the
boundary must be erosional. Only locally, at Långvattnet
(Fig. 60.3), the Kalvberget Fm. dolomites display a gradual tran-
sition up to a diamictite (of uncertain origin). The diamictite, in
turn, grades upwards into a finer-grained clastic sediment with
coarse framework grains of quartz in a dolomite matrix, and then
 CENTRAL SCANDINAVIAN CALEDONIDES
to the Gärdsjön Fm. quartzite (Fig. 60.3); the lonestone-bearing
mudstone is missing here. With this exception, the upper contact
of the Långmarkberg Fm. with the Gärdsjön Fm. is sharp. Ero-
sional features are encountered locally.
Diamictite units older than the Risbäck Group
The diamictite units at the base of the Risbäck Group rest on crys-
talline basement rocks; the exact contact is not exposed, but was
probably deposited in shallow depressions of the basement
surface. The upper boundary of the diamictite is gradational to
red arkose of the Risbäck Group. This diamictite is distinctly
red, which implies staining by iron oxides, mostly hematite and
goethite, similar to the overlying arkose and feldspathic sandstone
of the Risbäck Group. However, magnetic susceptibility in the dia-
mictite is exceptionally high, in the order of 5–10  1023 SI
units (Febbroni 1997, our data). Such high susceptibilities point
to an additional magnetite content of up to 1%. The magnetite is
distributed both in the clasts (boulders) of crystalline basement
rocks and in the matrix of the diamictite.
Chemostratigraphy
No data are available.
Palaeolatitude and palaeogeography
Restorations by Greiling & Smith (2000), Hartz & Torsvik (2002)
or Cawood & Pisarevsky (2006) imply high southern latitudes for
the present area at the time of Långmarkberg Fm. deposition.
Geochronological constraints
No data are available.
Discussion
There are potentially two different glaciogenic diamictite units in
this region, one on top of the Risbäck Group and the other below
the Risbäck Group.
Långmarkberg Fm.
Although thin and discontinuous at a regional scale, the Långmark-
berg Fm. can be traced all across the Cryogenian to Cambrian suc-
cession from the Autochthon, Lower and Middle Allochthons of
the central Scandinavia Caledonides and further north and south.
The interpretations of the various facies types within the Lång-
markberg Fm. rely essentially on Hambrey & Harland (1981)
and Hambrey (1994).
The deposition of the Långmarkberg Fm. has been interpreted
(Thelander 1981) in the framework of one glacial–deglacial
cycle. The various sections through the formation vary greatly in
facies, although generally poorly sorted matrix-supported diamic-
tite occupies the basal part of the sections. The massive units may
have been deposited by meltout, dumping or redeposition subgla-
cially or in a proglacial marine setting. The poorly stratified dia-
mictite may have been deposited subglacially from a grounded
ice-sheet. The stratified, sandy to gravelly beds in or between dia-
mictite beds could have been deposited from glaciofluvial traction
currents beneath temporarily floating ice or close to the sub-
aqueous grounding line of a glacier.
The upper, mudstone-dominated part displays distinct lami-
nation originating from subaqueous deposition at some distance
from floating ice or a grounded ice margin. Melting icebergs trans-
ported clastic debris, dropstones up to boulder size, to this muddy
627
environment. Occasional lensoidal bodies of diamictite with
muddy matrix within this upper laminated unit are probable mass-
flow deposits or were dumped from icebergs.
The lower boundary of the formation is a regional, erosional,
low-angle disconformity, which is substantiated by the presence
(side views) of grooves and other minor linear irregularities.
The diamictite units and their substratum are so indurated that
the possible glacially striated contact surface is only rarely avail-
able. This also limits the findings of striated lonestones. The
upper boundary is commonly sharp and erosional and overlain
by the basal quartz arenite (occasionally conglomerate-bearing)
of the Gärdsjön Fm.
The Långmarkberg Fm. has been correlated stratigraphically
with the upper tillite horizon, the Mortensnes Fm. in northern
Norway (Siedlecka et al. 2004) and with the Moelv Fm. in south-
eastern Norway (Asklund & Thosrslund 1934; Thelander 1981;
Siedlecka et al. 2004; Nystuen et al. 2008; Nystuen & Lamminen
2011), but the scarce isotopic evidence derived from this particular
region of Scandinavia precludes such correlations and remains
speculative.
Possible glaciogenic diamictite at the base of the
Risbäck Group
The ‘red conglomerate’ at the base of the Risbäck Group was inter-
preted by Kulling (1942) as an ‘in water redeposited conglomerate,
which originally may have been a till’. In addition, it is peculiar,
because of its characteristic magnetite content (Febbroni 1997)
which has not been observed in the overlying sedimentary succes-
sion. Magnetite is not stable under present atmospheric or shallow-
water conditions (e.g. Piper 1987; Chan et al. 2005). It is therefore
speculated that magnetite formation and/or stability in this sedi-
mentary environment may be due to exceptional atmospheric/cli-
matic conditions and perhaps related to a Snowball Earth situation
(e.g. Hoffman & Schrag 2002). Therefore, the magnetite contents,
if related to a Snowball Earth situation, may also be a further indi-
cation for a diamictite horizon at the base of the Risbäck Group in
other areas. The red conglomerates at the base of the Risbäck
Group may then correspond to the lower tillite horizon (Smalfjord
Fm.) in northern Norway.
R.O.G. acknowledges support from the Geological Survey of Sweden. Some
of the data presented here were acquired during the PNASTINA project.
Sorsele Kommun supported the work in the Jillesnåle/Vindelälven areas.
R.A.K. acknowledges support from Stockholm University. The manuscript was
improved by the comments of E. Arnaud, an anonymous reviewer and
G. A. Shields. This represents a contribution of the IUGS- and UNESCO-
funded IGCP (International Geoscience Programme) Project #512.
References
Asklund, B. 1938. Hauptzüge der Tektonik und Stratigraphie der mittle-
ren Kaledoniden in Schweden. Sveriges geologiska undersökning C,
417, 99.
Asklund, B. 1960. Studies of the thrust region of the southern part of the
Swedish mountain chain. International Geological Congress, XXI
Session Norden, Guide to the Excursions, A24, C19.
Asklund, B. & Thorslund, P. 1934. Fjällkedjerandens bergbyggnad i
norra Jämtland och Ångermanland. Sveriges geologiska undersökn-
ing C, 382, 110.
Cawood, P. A. & Pisarevsky, S. A. 2006. Was Baltica right-way-up or
upside down in the Neoproterozoic? Journal of the Geological
Society London, 163, 753–759.
Chan, M. A., Bowen, B. B., Parry, W. T., Ormö, J. & Komatsu, G.
2005. Red rock and red planet diagenesis: comparisons of Earth
and Mars concretions. GSA Today, 15, 4 – 10.
Crowell, J. C. 1964. Climate significance of sedimentary deposits
containing disperced megaclasts. In: Nairn, A. E. M. (ed.) Problems
in Palaeoclimatology. Interscience, London, 86– 99.
628 R. A. KUMPULAINEN & R. O. GREILING
Eliasson, T., Greiling, R. O. & Triumf, C.-A. 2003. Bedrock map 24H
Sorsele SV, 1:50,000. Sveriges geologiska undersökning Ai, 188.
Febbroni, S. 1997. Rilevamento geologico e dei valori della suscettivita
magnetica nelle Caledonidi della Lapponia svedese. AZ Marmi,
131, 50.
Gayer, R. A. & Greiling, R. O. 1989. Caledonian nappe geometry in
north-central Sweden and basin evolution on the Baltoscandian
margin. Geological Magazine, 126, 499– 513.
Gayer, R. A., Rice, A. H. N., Roberts, D., Townsend, C. & Welbon, A.
1987. Restoration of the Caledonian Baltoscandian margin from
balanced cross-sections: the problem of excess continental crust.
Transactions of the Royal Society of Edinburgh: Earth Sciences,
78, 197– 217.
Gee, D. G. 1975. A geotraverse through the Scandinavian Caledonides –
Östersund to Trondheim. Sveriges geologiska undersökning C,
717, 66.
Gee, D. G., Karis, L., Kumpulainen, R. & Thelander, T. 1974. A
summary of the Caledonian front stratigraphy, northern Jämtland/
southern Västerbotten, central Swedish Caledonides. Geologiska
Föreningens i Stockholm Förhandlingar, 96, 389– 397.
Gee, D. G., Kumpulainen, R. & Thelander, T. 1978. The Tåsjön
Décollement, central Swedish Caledonides. Sveriges geologiska
undersökning C, 742, 35.
Gee, D. G., Kumpulainen, R., Roberts, D., Stephens, M. B., Thon, A.
& Zachrisson, E. 1985. Scandinavian Caledonides –Tectonostrati-
graphic Map, Scale 1:2,000,000. Sveriges geologiska undersökning
Ba, 35.
Gee, D. G., Kumpulainen, R. & Karis, L. 1990. Fjällrandens berggrund.
In: Lundqvist, T., Gee, D. G., Kumpulainen, R., Karis, L. &
Kresten, P. Beskrivning till berggrundskartan över Västernorrlands
län. Sveriges geologiska undersökning Ba, 31, 206–237.
Greiling, R. O. 1985. Strukturelle und metamorphe Entwicklung an der
Basis grosser, weittransportierter Deckeneinheiten am Beispiel des
Mittleren Allochthons in den zentralen Skandinavischen Kaledoniden
(Stalon-Deckenkomplex in Västerbotten, Schweden). Geotekto-
nische Forschungen, 69, 129.
Greiling, R. O. 1988. Ranseren, berggrunnskart 2025/3, 1:50,000,
foreløpig utgave. Norges geologiske undersökelse.
Greiling, R. O. & Zachrisson, E. 1999a. Bedrock map 23 G Dikanäs
NW, 1:50,000. Sveriges geologiska undersökning Ai, 122.
Greiling, R. O. & Zachrisson, E. 1999b. Bedrock map 23 G Dikanäs
SW, 1:50,000. Sveriges geologiska undersökning Ai, 123.
Greiling, R. O. & Smith, A. G. 2000. The Dalradian of Scotland: missing
link between the Vendian of northern and southern Scandinavia?
Physics and Chemistry of the Earth, 25, 495– 498.
Greiling, R. O. & Kumpulainen, R. 2004. Spår av världens största
inlandsis i vilhelminafjällen? Geologiskt Forum, 43, 8– 11.
Greiling, R. O., Gayer, R. A. & Stephens, M. B. 1993. A basement
culmination in the Scandinavian Caledonides formed by antiformal
stacking (Bångonåive, northern Sweden). Geological Magazine,
130, 471– 482.
Greiling, R. O., Jensen, S. & Smith, A. G. 1999a. Vendian– Cambrian
subsidence of the passive margin of western Baltica – application of
new stratigraphic data from the Scandinavian Caledonian margin.
Norsk Geologisk Tidsskrift, 79, 133– 144.
Greiling, R. O., Zachrisson, E., Thelander, T. & Sträng, T. 1999b.
Bedrock map 23 G Dikanäs NE, 1:50,000. Sveriges geologiska under-
sökning Ai, 124.
Greiling, R. O., Zachrisson, E., Thelander, T. & Sträng, T. 1999c.
Bedrock map 23 G Dikanäs SE, 1:50,000. Sveriges geologiska under-
sökning Ai, 125.
Hambrey, M. J. 1994. Glacial Environments. University College London
Press, London.
Hambrey, M. J. & Harland, W. B. 1981. Part I Introduction. In:
Hambrey, M. J. & Harland, W. B. (eds) Earth’s Pre-Pleistocene
Glacial Record. Cambridge University Press, Cambridge, 3 – 28.
Hartz, E. H. & Torsvik, T. H. 2002. Baltica upside down: a new plate
tectonic model for Rodinia and the Iapetus Ocean. Geology, 30,
255– 258.
Hoffman, P. F. & Schrag, D. P. 2002. The snowball Earth hypothesis:
testing the limits of global change. Terra Nova, 14, 129–155.
Holtedahl, O. 1922. A tillite-like conglomerate in the ‘Eo-Cambrian’
Sparagmite of Southern Norway. American Journal of Science, 4,
165– 173.
Kulling, O. 1942. Grunddragen av fjällkedjerandens bergbyggnad
inom Västerbottens län. Sveriges geologiska undersökning C,
445, 320.
Kumpulainen, R. & Thelander, T. 1978. Geological map of the
area between Ströms Vattudal and Malgomaj, northern Jämtland–
southern Västerbotten, central Swedish Caledonides. Sveriges geo-
logiska undersökning BRAP, 97001 (unpublished map 1:100,000).
Kumpulainen, R. & Nystuen, J. P. 1985. Late Proterozoic basin
evolution and sedimentation in the westermost part of Baltoscandia.
In: Gee, D. G. & Sturt, B. A. (eds) The Caledonide Orogen – Scan-
dinavia and Related Areas. John Wiley & Sons, Chichester,
213– 232.
Nystuen, J. P. & Lamminen, J. T. 2011. Neoproterozoic glaciation of
South Norway: from continental interior to rift and pericratonic
basins in western Baltica. In: Arnaud, E., Halverson, G. P. &
Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic
Glaciations. Geological Society, London, Memoirs, 36, 613–622.
Nystuen, J. P., Andresen, A., Kumpulainen, R. & Siedlecka, A. 2008.
Neoproterozoic basin evolution in Fennoscandia, East Greenland and
Svalbard. Episodes, 31, 35 –43.
Piper, J. D. A. 1987. Palaeomagnetism and the Continental Crust. Open
University Press, Milton Keynes.
Siedlecka, A., Roberts, D., Nystuen, J. P. & Olovyanishnikov, V. G.
2004. Northeastern and northwestern margins of Baltica in Neopro-
terozoic time: evidence from the Timanian and Caledonian
Orogens. In: Gee, D. G. & Pease, V. (eds) The Neoproterozoic Tima-
nide Orogen of Eastern Baltica. Geological Society, London,
Memoir, 30, 169– 190.
Stephens, M. B. 1977. Stratigraphy and relationship between folding,
metamorphism and thrusting in the Tärna-Björkvattnet area, northern
Swedish Caledonides. Sveriges geologiska undersökning C, 726,
1 –146.
Stodt, F., Rice, A. H. N., Björklund, L., Bax, G., Halverson, G. P. &
Pharaoh, T. C. 2011. Evidence of Neoproterozoic glaciation in the
Caledonides of NW Scandinavia. In: Arnaud, E., Halverson,
G. P. & Shields-Zhou, G. (eds) The Geological Record of Neopro-
terozoic Glaciations. Geological Society, London, Memoirs, 36,
603– 611.
Thelander, T. 1981. The late Precambrian Långmarkberg Formation,
central Swedish Caledonides. In: Hambrey, M. J. & Harland, W.
B. (eds) Earth’s Pre-Pleistocene Glacial Record. Cambridge Univer-
sity Press, Cambridge, 615–619.
Thelander, T. 1982. The Torneträsk Formation of the Dividal Group,
northern Swedish Caledonides. Sveriges geologiska undersökning
C, 789, 1 –41.
Warr, L. N., Greiling, R. O. & Zachrisson, E. 1996. Thrust-
related, very low-grade metamorphism in the marginal part of
an orogenic wedge, Scandinavian Caledonides. Tectonics, 15,
1213– 1229.
Willdén, M. Y. 1980. Paleoenvironment of the autochthonous sedimen-
tary rock sequence at Laisvall, Swedish Caledonides. Stockholm Con-
tributions in Geology, 33, 1– 100.
Zachrisson, E. 1997a. Bedrock map 22F Risbäck NW, 1:50,000. Sveriges
geologiska undersökning Ai, 102.
Zachrisson, E. 1997b. Bedrock map 22F Risbäck SW, 1:50,000. Sveriges
geologiska undersökning Ai, 103.
Zachrisson, E. 1997c. Bedrock map 22F Risbäck NE, 1:50,000. Sveriges
geologiska undersökning Ai, 104.
Zachrisson, E. 1997d. Bedrock map 22F Risbäck SE, 1:50,000. Sveriges
geologiska undersökning Ai, 105.
Zachrisson, E. & Greiling, R. O. 1993a. Bedrock map 23F Fatmomakke
NE, 1:50,000. Sveriges geologiska undersökning Ai, 77.
Zachrisson, E. & Greiling, R. O. 1993b. Bedrock map 23F Fatmomakke
SE, 1:50,000. Sveriges geologiska undersökning Ai, 78.
 Chapter 61

The Neoproterozoic glaciogenic Lillfjället Formation, southern Swedish Caledonides

RISTO A. KUMPULAINEN
Department of Geological Sciences, Stockholm University, SE-106 91 Stockholm, Sweden (e-mail: risto.kumpulainen@geo.su.se)

Abstract: The Tossåsfjället Group, which crops out in the Middle Allochthon of the Scandinavian Caledonides, contains the glaciogenic
Lillfjället Fm. Evidence that supports the interpretation of the Lillfjället formations as glacially related include the presence of diamictite
beds, the presence of lonestones, although few, in laminated units, one possible faceted clast, involution-like deformation structures, and
the presence of sandstone wedges cutting 2.5– 3 m down into two different diamictite beds. Similar to the other glaciogenic formations in
southern and central Scandinavia (Moelv and Långmarkberg formations), the Lillfjället Fm. rests on a unit of peritidal, sabkha-type dolo-
mite of the Storån Fm. Poor exposure contributes to the uncertainty about the thickness of the formation. In this account, the Lillfjället
Fm. stratigraphy has been divided into three subunits: (i) a lower diamictite-dominated unit, (ii) a middle unit composed of distinctly
laminated grey sandy mudstone and (iii) an upper diamictite-dominated unit. There is no information available from the Tossåsfjället
Group concerning palaeolatitudes and chemostratigraphy. Poor isotopic evidence (40Ar/39Ar) from the Ottfjället Dolerites, which cut
the Tossåsfjället Group, indicates that the succession is older than c. 665 Ma, so a reliable correlation with global glaciation events is
not yet possible.

Glaciogenic diamictite and associated sandy mudstone occur
in the Neoproterozoic Tossåsfjället Group of the Särv Nappe of
the Caledonian Middle Allochthon in west-central Sweden
(Fig. 61.1; Gee et al. 1985). This glaciogenic unit, which was
named the Lillfjället Fm. by Kumpulainen (1980, 1981), is rep-
resented by less than a dozen, relatively small outcrops (stars in
Fig. 61.1), arranged in a curvilinear manner across the nappe
over a distance of c. 70 km.
Along the southern margin of the nappe, close to the basal mylo-
nites (Fig. 61.1), Strömberg (1961) and Röshoff (1975) described a
possible glaciogenic diamictite, which was given the names Grön-
stack and Ulvberget complexes, respectively. These outcrops are
now included in the Lillfjället Fm.
The Lillfjället Fm. has been correlated with other glacio-
genic diamictites and related rocks in the Caledonides of south-
ern and central Scandinavia. Descriptions of these deposits are
given by Nystuen & Lamminen (2011) and Kumpulainen &
Greiling (2011). A summary of other correlatives of these glacio-
genic successions further north in Scandinavia is given by Stodt
et al. (2011).
This chapter is based essentially on the first descriptions by
Kumpulainen (1980, 1981). The descriptions derive from two
areas, the Lillfjället area in the south and the Stor-Lövsjön area
in the north (Fig. 61.1). The other exposures are more deformed
and are not included here.
Structural relationships
over a distance of a few metres to zero in the basal mylonitic foli-
ation. The arkose and dolerite dykes were altered to banded, felsic
and mafic mylonite, respectively. The nappe was subjected to
biotite-grade metamorphism, which recrystallized the rocks, and
blurs the micro-textures.
The tectono-sedimentary development of the Tossåsfjället
Group has been correlated with that of the other Neoproterozoic
successions in southern and central Scandinavia (Nystuen et al.
2008). One of the features used to correlate these successions is
the ‘carbonate –tillite’ couplet, which occurs in most of the succes-
sions in this region. Other correlative features include the gradual
facies change from proximal to distal depositional sites and
systematic change in thickness of the successions below the car-
bonate –tillite couplet.
These Neoproterozoic sedimentary successions were involved
in the Caledonian orogenic shortening and the various allochthons
were translated towards ESE across the western Baltica margin to
their present locations. Being the uppermost tectonic unit in central
and southern Scandinavia to contain an assemblage of well pre-
served rocks, the Tossåsfjället Group provides the most reliable
information from the outermost possible succession more than
500 km WNW of its present location (Kumpulainen & Nystuen
1985; Gayer & Greiling 1989; Nystuen et al. 2008). A pre-Iapetus
plate tectonic reconstruction of Baltica and adjacent crustal units
(Greiling & Smith 2000) places Scotland close to the western
(present coordinates) Baltica margin, suggesting a juxtaposition(?)
of the Tossåsfjället succession with the Dalradian succession
(including the Port Askaig Fm.) in southwestern Scotland.

The Särv Nappe is the uppermost tectonic unit in the Scandinavian

Caledonides that hosts well-preserved glaciogenic rocks of Neo-
proterozoic age. The high-quality preservation is a consequence
of the intrusion of tens of thousands of dolerite dykes, the Ottjället
dolerite (Holmqvist 1894), into the c. 4.5-km-thick Tossåsfjället
Group. This dyke swarm reinforced the sedimentary succession
and limited Caledonian deformation to only a few shear zones
within the nappe. Owing to this, most of the original sedimentary
structures survived. The whole nappe body, that is, the sedimen-
tary succession and the dyke swarm together, was folded into a
regional fold. The sedimentary bedding in the western limb of
this fold is gently west-dipping but is slightly overturned in the
Stor-Lövsjön area in the east (Kumpulainen 1980; Gilotti & Kum-
pulainen 1986). In the overturned sequence the bedding-dyke
angle still remains close to 908. Approaching the basal shear
zone, the bedding-dyke angle decreases gradually and reduces
Stratigraphy
The unfossiliferous, Neoproterozoic Tossåsfjället Group (Fig. 61.2)
was divided by Kumpulainen (1980) into the Lunndörrsfjällen
(alluvial), Kråkhammaren (alluvial-marine), Storån (peritidal),
Lillfjället (glaciogenic), and Lövan (marine/lacustrine-alluvial)
formations. The stratigraphic position of the Lundörrsfjällen Fm.
is uncertain, because it is separated from the main nappe body
by a fault zone (Gilotti & Kumpulainen 1986, fig. 5) and hence
correlation between the successions is currently not possible.
Otherwise the other formations may be traced across the nappe
from SW to NE (Fig. 61.1). Gilotti & Kumpulainen (1986) ident-
ified in the Stor-Lövsjön area one more stratigraphic unit, an
‘unnamed unit’ (UNU). On the map (Fig. 61.3), the UNU
exceeds 500 m in thickness and occupies the area between the

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 629– 634. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.61
630
Fig. 61.1. Generalized map of the main outcrop area of the Särv nappe in
the southern Swedish Caledonides, distinguishing between pre-glacial rocks, on
the left, and post-glacial rocks, on the right, separated by the Lillfjället Fm.
(dotted line) (modified from Kumpulainen 1980). The inset map shows the
location of the study area within the Caledonian Orogen in Scandinavia.
poorly exposed Storån Fm. in the west and Lillfjället Fm. diamic-
tite beds in the east. Bedding in this unit conforms to that of the
regional structure (Gilotti & Kumpulainen 1986, figs 4 & 5), justi-
fying the interpretation that the UNU is part of the Särv Nappe
stratigraphy. This new stratigraphic unit warrants a modification
of the original definition of the Lillfjället Fm. The new interpret-
ation of the Lillfjället Fm. stratigraphy (Fig. 61.4) places the
whole Lillfjället section in the base of the formation (on
Fig. 61.3) between the Storån Fm. and the UNU. However, to
confirm this hypothesis requires a drilling operation. Resting on
top of the UNU, the diamictite units in the Stor-Lövsjön area are
placed in the upper part of the formation. Hence, in the present
account the UNU is tentatively interpreted as a member within
the Lillfjället Fm. (Figs 61.2 & 61.4).
In this chapter, the discussion concerns the development of the
Lillfjället Fm. and its relation to the Storån (dolomite) and Lövan
(sandstone and conglomerate) formations (Figs 61.2 & 61.3).
Glaciogenic and associated deposits
The Lillfjället Fm.
On the northern slope of the Lillfjället hill (628330 N/28500 E), five
diamictite units and four sandy mudstone and sandstone units
(Figs 61.1 & 61.5) occur with a collective thickness of c. 120 m.
The thickness of the individual diamictite units (informally D1
to D5) varies from c. 0.05 m to c. 50 m, whereas the sandy mud-
stone units are less than 2 m thick, except for the uppermost,
which is c. 16 m thick. The diamictite is massive, and stratification
has been observed only at the base of the D1 and D4 units. The
matrix in the various diamictite units is sand-dominated, poorly
sorted, and contains, particularly in the lowermost diamictite, a
significant proportion of sand-grade clastic dolomite debris. The
R. A. KUMPULAINEN
Fig. 61.2. Stratigraphy of the Tossåsfjället Group (modified from Kumpulainen
1980). The location of the Lunndörrsfjällen Fm. in the Särv Nappe stratigraphy
is uncertain as it may compose its own tectonic subunit within the nappe.
s, silt; m, medium sand; g, gravel.
clast content in the lowermost diamictite bed is c. 25% dolomite,
38% other sedimentary rocks and 38% igneous rocks (granite,
felsic porphyry and occasional greenstone). Higher up, the content
of sedimentary clasts (including subordinate dolomite) varies
between c. 10% and 25%, whereas crystalline clasts vary between
75% and 90%.
The diamictite units are interrupted by units of horizontally
laminated sandy mudstone and weakly stratified dolomitic sand-
stone. In the thick mudstone unit between D4 and D5, lamination
disappears gradually towards the upper part of the unit, where a
few lonestones are found, the largest being c. 0.4  0.6 m in size.
The lower and upper boundaries of D1 are not exposed. The first
mudstone unit grades over a short interval into D2. The top of D2
is undulating and sharply overlain by the second mudstone –
sandstone. The base and the top of the thin D3 are sharp, as are
the base and the top of D4; the top of D4 is cut erosively by a thin
conglomerate at the base of the uppermost mudstone unit. The
base of D5 is sharp.
No striated stones have been observed in this section, which may
be because these rocks, including the clasts, split across, rather
than along the matrix –clast boundary.
The UNU in the Stor-Lövsjön area is more than 500 m thick and
is composed of grey, distinctly parallel-laminated, sandy mud-
stone, in which neither lonestones nor cross-bedded laminae
have been observed.
In the same area (628580 N/138210 E), the UNU grades gradually
upwards to the lower of the two diamictite beds (Fig. 61.4). These
two diamictite beds, c. 5–8 m (lower) and c. 5 m (upper), are
separated by a unit of laminated mudstone. The succession here
displays a stretching lineation, which to some extent makes examin-
ation difficult. The diamictite units are dominated by igneous clasts;
 NEOPROTEROZOIC GLACIOGENIC LILLFJÄLLET FORMATION
Fig. 61.3. Geological map of the Stor-Lövsjön area. The Tossåsfjället Group
stratigraphy occurs in the steeply standing succession around Stor-Lövsjön
(modified from Gilotti & Kumpulainen 1986). The Storån Fm. occurs west of
Buavan as a poorly exposed narrow belt resting on top of the shallow-marine
Kråkhammaren Fm.; neither the lower nor the upper boundary of the Storån Fm.
were observed in this area. The next belt to the east is composed of laminated
sandy mudstone, the ‘unnamed unit’ (UNU). The UNU grades upwards to
diamictite in the uppermost part of the Lillfjället Fm. In this account the UNU
is interpreted as a member of the Lillfjället Fm. Key: 1, Kråkhammaren Fm.
(sandstone and shale); 2, Storån Fm. (dolomite, shale and sandstone); 3, UNU
(sandy mudstone); 4, Lillfjället Fm. (diamictite); 5, lower part of the Lövan Fm.
(mudstone to sandstone); 6, upper part of the Lövan Fm. (gravelly sandstone); 7,
sandstone of uncertain affiliation (Lundörrsfjällen Fm.). The inferred lower
dolomite-dominated unit in the lower part of the Lillfjället Fm. is not shown
on this map.
the largest observed, which is composed of syenite, measures
c. 1.6  0.8 m in size. These two diamictite beds are separated
by a unit of thinly laminated sandstone and mudstone with some
minor soft-sediment deformation structures. The upper boundary
of the mudstone with the upper diamictite is not exposed.
The most spectacular details exposed at this locality are clastic
wedges cutting into the top of each of the two diamictite beds; at
least four in the lower and five in the upper have been observed.
One wedge from each unit is described in more detail
(Fig. 61.6). The bowl-shaped upper end of the wedge in the
lower diamictite bed (Fig. 61.6a) is c. 0.8 –0.9 m wide. Its thick-
ness decreases rapidly downwards and at c. 1 m depth it is less
than 0.2 m wide. Its total depth is almost 3.5 m. The grain size
in the wedge decreases downwards from very poorly sorted
coarse sandy to gravely infill to fine-grained sandstone. Other
bowl-shaped depressions occur on top of this particular diamictite
unit and appear to form a semi-continuous bed on top of the lower
diamictite. Laminated mudstone and a bed of deformed coarse
sand succeed this diamictite. No lonestones have been observed
in this laminated unit.
The other clastic wedge, which cuts the top of the upper diamic-
tite, is more wedge-shaped (Fig. 61.6b). Its upper end is less
than 0.4 m wide and extends downwards at least 3 m. The infill of
this wedge is vertically laminated. Similar to the other wedge, the
grain size here also decreases downwards from coarse to fine sand.
The upper end of the wedge is sharply cut by a poorly sorted
631
Fig. 61.4. Stratigraphy of the Lillfjället and adjacent formations (not to scale).
The Kråkhammaren Fm. subjacent to the Storån Fm. and the upper part of the
Lövan Fm. are not shown here. The two sections, Lillfjället section and the
Stor-Lövsjön section, respectively, indicate in this diagram what stratigraphic
intervals they represent in the Tossåsfjället Group, respectively. The UNU was
probably never deposited in the Lillfjället area.
coarse-grained sandstone bed, c. 0.1–0.2 m thick. Within this
sandstone bed and resting on the diamictite, there is a faceted,
flat-based clast (c. 0.1  0.25 cm) with a convex upper side.
This diamictite unit is also overlain by a laminated and small-scale
cross-laminated sandy mudstone (0.6 m thick) and a soft-sediment
deformed sandstone bed (0.2 m thick). One small lonestone of
sand grade occurs in the laminated mudstone c. 0.4 m above the
diamictite top.
Boundary relationships to the underlying and overlying
non-glacial units
In the Lillfjället section, the glaciogenic Lillfjället Fm. (Kumpulai-
nen 1980) rests on thinly bedded dolomite of the Storån Fm.
(Fig. 61.5). The exact boundary is hidden in a c. 1–2 m large unex-
posed interval between the formations. The Storån Fm. contains
casts after evaporitic anhydrite that was replaced by chalcedony
during diagenesis. The chalcedony was then recrystallized to
microcrystalline quartz by the Caledonian metamorphism (Kum-
pulainen 1980, fig. 11). The Storån Fm. also contains occasional
sandstone and mudstone beds. In this section, the Lillfjället Fm.
is overlain by coarse-grained sandstones of the Lövan Fm., but
the boundary area is unexposed and corresponds to a stratigraphic
thickness of a few tens of metres.
In the Stor-Lövsjön area (Fig. 61.3), a narrow belt of locally
derived dolomite debris suggests the presence of the Storån Fm.
striking parallel and next to the underlying Kråkhammaren Fm.
632 R. A. KUMPULAINEN
Neither the lower nor the upper boundary of the Storån Fm. has
been observed along this belt. The next exposed rock unit to the
east of the Storån Fm. is the sandy mudstone unit, UNU. The
unexposed ground on the map between these two belts of rocks
is wide enough to accommodate a hypothetical lower diamictite-
dominated unit corresponding to the Lillfjället section as suggested
Fig. 61.6. Two clastic wedges cutting tops of two diamictite beds, lower (a) and
upper (b), respectively, on the western shore of Storavan. Note that the
framework clasts in both diamictites are parallel to the wedges, being probably
the result of the approximately east–west-trending stretching of the nappe body
and simultaneous rotation of the clasts. This implies also that the wedges have
been stretched and elongated. The original depth of the wedges is estimated at
c. 2.5– 3 m.
Fig. 61.5. (a) Profile along the northern
Lillfjället hill side, where a dolerite dyke
embraces lenses of Storån and Lillfjället
formations. (b) Logged section of the
sedimentary succession in the hill side. No
pattern in the top of the upper mudstone
indicates massive mudstone. (c) Detail of
(b). Modified from Kumpulainen (1980).
above in the new interpretation of the Lillfjället Fm. stratigraphy
(Figs 61.2 & 61.4; see ‘Discussion’).
The upper boundary of the Lillfjället Fm. is located on top of the
sandstone bed 1 m above the second diamictite (Fig. 61.6b). A
c. 4-m-thick unit of laminated sandy mudstone resting sharply
on top of the Lillfjället Fm. is referred to the base of the Lövan
Fm. This mudstone bed grades upwards to thinly bedded sandstone
(Fig. 61.4), which coarsens gradually into trough-cross-bedded
coarse-grained sandstone and conglomerate.
Chemostratigraphy
No data are available from this succession.
Palaeolatitudes
No data are available from Tossåsfjället Group. Data from other
successions (Greiling & Smith 2000; Hartz & Torsvik 2002;
Cawood & Pisarevsky 2006) imply high southern latitudes for
this area at the time of deposition of the Lillfjället Fm.
Geochronology
Although no data are available from the Tossåsfjället sedimentary
succession itself, the group is cut by (Fig. 61.4a) the Ottjället doler-
ite dyke swarm, which has given an 40Ar/39Ar age of 665 + 10 Ma
(Claesson & Roddick 1983). These dolerites are commonly corre-
lated with extensive dyke swarms in other parts of the Scandina-
vian Caledonides, providing ages in the interval c. 610 –550 Ma
(Svenningsen 2001; Paulsson & Andréasson 2002) and also with
the Egersund dolerite dykes in southwestern Norway (c. 616 Ma;
Bingen et al. 1998). New dating of the Ottfjäll dolerites is necess-
ary to better correlate successions from one region to another in
Scandinavia.
Discussion
A striking feature of the Neoproterozoic sedimentary successions
in the central and southern Scandinavian Caledonides is that peri-
tidal, evaporitic, dolomite extends over the whole region and is
succeeded by presumably glaciogenic sediments attesting to a
 NEOPROTEROZOIC GLACIOGENIC LILLFJÄLLET FORMATION
significant climate change in Neoproterozoic times (Nystuen &
Lamminen 2011; Kumpulainen & Greiling 2011). The extensive
dolomite unit additionally supports the interpretation that the Neo-
proterozoic landscape was close to peneplained and just above the
sea level of that time.
There are a range of criteria that have been described to support
the glaciogenic origin of a certain rock unit or a stratigraphic
section (e.g. Hambrey & Harland 1981; Hambrey 1994). A
number of these criteria have been described from the Lillfjället
Fm., suggesting that it is of glaciogenic origin. In the Lillfjället
section, the glaciogenic evidence includes (i) the presence of dia-
mictite and (ii) the presence of lonestones, that is, probable
ice-rafted debris, although they are few (see below). The Stor-
Lövsjön area additionally offers (iii) periglacial clastic wedges,
(iv) periglacial involution-like structures, and (v) a flat stone on
top of the upper diamictite (Fig. 61.6b) may be glacially faceted
and subsequently reworked by wind action. No striated pavements
and striated stones have been found to date.
Soft-sediment deformation in the stratified units in Lillfjället is
very subordinate (Fig. 61.5b,c). The absence of lamination in the
upper part of the thick sandy mudstone unit may be due to
partial liquefaction, possibly due to rapid deposition of the next
diamictite unit and rapid increase of the pore-water pressure.
The interpretation of the origin of the large lonestone in the
massive upper part of this sandy mudstone is not straightforward,
because the possible deformation structures around that boulder
are obliterated.
The interpretation that the Lillfjället succession was overrun by
five consecutive glaciers is not justified. More likely, the Lillfjället
succession was formed sub-aqueously, the diamictites being
meltout tills or redeposited tills (debris flows), rather than having
been in contact with glacier ice.
Also in the Stor-Lövsjön area, the two diamictite units are
massive and may have been deposited sub-aqueously; the basal
contact of the lower unit is gradual and may indicate that this
diamictite unit was dumped on its laminated muddy substratum.
The base of the upper diamictite is not exposed, providing no
further evidence as to how the upper diamictite was formed. The
tops of both of the diamictite units are sharp, and both are cut by
sandstone wedges forming a network on both upper surfaces.
The lower top surface additionally displays bowl-shaped struc-
tures, possible involution structures (cf. French 1976; Miller
1996). These features support the interpretation that, although
they may have been formed sub-aqueously, they were later
subject to sub-aerial conditions.
The clastic wedges in the Stor-Lövsjön area indicate permafrost
conditions. Wedge (a), in the lower diamictite, is a probable peri-
glacial ice-wedge cast: a soil wedge (cf. Washburn 1980). With
no evidence of erosion, this frost-reworked land surface was sub-
merged before the deposition of the laminated mudstone.
Wedge (b) (Fig. 61.6b) is different in origin, although it too
formed sub-aerially. The vertically laminated sand inside this
wedge indicates that the wedge is a sand wedge (Péwé 1959),
which formed in permafrost areas with very low precipitation
and without snow cover. It was not filled originally with ice, but
with dry sand as the frost crack opened. The vertical lamination
is produced after a series of openings and fillings. After the for-
mation of wedge (b), erosion removed some of the land surface
including a sand-filled depression that is commonly located
above a sand wedge (Berg & Black 1966). A poorly sorted gravelly
sandstone was formed possibly by winnowing on the erosional
surface. Subsequent submergence of the land surface allowed the
deposition of a laminated mudstone and the succeeding sandstone
in the top of the Lillfjället Fm. Permafrost-related sand wedges
may penetrate several metres into the ground (Berg & Black
1966). In the Stor-Lövsjön area, the estimated original depth of
the wedges is c. 2.5– 3 m, which is probably too deep for a
wedge to be formed in a perennially active frozen layer. This
suggests that the glaciogenic Stor-Lövsjön succession was
633
subjected twice to permafrost conditions, due to temporary emer-
gence of the depositional site.
The major difference in stratigraphy between our two sub-areas
is the presence of the UNU in the Stor-Lövsjön area and its appar-
ent absence in the Lillfjället area. This could be for tectonic or
depositional reasons. The Särv nappe is composed of coherent tec-
tonic lenses separated by shear zones. No such deformation zone is
observed to cut the stratigraphy in the Lillfjället area, which may
be due to limited exposure, but the inferred absence of such a
deformation zone is corroborated by the fact that the dyke
swarm preserves its parallel trend over Lillfjället. A shear zone
would be identified by an offset in the dykes’ trend. Hence, tec-
tonic thinning or cutting is less probable here.
The UNU is a more than 500-m-thick monotonous unit of grey
parallel-laminated sandy mudstone, where neither lonestones nor
cross-laminated units have been observed. The distinct, parallel
lamination suggests that this unit was deposited in standing body
of water, probably a pro-glacial lake, where currents were too
weak to produce traction deposits. Also, no mass-flow deposits
have been observed. The thickness of the unit requires a significant
accumulation space or a regular subsidence of the basin floor.
Evidence described above suggests that the Lillfjället Fm. is gla-
ciogenic, and its depositional development may be summarized as
follows. The lower part of the formation was deposited on top of a
peritidal dolomite unit, the Storån Fm., as described from the Lillf-
jället section (Figs 61.4 & 61.5). The various diamictite units were
deposited either as sub-aqueous meltout tills or debris-flow units.
The diamictite deposition does not require repeated ice advance
and retreat events. A paucity of lonestones (and possible drop-
stones) suggests that icebergs were not able to travel these
waters. The present interpretation also suggests that a correspond-
ing lower diamictite-dominated section was deposited on top of the
Storån Fm. in the Stor-Lövsjön area (Fig. 61.3); this section could
be thinner or thicker than that in the Lillfjället section. For a con-
firmation of this hypothesis, drilling is required in this area.
The UNU occupies the middle part of the Lillfjället Fm. strati-
graphy (Figs 61.2 & 61.4). It occurs as a thick unit in the Stor-
Lövsjön area, but appears to be absent in the Lillfjället section.
This thickness change is interpreted here as a difference in the sedi-
mentary system, a large probable pro-glacial lake, where the Stor-
Lövsjön area was the depocentre of the UNU and the Lillfjället
section was located outside that depocentre. The fine-grained clastic
material in this unit was provided by glacial melt waters. The scar-
city of dropstones suggests that the glacial margin at that time was
located on land or otherwise isolated from this depositional area.
The Lillfjället Fm. has been correlated previously (Kumpulai-
nen 1980) with the glaciogenic Moelv Fm. in southern Norway,
the Långmarkberg Fm. in central Scandinavia and the glaciogenic
deposits in northern Norway. Since the group carries no fossils, the
Ottfjället dolerite dyke swarm remains the only significant basis
for a correlation of the Lillfjället Fm. with other successions in
Scandinavia. Earlier dating methods for mafic rocks have given
very variable ages, and the 40Ar/39Ar age of 665 + 10 Ma
(Claesson & Roddick 1983) may contain a principal methodologi-
cal error. Speculating that the various dyke swarms that produce
ages of c. 600–615 Ma could be correlated with the Ottfjället
dolerites, then the Lillfjället Fm. must be older than that age inter-
val, meaning that this probably glaciogenic formation can be
correlated with neither the Gaskiers glaciation event (c. 580 Ma)
nor the Marinoan glaciation event (c. 635 Ma). Modern dating
techniques are required.
The author wishes to thank T. Thelander and J. P. Nystuen for discussions about
glacial deposits. N. Johansson provided valuable geographical information from
the Stor-Lövsjön area. The manuscript was improved by the comments of
E. Arnaud, an anonymous reviewer and G. A. Shields. The author is grateful
for support from Stockholm University. This represents a contribution of the
IUGS- and UNESCO-funded IGCP (International Geoscience Programme)
Project #512.
634 R. A. KUMPULAINEN
References
Bingen, B., Demaiffe, D. & van Breemen, O. 1998. The 616 Ma old
Egersund basalt dike swarm, SW Norway, in the context of late-
Neoproterozoic opening of the Iapetus Ocean. Norsk Geologisk
Tidsskrift, 79, 69 –86.
Berg, T. E. & Black, R. F. 1966. Preliminary measurements of the growth
of nonsorted polygons, Victoria Land, Antarctica. American Geophy-
sical Union, National Academy of Sciences – National Research
Council Publication, 1418, 61– 108.
Cawood, P. A. & Pisarevsky, S. A. 2006. Was Baltica right-way-up or
upside down in the Neoproterozoic? Journal of the Geological
Society London, 163, 753– 759.
Claesson, S. & Roddick, J. C. 1983. 40Ar/39Ar data on the age and meta-
morphism of the Ottfjället Dolerites, Särv Nappe, Swedish Caledo-
nides. Lithos, 16, 61– 73.
French, H. M. 1976. The Periglacial Environment. Longman Group,
New York.
Gayer, R. A. & Greiling, R. O. 1989. Caledonian nappe geometry in
north-central Sweden and basin evolution on the Baltoscandian
margin. Geological Magazine, 126, 499– 513.
Gee, D. G., Kumpulainen, R., Roberts, D., Stephens, M. B., Thon, A.
& Zachrisson, E. 1985. Scandinavian Caledonides –Tectonostrati-
graphic Map, Scale 1:2,000,000. Sveriges geologiska undersökning
Ba, 35.
Gilotti, J. A. & Kumpulainen, R. 1986. Strain softening induced ductile
flow in the Särv thrust sheet, Scandinavian Caledonides. Journal of
Structural Geology, 8, 441– 455.
Greiling, R. O. & Smith, A. G. 2000. The Dalradian of Scotland: missing
link between the Vendian of northern and southern Scandinavia?
Physics and Chemistry of the Earth, 25, 495– 498.
Hambrey, M. J. 1994. Glacial Environments. University College London
Press, London.
Hambrey, M. J. & Harland, W. B. 1981. Part I Introduction. In:
Hambrey, M. J. & Harland, W. B. (eds) Earth’s Pre-Pleistocene
Glacial Record. Cambridge University Press, Cambridge, 3 – 28.
Hartz, E. H. & Torsvik, T. H. 2002. Baltica upside down: a new plate
tectonic model for Rodinia and the Iapetus Ocean. Geology, 30,
255– 258.
Holmqvist, P. J. 1894. Om diabasen på Ottfjället. Geologiska Förenin-
gens i Stockholm Förhandlingar, 16, 175– 192.
Kumpulainen, R. 1980. The Upper Precambrian stratigraphy and deposi-
tional environments of the Tossåsfjället Group, Särv Nappe, southern
Swedish Caledonides. Geologiska Föreningens i Stockholm Förhan-
dlingar, 102, 531–550.
Kumpulainen, R. 1981. The Late Precambrian Lillfjället Formation
in the southern Swedish Caledonides. In: Hambrey, M. J. &
Harland, W. B. (eds) Earth’s Pre-Pleistocene Glacial Record.
Cambridge University Press, Cambridge, 620– 623.
Kumpulainen, R. & Nystuen, J. P. 1985. Late Proterozoic basin
evolution and sedimentation in the westermost part of Baltoscandia.
In: Gee, D. G. & Sturt, B. A. (eds) The Caledonide Orogen –
Scandinavia and Related Areas. John Wiley & Sons, Chichester,
213– 232.
Kumpulainen, R. A. & Greiling, R. O. 2011. Evidence for late Neopro-
terozoic glaciation in the central Scandinavian Caledonides. In:
Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geo-
logical Record of Neoproterozoic Glaciations. Geological Society,
London, Memoirs, 36, 623–628.
Miller, J. M. G. 1996. Glacial sediments. In: Reading, H. G. (ed.) Sedi-
mentary Environments, Facies and Stratigraphy. Blackwell Science,
Oxford, 454– 484.
Nystuen, J. P. & Lamminen, J. T. 2011. Neoproterozoic glaciation of
South Norway: from continental interior to rift and pericratonic
basins in western Baltica. In: Arnaud, E., Halverson, G. P. &
Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic
Glaciations. Geological Society, London, Memoirs, 36, 613–622.
Nystuen, J. P., Andresen, A., Kumpulainen, R. & Siedlecka, A. 2008.
Neoproterozoic basin evolution in Fennoscandia, East Greenland and
Svalbard. Episodes, 31, 35 –43.
Péwé, T. L. 1959. Sand-wedge polygons (tessellations) in the McMurdo
Sound region, Antarctica – in progress report. American Journal of
Science, 257, 545– 552.
Paulsson, O. & Andréasson, P. G. 2002. Attempted break-up of Rodinia
at 850 Ma; geochronological evidence from the Seve – Kalak Super-
terrane, Scandinavian Caledonides. Journal of the Geological Society
London, 159, 751–761.
Röshoff, K. 1975. A probable glaciogenic sediment in the Särv Nappe,
central Swedish Caledonides. Geologiska Föreningens i Stockholm
Förhandlingar, 97, 192– 195.
Stodt, F., Rice, A. H. N., Björklund, L., Bax, G., Halverson, G. P. &
Pharaoh, T. 2011. Evidence for late Neoproterozoic glaciation in the
Caledonides of NW Scandinavia. In: Arnaud, E., Halverson, G. P.
& Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic
Glaciations. Geological Society, London, Memoirs, 36, 603–611.
Strömberg, A. 1961. On the tectonics of the Caledonides in the south-
western part of the county of Jämtland, Sweden. Bulletin of the Geo-
logical Institutions of the University of Uppsala, 39, 1– 92.
Svenningsen, O. M. 2001. Onset of seafloor spreading in the
Iapetus Ocean at 608 Ma; precise age of the Sarek Dyke Swarm,
northern Swedish Caledonides. Precambrian Research, 110,
251– 254.
Washburn, A. L. 1980. Cryogeology; A Survey of Periglacial Processes
and Environments. Edward Arnold, London.
 Chapter 62

The Port Askaig Formation, Dalradian Supergroup, Scotland

EMMANUELLE ARNAUD1* & IAN J. FAIRCHILD2

1
School of Environmental Science, University of Guelph, Guelph, Ontario, N1G 2W1, Canada
2
School of Geography, Earth and Environmental Sciences, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK
*Corresponding author (e-mail: earnaud@uoguelph.ca)

Abstract: The Port Askaig Formation (Fm.) is a thick glaciogenic succession within the Dalradian Supergroup that consists of over
700 m of variably dolomitic diamictite, conglomerate, sandstone mudstone and minor dolomite, and is bounded by mixed siliciclas-
tic –carbonate successions of the Islay (Lossit) and Bonahaven formations. These strata are exposed in the metamorphic Caledonides
of Scotland, although excellent preservation of sedimentary structures can be found at several sites. An extensional setting for this suc-
cession has been proposed based on stratigraphic and structural arguments. The available chemostratigraphic data include the Chemical
Index of Alteration, d13C and Sr-isotope values. Palaeomagnetic analyses have been shown to be subject to post-depositional Caledonian
overprinting. There is also continued debate over the regional palaeogeographical reconstructions of the Scottish promontory for this time
period. The succession is chronologically poorly constrained with U– Pb analyses of stratigraphically much higher or lower deposits. The
thick succession is thought to record glacially influenced marine sedimentation and reworking of unstable sediments in a tectonically
active setting with evidence of ice-margin fluctuations. Alternative palaeoenvironmental interpretations that focus on glacial terrestrial
processes and emphasize climatic influence instead of tectonic activity have also been proposed. The overlying carbonate is a lithologi-
cally diverse coastal complex and so does not fit the Neoproterozoic norm. Research has to date focused on the stratigraphic and sedi-
mentological aspects of this succession, as well as some of the broader palaeogeographical and structural features of the Dalradian basin.
Future efforts should focus on the chronological, structural and palaeogeographical constraints of this succession.

The Port Askaig Fm. is exposed at several sites in Scotland, includ-
ing Islay, the Garvellach Islands, Schichallion, Braemar, Muckle
Fergie Burn and Fordyce (Spencer & Pitcher 1968; Spencer
1971; Litherland 1980) as well as in Donegal, Mayo and Conne-
mara in Ireland (Howarth 1971; Tanner & Shackleton 1979; Max
1981). The type section is located at Port Askaig on the island of
Islay, with several good along-dip outcrops over several kilometres
of shoreline and nearby moorland (Spencer 1971). The best out-
crops can be seen on the Garvellach Islands, with lateral and vertical
exposures of hundreds of metres to several kilometres, though only
the lowermost three members of the formation are exposed there
(Fig. 62.1). These exposures are of phenomenal quality and, com-
pared with other Neoproterozoic successions, relatively easy to
access. They provide abundant sedimentological information as
well as an excellent sense of lateral and vertical facies variability
and enable relatively detailed palaeoclimatic reconstructions.
Other sites in Scotland and Ireland tend to have limited outcrop
exposure, complicated structural relationships, and higher degree
of metamorphism. The Port Askaig Fm. is also known as the Port
Askaig Tillite in the literature, but this genetic term is avoided
here in accordance with modern stratigraphic practice.
The Port Askaig Fm. outcrops are significant in providing the
first record of glaciogenic deposits now known to be Precambrian
(Thomson 1871). Subsequent Survey mapping and other works
carried out until the early 1930s (references in Spencer 1971) pro-
vided more geological context, but with little consciousness of
their international significance. Bailey (1916) was the first to use
way-up indicators in these rocks and to demonstrate the simplicity
of the structure. Later, the stratigraphy and origin of these deposits
became the focus of several detailed studies (Kilburn et al. 1965;
Spencer 1971; Eyles & Eyles 1983; Eyles 1988; Arnaud & Eyles
2006). The most comprehensive study was Spencer’s (1971)
work, in which the variable sedimentology and regional stratigra-
phy were documented at various sites in Scotland and Ireland. The
majority of other studies have focused on the excellent exposures
on the Garvellach Islands.
Additional papers have focused on specific aspects of this thick
succession such as provenance (Anderton 1980; Fitches et al. 1996;
Evans et al. 1998; Cawood et al. 2003), the sandstone intrusions or
wedges (Eyles & Clark 1985), the Great Breccia (Arnaud & Eyles
2002; Benn & Prave 2006), the giant cross-bedded sandstone
(Arnaud 2004) and the origin of associated carbonate strata (Fair-
child 1980a). Anderton (1982, 1985), Yardley et al. (1982), Harris
et al. (1978, 1993) and Prave (1999) provide valuable information
on the depositional setting and tectonic evolution of the Dalradian
Supergroup, while discussing ongoing controversies. Harris et al.
(1993) also provides a useful set of lithostratigraphic columns of
the Dalradian Supergroup at selected sites in Ireland, Scotland
and Shetland. Field-trip guidebooks are available (itineraries I, II
and III in Hambrey et al. 1991; Arnaud & Shields 2005). Sedimen-
tological studies of the overlying Bonahaven Fm. (Spencer &
Spencer 1972; Fairchild 1977, 1980a, b, 1985b) have recently
been supplemented with chemostratigraphic studies of the bound-
ing formations that constrain their global context (Brasier &
Shields 2000; McCay et al. 2006; Prave et al. 2009). The under-
lying carbonates have been described on the Garvellachs
(Spencer 1971), but are less well exposed and less studied on
Islay, although an updated map of northern Islay has been pub-
lished (British Geological Survey 1994).
Structural framework
The Port Askaig Fm. and associated strata are part of the thick Dal-
radian Supergroup exposed within the metamorphic Caledonides
between the Great Glen Fault and the Highland Boundary Fault.
Regionally, the Dalradian Supergroup rocks were affected by
several phases of deformation and folding associated with the
Caledonian orogeny (Treagus 1987). Slides or thrust faults that
lie sub-parallel to bedding and are thought to have last moved in
Ordovician times, and Devonian-age granitic intrusions are also
present in the region (Harris et al. 1993). The outcrops at Port
Askaig are exposed in the relatively structurally simple NNE-
trending Islay anticline (Bailey 1916; Fairchild 1980c). On the
Garvellach Islands, the beds dip uniformly to the south and SE
(c. 358). Tertiary-age dolerite dykes cross-cut the strata.
The Argyll Group, which includes the Port Askaig Fm. at its
base (Fig. 62.2), is thought to have accumulated in an extensional

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 635– 642. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.62
636 E. ARNAUD & I. J. FAIRCHILD
basin experiencing increasing tectonic activity and development of
fault-bounded sub-basins prior to the opening of the Iapetus ocean
(Anderton 1982; Harris et al. 1978, 1993). Evidence that basin
extension started during accumulation of Port Askaig Fm. sedi-
ments includes rapid lateral thickness changes over short distances
(Anderton 1982, 1985). For example, the Port Askaig Fm. appears
to thin rapidly westwards on Islay to the limit of its outcrop (British
Geological Survey 1994) and the overlying Bonahaven Fm. thins
significantly west across the Bolsa fault in this region (Fairchild
1980c; Anderton 1985). There is also a coarse conglomerate
facies of reworked limestone in the Lossit Limestone in the
western part of its outcrop (British Geological Survey 1994). It
is within this context that the Ordovician-age Grampian slides
were proposed to be Proterozoic synsedimentary listric faults
that were later re-activated, based on their association with rapid
thickness changes at various stratigraphic levels within the Dalra-
dian Supergroup (Soper & Anderton 1984; Anderton 1985, 1988).
Most recently, some of the diamictite units as well as repeated
horizons of soft-sediment deformation structures have been inter-
preted as additional indicators of early extensional tectonic activity
in Argyll times (Arnaud & Eyles 2002, 2006).
The rocks have undergone deformation and metamorphism as a
result of the Caledonian orogeny, with open folding and upright
cleavage developed in the least affected areas of Islay, Garvellachs
and Northern Donegal (Spencer 1971; Fairchild 1980b). Locally
there are high-strain areas (Borradaile 1979; Fairchild 1980c) par-
ticularly nearing bounding thrusts, re-activated faults and some
major lithological boundaries (Anderton 1985; Fairchild 1985b;
Treagus & Treagus 2002). Pressure-solution effects are wide-
spread more generally, with limestone and diamictite much more
strongly deformed than dolostone or quartz sandstone (Fairchild
1985b). In the most-studied locations at Port Askaig and in the
Fig. 62.2. Generalized stratigraphy and tectonic setting of the Port Askaig Fm.
and associated strata. P, Port Askaig Fm.; I, Islay Limestone; B, Bonahaven
Fm.; J, Jura Quartzite; S, Scarba conglomerate; *earthquake-induced
liquefaction features (modified from Arnaud & Eyles 2006).
Fig. 62.1. Map showing location of Port
Askaig Fm. outcrops on the Garvellach
Islands (modified from Spencer 1971).
Detailed map of outcrops on Islay can be
found in Spencer (1971) and Hambrey et al.
(1991). Note the lithology ‘sandstone’
includes from 0 to 90þ% dolomite at
different horizons.
Garvellach Islands, sedimentary structures are well preserved,
although noticeably deformed in cleaved lithologies.
Stratigraphy
The Port Askaig Fm. is found within the predominantly marine
succession of the Argyll Group in the Dalradian Supergroup
(Fig. 62.2), being underlain by the Islay (Lossit) Limestone
(mixed dolomitic-siliciclastic at its top) and overlain by the
mixed dolomitic –siliciclastic Bonahaven Fm.
The Port Askaig Fm. is over 700 m thick. Spencer (1971)
defined five members based on predominant facies and clast lithol-
ogy (Fig. 62.3). Members I to III are best exposed on the Garvel-
lach Islands, whereas Members IV and V are only exposed on
the island of Islay. The lowermost member (Member I, Beannan
Buidhe) consists primarily of stacked carbonate-rich diamictite
with commonly discontinuous sandstone and conglomerate inter-
beds. Member I also contains a thick diamictite bed called the
Great Breccia, with very large clasts in a muddy sandstone
matrix and a series of folded and boudinaged interbedded units
of variable lithologies (mudstone, sandstone, dolomite and diamic-
tite) commonly referred to as the ‘Disrupted Beds’. Member II (An
Tamhanachd) consists of diamictite associated with sandstone and
mudstone interbeds with a base defined by the appearance of extra-
basinal clasts. Member III (Creagan Loisgte) consists of thick
packages of laterally continuous sandstone interbedded with dia-
mictite. Member IV (Ruahd Phort Beag) is again dominated by
diamictite, whereas Member V (Con Tom) is dominated by sand-
stone with minor diamictite.
Significant lateral changes to the west on Islay were mentioned
in the previous section. Within the Port Askaig Fm., the succession
on the island of Islay is thin at the base with the unique facies of the
overlying Disrupted Beds being the first unequivocal lithostrati-
graphic tie to the Garvellachs, but which is separated from the
underlying carbonates by ,6 m of sandstone and dolomitic con-
glomerate correlated with the Great Breccia by Spencer (1971).
The remainder of Members I, II and III are broadly similar
between Islay and the Garvellachs.
Glaciogenic deposits and associated strata
The Port Askaig Fm.
The following is a summary of a relatively thick, complex and
superbly exposed succession based on work by Kilburn et al.
(1965), Spencer (1971), Eyles & Clark (1985), Eyles (1988),
Arnaud (2002), Arnaud & Eyles (2002, 2006), Arnaud (2004)
and Benn & Prave (2006). The Port Askaig Fm. consists of a
thick succession of diamictite (44 distinct units), interbedded
with sandstone and minor units of mudstone and conglomerate.
Diamictite are massive to stratified with clasts up to several
metres in diameter floating in a siltstone to silty sandstone
matrix. Some diamictite beds exhibit coarse-tail inverse grading,
and many contain sandstone stringers or inclusions of variable
lithologies (Arnaud & Eyles 2006). Some of the diamictite units
 PORT ASKAIG FORMATION, SCOTLAND
Fig. 62.3. Stratigraphic log of the Port Askaig Fm. based on outcrops on the
Garvellach Islands (Members I to III) and on the island of Islay (Members IV
and V). GB, Great Breccia; DB, Disrupted Beds; XB, Giant cross-bedded
sandstone (modified from Arnaud & Eyles 2002). Note the dolomite interbeds
are considered detrital in origin (see text for discussion).
contain lenses of bedded siltstone sandstone and/or conglomerate
(Spencer 1971). Clast and matrix lithology change up-section from
predominantly intrabasinal and dolomitic at the base to extrabas-
inal and siliciclastic at the top. Pink granitoid extrabasinal clasts
appear to have come from Palaeoproterozoic plutonic rocks of
the Svecofennian-Makkovik-Ketilidian province (Evans et al.
1998). Beds are typically several metres to 10 m in thickness
with sharp conformable to erosional basal contacts. Some basal
contacts are gradational with associated laminated mudstone
(Arnaud 2002; Arnaud & Eyles 2006). The Great Breccia is a par-
ticularly thick diamictite unit (up to 50 m), with megaclasts
ranging from several metres to over 100 m in diameter (see
637
Spencer 1971; Arnaud & Eyles 2002; Benn & Prave 2006 for
detailed descriptions).
Sandstone in the Port Askaig Fm. exhibits a variety of character-
istics including a wide range of textures (fine to very coarse), sorting
(poorly to well sorted), lithologies (dolomitic to quartzitic) and struc-
tures (massive, horizontally bedded, cross-bedded and deformed).
Thick packages of sandstone within Member III are particularly
notable as they contain thick to very thick sets of cross-bedded sand-
stone (individual set thickness averages 3 m, maximum is 11 m;
Arnaud 2004). The lowermost package of these giant cross-bedded
sandstone exhibits a preferred southerly palaeocurrent direction,
whereas cross-bedded sandstone in the rest of the succession is
more variable (Spencer 1971; Arnaud 2004). Other interbeds within
the Port Askaig Fm. include finely laminated mudstone (rhythmites),
(clastic) dolomite and massive to stratified conglomerate. Conglom-
erate often overlies and loads into diamictite, although some also
occurs interbedded with sandstone and mudstone. Outsized clasts
in laminated mudstones or diamictites occur at various horizons,
but only limited examples displaying clear deflection of underlying
laminae occur, especially in the Disrupted Beds (Spencer 1971;
Hambrey et al.1991). Petrographic study of the carbonate in the
Port Askaig Fm. indicates that it appears to be largely detrital in
origin, with that in the rhythmite facies being secondary (Fairchild
1985a; Hambrey et al. 1991, p. 27, 34).
Deformation structures have been documented at multiple dis-
tinct stratigraphic horizons. Several of these, namely sandstone
wedges, sandstone dykes and sandstone downfold structures, have
been examined in detail (Spencer 1971; Eyles & Clark 1985). The
sandstone wedges are predominantly ,10 cm wide, have sharp
irregular outer geometries and commonly penetrate up to several
metres downward into diamictite units or into carbonate units
within the underlying Islay Limestone. Some exhibit branching or
polygonal patterns on bedding plane surfaces. Sandstone dykes are
more tabular in form and intrude diamictite and siltstone. Sand-
stone downfold structures, refered to as load casts or ball and
pillow structures by Arnaud & Eyles (2006), are also common on
top of diamictite units or within interbedded sandstone and mud-
stone. These features, together with horizons of convolute and con-
torted bedding and pseudonodules, were shown to be found most
commonly within Member II sediments on the Garvellach
Islands (Arnaud & Eyles 2006).
The Lossit Limestone
Strata underlying the Port Askaig Fm. have been known as the
Islay Limestone and on the Garvellachs a succession of 72 m of
predominantly mixed dolomitic –siliciclastic sediments with
some stromatolitic and other limestone is present (Spencer
1971). There are some pure dolomicrite and dolomitic stromato-
lites, but for the most part the facies contain both terrigenous
debris and reworked intrabasinal carbonate (Hambrey et al.
1991). The most interesting structures are crystal pseudomorphs
originally illustrated by Spencer (1971). Similar structures from
the Irish Dalradian were interpreted as glendonites, which are
pseudomorphs after ikaite (Johnston 1995).
On Islay, the top 300 m of the Islay Limestone have been rede-
signated as the Lossit Limestone Fm. (British Geological Survey
1994). Its topmost (Persabus) member (c. 80 m thick) consists of
interbedded dolostone (locally stromatolitic or intraclastic), quart-
zite, slate and mixed lithologies and overlies a c. 70 m pure (Kiells)
limestone member, locally oolitic, which in turn overlies several
thick slate and limestone units.
The Bonahaven Fm.
The lithostratigraphy and geological context of the overlying
mixed carbonate –siliciclastic Bonahaven Fm. has been documen-
ted by Spencer & Spencer (1972), who divided it into four
638 E. ARNAUD & I. J. FAIRCHILD
members (Fairchild 1977, 1980a, b, 1985b; Hambrey et al. 1991).
On the east coast of Islay, the sub-arkosic arenite at the top of the
Port Askaig Fm. is succeeded by c. 65 m of sandy facies with mud-
stone and local dolomite at the top (Member 1), 25 m of quartzite
(Member 2), c. 200 m of dolomitic sandstone, oolitic facies and
mudrock (Member 3) and 55 m of heterolithic siliclastic rocks
with a central 12 m of pure dolostone (Member 4). In turn this is
succeeded by several kilometres of pure quartzite (Jura Quartzite).
The structures at the base of the Bonahaven Fm., taken by pre-
vious workers to be desiccation cracks, have been reinterpreted as
interstratal dewatering features by Tanner (1998). Primary carbon-
ate is first encountered near the top of Member 1, where displacive
dolomite fabrics in terrigenous mudstone with genuine desiccation
cracks (Fairchild 1977, 1980b) are suggestive of dolocrete. The
main carbonate unit (Member 3) consists predominantly of three
dolomite facies (stromatolitic, bimodally cross-stratified intraclas-
tic and/or oolitic sandstone, and mudstone with thin graded and/or
wave-rippled sand) (Fairchild 1980a).
Boundary relations with overlying and underlying
non-glacial units
In eastern Islay, basal sandstone, clastic dolostone and diamictite
of the Port Askaig Fm. variably rest erosionally on the underlying
dolomite (at times stromatolitic) and thin-bedded quartzite and
shale, which in turn rest on pure limestone (Spencer 1971;
British Geological Survey 1994). Over 90 m of Member I
exposed on the Garvellachs is missing here on Islay. Further
west, Port Askaig Fm. diamictite units are less poorly exposed,
but commonly are limited by sedimentary limestone breccias
(British Geological Survey 1994). On the Garvellach Islands,
diamictite sharply and erosionally overlies the Islay Limestone
in places (Spencer 1971). In other areas of the Garvellach
Islands (Dun Chonnuill & Garbh Eileach), the contact is grada-
tional as shown by the appearance of rare clasts and the presence
of siltstone (typical of the siltstone matrix of the overlying diamic-
tite) interbeds in the upper part of the Islay limestone (Spencer
1971; Arnaud 2002; Arnaud & Eyles 2006).
The upper contact of the Port Askaig Fm. is seen on Islay, where
it is conformable with the Bonahaven Fm., but it is difficult to
define where the last evidence of glacial phenomena occur as the
dominant facies near the top is quartzite, with only rare thin dia-
mictite horizons. The highest occurrence of chess-board albite,
which is diagnostic of the granitic pebbles of the Port Askaig
Fm. (Spencer 1971), is within a channelled conglomerate
horizon, located c. 25 m above the base of the Bonahaven Fm.
(Hambrey et al. 1991, p. 39, 40). The first occurrence of dolomite
is found near the top of member 1 of the Bonahaven Fm. where it
occurs in tidal flat facies (Fairchild 1977, 1980c) and is succeeded
by more extensive dolomite facies in Member 3, with local undo-
lomitized oolitic and micritic limestone (Fairchild 1980a).
Chemostratigraphy
Panahi & Young (1997) carried out a study of major and trace
element geochemistry, specifically utilizing the Chemical Index of
Alteration (CIA). Most of the 21 samples were taken from the
matrix of diamictite exposed on Garbh Eileah, Garvellach Islands.
Analysis showed a decrease in CIA values from the base of the
Port Askaig Fm. (values range from 68 to 77) to the upper part of
Member III (values range from 60 to 68). The CIA values are an indi-
cation of the extent of weathering based on the relative proportions of
alkali and alkaline earth elements and thus the decrease was inter-
preted as indicating a change in sediment source. The high CIA
values reflect erosion of underlying sedimentary rocks, which had
already experienced weathering, whereas the lower CIA values at
the top of the section indicate incorporation of sediments from
relatively unweathered basement rocks (Panahi & Young 1997). In
addition, analysis of the trace element geochemistry suggests
erosion of shale developed on a post-Archaean crystalline basement,
although a specific source area could not be identified.
Brasier & Shields (2000) provided the first isotopic chemostra-
tigraphic constraints, with reliable Sr-isotope values as low as
0.7067 being obtainable from the Sr-rich facies of pure Lossit
(Islay) Limestone underlying the Port Askaig Fm. These
Sr-isotope values are comparable with facies underlying the
earliest evidence of Neoproterozoic glaciation in other regions.
Thomas et al. (2004) verified this result at a slightly lower
horizon in the Islay limestone (Storakaig limestone of Ballygrant)
(0.706651 –0.706902) using a slightly more careful sample prep-
aration protocol. Sawaki et al. (2010) carried out a careful chemo-
stratigraphic study on both Islay and the Garvellachs. The samples
closest to the onset of glaciation, which also passed a stringent test
for preservation (Mn/Sr , 0.2) were found 40 m below the base of
the Port Askaig Fm. on Garbh Eilach with values as low as
0.70640. These are close to the lowest values found in East Green-
land immediately prior to glaciation (Fairchild et al. 2000) and
suggest that the Garbh Eilach section is similarly complete.
McCay et al. (2006) described evidence for a third Dalradian
glacial from Ireland and also built on earlier d13C results of
Brasier & Shields (2000). Values of þ5‰PDB in a lower (Bally-
grant) limestone were followed by a decline to weakly positive to
negative values in the Lossit Limestone (Brasier & Shields 2000).
McCay et al. (2006) and Prave et al. (2009) show that in the Garvel-
lachs section, both dolomitic and limestone facies show a change
upwards from negative values (–4 to –6) in the strata in which
Sawaki et al. (2010) measured the lightest Sr-isotope signatures,
to weakly positive values. In terms of the carbonate rocks overlying
the Port Askaig Fm., all these publications show that the main
ferroan dolomitic part (Member 3) of the Bonahaven Fm. displays
negative values, and Brasier & Shields (2000) document values in
the Member 4 dolomite horizon exceeding þ10‰.
Other characteristics (e.g. economic deposits, biomarkers)
Mining of zones of epigenetic Pb–Zn mineralization formerly
occurred on Islay in the Ballygrant and Lossit limestones
(British Geological Survey 1994). The Port Askaig Fm. is
locally rich in detrital magnetite and in the Disrupted Beds there
are occasional massive layers of magnetite.
Spencer (1971) reported possible organic traces in Member 1
(Spencer & Spencer 1972) of the Bonahaven Fm. A detailed
description of sole structures interpreted as representing chains
of faecal pellets was made by Brasier & McIlroy (1998), although
Brasier and Shields (2000) conceded that they could well be of
inorganic origin.
Fairchild (1977) described clear 0.1-mm-sized mica spheres
within a 2 m stratigraphic interval of carbonaceous mudstones near
the top of Member 1 of the Bonahaven Fm. Petrographic evidence
indicates that they were delicate enough to collapse when mud desic-
cated, yet were mineralized (perhaps by glauconite) when eroded as
intraclasts. They were interpreted as an unusual form of preservation
of acritarch fossils. However, their great similarity with Triassic mica
spheres from SW England was noted and the latter were sub-
sequently interpreted as tektites by Walkden et al. (2002); this
seems a more likely explanation for the Islay occurrence. Only one
other example of impact-related phenomena has been reported in
the Neoproterozoic of the British Isles (Amor et al. 2008).
Palaeolatitude and palaeogeography
Earlier palaeomagnetic work by Tarling (1974) and Urrutia-
Fucugauchi & Tarling (1983) based on large sample suites on
the Garvellachs appeared to be consistent with low palaeolatitudes,
 PORT ASKAIG FORMATION, SCOTLAND
although difficulties were encountered due to the unknown age of
the deposits and the inability to eliminate the possibility of Caledo-
nian overprinting. Stupavsky et al. (1982) specifically addressed
the overprinting issues using clasts (79 specimens from 36 cores)
and matrix (2 to 3 specimens from 20 cores) of the diamictite
units as well as siltstone (2 to 3 specimens from 6 cores) from
Garbh Eileach, Garvellach Islands. Remanent magnetization
measurements were used to calculate remanence angular standard
deviation in order to evaluate within-specimen homogeneity and
reliability of specimens that were sampled. Specimens were also
subjected to alternating field and thermal demagnetization. All
samples yielded similar results, thus failing the conglomerate
test, and suggesting that these deposits have been remagnetized
by Ordovician-age overprinting (Stupavsky et al. 1982).
The Dalradian Supergroup is generally associated with the
Proto-Iapetus Ocean, although palaeogeographical reconstructions
are uncertain, in part because of an incomplete understanding of
Dalradian basin development (Soper 1994; Tanner & Bluck
1999; Prave 1999; Dalziel & Soper 2001; Demspter et al. 2002;
Hutton & Alsop 2004; Tanner et al. 2005). Some have suggested
the Dalradian experienced orogenesis prior to rifting and
opening up of the Iapetus Ocean in Argyll-Southern Highland.
Such an orogenic event would suggest an affinity with Gondwana
and a palaeogeographical location either off NW Gondwana
(Bluck & Dempster 1991) or off Amazonia (Dalziel 1994). Others
have suggested the Dalradian experienced prolonged rifting and
extension throughout its history consistent with a palaeogeogra-
phical location on the margin of Laurentia (Dalziel & Soper
2001). Others still have suggested the Dalradian was an extension
of Baltica (Greiling & Smith 2000) based on the similarity between
granitic clasts in the Port Askaig Fm. and Scandinavian intrusions.
Recent work by Cawood et al. (2003) demonstrates the similar
characteristics of the detrital zircon populations throughout
the Dalradian and their close match to Laurentian sources, with
the Port Askaig Fm. clasts having closest matches in the North
Atlantic Borderlands (Makkovik, Ketilidian and Svecofennian
provinces). In this regard, the distinctive provenance represents
along-basin rather than cross-basin transport.
Geochronological constraints
Radiometrically, the Port Askaig Fm. is relatively poorly con-
strained. A maximum age of c. 806 Ma comes from the underlying
Grampian Shear Zone located at the base of the Dalradian Super-
group (Noble et al. 1996). This finding is based on three dates
(806 + 3, 808 þ 11/–9 and 804 þ 13/–12 Ma) obtained from
U –Pb isotope analyses of primary monazite in pegmatite and
neocrystalline monazite associated with mylonitic host rocks.
Samples were collected from the Grampian Shear Zone at Lochin-
dorb and A’Bhuidheanaich, East of the Great Glen Fault and
Inverness, Scotland.
A minimum age for the Port Askaig Fm. is provided by two
dates from the Tayvallich volcanic rocks, which are stratigraphi-
cally 8 km above the Port Askaig Fm. at the top of the Argyll
Group in SW Scotland (Fig. 62.2; Prave 1999). Halliday et al.
(1989) present various data from U –Pb, 207Pb/206Pb and
Sm –Nd isotopic analyses of zircons from a keratophyre sampled
from a small laccolithic body on the Tayvallich Peninsula. They
conclude that the most likely age of the keratophyre is
595 þ /24 based on their Pb/Pb analyses. Field relationships
between the small laccolithic body and the Tayvallich volcanic
rocks are somewhat unclear, and Dempster et al. (2002) have
suggested that this age may be younger than the Tayvallich volca-
nic rocks. In an attempt to refine this age, Dempster et al. (2002)
analysed 14 zircons from a felsic tuff collected at Port a’ Bhuailteir
on the Tayvallich Peninsula. Concordia diagrams show a mean
206
Pb – 238U age of 601.4 þ /23.7 Ma (n ¼ 13/14; 2s;
MSWD ¼ 0.82).
639
Lithological comparisons with other diamictite-bearing succes-
sions in Scotland and Ireland and throughout the North Atlantic, as
well as chemostratigraphic studies, have led to various regional
and global stratigraphic correlation schemes (Spencer 1975;
Hambrey 1983; Prave 1999; Brasier & Shields 2000; Halverson
et al. 2005; McCay et al. 2006), with the most recent work
suggesting that the Port Askaig Fm. likely represents the oldest
of several Neoproterozoic glacial periods within the Dalradian
basin and the North Atlantic region. A particularly distinctive
characteristic is the low Sr-isotope ratios of the underlying
limestone (Brasier & Shields 2000; Thomas et al. 2004; Sawaki
et al. 2010). However, this has recently been called into question
based on new Re-Os dates that provide a maximum age of
659.6 + 9.6 Ma for the Port Askaig Fm. (Rooney et al. 2011).
Discussion
Although there is broad agreement that the Port Askaig Fm.
records environmental conditions during one of the Neoprotero-
zoic glacial periods, several palaeoenvironmental models have
been proposed with differences hinging largely on the interpret-
ation of diamictite units as directly deposited by ice (Kilburn
et al. 1965; Spencer 1971; Benn & Prave 2006) or as primarily
deposited in a marine setting influenced by ice-rafting and tectonic
instability (Eyles 1988; Arnaud & Eyles 2006).
In the models that emphasize a subglacial or ice marginal origin,
diamictite units are interpreted as tills recording multiple grounded
ice advances based on the lateral extent of diamictite, the discon-
tinuous nature of some of the interbeds in the lower members, the
presence of large extrabasinal clasts and faceted clasts, the sand-
stone wedges that are interpreted as periglacial, and the siltstone
with outsized clasts that are interpreted as varves with ice-rafted
debris. In these models, the Great Breccia is thought to record
glaciotectonic deformation of sediments based on its similarity
to deformed chalk rafts in tills of Norfolk (UK) and its correlative
erosional unconformity on Islay. The sandstone wedges are
thought to be periglacial because of their similarity to those
formed by repeated freeze and thaw. The associated interbeds
are thought to record either terrestrial or marine depositional
conditions preserved between successive ice advances.
Benn & Prave (2006) took this argument further by suggesting
that the Great Breccia and associated Disrupted Beds recorded
proglacial and subglacial phases of a single glaciotectonic defor-
mation cycle. Although they acknowledged that the sedimentary
characteristics of the Great Breccia were consistent with either a
glaciotectonic or non-glacial sediment gravity flow origin, they
preferred a glacial origin based on their interpretation of several
features within the associated Disrupted Beds as indicative of sub-
glacial deformation; namely laminae that resemble glaciotectonic
laminae, deformation indicative of shear, an increase up-section in
the number of extrabasinal clasts, and evidence of increasing
upwards cumulative strain in diamictite.
In the tectonically influenced glaciomarine models, many dia-
mictite units are thought to record sediment instability associated
with basin development, whereas diamictite units in the uppermost
part of the succession are thought to record reworking of sediments
and rainout of fine-grained sediments and ice-rafted debris in a gla-
cially influenced basin (Arnaud & Eyles 2006). This interpretation
is based on the presence of coarse-tail inverse grading, gradational
basal contacts and the close association with other sediment
gravity flow deposits (Boulton 1972; Nardin et al. 1979; Mulder
& Alexander 2001; Arnaud & Eyles 2006). The Great Breccia is
interpreted as a catastrophic subaqueous landslide associated
with local tectonic activity (Arnaud & Eyles 2002) based on
mapping that revealed (i) the Great Breccia to be a ‘composite
graded sequence’, (ii) an intimate association with undeformed
subaqueous sediment gravity flow and traction current deposits
and (iii) a similarity to published studies of allochtonous carbonate
640 E. ARNAUD & I. J. FAIRCHILD
megabreccia (Arnaud & Eyles 2002). The stratigraphic horizons
of deformation structures, including the overlying Disrupted
Beds, are interpreted as seismites indicative of local tectonic
instability based on their form, geographical extent, and tectonic
setting in which they are found (Arnaud & Eyles 2006). Strati-
graphic analysis of the sedimentary facies and indicators of
glacial and tectonic activity suggest that repeated ice-margin fluc-
tuation and tectonically quiet conditions occurred during depo-
sition of Member III sediments (Arnaud & Eyles 2006).
The sandstone, mudstone and conglomerate interbeds record
shallow marine conditions affected by sediment gravity flows
and traction currents (Arnaud & Eyles 2006). The giant cross-beds
are thought to result from the migration of large dunes under strong
tidal currents considering the Dalradian basin is thought to be
narrow at this time (Eyles 1988; Arnaud 2004). In terms of the
dolomite interbeds, there is no specific evidence of chemical depo-
sition and much evidence of detrital dolomite (Fairchild 1985a;
Hambrey et al. 1991).
The superb exposures in the Garvellach Islands have resulted in
numerous detailed studies, yet the resulting depositional models
proposed for the Port Askaig Fm. have some significant differences
in their climatic and palaeoenvironmental implications. A full dis-
cussion of these models is beyond the scope of this paper and the
reader is referred to the original works for more details. Although
recent developments in glacial geology have allowed some of the
earlier interpretations to be discounted, there are still instances
where the exact nature of glacial influence over these deposits is
debatable and difficult to establish unequivocally.
The carbonate succession underlying the Port Askaig Fm. on
Islay is consistent with marine regression from offshore shale
to coastal facies. On the Garvellachs, the mixed carbonate-
siliciclastic sediments bear local probable ikaite pseudomorphs
near the top, suggestive of cool marine conditions. Facies resem-
bling modern carbonate tropical platforms are limited to local
intraclastic dolostones in parts of the Persabus Member at the
top of the Lossit Limestone, whereas more distinctive oolitic lime-
stone facies occur below in the Kiells member. The relative strati-
graphic positionings of these occurrences between the Garvellachs
and Islay cannot be resolved because of the lack of marker hor-
izons between the two.
The mixed carbonate-siliciclastic succession that overlies the
Port Askaig Fm. is thought to record shallow marine sedimen-
tation. No climatically distinctive facies occur in the basal part
of the Bonahaven Fm. The most prominent carbonate (Member
3 of the Bonahaven Fm.) records a lagoonal and tidal complex
(Fairchild 1980a). It is generally agreed that the Bonahaven Fm.
is not a typical cap-carbonate sequence (Halverson et al. 2005;
McCay et al. 2006). It is hard to pinpoint the last glacial influence
(the highest characteristic conglomerate is argued to be reworked
and represents a tidal channel deposit, Hambrey et al. 1991)
and the facies succession does not resemble that of carbonate
successions found elsewhere. Member 3 carbonates have only
one exact facies counterpart in the geological record – the upper
Canyon Fm. of East Greenland – and the latter occurs some
distance above the upper of two glacial deposits in that area
(Fairchild 1989).
A difference of viewpoint exists about the significance of the
negative C-isotope anomalies below the Port Askaig Fm. and in
the overlying Bonahaven Fm. Prave et al. (2009) argue that the
C-isotope stratigraphy recorded in the Dalradian succession as a
whole bears close comparison with global trends and that the pub-
lished signatures can be taken at face value as primary signatures.
The alternative view is that wherever impure ferroan dolomites are
present, it would be expected that negative deviations from marine
signals would be present. This applies to some (but not all) of the
pre-glacial facies of the Lossit Limestone on the Garvellachs and
to the samples of Member 3 of the Bonahaven Fm. that have
been studied so far. The exact facies equivalents of the Bonahaven
rocks in East Greenland show coherent variations in negative d13C
values over vertical distances of 10 –20 m (Fairchild 1991). There
are limestone facies in the Bonahaven Fm. that could be studied to
help resolve this issue. The extraordinarily high d13C signature in
the pure dolostone of Member 4 is an enigma; a similar strong
signal appears rather soon after glaciation in deposits of NW
Canada, known as the Keele peak (McCay et al. 2006). Whereas
Halverson et al. (2005) chose to trust the pre-Port Askaig d13C
negative anomaly to correlate with late Cryogenian glaciations
elsewhere, it is now clear that such an anomaly is present below
both early and late Cryogenian glacial deposits (Prave et al.
2009). Most authors believe that the low Sr-isotope ratio is a
more specific feature with which to establish an early- to mid-
Cryogenian age for the Port Askaig Fm. The minimum value
obtained a short distance below the Port Askaig Fm. of 0.70640
(Sawaki et al. 2010) is close to the lowest value of 0.7063 found
in East Greenland immediately prior to glaciation (Fairchild
et al. 2000), and suggest that the Garbh Eilach section is similarly
complete. There are difficulties in reconstructing stratigraphic pro-
files through the entire thickness of the Dalradian succession and
there is also evidence of diachroneity of widely distributed glacial
deposits thought previously to be correlative (e.g. Kendall et al.
2006; Fanning & Link 2004), but Prave et al. (2009) were optimis-
tic that there are sufficient diagnostic chemostratigraphic results
to allow the global stratigraphic context of the Port Askaig Fm.
as early- to mid Cryogenian (Sturtian in their terminology) to
be confirmed. Rooney et al.’s (2011) findings will ensure contin-
ued debate and suggest additional work is needed to constrain
the timing of glaciation.
The authors would like to thank A. Spencer and G. Shields for helpful comments
on earlier drafts of this paper. E. A.’s research on Neoproterozoic glaciogenic suc-
cessions is supported by the Natural Sciences and Engineering Research Council
of Canada. This represents a contribution of the IUGS- and UNESCO-funded
IGCP (International Geoscience Programme) Project #512.
References
Amor, K., Hesselbo, S. P., Porcelli, D., Thackrey, S. & Parnell, J.
2008. A Precambrian proximal ejecta blanket from Scotland.
Geology, 36, 303–306.
Anderton, R. 1980. Distinctive pebbles as indicators of Dalradian prov-
enance. Scottish Journal of Geology, 16, 143–152.
Anderton, R. 1982. Dalradian deposition and the late Precambrian –
Cambrian history of the North Atlantic region: a review of the early
evolution of the Iapetus Ocean. Journal of the Geological Society,
London, 139, 421–431.
Anderton, R. 1985. Sedimentation and tectonics in the Scottish Dalra-
dian. Scottish Journal of Geology, 21, 407–436.
Anderton, R. 1988. Dalradian slides and basin development: a radical
interpretation of stratigraphy and structure in the SW and Central
Highlands of Scotland. Journal of the Geological Society, London,
145, 669– 678.
Arnaud, E. 2002. Sedimentological analysis of Neoproterozoic glacio-
genic successions in Norway and Scotland. Unpublished PhD thesis,
School of Geography and Geology, McMaster University.
Arnaud, E. 2004. Giant cross-beds in the Neoproterozoic Port Askaig
Formation, Scotland: implications for snowball Earth. Sedimentary
Geology, 165, 155–174.
Arnaud, E. & Eyles, C. H. 2002. Catastrophic mass failure of a
Neoproterozoic glacially-influenced continental margin, the Great
Breccia, Port Askaig Formation, Scotland. Sedimentary Geology,
151, 313– 333.
Arnaud, E. & Shields, G. 2005. The sedimentary record of a Neoproter-
ozoic glaciation. (IGCP# 512 International Field workshop guide-
book), International Association of Sedimentologists Conference on
Glacial Sedimentary Processes and Products, University of Wales,
Aberystwyth, 29 August– 3 September 2005, 60.
Arnaud, E. & Eyles, C. H. 2006. Neoproterozoic environmental change
recorded in the Port Askaig Formation, Scotland: climatic and tec-
tonic controls on sedimentation. Sedimentary Geology, 183, 99 – 124.
 PORT ASKAIG FORMATION, SCOTLAND
Bailey, E. B. 1916. The Islay Anticline (Inner Hebrides). Quaterly
Journal of the Geological Society of London, 72, 132– 164.
Benn, D. I. & Prave, A. R. 2006. Subglacial and proglacial glacitectonic
deformation in the Neoproterozoic Port Askaig Formation, Scotland.
Geomorphology, 75, 266–280.
Bluck, B. J. & Dempster, T. J. 1991. Exotic metamorphic terranes in the
Caledonides: tectonic history of the Dalradian block, Scotland.
Geology, 19, 1133– 1136.
Borradaile, G. J. 1979. Strain study of the Caledonides in the Islay
region, SW Scotland: implications for strain histories and defor-
mation mechanisms in greenschists. Journal of the Geological
Society, London, 136, 77 –88.
Boulton, G. S. 1972. Modern arctic glaciers as depositional models for
former ice sheets. Journal of the Geological Society, London, 128,
361– 393.
Brasier, M. D. & McIlroy, D. 1998. Neonereites uniserialis from
c. 600 Ma year old rocks in western Scotland and the emergence of
animals. Journal of the Geological Society, London, 155, 5 –12.
Brasier, M. D. & Shields, G. 2000. Neoproterozoic chemostratigraphy
and correlation of the Port Askaig glaciation, Dalradian Supergroup
of Scotland. Journal of the Geological Society, London, 157,
909– 914.
BRITISH GEOLOGICAL SURVEY. 1994. North Islay. Scotland Sheet 27. Solid
and drift geology. 1:50000 provisional series. British Geological
Survey, Keyworth.
Cawood, P. A., Nemchin, A. A., Smith, M. & Loewy, S. 2003. Source of
the Dalradian Supergroup constrained by U– Pb dating of detrital
zircon and implications for the East Laurentian margin. Journal of
the Geological Society, London, 160, 231–246.
Dalziel, I. W. D. 1994. Precambrian Scotland as a Laurentia-Gondwana
link: Origin and signficance of cratonic promontories. Geology, 22,
589– 592.
Dalziel, I. W. D. & Soper, N. J. 2001. Neoproterozoic extension on the
Scottish Promontory of Laurentia: paleogeographic and tectonic
implications. Journal of Geology, 109, 299–317.
Dempster, T. J., Rogers, G. et al. 2002. Timing of deposition, orogen-
esis and glaciation within the Dalradian rocks of Scotland: constraints
from U– Pb zircon ages. Journal of the Geological Society, London,
159, 83– 94.
Evans, J. A., Fitches, W. R. & Muir, R. J. 1998. Laurentian clasts in
a Neoproterozoic tillite from Scotland. Journal of Geology, 106,
361– 366.
Eyles, C. H. 1988. Glacially and tidally-influenced shallow marine sedi-
mentation of the Late Precambrian Port Askaig Formation, Scotland.
Palaeogeography, Palaeoclimatology, Palaeoecology, 68, 1 –25.
Eyles, C. H. & Eyles, N. 1983. Glaciomarine model for upper Precam-
brian diamictites of the Port Askaig Formation, Scotland. Geology,
11, 692– 696.
Eyles, N. & Clark, B. M. 1985. Gravity-induced soft-sediment defor-
mation in glaciomarine sequences of the Upper Proterozoic Port
Askaig Formation, Scotland. Sedimentology, 32, 789–814.
Fairchild, I. J. 1977. Phengite spherules from the Dalradian Bonahaven
Formation, Islay, Scotland: glauconitized microfossils? Geological
Magazine, 114, 355– 364.
Fairchild, I. J. 1980a. Sedimentation and origin of a late Precambrian
‘dolomite’ from Scotland. Journal of Sedimentary Petrology, 50,
423– 446.
Fairchild, I. J. 1980b. Stages in Precambrian dolomitization, Scotland:
cementing v. replacement textures. Sedimentology, 27, 631–650.
Fairchild, I. J. 1980c. The structure of NE Islay. Scottish Journal of
Geology, 16, 189–197.
Fairchild, I. J. 1985a. Comment on ‘Glaciomarine model for upper
Precambrian diamictites of the Port Askaig Formation, Scotland’.
Geology, 13, 89 –90.
Fairchild, I. J. 1985b. Petrography and carbonate chemistry of
some Dalradian dolomitic metasediments: preservation of
diagenetic textures. Journal of the Geological Society, London,
142, 167– 185.
Fairchild, I. J. 1989. Dolomitic stromatolite-bearing units with storm
deposits from the Vendian of East Greenland and Scotland: a case
of facies equivalence. In: Gayer, R. A. (ed.) The Caledonide Geology
of Scandinavia. Graham & Trotman, London, 275– 283.
641
Fairchild, I. J. 1991. Origins of carbonate in Neoproterozoic stro-
matolites and the identification of modern analogues. Precambrian
Research, 53, 281– 299.
Fairchild, I. J., Spiro, B., Herrington, P. M. & Song, T. 2000. Controls
on Sr and C isotope compositions of Neoproterozoic Sr-rich lime-
stones of East Greenland and North China. In: Grotzinger, J. P &
James, N. P. (eds) Carbonate Sedimentation and Diagenesis in the
Evolving Precambrian World. SEPM Special Publication, 67,
297– 313.
Fanning, C. M. & Link, P. K. 2004. U–Pb SHRIMP ages of Neoproter-
ozoic (Sturtian) glaciogenic Pocatello Formation, southeastern Idaho.
Geology, 32, 881– 884.
Fitches, W. R., Pearce, N. J. G., Evans, J. A. & Muir, R. J. 1996. Pro-
venance of late Proterozoic Dalradian tillite clasts, Inner Hebrides,
Scotland. In: Brewer, S. T. (ed.) Precambrian Crustal Evolution
in the North Atlantic Region. Geological Society, Special Publi-
cations, London, 112, 367– 377.
Greiling, R. O. & Smith, A. G. 2000. The Dalradian of Scotland: Missing
link between the Vendian of Northern and Southern Scandinavia?
Physics and Chemistry of the Earth, Part A, 25, 495– 498.
Halliday, A. N., Graham, C. M., Aftalion, M. & Dymoke, P. 1989.
Short paper: The depositional age of the Dalradian Supergroup:
U–Pb and Sm–Nd isotopic studies of the Tayvallich Volcanics,
Scotland. Journal of the Geological Society, London, 146, 3– 6.
Halverson, G. P., Hoffman, P. F., Schrag, D. P., Maloof, A. C. &
Rice, A. H. N. 2005. Toward a Neoproterozoic composite carbon-
isotope record. Geological Society of America Bulletin, 117,
1181–1207.
Hambrey, M. J. 1983. Correlation of Late Proterozoic tillites in the North
Atlantic region and Europe. Geological Magazine, 120, 209–232.
Hambrey, M. J., Fairchild, I. J., Glover, B. W., Stewart, A. D.,
Treagus, J. E. & Winchester, J. A. 1991. The Late Precambrian
Geology of the Scottish Highlands and Islands. Geologists Associ-
ation Guide No. 44. The Geologists’ Association, London, England.
Harris, A. L., Baldwin, C. T., Bradbury, H. J., Johnson, H. D. &
Smith, R. A. 1978. Ensialic basin sedimentation: the Dalradian
Supergroup. In: Bowes, D. R. & Leake, B. L. (eds) Crustal Evolution
in Northwestern Britain and Adjacent Regions. Seel House Press,
Liverpool, 115–138.
Harris, A. L., Haselock, P. J., Kennedy, M. J., Mendum, J. R., Long,
C. B., Winchester, J. A. & Tanner, P. W. G. 1993. The Dalradian
Supergroup in Scotland, Shetland, and Ireland. In: Gibbons, W. &
Harris, A. L. (eds) A Revised Correlation of Precambrian Rocks in
the British Isles. Geological Society, Special Report, London, 33– 53.
Howarth, R. J. 1971. The Port Askaig tillite succession (Dalradian) of
Co. Donegal. Proceedings of the Royal Irish Academy, 71, 1 –36.
Hutton, D. H. W. & Alsop, G. I. 2004. Evidence for a major Neopro-
terozoic orogenic unconformity within the Dalradian Supergroup
of NW Ireland. Journal of the Geological Society, London, 161,
629– 640.
Johnston, J. D. 1995. Pseudomorphs after ikaite in a glaciomarine
sequence in the Dalradian of Donegal, Ireland. Scottish Journal of
Geology, 31, 3 –9.
Kendall, B. S., Creaser, R. A. & Selby, D. 2006. Re– Os geochrono-
logy of postglacial black shales in Australia: constraints on the
timing of the Sturtian glaciation. Geology, 34, 729–732.
Kilburn, C., Pitcher, W. S. & Shackleton, R. M. 1965. The stratigra-
phy and origin of the Port Askaig boulder bed series (Dalradian).
Geological Journal, 4, 343– 360.
Litherland, M. 1980. The stratigraphy of the Dalradian rocks around
Loch Creran, Argyll. Scottish Journal of Geology, 16, 105–123.
Max, M. D. 1981. E20: Dalradian Tillite of northwestern Ireland. In:
Hambrey, M. J. & Harland, W. B. (eds) Earth’s Pre-Pleistocene
Glacial Record. Cambridge University Press, Cambridge, 640–642.
McCay, G. A., Prave, A. R., Alsop, G. I. & Fallick, A. E. 2006. Glacial
trinity: Neoproterozoic Earth history within the British-Irish Caledo-
nides. Geology, 34, 909– 912.
Mulder, T. & Alexander, J. 2001. The physical character of subaqueous
sedimentary density flows and their deposits. Sedimentology, 48,
269– 299.
Nardin, T. R., Hein, F. J., Gorsline, D. S. & Edwards, B. D. 1979. A
review of mass movement processes, sediment and acoustic
642 E. ARNAUD & I. J. FAIRCHILD
characteristics and contrasts in slope and base of slope systems
v. canyon fan basin floor systems. In: Doyle, L. J. & Pilkey, O.
H. (eds) Geology of Continental Slopes. Society of Economic Paleon-
tologists and Mineralogists, Special Publication, 61 –73.
Noble, S. R., Hyslop, E. K. & Highton, A. J. 1996. High precision U –Pb
monazite geochronology of the c. 806 Ma Grampian Shear Zone and
the implications for the evolution of the Central Highlands of
Scotland. Journal of the Geological Society, London, 153, 511– 514.
Panahi, A. & Young, G. M. 1997. A geochemical investigation into the
provenance of the Neoproterozoic Port Askaig tillite, Dalradian
Supergroup, western Scotland. Precambrian Research, 85, 81– 96.
Prave, A. R. 1999. The Neoproterozoic Dalradian Supergroup of Scot-
land: an alternative hypothesis. Geological Magazine, 136, 609– 617.
Prave, A. R., Fallick, A. E., Thomas, C. W. & Graham, C. M. 2009. A
composite C-isotope profile for the Neoproterozoic Dalradian Super-
group of Scotland and Ireland. Journal of the Geological Society,
London, 166, 845–857.
Rooney, A. D., Chew, D. M. & Selby, D. 2011. Re-Os geochronology of
the Neoproterozoic-Cambrian Datradian Supergroup of Scotland and
Ireland: implications for Neoproterozoic stratigraphy, glaciations and
Re-Os systematics. Precambrian Research, 185, 202– 214.
Sawaki, Y., Kawai, T. et al. 2010. 87Sr/86Sr chemostratigraphy of
Neoproterozoic Dalradian carbonates below the Port Askaig glacio-
genic Formation, Scotland. Precambrian Research, 179, 150– 164.
Soper, N. J. 1994. Neoproterozoic sedimentation on the northeast margin
of Laurentia and the opening of Iapetus. Geological Magazine, 131,
291– 299.
Soper, N. J. & Anderton, R. 1984. Did the Dalradian slides originate as
extensional faults? Nature, 307, 357–360.
Spencer, A. M. 1971. Late Precambrian glaciation in Scotland. Memoirs
of the Geological Society of London, 6, 1 –100.
Spencer, A. M. 1975. Late Precambrian glaciation in the North Atlantic
region. In: Wright, A. E. & Moseley, F. (eds) Ice Ages, Ancient
and Modern. Seel House Press, Liverpool, 217– 240.
Spencer, A. M. & Pitcher, W. S. 1968. Occurrence of the Port Askaig
Tillite in north-east Scotland. Proceedings of the Geological
Society of London, 1650, 195– 198.
Spencer, A. M. & Spencer, M. O. 1972. The late Precambrian/Lower
Cambrian Bonahaven Dolomite of Islay and its stromatolites. Scottish
Journal of Geology, 8, 269–282.
Stupavsky, M., Symons, D. T. A. & Gravenor, C. P. 1982. Evidence
for metamorphic remagnetisation of upper Precambrian tillite in the
Dalradian Supergroup of Scotland. Transactions of the Royal
Society of Edinburgh, Earth Sciences, 73, 59 –65.
Tanner, P. W. G. 1998. Interstratal dewatering origin for polygonal
patterns of sand-filled cracks; a case study from late Proterozoic meta-
sediments of Islay, Scotland. Sedimentology, 45, 71 –89.
Tanner, P. W. G. & Shackleton, R. M. 1979. Structure and stratigraphy
of the Dalradian rocks of the Bennabeola area; Connemara, Eire. In:
Harris, A. L., Holland, C. H. & Leake, B. E. (eds) The Caledo-
nides of the British Isles – Reviewed. Geological Society, London,
Special Publications, 8, 243 – 256.
Tanner, P. W. G. & Bluck, B. J. 1999. Current controversies in the
Caledonides. Journal of the Geological Society, London, 156,
1137– 1141.
Tanner, P. W. G., Alsop, G. I. & Hutton, D. 2005. Discussion on
evidence for a major Neoproterozoic orogenic unconformity within
the Dalradian Supergroup of NW Ireland. Journal of the Geological
Society, London, 162, 221–224.
Tarling, D. H. 1974. A paleomagnetic study of Eocambrian tillites in
Scotland. Journal of the Geological Society, London, 130, 163– 177.
Thomas, C. W., Graham, C. M., Ellam, R. M. & Fallick, A. E. 2004.
87
Sr/86Sr chemostratigraphy of Neoproterozoic Dalradian lime-
stones of Scotland and Ireland: constraints on depositional ages
and time scales. Journal of the Geological Society, London, 161,
229– 242.
Thomson, J. 1871 (for 1870). On the occurrence of pebbles and boulders
of granite in schistose rocks in Islay, Scotland. 40th Meeting British
Association, Liverpool, Transactions, 88.
Treagus, J. E. 1987. The structural evolution of the Dalradian of the
Central Highlands of Scotland. Transactions of the Royal Society of
Edinburgh, 78, 1– 15.
Treagus, S. H. & Treagus, J. E. 2002. Studies of strain and rheology of
conglomerates. Journal of Structural Geology, 24, 1541–1567.
Urrutia-Fucugauchi, J. & Tarling, D. H. 1983. Paleomagnetic proper-
ties of Eocambrian sediments in Northwestern Scotland: implications
for world wide glaciation in the Late Precambrian. Palaeogeography,
Palaeoclimatology, Palaeoecology, 41, 325– 344.
Walkden, G., Parker, J. & Kelley, S. 2002. A late Triassic impact
ejecta layer in southwestern Britain. Science, 298, 2185–2188.
Yardley, B. W. D., Vine, F. J. & Baldwin, C. T. 1982. The plate tec-
tonic setting of NW Britain and Ireland in late Cambrian and early
Ordovician times. Journal of the Geological Society of London,
139, 455– 463.
 Chapter 63

The Neoproterozoic glaciogenic deposits of Scotland and Ireland

A. R. PRAVE1 * & A. E. FALLICK2

1
Department of Earth Science, University of St Andrews, St Andrews, KY16 9AL, UK
2
Scottish Universities Environmental Research Centre, East Kilbride, G75 0QF, UK
*Corresponding authors (e-mail: ap13@st-andrews.ac.uk)

Abstract: Of the three major Neoproterozoic supracrustal units in the Scottish and Irish Highlands (the Torridonian, Moine and
Dalradian Supergroups), only the latter contains evidence of Neoproterozoic glaciations. The Dalradian is siliciclastic-dominated and
constitutes much of the Scottish–Irish Highlands between the Great Glen and Highland Boundary Fault Zones, and their correlatives
in Ireland. At the time of writing, three stratigraphically distinct glacial intervals in the Dalradian have been documented in the literature.
The oldest is the Port Askaig Formation (Fm.) at the base of the Argyll Group (see Arnaud & Fairchild 2011). It ranges from several tens
to many hundreds of metres in thickness and occurs in numerous localities in Scotland and the north of Ireland. A second glacial is
recorded in the middle part of the Argyll Group (Easdale Subgroup) and consists of localised sedimentary breccias as well as pelites
and schists containing dropstone/lonestone units inferred to be ice-rafted debris; these rocks are patchily preserved and typically a
few metres or less in thickness. It is sharply overlain by a variably developed carbonate unit that is marked by a 1 –7-m-thick, light-
coloured, basal dolostone or dolomitic limestone interpreted as a cap carbonate. This succession is best preserved in Donegal,
Ireland, as the Stralinchy–Reelan glacial and Cranford cap-carbonate sequence. A correlative cap carbonate, the Whiteness Limestone,
has been identified in the Shetland Islands. The third and youngest glacial is represented by locally preserved dropstone and polymict
diamictite beds ranging in thickness from several to a few tens of metres in thickness in the lower Southern Highland Group. These
include the MacDuff and Loch na Cille Boulder Beds in, respectively, NE and SW Scotland, and the Inishowen Beds in Donegal, Ireland.

The late Proterozoic geological framework of the Scottish –Irish
Highlands consists largely of three (meta)sedimentary succes-
sions (each many kilometres thick): the Torridonian, preserved
NW of the Moine Thrust; the Moine, in the upper plate of the
Moine Thrust zone southeastward to the Great Glen Fault; and
the Dalradian, which is sandwiched between the Great Glen
and Highland Boundary Fault Zones (Fig. 63.1). Only the Dalra-
dian contains a record of Neoproterozoic glaciations. A purported
glacial deposit was inferred for a unit in the lower part of the
Torridonian (Davison & Hambrey 1996) but this has been
shown to be incorrect (Stewart 1997, 2002). Along most of the
Moine outcrop belt the rocks are too strongly deformed to
obtain sedimentological information, but even in areas of low
strain no evidence has been found for glaciogenic deposits.
This is not surprising given that both the Moine and Torridonian
pre-date Cryogenian time: the Moine was deposited prior to
c. 840 Ma, the age of cross-cutting igneous intrusions and mag-
matic overgrowths on detrital zircons (e.g. Vance et al. 1998;
Millar 1999; Kirkland et al. 2008), and deposition of the Torrido-
nian occurred before c. 950 Ma, as based on Rb – Sr and Pb/Pb
diagenetic ages and U –Pb ages on the youngest detrital zircons
(Turnbull et al. 1996; Rainbird et al. 2001).
The Dalradian is a mainly metasedimentary succession sub-
divided into four major units (Harris et al. 1994): the Grampian,
Appin, Argyll and Southern Highland Groups (Figs 63.1 &
63.2). Greenschist- to amphibolite-facies metamorphism and pen-
etrative tectonic fabrics associated with the Neoproterozoic Knoy-
dartian (c. 840 –720 Ma; e.g. Piasecki & van Breeman 1983;
Vance et al. 1998; Tanner & Evans 2003) and Ordovician Gram-
pian (c. 465 Ma; Oliver et al. 1998) tectonothermal events in
many places obscure original sedimentary features and render pro-
blematic any assessment of depositional processes. Areas of low
strain are geological rosetta stones from which to establish strati-
graphic relationships and make palaeoenvironmental interpret-
ations, and to extrapolate those to the more deformed portions of
the Dalradian outcrop belt. It needs to be noted that in much of
the older literature on the Dalradian, the term ‘Boulder Bed’
appears. This term was applied to schists and pelites containing
polymict, variably shaped clasts in matrix-supported textures
with poor sorting. Many of these deposits are not glaciogenic in
origin; research is currently being undertaken by the authors to
document those that are.
The best-known, longest-studied ancient glacial deposit in the
Dalradian is the Port Askaig Fm. at the base of the Argyll Group
(Fig. 63.2). It is present in numerous localities scattered across
the Dalradian outcrop belt and is more than 500 m thick at its
type locality on the Inner Hebridean island of Islay (Fig. 63.1).
The seminal study of Spencer (1971), building upon the efforts
of earlier workers, confirmed its glaciogenicity. Another, but
younger glaciogenic interval to be well documented is the
12-m-thick MacDuff Boulder Bed, exposed near the eponymous
town along the Banffshire coastline in NE Scotland (Figs 63.1 &
63.2). Sutton & Watson (1954) initially, and Stoker et al. (1999)
subsequently, showed that this unit contains ice-rafted dropstones.
Condon & Prave (2000) reported on other ice-rafted-debris beds a
few decimetres to several metres thick along the Inishowen Penin-
sula of Donegal, Ireland (Figs 63.1 & 63.2). They proposed corre-
lations of these beds to the MacDuff Boulder Bed, as well as
another Boulder Bed, the Loch na Cille Boulder Bed, on the
Tayvallich Peninsula in southwestern Scotland (Fig. 63.1); all of
these occurrences are in the lower Southern Highland Group.
McCay et al. (2006) documented the Stralinchy –Reelan glacial
and Cranford Limestone cap-carbonate sequence in the middle
of the Argyll Group (near the top of the Easdale Subgroup) in
Donegal (Figs 63.1 & 63.2). This work confirmed that the Dalra-
dian contained three, stratigraphically distinct glacial intervals:
(i) the Port Askaig Fm. (base of the Argyll Group); (ii) the
glacial –cap carbonate couplet of the Stralinchy –Reelan and Cran-
ford Limestone formations (Easdale Subgroup); and (iii) the
MacDuff –Inishowen – Loch na Cille Beds (lower Southern High-
land Group). Subsequently, Prave et al. (2009a) interpreted the
Whiteness Limestone in the Shetland Islands, some 200 km
north of the Scottish mainland, as a cap-carbonate correlative
with the Cranford Limestone.
Structural framework
The rather severe tectonothermal overprinting of the Dalradian
rocks during early Palaeozoic Caledonian orogenesis resulted in

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 643– 648. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.63
644 A. R. PRAVE & A. E. FALLICK

Fig. 63.1. Highly generalized geological map of the Scottish– Irish Dalradian Highlands highlighting the outcrop belts of the Argyll and Southern Highlands Groups.
Small inset map shows simplified tectonic template of the Highlands. GGF, Great Glen Fault; HBF, Highland Boundary Fault; MT, Moine Thrust. These fault
zones demarcate the three Proterozoic Supergroups of the Highlands: the Torridonian occurs NW of the Moine Thrust, the Moine is between the Moine Thrust and Great
Glen Fault, and the Dalradian (grey-shading) is bounded by the Great Glen and Highland Boundary Fault Zones. Main map, locations discussed in the text: C, Cranford;
LC, Loch na Cille; M, MacDuff; P, Port Askaig; R, Reelan; S, Stralinchy.

structural disruption of the Dalradian outcrop belt by a series of 1984). This structural and metamorphic overprinting makes
so-called tectonic ‘slides’. Most of these are interpreted as exten- assessment of original depositional settings problematic. Never-
sional collapse structures reactivated along original reverse or theless, the Dalradian Supergroup is often interpreted as recording
thrust faults (Harris et al. 1994), but several are thought to rep- deposition in a succession of rift basins (Anderton 1982, 1985;
resent original basin-bounding faults (e.g. Soper & Anderton Robertson & Smith 1999; Dalziel & Soper 2001; Arnaud &
Eyles 2006). This view, however, has been questioned, and the
exact nature of basinal genesis remains an outstanding problem
Cambrian Leny Limestone Ma
to resolve (Prave 1999). The terminal Neoproterozoic geodynamic
542
Southern Highland Gp

evolution of the Dalradian succession is, however, relatively

P

uncontroversial and records syn-depositional extensional tecton-

U

ism and continental rifting (Anderton 1979, 1982, 1985, 1988;

O

582 Soper & Anderton 1984). This phase of extensional tectonics

MacDuff–Inishowen–Loch na Cille Boulder Beds began prior to deposition of the middle glacial –cap carbonate
R

couplet (Stralinchy –Reelan and Cranford formations) and pro-

Ediacaran

gressed to active rifting through the lower Southern Highland

G

601 Ma Tayvallich volcanics

Group. Antecedent to this, the lithostratigraphic framework of
Tayvallich Sgps
R

the underlying parts of the Dalradian is not readily attributable

Crinan &

to deposition in rift basins and any such interpretation should be

E

viewed with a healthy scepticism (Prave 1999).

P
Argyll Group

‘upper’ Ardrishaig–Craignish & Port Ellen Phyllites

U

Subgroup

Cranford–Whiteness Limestones (cap carbonates)

Easdale

IRD
635 Stratigraphy
S

‘lower’ Ardrishaig–Craignish Phy.

Stralinchy
Congl. Easdale Slates
Scarba Congl. Group and subgroup correlations across the Dalradian outcrop belt
Atholl Subgroups Subgroup

Jura – Slieve Tooey Quartzites

are achievable but, because of variable exposure quality and perva-
Islay

Bonahaven Dolostone
sive deformational fabrics, finer-resolution correlations between
N

Port Askaig Formation 713

many areas are seldom attainable. Consequently, correlations
A

Ballachulish & Blair

Islay – Glencolumbkille Limestones from one locality to the next rely largely on the recognition of
Cryogenian
Appin Gp
I

packages of associated lithologies, rather than single units or

D

stratal surfaces (such as those utilized in sequence stratigraphy).

Nevertheless, over 150 years of mapping has provided a reliable
A

geological framework from which to construct a stratigraphy for

R
Grampian Gp

the Dalradian’s glaciogenic units. Like many long-studied succes-

L

sions, the Dalradian is rife with stratigraphic nomenclature. To

A

avoid getting bogged down in a morass of names, discussion is

restricted to the stratigraphic units directly pertinent to the glacio-
D

? genic rocks and their associated strata (see Fig. 63.2).

? ? (nature of base, whether unconformable or stratigraphic, remains unresolved)
metamorphic rocks having 840–800 Ma deformational ages (U-Pb zircon & monazite)

Fig. 63.2. Simplified stratigraphic framework of the Dalradian Supergroup. Glaciogenic deposits and associated strata
The three recognized glaciogenic intervals in the Dalradian (shaded) and their
inferred correlation to geochronological ages and subdivisions of The following focuses on the two younger Neoproterozoic glacial
Neoproterozoic Earth history are shown, as are other stratigraphic units units and their associated strata. The oldest glacial, the Port
mentioned in the text. IRD, ice-rafted debris. See text for details. Askaig, is discussed by Arnaud & Fairchild (2011).
 NEOPROTEROZOIC GLACIOGENIC DEPOSITS OF SCOTLAND AND IRELAND
The Stralinchy – Reelan formations and the Cranford and
Whiteness cap carbonates
McCay et al. (2006) showed that the Stralinchy Conglomerate and
its basinal equivalent containing isolated dropstones, the Reelan
Fm., together with the overlying Cranford Limestone constitute a
glacial –cap carbonate succession in the middle part of the
Argyll Group in Donegal, Ireland (Figs 63.1 & 63.2). The glacial
rocks occur as lenses (from 0 to several tens of metres in thickness)
of mono- to poly-mict, poorly sorted sedimentary breccia (Stralin-
chy Conglomerate) and laterally equivalent thin-bedded pelite and
limestone containing pebble- to cobble-sized dropstones and
lonestones (Reelan Fm.). Clasts consist of quartzo-feldspathic,
quartzitic, schistose and carbonate lithologies. The sedimentary
breccias alter from clast-supported to diamictic in texture, clasts
range in size from granule to metre-sized blocks, and matrix
composition varies along strike from siliciclastic- to carbonate-
dominated. Lonestones and dropstones pierce and distort
laminae in decimetre-thick intervals of finely compositionally
layered pelite and limestone; on-lap and splash-up structures are
associated with some of the larger clasts. Sitting sharply on these
glaciogenic units is the Cranford Limestone. Its base is marked
by a 1– 7-m-thick, laterally continuous tan-coloured dolomicrite,
that is, a cap carbonate. At the type locality, the cap is overlain
by 200–300 m of limestone-rhythmite and rhythmite breccia. In
basinal settings, the Cranford Limestone shales out (metamor-
phosed to phyllite; Alsop & Hutton 1990) and only the basal cap
dolostone remains.
As documented by Prave et al. (2009a), an inferred correlative
cap carbonate, the Whiteness Limestone, is present in the Shetland
Islands (these islands lie 200 km north of mainland Scotland).
Similar to the Cranford Limestone (and to cap carbonates else-
where; e.g. Hoffman & Schrag 2002; Fairchild & Kennedy
2007), the Whiteness Limestone has a basal, 2– 5-m-thick, light-
coloured dolostone to dolomitic limestone overlain by 20– 40 m
of thin-bedded limestone-pelite rhythmites. These, in turn, pass
upward into a tens-of-metres-thick pelitic interval, itself overlain
by many hundreds of metres of thick-bedded calcitic marbles.
The basal dolostone sits sharply on a thick (1– 2 km) sequence
of siliciclastic units devoid of carbonate rocks. Within 5– 10 m
of the base of the cap, the underlying schistose rocks become con-
spicuously finer-grained and darker-coloured. It is within this
interval that rare quartzo-feldspathic lonestone clasts can be
found. The absence of any physical or textural evidence for depo-
sition of the lonestones by traction currents, sediment-gravity-flow
or volcanic ejecta processes, and their stratigraphic restriction to
subjacent to the base of the Whiteness cap, led Prave et al.
(2009a) to interpret the lonestones as ice-rafted debris, that is, a
glacially influenced phase of sedimentation pre-dating deposition
of the cap carbonate.
The presence of glaciogenic rocks and cap carbonates at the
southwestern (Donegal) and northeastern (Shetland) edges of the
Dalradian outcrop belt has focused research efforts on mainland
Scotland to document correlative glacial –cap carbonate couplets
there, or find a reason for their absence.
Boundary relations with overlying and underlying
non-glacial units
In many places across the Dalradian outcrop belt, Caledonian
strain compromises contacts, but in areas of low strain, and
where the glaciogenic units consist of sedimentary breccia or dia-
mictite (e.g. the Stralinchy type locality in Donegal), the contact is
sharp and erosive. Where the glaciogenic units consist of drop-
stone or lonestone intervals, defining contacts relies on finding dis-
persed clasts and/or the contact with the overlying cap carbonate
to confirm being in or near glacial strata. The base of the cap
645
carbonate is everywhere sharp. Commonly, only the basal cap
dolostone is present and in these localities the top also appears
sharp. Where the entire cap-carbonate sequence is developed
(the eponymous type localities of the Cranford and Whiteness
Limestones), the basal cap displays a gradational contact into over-
lying carbonate rhythmites and/or a thin calc-pelite unit, itself
transitioning into carbonate rhythmites. Thick siliciclastic-
dominated intervals overlie the cap sequences.
MacDuff, Inishowen and the Loch na Cille Boulder Beds
The MacDuff Boulder Bed occurs in the core of a syncline of the
lower Southern Highland Group rocks along the coast adjacent to
the eponymous Banffshire town (Fig. 63.1); it is c. 12 m thick and
has long been recognized as glaciogenic in origin (Sutton &
Watson 1954; Stoker et al. 1999). It was speculated to be as
young as Ordovician in age (Hambrey 1983; Molyneux 1997),
but this is wrong: regional mapping leaves no doubt that it is
much older and in the Southern Highland Group (e.g. Read
1923). The MacDuff Boulder Bed is part of a thick succession of
siliciclastic rocks interpreted as a deep-marine fan complex (e.g.
Trewin 1987). What distinguishes it is the presence of several
dropstone intervals in which variably shaped clasts, consisting of
quartzite, gneiss, igneous and pale-coloured carbonate rocks
ranging from angular to rounded and from granule to boulder in
size, can be seen to pierce laminae, be on-lapped by subsequent
laminae and, in some places, show splash-up structures. They
are concentrated in discrete, decimetre-thick units indicating that
they originated during periods of rainout, rather than as part of
some continuous background sedimentation process, and an
origin as ice-rafted debris, not tills (suggested by Hambrey &
Waddams 1981), is certain (Sutton & Watson 1954; Stoker et al.
1999). Stoker et al. (1999) also detailed aspects of seven
decimetre-thick diamictite beds. These contain a variety of clasts
of varying sizes and shapes in a mud matrix and were interpreted
as resedimented glaciomarine deposits.
Condon & Prave (2000) documented glaciogenic deposits in the
Southern Highland Group along the northern Inishowen Peninsula
(Fig. 63.2), Donegal, Ireland. These consist of five dropstone units,
each many centimetres to several decimetres thick, spaced irregu-
larly through a c. 500-m-thick arkosic psammite-pelite succession.
The dropstone units are characterized by granule- to pebble-sized
clasts of quartzite, gneiss, schist and igneous rocks; in many places
the clasts can be observed to pierce laminae and be on-lapped by
successive laminae. These units are separated by decimetre- to
metre-thick intervals of fine-grained pelite devoid of lonestones
or gritty horizons. The arkosic nature of these strata is noteworthy
because, for hundreds of metres above and below this interval,
units are quartzitic in composition. Condon & Prave (2000) and
Condon et al. (2002) interpreted the dropstone beds as ice-rafted
debris and speculated that the arkosic composition of the encasing
strata was due to cold-climate weathering processes. Condon et al.
(2002) also documented an iceberg dump structure, a poorly
sorted, coarse-grained dome-shaped sedimentary feature having
a flat basal surface moulded onto the top of a pelite bed. It is
c. 1–2 m in length along its longest exposed axis and c. 10–
20 cm thick along its crest before tapering off irregularly to its
thinned edges.
A distinctive unit occurs in the lower Southern Highland Group
in SW Scotland along the Tayvallich Peninsula (Fig. 63.2), the
Loch na Cille Boulder Bed (Elles 1934). It occurs as discontinuous
lenses, typically several metres thick, and consists of matrix-
supported polymict clasts, including felsite and mafic volcanites,
granitoid, sedimentary and metamorphic rock fragments (Alsop
et al. 2000). It was recognized and mapped separately because of
its distinctiveness from the associated hundreds of metres of volca-
nic and pelitic rocks. Its matrix is finer-grained, lighter-coloured
and more quartzo-feldspathic than the encasing rocks, and its
646 A. R. PRAVE & A. E. FALLICK

clast content is polymict, rather than monomict volcanic detritus.
These observations, and compatible stratigraphic position of the
Loch na Cille Boulder Bed with the MacDuff and Inishowen
beds, indicate that it, too, is a glaciogenic deposit.
Boundary relations with overlying and underlying
non-glacial units
The lowest observed, glacially influenced deposits in the Southern
Highland Group rocks occur abruptly. They are eye-catching
because they are compositionally distinct, both in clast content
and matrix, from encasing strata whether those are sedimentary,
as in the case of the MacDuff and Inishowen rocks, or volcaniclas-
tic, as in the Loch na Cille Beds. The upper contacts of the Inisho-
wen and Loch na Cille Beds are placed at the last occurrence of
ice-rafted debris; the MacDuff Boulder Bed has no top, it is the
youngest unit in the core of an exposed syncline. The impression
one gets from all three units is that glaciogenic deposition was
intermittent, discrete events superimposed on the overall
background sedimentation.
confirm that negative C-isotopic excursions are antecedent to
two stratigraphically and temporally distinct glaciations and urge
caution when attempting to use such excursions as chronostrati-
graphic markers (see Halverson et al. 2005). The inferred cap
carbonates exhibit a trend commonly attributed to post-Mar-
inoan-equivalent successions (e.g. Kennedy et al. 1998), namely
values declining from c. –3‰ in the basal cap, reaching a nadir
at c. – 6‰ in overlying limestone rhythmites, and then recovering
back towards positive values up-section. It should be noted that the
basal Whiteness Limestone on the Shetland Islands has initial
values between 1‰ and 2‰. A more detailed C-isotopic study
of the Dalradian is given in Prave et al. (2009b).
Other characteristics (e.g. economic deposits, biomarkers)
There are no economic deposits associated with the Dalradian’s
glacial rocks. The cap carbonates have, in many places, been quar-
ried, but those operations were small and not commercially viable.
Likewise, no depositional biomarkers have been found in any of
these rocks, not surprising given the metamorphic grade.

Chemostratigraphy
Palaeolatitude and palaeogeography
Figure 63.3 provides a summary diagram of our C-isotopic data
associated with the Dalradian’s glaciogenic intervals. Dramatic There are no reliable depositional palaeomagnetic data on the
declines, from values of þ2‰ to þ6‰ to as low as –10‰ to Dalradian rocks; this renders palaeolatitudinal placements, at
–12‰, and recoveries mark the carbonate strata predating both best, speculative. Generalized palaeogeographies envisage east
the Port Askaig and Stralinchy-Reelan formations. These data to west (current-day coordinates) transitions from shallow-marine
shelf to deeper-marine settings in rift basins along the eastern
margin of Laurentia for most of Dalradian time (Anderton 1982,
1985; Harris et al. 1994). Although several nicely detailed local
studies have been undertaken (Klein 1970; Anderton 1976; Fair-
INISHOWEN–MACDUFF–LOCH NA CILLE GLAC. child 1980), much additional work is needed to help refine recon-
structions. Complicating matters are the wildly varying estimates
Marinoan–equivalent for the magnitude of pre-Devonian displacement along the Great
cap carbonates Glen Fault Zone; restoration of the Dalradian block can vary by
Whiteness Limestone
Cranford Limestone
as much as 1000 km (e.g. Dewey & Strachan 2003). Most recon-
structions incorporate the Dalradian as part of eastern Laurentia,
the position of which within a disassembling Rodinia is often
depicted as moving along a broadly low- to mid-latitude arc
during Cryogenian time (e.g. Dalziel 1997; Pisarvesky et al. 2008).
STRALINCHY – REELAN GLACIAL
Ardrishaig Phyllite (& correlatives) Geochronological constraints
pre–glacial decline
and recovery Geochronological constraints are frustratingly sparse for the
(Trezona anomaly Dalradian:
equivalent?)
Easdale Subgroup † The only direct depositional age for the Dalradian is from
carbonate rocks an ash bed in the Tayvallich Volcanics near the base of the
Southern Highland Group. It has a U –Pb zircon age of
601 + 4 Ma (Dempster et al. 2002) and occurs more than
Bonahaven 500 m beneath the youngest Dalradian glacial beds (in this
Formation
instance, the Loch na Cille Boulder Bed). The top of the Dalra-
PORT ASKAIG GLACIAL dian, well above the youngest glacial rocks, passes stratigraphi-
cally into the Cambrian Leny Limestone (Tanner 1995).
pre–glacial decline † The nature of the base of the Dalradian, whether unconformable
and recovery Islay–Glencolumbkille–
Limestones or depositional on rocks containing U –Pb zircon and monazite
deformational ages ranging from c. 840–800 Ma (Highton
et al. 1999; Noble et al. 1996), remains unresolved. If the
base defines an unconformity, then these are maximum ages
-15 -10 -5 0 5 10 for the initiation of Dalradian sedimentation; if the base is tran-
sitional (albeit tectonized), then the lower parts of the Dalradian
δ C‰ (V-PDB)
13
could be much older. Regardless, the earliest record of glacia-
Fig. 63.3. C-isotopic trends for part of the Dalradian Supergroup with inferred tion in the Dalradian, the Port Askaig Fm. at the base of the
correlations to known Neoproterozoic chemostratigraphic events. Data are Argyll Group, occurs some 10 km above these rocks.
from McCay et al. (2006), Prave et al. (2009a) and Prave & Fallick † The youngest concordant detrital zircon U –Pb ages for the
(unpublished). See text for discussion. Argyll Group are c. 1000 Ma (Cawood et al. 2003). These are
 NEOPROTEROZOIC GLACIOGENIC DEPOSITS OF SCOTLAND AND IRELAND
not robust constraints for depositional ages, but do confirm a
post-early Neoproterozoic age for the initiation of Argyll
Group sedimentation.
† Time lines for the Dalradian have been proposed using chron-
ostratigraphic constraints established for global climatic
events and secular variations in the isotopic compositions of
Neoproterozoic oceans. The proposed linkage by McCay
et al. (2006) of the three Dalradian glacials to Neoproterozoic
glaciations known from elsewhere is viable and testable. In
this framework, the Port Askaig Fm. would be chronostratigra-
phically tied to the mid-late Neoproterozoic Sturtian glacia-
tion(s). We recognize that the age and number of glaciations
during this climatic phase remain contentious, but Brasier &
Shields (2000) have argued, based on C and Sr isotopes, that
the Port Askaig is best correlated to the c. 713 Ma Gubrah
glacial in Oman (Bowring et al. 2007). This, then, would be
the age of (or close to) initiation of Argyll Group sedimentation.
The Cranford –Whiteness Limestones are interpreted as
Marinoan-equivalent cap carbonates (McCay et al. 2006;
Prave et al. 2009a), thereby making this part of the Easdale
Subgroup 635 Ma, that is the age for the end of the Marinoan
glaciation (Hoffmann et al. 2004; Condon et al. 2005). The
MacDuff – Inishowen –Loch na Cille beds, which occur above
the 601 Ma Tayvallich Volcanics and their correlatives, can
be reasonably linked to the 582 Ma Gaskiers glaciation
(Bowring et al. 2003, 2007).
The best these data permit is to indicate that the post-Easdale Sub-
group portion of the Dalradian belongs to the Ediacaran Period and
that the underlying Dalradian is Cryogenian in age (Fig. 63.2).
Such broad age constraints are unsatisfying. The irony is that,
although it is one of the most-studied and mapped regions in the
world, the Scottish –Irish Highlands is one of the most poorly
constrained, geochronologically.
Discussion
Two main data sets have enabled construction of a glacial-based
stratigraphic framework for the Scottish –Irish Dalradian. The
first is the presence of discrete glacially influenced deposits
ranging from thick diamictites to thin, discontinuous ice-rafted-
debris beds (in areas of higher metamorphic grade, these are recog-
nized as fine-grained schists with lonestones). Excepting the Port
Askaig Fm., with its impressive, laterally extensive diamictite
beds, the record of Neoproterozoic glaciation in the Dalradian
is modest and mostly represented by thin ice-rafted-debris
intervals. The second data set comes from carbonate units that
are lithologically and C-isotopically compatible with Neoprotero-
zoic cap carbonates.
The Scottish – Irish Dalradian succession contains evidence for
three distinct glaciations, from oldest to youngest, (i) the Port
Askaig Fm. at the base of the Argyll Group (note that no associated
cap carbonate has been recognized), (ii) the Stralinchy –
Reelan formations (and correlative ice-rafted-debris beds) and
the Cranford and Whiteness cap carbonates in the Easdale
Subgroup, and (iii) the MacDuff –Inishowen –Loch na Cille
Boulder Beds and ice-rafted-debris units in the lower Southern
Highland Group. Geochronological constraints are sparse for the
Dalradian, but those that exist, combined with C-isotope chemo-
stratigraphy, enable placing the base of the Ediacaran Period in
the Easdale Subgroup and the underlying Dalradian units in the
Cryogenian.
Numerous colleagues have provided insight and help over the years and we would
like to particularly acknowledge I. Alsop, D. Benn, D. Condon, G. McCay,
G. Oliver, C. Rose, J. Soper, R. Strachan and G. Thomas. We thank The Carnegie
Trust for The Universities of Scotland, The Russell Trust and The Schroder Foun-
dation (Islay) for financial support. This represents a contribution of the IUGS-
and UNESCO-funded IGCP (International Geoscience Programme) Project #512.
647
References
Alsop, G. I. & Hutton, D. W. 1990. A review and revision of Dalradian
stratigraphy in central and southern Donegal, Ireland. Irish Journal of
Earth Sciences, 10, 181–198.
Alsop, G. I., Prave, A. R., Condon, D. J. & Phillips, C. A. 2000.
Cleaved clasts in Dalradian conglomerates: possible evidence for
Neoproterozoic compressional tectonism in Scotland and Ireland?
Geological Journal, 35, 87– 98.
Anderton, R. 1976. Tidal shelf sedimentation: an example from the
Scottish Dalradian. Sedimentology, 23, 429–458.
Anderton, R. 1979. Slope, submarine fans, and syn-depositional faults:
sedimentology of parts of the Middle and Upper Dalradian in the
SW Highlands of Scotland. In: Harris, A. L., Holland, C. H. &
Leake, B. E. (eds) The Caledonides of the British Isles – Reviewed.
Geological Society, London, Special Publications, 8, 483–488.
Anderton, R. 1982. Dalradian deposition and the late Precambrian –
Cambrian history of the North Atlantic region: a review of the early
evolution of the Iapetus Ocean. Journal of the Geological Society,
London, 139, 421– 431.
Anderton, R. 1985. Sedimentation and tectonics in the Scottish Dalra-
dian. Scottish Journal of Geology, 21, 407– 436.
Anderton, R. 1988. Dalradian slides and basin development: a radical
reinterpretation of stratigraphy and structure in the SW and Central
Highlands of Scotland. Journal of the Geological Society, London,
145, 669–678.
Arnaud, E. & Eyles, C. H. 2006. Neoproterozoic environmental change
recorded in the Port Askaig Formation, Scotland: climatic and
tectonic controls on sedimentation. Sedimentary Geology, 183,
99– 124.
Arnaud, E. & Fairchild, I. J. 2011. The Port Askaig Formation, Dalra-
dian Supergroup, Scotland. In: Arnaud, E., Halverson, G. P. &
Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic
Glaciations. Geological Society, London, Memoirs, 36, 635– 642.
Bowring, S., Myrow, P., Landing, E., Ramezani, J. & Grotzinger, J.
2003. Geochronological constraints on terminal Neoproterozoic
events and the rise of metazoans. Geophysical Research Abstracts,
5, 13219.
Bowring, S. A., Grotzinger, J. P., Condon, D. J., Ramezani, J.,
Newall, M. & Allen, P. A. 2007. Geochronologic constraint of
the chronostratigraphic framework of the Neoproterozoic Huqf
Supergroup, Sultanate of Oman. American Journal of Science, 307,
1097–1145.
Brasier, M. D. & Shields, G. 2000. Neoproterozoic chemostratigraphy
and correlation of the Port Askaig glaciation, Dalradian Supergroup
of Scotland. Journal of the Geological Society, London, 157,
909– 914.
Cawood, P. A., Nemchin, A. A., Smith, M. & Loewy, S. 2003. Source of
the Dalradian Supergroup constrained by U –Pb dating of detrital
zircon and implications for the East Laurentian margin. Journal of
the Geological Society, London, 160, 231–246.
Condon, D. J. & Prave, A. R. 2000. Two from Donegal: Neoproterozoic
glacial episodes on the NE margin of Laurentia. Geology, 28,
951– 954.
Condon, D. J., Prave, A. R. & Benn, D. 2002. Neoproterozoic glacial –
rainout intervals: observations and implications. Geology, 30, 35– 38.
Condon, D., Zhu, M., Bowring, S., Wang, W., Yang, A. & Jin, Y. 2005.
U–Pb ages from the Neoproterozoic Doushantuo Formation, China.
Science, 308, 95– 98.
Dalziel, I. W. D. 1997. Neoproterozoic-Paleozoic geography and tec-
tonics: review, hypothesis, environmental speculation. Geological
Society of America Bulletin, 109, 16– 42.
Dalziel, I. W. D. & Soper, N. J. 2001. Neoproterozoic extension on the
Scottish Promontory of Laurentia: paleogeographic and tectonic
implications. Journal of Geology, 109, 299–317.
Davison, S. & Hambrey, M. J. 1996. Indications of glaciation at the base
of the Proterozoic Stoer Group (Torridonian), NW Scotland. Journal
of the Geological Society, London, 153, 139–149.
Dempster, T. J., Rogers, G. et al. 2002. Timing of deposition, orogen-
esis and glaciation within the Dalradian rocks of Scotland: Con-
straints from U –Pb zircon ages. Journal of the Geological Society,
London, 159, 83 – 94.
648 A. R. PRAVE & A. E. FALLICK
Dewey, J. F. & Strachan, R. A. 2003. Changing Silurian –Devonian
relative plate motion in the Caledonides: sinistral transpression to
sinistral transtension. Journal of the Geological Society, London,
160, 219– 229.
Elles, G. L. 1934. The Loch na Cille Boulder Bed and its place in the
Highland succession, Geological Society of London Quarterly
Journal, 91, 111– 147.
Fairchild, I. 1980. Sedimentation and origin of a late Precambrian ‘dolo-
mite’ from Scotland. Journal of Sedimentary Petrology, 50, 423–446.
Fairchild, I. & Kennedy, M. J. 2007. Neoproterozoic glaciation in the
Earth System. Journal of the Geological Society, London, 164,
895– 921.
Hambrey, M. J. 1983. Correlation of Late Proterozoic tillites in the North
Atlantic region and Europe. Geological Magazine, 120, 209– 232.
Hambrey, M. J. & Waddams, P. 1981. Glaciogenic boulder-bearing
deposits in the Upper Dalradian MacDuff Slates, northeastern
Scotland. In: Hambrey, M. J. & Harland, W. W. (eds) Earth’s
Pre-Pleistocene Glacial Record. Cambridge University Press,
Cambridge, 571– 575.
Harris, A. L., Haselock, P. J., Kennedy, M. & Mendum, J. R. 1994. The
Dalradian Supergroup in Scotland, Shetland and Ireland. In:
Gibbons, W. & Harris, A. L. (eds) A Revised Correlation of Precam-
brian Rocks in British Isles. Geological Society, London, Special
Report, 22, 33 –53.
Highton, A. J., Hyslop, E. K. & Noble, S. R. 1999. U– Pb zircon geo-
chronology of migmatisation in the northern-central Highlands: evi-
dence for pre-Caledonian (Neoproterozoic) tectonometamorphism
in the Grampian block, Scotland. Journal of the Geological Society,
London, 156, 1195–1204.
Hoffman, P. F. & Schrag, D. P. 2002. The snowball Earth hypothesis:
testing the limits of global change. Terra Nova, 14, 129–155.
Hoffmann, K. H., Condon, D. J., Bowring, S. A. & Crowley, J. L. 2004.
A U–Pb zircon date from the Neoproterozoic Ghaub Formation,
Namibia: constraints on Marinoan glaciation. Geology, 32, 817–820.
Kennedy, M. J., Runnegar, B., Prave, A. R., Hoffmann, K. H. &
Arthur, M. 1998. Two or four Neoproterozoic glaciations?
Geology, 26, 1059–1063.
Kirkland, C. L., Strachan, R. A. & Prave, A. R. 2008. Detrital zircon
signature of the Moine Supergroup, Scotland: contrasts and compari-
sons with other Neoproterozoic successions within the circum-North
Atlantic region. Precambrian Research, 163, 332–350.
Klein, G. De V. 1970. Tidal origin of a Precambrian quartzite: lower fine-
grained quartzite (Middle Dalradian) of Islay, Scotland. Journal of
Sedimentary Petrology, 40, 973– 985.
McCay, G. A., Prave, A. R., Alsop, G. I. & Fallick, A. E. 2006. Glacial
trinity: Neoproterozoic Earth history within the British– Irish Caledo-
nides. Geology, 34, 901–912.
Millar, I. L. 1999. Neoproterozoic extensional basic magmatism associ-
ated with the West Highland granite gneiss in the Moine Supergroup
of NW Scotland. Journal of the Geological Society, London, 156,
1153– 1162.
Molyneux, S. G. 1997. An upper Dalradian microfossil reassessed.
Journal of the Geological Society, London, 155, 741–743.
Noble, S. R., Hyslop, E. K. & Highton, A. J. 1996. High precision U –Pb
monazite geochronology of the c.806 Ma Grampian Shear Zone and
implications for the evolution of the Central Highlands of Scotland.
Journal of the Geological Society, London, 153, 511–514.
Oliver, G. J. H., Chen, F., Buchwald, R. & Hegner, E. 1998. Fast tec-
tonometamorphism and exhumation in the type area of the Barrovian
and Buchan zones. Geology, 28, 459– 462.
Piasecki, M. A. J. & van Breemen, O. 1983. Field and isotope evidence
for a c.750 Ma tectonothermal event in Moine rocks in the Central
Highland region of the Scottish Caledonides. Transactions of the
Royal Society of Edinburgh: Earth Sciences, 73, 119– 134.
Pisarvesky, S. A., Murphy, J. B., Cawood, P. A. & Collins, A. S.
2008. eeLate Neoproterozoic and early Cambrian palaeogeography:
models and problems. Journal of the Geological Society, London,
294, 9– 31.
Prave, A. R. 1999. The Neoproterozoic Dalradian Supergroup
of Scotland: an alternative hypothesis. Geological Magazine, 136,
609– 617.
Prave, A. R., Strachan, R. A. & Fallick, A. E. 2009a. Global C cycle
perturbations recorded in marbles: a record of Neoproterozoic Earth
history within the Shetland Islands, Scotland. Journal of the Geologi-
cal Society, London, 166, 129–135.
Prave, A. R., Fallick, A. E., Thomas, C. W. & Graham, C. M. 2009b. A
composite C-isotopic profile for the Neoproterozoic Dalradian Super-
group of Scotland and Ireland. Journal of the Geological Society,
London, 166, 845–857.
Rainbird, R. H., Hamilton, M. A. & Young, G. M. 2001. Detrital zircon
geochronology and provenance of the Torridonian, NW Scotland.
Journal of the Geological Society, London, 158, 15 –27.
Read, H. H. 1923. The geology of the country around Banff, Huntly and
Turriff. Memoir Geological Survey Scotland, 86 & 96, 240.
Robertson, S. & Smith, M. 1999. The significance of the Geal Charn–
Ossian Steep Belt in basin development and inversion in the
Central Scottish Highlands. Journal of the Geological Society,
London, 156, 1175– 1182.
Soper, N. J. & Anderton, R. 1984. Did the Dalradian slides originate as
extensional faults? Nature, 307, 357–360.
Spencer, A. M. 1971. Late Pre-cambrian Glaciaion in Scotland. Geo-
logical Society, London, Memories, 6, 100.
Stewart, A. D. 1997. Discussion on: Indications of glaciation at the base
of the Proterozoic Stoer Group (Torridonian), NW Scotland. Journal
of the Geological Society, London, 154, 375– 376.
Stewart, A. D. 2002. The Later Proterozoic Torridonian Rocks of Scot-
land: their Sedimentology, Geochemistry and Origin. Geological
Society, London, Memoirs, 24, 130.
Stoker, M. S., Howe, J. A. & Stoker, S. J. 1999. Late Vendian? Cam-
brian glacially influenced deepwater sedimentation, MacDuff Slate
Formation (Dalradian), NE Scotland. Journal of the Geological
Society, London, 156, 55 –61.
Sutton, J. & Watson, J. V. 1954. Ice-borne boulders in the MacDuff
Group of the Dalradian of Banffshire. Geological Magazine, 91,
391– 398.
Tanner, P. W. G. 1995. New evidence that the Lower Cambrian Leny
Limestone at Callender, Perthshire, belongs to the Dalradian Super-
group, and a reassessment of the ‘exotic’ status of the Highland
Border Complex. Geological Magazine, 132, 473– 483.
Tanner, P. W. G. & Evans, J. A. 2003. Late Precambrian U– Pb titanite
age for peak regional metamorphism and deformation (Knoydartian
orogeny) in the western Moine, Scotland. Journal of the Geological
Society, London, 160, 555–564.
Trewin, N. H. 1987. MacDuff, Dalradian turbidite fan and glacial depos-
its. In: Trewin, N. H., Kneller, B. C. & Gillen, C. (eds) Geology of
the Aberdeen Area. Geological Society of Aberdeen. Scottish Aca-
demic Press, Edinburgh, 79 –88.
Turnbull, M. J. M., Whitehouse, M. J. & Moorbath, S. 1996. New
isotopic age determinations for the Torridonian, NW Scotland.
Journal of the Geological Society, London, 153, 955– 964.
Vance, D., Strachan, R. A. & Jones, K. A. 1998. Extensional
versus compressional settings for metamorphism: garnet chrono-
metry and pressure– temperature –time histories in the Moine Super-
group, NW Scotland. Geology, 26, 927– 930.
 Chapter 64

Neoproterozoic glacial deposits of Tasmania

CLIVE R. CALVER
Mineral Resources Tasmania, P.O. Box 56, Rosny Park, Tasmania, Australia 7018 (e-mail: ccalver@mrt.tas.gov.au)

Abstract: In Tasmania, Neoproterozoic glaciogenic deposits were laid down in one or more epicratonic basins, probably situated at the
eastern margin of the Australian– Antarctic craton. Rifting and volcanism took place in the late Cryogenian to early Ediacaran. On King
Island, north of Tasmania, the Cottons Breccia consists of 50–200 m of diamictite, conglomerate and sandstone. Limestone and dolo-
stone clasts are abundant in the diamictite, although carbonate is unknown in the underlying successions. The Cottons Breccia is overlain
by 10 m of laminated dolostone and limestone with a negative, upward-decreasing d13C profile. Rift volcanics and shallow intrusives
higher in the sequence are dated at c. 575 Ma. In NW Tasmania, two diamictite units are found in the Togari Group. The Julius
River Member, 200 m thick, contains dominantly dolostone clasts and overlies a shallow-marine dolostone unit with vase-shaped micro-
fossils and C-isotopes consistent with a mid-Cryogenian age. Some clasts in the Julius River Member contain a stromatolite (Baicalia cf.
B. burra) very similar to a form that is abundant in the middle part of the Burra Group, Adelaide rift basin. The Julius River Member is
immediately overlain by black shale and impure carbonate dated by Re–Os at 641 + 5 Ma. The younger diamictite in the Togari Group
is the Croles Hill Diamictite, 70 m thick, with predominantly volcanic clasts, underlain by a shale and mafic-volcaniclastic succession
and overlain by thin mudstone followed by thick rift tholeiites. At one locality this diamictite is underlain by a rhyodacite flow dated at
582 + 4 Ma. In southern Tasmania, diamictites are found in the Wedge River Beds and in the Cotcase Creek Formation (Fm.) (Weld
River Group). Laminated siltstone with dropstones is associated with the diamictites in the Cotcase Creek Fm. The southern
Tasmanian deposits are poorly constrained in age.
Chronometric and other evidence suggests correlation of the Julius River Member, Cottons Breccia and Croles Hill Diamictite with the
Sturt, Elatina and Gaskiers glacial phases, respectively. However, a glacial origin for the Julius River Member and Croles Hill Diamictite
remains uncertain.

Proterozoic rocks of King Island and western Tasmania comprise
a continental fragment that lay at the eastern margin of the
Australian/east Antarctic craton in the Neoproterozoic (Burrett
& Berry 2000; Li et al. 2008). This fragment probably rifted at
c. 580 Ma and accreted back to the craton in a late phase
of the Cambrian, Delamerian/Ross Orogeny (Berry et al. 2008).
Alternatively, King Island (and perhaps western Tasmania as
well) comprised a rifted margin that never completely separated
from the craton (Meffre et al. 2000, 2004). Cryogenian to
Ediacaran successions unconformably overlie Mesoproterozoic
to early Neoproterozoic sediments and metasediments in many
places, but Cambrian and later tectonism, as well as significant
lateral variability in the Cryogenian –Ediacaran successions
themselves, have made correlation across Tasmania problematic.
Consequently, three regions with known or probable Neoprotero-
zoic glaciogenic deposits are described separately below: King
Island, NW Tasmania and southern Tasmania.
King Island: the Cottons Breccia
Introduction
The Cottons Breccia (Jago 1974) occurs near the base of the Grassy
Group on King Island (Fig. 64.1). It consists of up to 200 m of
diamictite and minor conglomerate and sandstone, and crops out
over a strike length of about 8 km along or near the SE coast of
the island from Cottons Flat to north of Cumberland Creek
(Fig. 64.1b). In a few places there is excellent exposure on the
coast, although the basal part of the unit is relatively poorly
exposed inland. Waterhouse (1916) first described the unit as a
lithified ‘glacial till’; Carey (1947) suggested it was an Adelaidean
tillite. Jago (1974) named and described the Cottons Breccia in
some detail, but left its mode of origin unresolved. No type
section was nominated. Further description and discussion were
provided by Jago (1981) and Waldron & Brown (1993), the
latter authors favouring a debris-flow origin. Calver & Walter
(2000) gave evidence, including d13C stratigraphy of the overlying
dolostone, for correlation of the Cottons Breccia with the Elatina
glacials of South Australia. Meffre et al. (2004) and Calver et al.
(2004) isotopically dated the overlying volcanics and shallow
intrusives respectively, the latter authors specifically in the context
of correlation and age constraints for the Cottons Breccia. Direen
& Jago (2008) contended that the Cottons Breccia is a non-
glaciogenic mass flow deposit laid down in a developing rift.
Structural framework
The Grassy Group unconformably overlies a thick, Mesopro-
terozoic to early Neoproterozoic siliciclastic shelf succession
(Black et al. 2004). The Grassy Group was deposited in an epi-
cratonic rift basin that may have evolved into an east-facing
passive margin soon after deposition of the Cottons Breccia
(Direen & Crawford 2003; Meffre et al. 2004; Direen & Jago
2008). The thick mafic volcanics (rift tholeiites and picrites) that
begin c. 100 m above the Cottons Breccia have been interpreted
as analogous to the mafic volcanic packages (‘seaward dipping
reflector sequences’) at the base of Mesozoic –Cenozoic volcanic
continental margins (Direen & Crawford 2003; Meffre et al.
2004). The distribution of these volcanics can be traced as an
offshore magnetic anomaly that extends southward to offshore
western Tasmania, but it is not continuous with the similar-aged
Spinks Creek Volcanics (next section), suggesting the Grassy
Group and Togari Group filled separate structural basins. King
Island was probably weakly deformed in the Cambrian, Tyennan
Orogeny (Delamerian/Ross Orogeny correlate). Further defor-
mation may have occurred in the Devonian, Tabberabberan
Orogeny, followed by local granitoid intrusion in the Early
Carboniferous. In the Cottons Flat –Cumberland Creek area, the
Grassy Group dips moderately east, and a weak, subvertical clea-
vage is developed in shales. Minor cross-faulting results in dif-
ferent parts of the succession being exposed along the coast.
Mineral assemblages in the volcanics indicate low greenschist-
facies metamorphism (Meffre et al. 2004). Near Grassy, how-
ever, the sequence has been strongly contact metamorphosed and
metasomatized by the Early Carboniferous Sandblow Granite
(Fig. 64.1b).

From: Arnaud, E., Halverson, G. P. & Shields-Zhou, G. (eds) The Geological Record of Neoproterozoic Glaciations. Geological Society, London,
Memoirs, 36, 649– 657. 0435-4052/11/$15.00 # The Geological Society of London 2011. DOI: 10.1144/M36.64
650
Fig. 64.1. (a) Distribution of Proterozoic rocks and major tectonic elements in Tasmania. (b) Simplified geological map of SE King Island with localities mentioned in
text. (c) Generalized stratigraphic section through Grassy Group in Bold Head–Cumberland Creek area.
Stratigraphy
The Grassy Group (Fig. 64.1c) unconformably overlies the Fraser
Fm., a thick succession of siliceous siltstone and mudstone of
Mesoproterozoic or early Neoproterozoic age (Black et al. 2004;
Direen & Jago 2008). The basal unit of the Grassy Group is the
Robbins Creek Fm., consisting of laminated siltstone and shale
with minor lavas and a thin basal conglomerate (Calver 2008a, b).
The overlying Cottons Breccia shows considerable lithological
variation, both laterally and vertically, and varies irregularly
in thickness from c. 50 m in the north (Cumberland Creek) to
c. 200 m in the south (City of Melbourne Bay) (Jago 1974;
Calver 2008b). The Cottons Breccia is overlain by the Cumberland
Creek Dolostone (c. 10 m thick), which passes up gradationally
into green and black, then red, shale (the Yarra Creek Shale:
Calver et al. 2004). The Yarra Creek Shale varies in thickness
from c. 10 m to 150 m, apparently in response to movement
on penecontemporaneous faults (Direen & Jago 2008; Calver
2008a). A differentiated sill of intermediate composition (Grimes
Intrusive Suite) intrudes the lower Yarra Creek Shale and older
units, and is locally amygdaloidal. The Yarra Creek Shale is con-
formably overlain by the tholeiitic City of Melbourne Volcanics
consisting of peperites and volcanic breccias overlain by pillow
lavas. The picritic Shower Droplet Volcanics follow, then the tho-
leiitic Grahams Road Volcanics (Waldron & Brown 1993; Meffre
et al. 2004; Calver 2008b). The top of the Grassy Group is
unknown and lies offshore. Aeromagnetic mapping shows that
the volcanics continue offshore and may be several kilometres
thick in total (Direen & Crawford 2003).
Glaciogenic deposits and associated strata
The Cottons Breccia is dominantly massive or crudely stratified
diamictite with minor conglomerate, sandstone and laminated
pebbly siltstone (Jago 1974). Clasts in the diamictite are pre-
dominantly sub-angular to sub-rounded, and the maximum size
in any outcrop is typically 300–500 mm, but commonly less
C. R. CALVER
(c. 100 mm). The larger clasts are commonly oriented with long
axes parallel to stratification. At the Gut (Fig. 64.1b) a 15-m-
thick interval contains carbonate boulders up to 3 m long, but
this horizon cannot be traced any distance laterally. Clast litholo-
gies comprise abundant carbonate (including cream and grey
dolostone, dark limestone, oolitic limestone and dolostone con-
glomerate), abundant fine-grained siliciclastics (mainly fine-
grained quartzarenite, siltstone, sandstone and mudstone), and
rare chert, red jasper, basic volcanics and dacite (Jago 1974;
Waldron & Brown 1993; Calver & Walter 2000). Common striated
clasts were noted by Solomon (1969) but have not been confirmed
by subsequent workers (Jago 1974; Waldron & Brown 1993).
Diamictite in the lower part of the Cottons Breccia has clasts of
predominantly carbonate and a grey muddy carbonate matrix,
while in the upper part, siliciclastic clasts become more abundant
and the matrix is a red-brown mudstone (Calver & Walter 2000).
Jago (1974) noted graded bedding (30 –300 mm) in southern out-
crops and a number of lenses of well-sorted sandstone and siltstone
up to 300 mm thick and 10 m long, within the diamictite. Between
Cottons Flat and the Gut there is a middle unit, up to 20 m thick, of
green mafic-volcaniclastic sandstone, with shards visible in thin
section. At Cottons Flat, diamictite in the lower part of the
formation includes minor intervals of well-laminated siltstone
crowded with dropstones up to 100 mm in size (Jago 1974, figs
23, 24; P. Hoffman, pers. comm.). At the Gut, there is an upward
transition from diamictite through 2 m of closed-framework con-
glomerate, then 0.8 m of plane-laminated red sandstone, pebbly
at the base, which in turn is conformably and gradationally over-
lain by the Cumberland Creek Dolostone.
The Cumberland Creek Dolostone is a pale grey to pale pinkish-
grey, fine-grained laminated dolostone (weathering to a pale
yellow-brown), passing up into thinly interbedded dolostone and
shale, then shale with thin, pale grey, fine-grained limestone
beds, c. 10 m thick in total. Thin sections of the dolostone show
dolomicrite or turbid dolomicrosparite, with scattered rounded
micritic peloids 0.1–0.25 mm in diameter, and irregular micro-
fenestrae (Calver & Walter 2000). Sharp-crested intrastratal anti-
clines, 100– 200 mm in amplitude, are found here and there in
 NEOPROTEROZOIC GLACIAL DEPOSITS OF TASMANIA
the lower, dolomitic part of the unit. The overlying Yarra Creek
Shale is massive or plane-laminated, and pale yellow-brown to
red except for a middle interval with beds of black shale that
have the characteristic microfabric of benthic microbial mats
(Calver & Walter 2000).
Boundary relations with overlying and underlying
non-glacial units
The base of the Cottons Breccia is not exposed, but its variable
thickness suggests erosional incision into the underlying Robbins
Creek Fm. (Calver 2008a). The top of the Cottons Breccia is, at
least locally, a conformable and gradational contact with the over-
lying Cumberland Creek Dolostone (previous section). However,
at a locality north of Conglomerate Creek, the contact with the
dolostone is abrupt and disconformable (Jago 1974).
Chemostratigraphy
Carbonate d13C and d18O data for six samples of the Cumberland
Creek Dolostone were presented by Calver & Walter (2000). The
unit undergoes a steep upward decline in d13C, from –1.9‰ at
the base to –5.0‰ near the top. O-isotopic compositions are
moderately to strongly depleted ( –7.9 to –16.7‰ VPDB).
Other characteristics
At Grassy, the Grassy Group hosts economically important calcic
scheelite skarns in the contact aureole of the Sandblow Granite
(Danielson 1975).
Palaeolatitude and palaeogeography
A palaeomagnetic study by T. Raub (pers. comm.) shows a prob-
able primary magnetization in the upper Cottons Breccia and
basal Cumberland Creek Dolostone, with both polarities preserved
and directed approximately to present-day north and south. These
units appear to have been deposited within 58 of the equator.
Geochronological constraints
Calver & Walter (2000) argued for correlation between the Cottons
Breccia and the Elatina (‘Marinoan’) glacials of the Adelaide
Geosyncline on the basis of, inter alia, the strong lithologic and
d13C-chemostratigraphic resemblance between the Cumberland
Creek Dolostone and the Nuccaleena Fm. (Williams et al. 2011).
Correlation of the Yarra Creek Shale with the lower Brachina
Fm. of the Adelaide Geosyncline is also supported by lithostrati-
graphy, including the presence of microbialite black shale beds
both in the Yarra Creek Shale and low in the Brachina Fm. and
equivalents (Logan et al. 1999; Calver 2000). There are two
radiometric minimum age constraints on the Cottons Breccia: (i)
a Nd –Sm isochron age of 579 + 16 Ma on the Shower Droplet
Volcanics and Grahams Road Volcanics (Meffre et al. 2004),
(n ¼ 5, 2s error limits, MSWD ¼ 1.6; initial 1Nd of 4.2); (ii) a
U –Pb (SHRIMP on zircon) age of 574.7 + 3.0 Ma on the
Grimes Intrusive Suite (Calver et al. 2004) (n ¼ 26, MSWD ¼
0.73; precision limit
651
NW Tasmania: Julius River Member and Croles
Hill Diamictite
Introduction
In the Smithton Synclinorium of NW Tasmania, two diamictite
units (the Julius River Member and the Croles Hill Diamictite)
are found in a Cryogenian to Early Cambrian succession known
as the Togari Group (Calver 1998; Everard et al. 2007). The
Julius River Member is found within the upper part of the
Black River Dolomite, and crops out at widespread localities
within the Smithton Synclinorium (Fig. 64.2). It was first briefly
described by Longman & Matthews (1962). Griffin & Preiss
(1976) described outcrops now recognized as belonging to the
Julius River Member, and the stromatolites contained in some of
the clasts. The Julius River Member was described under the
informal name ‘Trowutta Breccia’ by Jago (1981). The unit was
formally defined by Everard et al. (2007), with a type section,
c. 200 m thick, on the Arthur River (Fig. 64.2). Calver (1998)
reported C and Sr chemostratigraphic data for the Togari Group
and suggested correlation of the Julius River Member with the
Sturtian glacial deposits of the Adelaide Geosyncline. Kendall
et al. (2007) supported this correlation with a Re – Os age determi-
nation on black shale immediately overlying the Julius River
Member.
The Croles Hill Diamictite (Everard et al. 2007) is part of the
Kanunnah Subgroup, which overlies the Black River Dolomite
(Fig. 64.2). Up to 250 m thick, the formation is best developed
in the south of the Synclinorium, with significant localities also
near Forest in the NE and at Robbins Passage in the NW of the
Synclinorium (Fig. 64.2; Brown 1989, p. 25; Calver et al. 2004).
Everard et al. (2007) formally defined the Croles Hill Diamictite,
with a type section on the Arthur River (Fig. 64.2).
Structural framework
The Togari Group unconformably overlies a thick early Neoproter-
ozoic siliciclastic shelfal succession, the Rocky Cape Group. The
part of the Togari Group below the Julius River Member records
epicratonic, shallow-marine, predominantly carbonate sedimen-
tation. In the Kanunnah Subgroup above the Julius River
Member, basalts (dominantly rift tholeiites) and associated volca-
niclastics are abundant. The NE-trending Roger River Fault was an
east-side-down growth fault at this time (Everard et al. 2007).
NW –SE extension is also shown by the approximately coeval
intrusion of the Tayatea Dyke Swarm (588 – 600 Ma) into base-
ment east of the Smithton Synclinorium (Brown 1989). Rifting
was evidently aborted at this location as the Kanunnah Subgroup
is conformably followed by thick upper Ediacaran shallow-marine
carbonates of the Smithton Dolomite.
The succession was gently folded in the Cambrian Tyennan
Orogeny and again in the Devonian, Tabberabberan Orogeny.
The Smithton Synclinorium (a large, gently north-plunging syn-
clinal structure) is probably largely a Devonian structure. Dips
are mainly gentle to moderate except along the western margin,
where steep dips are associated with east-directed thrusting
(Everard et al. 2007). The rocks are undeformed to weakly
deformed except locally in the south, where pelitic lithologies
are moderately cleaved. Low-grade regional metamorphism (of
prehnite-pumpellyite grade, Griffin & Preiss 1976) affects the