CARBONATE CEMENTATION

IN SANDSTONES

Carbonate Cementation in Sandstones Edited by Sadoon Morad

© 1998 The International Association of Sedimentologists
ISBN: 978-0-632-04777-2
 SPECIAL PUBLICATION NUMBER 26 OF THE

INTERNATIONAL ASSOCIATION OF SEDIMENTOLOGISTS

Carbonate Cementation
in Sandstones
DISTRIBUTION PATTERNS AND
GEOCHEMICAL EVOLUTION

EDITED BY
SADOON MORAD

b
Blackwell
Science
� 1998 The International Association DISTRIBUTORS
of Sedimentologists
Marston Book Services Ltd
published by
PO Box 269
Blackwell Science Ltd
Abingdon, Oxon OXI4 4YN
Editorial Offices:
(Orders: Tel: 01235 465500
Osney Mead. Oxford OX2 OEL
Fax: 01235 465555)
25 John Street, London WC IN 2BL
23 Ainslie Place. Edinburgh EH3 6AJ USA
350 Main Street. Malden Blackwell Science, Inc.
MA 02148 5018. USA Commerce Place
54 University Street. Carlton 350 Main Street
Victoria 3053. Australia Malden, MA 02148 5018
I 0. rue Casimir Delavigne (Orders: Tel: 800 759 6102
75006 Paris, France 781 388 8250
Fax: 781 388 8255)
Other Editorial Offices:
Canada
Blackwell Wissenschafts-Verlag GmbH
Login Brothers Book Company
Kurfllrstendamm 57
324 Saulteaux Crescent
I 0707 Berlin, Germany
Winnipeg, Manitoba R3J 3T2
Blackwell Science KK (Orders: Tel: 204 224-4068)
MG Kodenmacho Building
Australia
7-10 Kodenmacho Nihombashi
Blackwell Science Pty Ltd
Chuo-ku. Tokyo I 04, Japan
54 University Street
Carlton, Victoria 3053
The right of the Authors to be
(Orders:Tel: 3 9347 0300
identified as the Authors of this Work
Fax: 3 9347 5001)
has been asserted in accordance
with the Copyright. Designs and
Patents Act 1988. A catalogue record for this title
is available from the British Library
All rights reserved. No pan of
ISBN 0-632-04975-8
this publication may be reproduced,
stored in a retrieval system, or
transmitted. in any form or by any Library of Congress
means, electronic, mechanical. Cataloging-in-publication Data
photocopying, recording or otherwise,
Carbonate cementation in sandstones/
except as permitted by the UK
edited by Sadoon Morad.
Copyright, Designs and Patents Act
p. ·em. - (Special publication
1988. without the prior permission
number 26 of the International Association
of the copyright owner.
of Sedimentologists)
Includes bibliographical references
First published 1998
and index.
ISBN 0-632-0497 5-8
Set by Semantic Graphics, Singapore
I. Sandstone. 2. Cementation (Petrology)
Printed and bound in Great Britain
3. Rocks, Carbonate.
at the University Press, Cambridge
I. Morad. Sadoon. II. Series:
Special publication .. . of the International
The Blackwell Science logo is a Association of Sedimentologists: no. 26.
trade mark of Blackwell Science Ltd. QE471.15.S25C37 1998
registered at the United Kingdom 552'.5-dc21 97-28570
Trade Marks Registry CIP
 Contents

1x Preface

Carbonate cementation in sandstones: distribution patterns and geochemical

evolution
S. Morad

27 Origin and spatial distribution of early vadose and phreatic calcite cements in the
Zia Formation, Albuquerque Basin, New Mexico, USA
J.R. Beckner and P.S. Mozley

53 Carbonate diagenesis and porosity evolution in sheet-flood sandstones: evidence

from the Middle and Lower Lunde Members (Triassic) in the Snorre Field,
Norwegian North Sea
S. Morad, L.F. de Ros, J.P. Nystuen and M. Bergan

87 Carbonate diagenesis in non-marine foreland sandstones at the western edge of the

Alleghanian overthrust belt, southern Appalachians
KL. Milliken

107 Palaeogeographical, palaeoclimatic and burial history controls on the diagenetic

evolution of reservoir sandstones: evidence from the Lower Cretaceous Serraria
sandstones in the Sergipe-Alagoas Basin, NE Brazil
A.J. V Garcia, S. Morad, L.F. de Ros and I.S. Al-Aasm

141 Carbonate cements in the Tertiary sandstones of the Swiss Molasse basin: relevance
to palaeohydrodynamic reconstruction
J. Matyas

163 Carbonate cement in the Triassic Chaunoy Formation of the Paris Basin:
distribution and effect on flow properties
R.H. Worden and J.M Matray

179 Calcite cement in shallow marine sandstones: growth mechanisms and geometry
0. Walderhaug and P.A. BjtJrkum

v
vi Contents

193 Origin of low-permeability calcite-cemented lenses in shallow marine sandstones and

CaC03 cementation mechanisms: an example from the Lower Jurassic Luxemburg
Sandstone, Luxemburg
N Molenaar

213 Geochemical history of calcite precipitation in Tertiary sandstones, northern

Apennines, Italy
K L. Milliken, E.F. McBride, W Cavazza, U. Cibin, D. Fontana, MD. Picard and
G. G. Zuffa

241 Diagenetic evolution of synorogenic hybrid and lithic arenites (Miocene), northern
Apennines, Italy
E. Spadafora, L.F. de Ros, G. G. Zuffa, S. Morad and I.S. Al-Aasm

261 Carbonate cementation in Tertiary sandstones, San Joaquin basin, California

J.R. Boles

285 Carbonate cementation i n the Middle Jurassic Oseberg reservoir sandstone, Oseberg
field, Norway: a case of deep burial-high temperature poikilotopic calcite
J.-P. Girard

309 Origin and timing of carbonate cementation of the Namorado Sandstone

(Cretaceous), Albacora Field, Brazil: implications for oil recovery
R.S. de Souza and C.M. de Assis Silva

327 Structural controls on seismic-scale carbonate cementation in hydrocarbon-bearing

Jurassic fluvial and marine sandstones from Australia: a comparison
J. Schulz-Rojahn, S. Ryan-Grigor and A. Anderson

363 Carbonate cementation-the key to reservoir properties of four sandstone levels

(Cretaceous) in the Hibernia Oilfield, Jeanne d'Arc Basin, Newfoundland, Canada
R. Hesse and I.A. Abid

39 5 The significance of 813 C of carbonate cements in reservoir sandstones: a regional

perspective from the Jurassic of the northern North Sea
C.!. Macaulay, A.E. Fallick, OM McLaughlin, R.S. Haszeldine and MJ. Pearson

409 Origin and significance of fracture-related dolomite in porous sandstones: an

example from the Carboniferous of County Antrim, Northern Ireland
R. Evans, J.P. Hendry, J. Parnell and R.M Kalin
 Contents VII

437 Saddle (baroque) dolomite in carbonates and sandstones: a reappraisal of a

burial-diagenetic concept
C. Spot! and J.K Pitman

461 Application of quantitative back-scattered electron image analysis in isotope

interpretation of siderite cement: Tirrawarra Sandstone, Cooper basin, Australia
MR. Rezaee and J.P. Schulz-Rojahn

483 Carbonate cement dissolution during a cyclic C02 enhanced oil recovery treatment
L.K Smith

501 Index
 Preface
Most special publications are proceedings of meet­
ings, and none covers specific topics of siliciclastic
diagenesis. It was, therefore, decided to invite
recognized experts from academia and industry to
contribute to this lAS special publication. Each
manuscript was examined by two independent
referees. This has resulted in volume that contains
papers covering fairly broad aspects of carbonate
cementation in sandstones in terms of the deposi­
tional, tectonic and diagenetic settings of the basins
studied. After my own opening review (Chapter 1),
contributions are arranged in the following order.
Chapters 2-7, which deal with carbonate cementa­
tion in continental sandstones, are followed by
others (Chapters 8-11) dealing with cementation in
marine sediments. Chapters 12-16 cover carbonate
IX
cementation and diagenetic evolution in oil-field
sandstones from USA, North Sea, Brazil, Australia
and Canada. Chapter 17 evaluates the large-scale
carbon isotopic signatures in Jurassic sandstones
from 13 North Sea oil fields. Chapter 18 discusses
fracture-related dolomite in porous sandstones,
whereas Chapter 19 presents a reappraisal of the
significance of saddle dolomite as an indicator of
burial diagenetic conditions in sandstones and car­
bonate rocks. Chapter 20 demonstrates the use of
quantitative back-scattered electron image analysis
in the interpretation of the isotopic signatures of
carbonate cements in sandstones. The closing chap­
ter discusses the dissolution of carbonate cement by
cyclic C02 enhanced oil recovery.
S. Morad
Spec. Pubis int. Ass. Sediment. ( 1 998) 26, 1 -26

Carbonate cementation in sandstones: distribution patterns

and geochemical evolution

S. M O R A D
Sedimentary Geology Research Group, Institute o fEarth Sciences,
Uppsala University, S-752 36 Uppsala, Sweden,
e-mail sadoon. morad@geo. uu.se

ABSTRACT

Carbonate cements in sandstones are dominated by calcite, dolomite, ankerite and siderite, whereas
magnesite and rhodochrosite are rare. The distribution patterns, mineralogy and elemental/isotopic
compositions of carbonate cements vary widely, both temporally and spatially. The most important
factors controlling these parameters during near-surface eodiagenesis include the depositional setting
(e.g. rate of deposition, pore water composition, hydrogeology, climate, latitude and sea-level
fluctuation), the organic matter content and the texture and detrital composition of the host sediments.
During burial (mesodiagenesis) the important controlling factors include the temperature, residence
time, chemistry and flow rates/pattern of subsurface waters, and the distribution patterns of eogenetic
carbonate cements. As a result of mass balance constraints, burial carbonates are thought to be formed
by the dissolution-reprecipitation (i.e. redistribution) of eogenetic carbonate cements and detrital
carbonates. However, cements may also be derived internally from the dissolution of carbonate
bioclasts, volcaniclastic material and calcium plagioclase, or externally from associated carbonate
rocks, evaporites and mudstones. During uplift and erosion, carbonate cements are subjected to
telogenetic alteration and dissolution. The imprints of eogenetic, mesogenetic and telogenetic
conditions might be unequivocally reflected in the mineralogy and geochemistry of carbonate cements.
However, eogenetic carbonates, particularly calcite and dolomite, may be subjected to recrystallization
and resetting of isotopic signatures, fluid inclusion thermometries and elemental compositions.

INTRODUCTION

Carbonates are among the predominant cements in
sandstones and thus an understanding of their
distribution patterns and geochemical evolution is
relevant to reservoir evaluation. Thorough studies
of the composition and origin of carbonate cements
in sandstones using modern analytical techniques
have attracted sedimentary petrologists only in the
past two decades. A proper study of carbonate
cementation should be carried out within the dia­
genetic context of the host sandstones and should
be based on as many analytical methods and as
many background data about the sedimentary basin
as possible. For instance, the timing and tempera­
ture of carbonate precipitation should not be de­
rived exclusively from thermometric measurements
of fluid inclusions because inclusions may re-
equilibrate subsequent to entrapment and give
anomalously high temperatures. Thus the measured
temperatures should be critically examined and
cross-checked against petrographic observations,
geochemical data on the carbonate and related
cements, and the thermal history of the basins.
Factors that control the geochemistry, abundance
and distribution of carbonate cements are of prime
importance in the understanding and prediction of
porosity-permeability variations and in tracing the
geochemical evolution of pore waters during the
burial of sandstones and associated sediments.
Moreover, the stable isotopic composition of near­
surface, eogenetic carbonates (e.g. in soil profiles)
provides important clues to the palaeoclimatic con­
ditions (e.g. Ceding, 1984).

Carbonate Cementation in Sandstones Edited by Sadoon Morad

© 1998 The International Association of Sedimentologists
ISBN: 978-0-632-04777-2
2 S. Morad
Water composition and flow pattern are of prime
importance in determining the distribution and
geochemical evolution of carbonate cements. These
water properties vary considerably between near­
surface to shallow eodiagenesis and deep mesodia­
genesis. During eodiagenesis, the pore water chem­
istry is strongly controlled by the composition of the
depositional waters, climate, detrital mineral com­
position and hydrology of the basin. Compared
with eodiagenesis, water migration in the deep
basinal regimes is limited by the decrease in poros­
ity and permeability of sandstones and associated
rocks. The amounts and distribution patterns of
mesogenetic carbonates, and hence the porosity­
permeability of the host sediments, are strongly
constrained by the chemistry as well as timing, rate
and extent of cross-formational water flow.
Carbonate cements either indirectly enhance or
deteriorate the reservoir properties of sandstones.
Enhancement of reservoir properties occurs when
(i) appreciable volumes of carbonate cements are
dissolved, causing the formation of secondary po­
rosity and (ii) small amounts of carbonate cement
are evenly distributed in the sandstones to support
the overburden weight and prevent the collapse of
framework grains and consequent elimination of
primary porosity. Souza et al. ( 1995) demonstrated
that a few per cent of dolomite cement is sufficient
to prevent the collapse of Aptian reservoir sand­
stones from Brazil despite the high content of
ductile lithic fragments.
The deterioration of reservoir properties occurs
when sandstones are massively cemented by car­
bonates. Although carbonate-cemented horizons
are thin (,;;;; c. 2 m) and form only a minor portion of
sandstone sequences, they may compartmentalize
reservoirs by acting as barriers to water (and hydro­
carbon) flow both during migration from the source
rocks to the reservoirs and during production (Kan­
torowicz et al., 1987; Carvalho et al., 1995). Com­
paction of sandstone sequences containing zones of
laterally continuous carbonate-cemented horizons
may lead to the development of overpressure in
underlying, weakly cemented zones. Laterally ex­
tensive carbonate-cemented sandstones occur both
in marine (Kantorowicz et al., 1987) and continen­
tal sequences (Arakel & McConchie, 1982). The
chemical composition and distribution pattern of
carbonate cements also has important implications
for secondary oil recovery. For instance, ferroan
carbonate reacts with injected acids to precipitate
iron oxides/oxyhydroxides along the pore throats of
sandstones, causing a deterioration in permeability
and oil recovery.
The aim of this paper is to discuss the following
topics: (i) the geochemical conditions of carbonate
cementation in terms of organic-inorganic interac­
tions; (ii) the petrological and geochemical charac­
teristics of facies-related carbonate cements; (iii) the
dissolution, recrystallization and replacement of
carbonate cements during progressive sediment
burial; and (iv) water-carbonate equilibrium states
in some reservoir sandstones and deep-sea sedi­
ments on which pore water analyses and mineralog­
ical data are available.
GEOCHEMICAL ZONES OF
CARBONATE CEMENTATION
Pore waters below the depositional surface undergo
systematic changes in chemical and isotopic compo­
sitions. These changes occur within zones which are
related to the availability of metabolizable organic
matter, Fe- and Mn-oxides/oxyhydroxides, alkalin­
2
ity and the concentration of dissolved 02 and so4 -
(Curtis, 1967, 1987; Claypool & Kaplan, 1974;
Froelich et al., 1979; Berner, 198 1; Coleman &
Raiswell, 1993). These geochemical changes (Fig. 1)
are likely to be imprinted in diagenetic carbonates to
an extent that recognition of the particular zone
within which they precipitated is possible. As sand­
stones are relatively poor in organic matter, it is
likely that the cementation related to the reactions
discussed in the following section occurs partly in
associated organic-rich mud.
Oxic carbonates
Pore waters in oxic zones are characterized by a
dissolved oxygen content greater than � 0.5 mill.
Oxic carbonates prevail in: (i) subaerial environ­
ments, such as the vadose zone where the pores are
periodically filled with gas, air and/or water; (ii)
immediately below the sediment-water interface in
aquatic environments; and (iii) in the phreatic zone
below the water table where all the pores are
regularly filled with water. The thickness of the oxic
zone depends on the penetration, by diffusion or
advection, of oxygen below the sediment surface.
Oxygen diffusion into pore waters is largely con­
trolled by the organic content and the rate of
deposition. In marine and lacustrine sediments the
 CH20 + HN03 --'»- C02 + N2+ H20
[02] s::: 0.5 mill
Mn-Fe rich calcite and dolomite
CH20 + Mn4+ --'»- Mn 2+aq + C02
rhodochrosite (613Cmarin• �
-6 %.)

·
CH20 +�pi·� HS"+ co;
� F.e3+ � Fe2+aq .
,, · �
F�pobr calclte:andldolomlte g.
'(1113c·:5>.- .2olto -10%o) ;. �
§'
-..
'1>
<5
�
c:;·
;:,:

�
8
;;.
a
;:,:
!::>
�
i;;
�
�
1:;'

Fig. 1. The geochemical zones of organic-inorganic interactions encountered during progressive burial of marine and continental siliciclastic sediments in
.
various depositional settings. The reactions are not balanced and aim to show the main reactants and products. These zones include: (i) oxic (OX); (ii) suboxic
which is composed of nitrate reduction (NR), manganese reduction (MnR) and iron reduction (FeR) subzones; (iii) bacterial sulphate reduction (SR); (iv)
microbial methanogenesis (Me); and (v) thermal decarboxylation of organic matter (D). The authigenic carbonates characteristic for each zone and their
o13Cp06 values are provided. Mg-siderite and Fe-magnesite are the more typical ferroan carbonates for burial diagenesis at elevated temperatures. Factors
controlling anoxity of the bottom waters, and hence the sediments below the water-sediment interface in semi-closed and open marine (left) and in lacustrine
(right) basins are illustrated too. Upwelling ofnutrient-rich waters (lower left) causes an increase in primary productivity, and hence higher organic matter
content in bottom sediments (black). However, some of the organic matter may be derived terrestrially. High organic matter content in such open-marine
sediments may lead to suboxic pore water compositions below the sediment-water interface. Anoxic non-sulphidic conditions in pore waters immediately
below the sediment-water interface in lacustrine environment can be enhanced by rapid rate of organic matter accumulation (lower right). See text for further w
explanation.
4 S. Morad
concentration of dissolved oxygen in pore waters,
and thus the thickness of the oxic zone, also de­
pends on the concentration of dissolved oxygen in
bottom waters and the extent of bioturbation.
Under oxic conditions, Mn- and Fe-oxyhydr­
oxides/oxides are stable and occur as discrete
phases or are adsorbed onto the surfaces of other
minerals such as clays. Therefore oxic carbonate
cements have low Mn and Fe contents and are
typical of near-surface, continental sediments with
a very low organic matter content. In these sedi­
ments dissolved carbon is derived from the decay of
plant remains in soil horizons and from atmo­
spheric C02 (Cerling, 1984). The 813C values of
authigenic carbonates forming in vadose and shal­
low phreatic zones mostly vary between -I Oo/oo and
-3o/oo, reflecting mixed sources of dissolved carbon
derivation from the decay of c and c4 plants and
3
from atmospheric C02. In continental settings the
8180 composition of meteoric waters, and hence of
carbonate cements, is strongly controlled by lati­
tude and climatic conditions (Suchecki et al., 1988;
Morad et al., 1995). Marine oxic carbonates precip­
itate in open diagenetic systems and thus have 813C
and 8180 compositions similar to those of unmod­
ified sea water. However, considerable variations in
oxygen isotopic values occur due to variations in
bottom temperature.
Suboxic carbonates
When pore waters in both marine and continental
sediments become significantly depleted in dis­
solved oxygen ( < 0.5 mill), three geochemical sub­
zones successively prevail (Fig. I ): (i) nitrate
reduction into nitrogen (NR); (ii) manganese reduc­
2
tion to Mn +aw (MnR); and, subsequently, (iii)
2
iron reduction to Fe + aq· (FeR). The type and
elemental composition of carbonate cement formed
are hence strongly controlled by the amount of Fe­
and Mn-oxides/oxyhydroxides.
An increase in carbonate alkalinity in the NR
subzone enhances the precipitation of carbonate
cements with 8180 compositions similar to oxic
carbonates, but with a slight enrichment in Mn and
Fe and depletion in 13C. Rhodochrosite and siderite
precipitate in the MnR and FeR subzones of sedi­
ments containing large amounts of Mn- and Fe­
oxides, respectively. Because the three subzones
overlap, it is common to observe, such as in
deep-sea sediments, that suboxic siderites and
rhodochrosite are enriched in Mn and Fe, respec-
tively (Chow et al., 1996). Separation of the sub­
zones occurs, however, in some settings of the deep
sea with very low sedimentation rates and a rela­
tively low organic content (Froelich et al., 1979).
As in the oxic zone, the 813C values of suboxic
carbonates in continental environments are con­
trolled by the 813C of atmospheric carbon and by
the oxidation of terrestrial organic matter in the soil
profile, whereas the 8180 values are mainly con­
trolled by latitude and climatic conditions.
The 813C values of suboxic marine carbonates are
influenced by carbon derived from sea water and
from the oxidation of organic matter. The extent of
2
1 C incorporation into the carbonates depends on
the amount and reactivity of the organic matter, the
depth of the suboxic zone below the seafloor and the
degree of bioturbation. The resultant 813C of dis­
solved carbon in the suboxic zone is �-6o/oo
(McArthur et al., 1986).
Carbonates from bacterial sulphate reduction
This process is most important in marine sediments
where the pore waters contain appreciable amounts
of dissolved sulphate. Bacterial sulphate reduction
(BSR) operates when the pore waters are devoid of
dissolved oxygen (i.e. anoxic). In euxinic basins the
sediment experiences BSR diagenesis directly at the
sediment-water interface (Fig. I); in other words,
no oxic and suboxic phases are encountered (Curtis,
1987). Sulphate reduction is aided by anaerobic
bacteria, as follows:
(I)
It is uncertain whether this reaction enhances car­
bonate cementation. Conversely, in the presence of
reactive iron, the precipitation of Fe-sulphide and a
considerable increase in alkalinity occur as follows:
2
4FeOOH + 4S04 - + 9CH20
goethite
4FeS + 9HC0 - +6H20 + H+ (2)
3
=
mackinawite
greigite
and
2
2Fe20 + 8S04 - + 15CH20
3
hematite
4FeS2 + 15HC0 - +7H20 +OW (3)
3
=
pyrite
The increase in alkalinity due to reactions (2) and
(3) enhances carbonate precipitation in the BSR
 Geochemical evolution of carbonate cements
zone (Sholkovitz, 1973; Berner, 1984). Increased
pore water alkalinity is recorded from organic-rich
sediments which are influenced by BSR and pyrite
formation (e.g. Berner et a/., 1970; Kastner et a/.,
1990).
Fez+ is incorporated into Fe-sulphides, thus cal­
cite and dolomite precipitating in the SR zone are
largely Fe-poor. However, the amount of Fe that is
incorporated into these carbonates depends on the
amounts and reactivity of organic matter and detri­
tal Fe-minerals and the diffusion rate of sulphate
from sea water. The latter is considerably influ­
enced by the degree of bioturbation, the sedimenta­
tion rate and the concentration of dissolved oxygen
in bottom waters. Moreover, Coleman et a/. (1993)
noted that some sulphate-reducing bacteria are
capable of reducing Fe3+ to Fez+ using Hz and
hence the availability of dissolved iron can be at
least partly independent of the flux rates of sulphide
ions. The decrease in concentration of sulphate
due to reduction into sulphide is believed to en­
hance the precipitation of dolomite (Baker & Kast­
ner, 1981). Indeed, dolomite is common in organic­
rich sediments (Garrison et a/., 1984; Burns et a/.,
1988; Slaughter & Hill, 1991; Baltzer et a/., 1994).
In addition to ml!diating BSR, the oxidation of
organic matter enhances dolomite formation by
increasing the alkalinity and pH of the pore waters
due to production of ammonia by the enzymatic
degradation of protein (Slaughter & Hill, I 991).
In marine sediments, the o13C signature of car­
bonate cements precipitated in the BSR zone is
dominated by dissolved carbon derived from the
oxidation of organic matter. However, mixing with
carbon derived from the other sources such as
marine pore waters and the dissolution of biogenic
carbonates are also common. Generally, BSR is
accomplished at shallow depths below the sedi­
ment-water interface to depths of a few hundred
metres. Bacterial sulphate reduction diagenesis oc­
curs either homogeneously distributed in the sedi­
ments or locally in sediments undergoing overall
oxic or suboxic diagenesis due to high local concen­
trations of organic matter, such as inside borings,
burrows and bioclasts.
Carbonates from microbial methanogenesis
This process prevails in anoxic marine and conti­
nental sediments and when sulphate is totally re­
duced in the BSR zone (Fig. I). Although the
precise mechanism is poorly understood, methano-
5
genesis (Me) is believed to occur by the fermenta­
tion of simple organic compounds, e.g. acetate
(4)
or via Hz production and subsequent C02 reduc­
tion:
(5)
The overall reaction of microbial methanogenesis
can be envisaged as follows:
(6)
Both reactions (4) and (5) probably occur in the Me
zone. The o13C values of C02 derived from these
reactions depend on the specific microbial process
involved. Where reaction (4) dominates, such as in
freshwater environments, C02 inherits the o13C of
the acetate, typically 5-l Oo/oo heavier than bulk
carbon in the precursor organic matter (o13C �
-I Oo/oo to -25o/oo), whereas the methane inherits the
o13C value (-55o/oo to -60o/oo) of the methyl groups
(Galimov, 1985; Whiticar et a/., 1986; Clayton,
1994). Reaction (5), which dominates in marine
sediments, involves a strong kinetic carbon isotopic
2
fractionation causing the enrichment of CH4 in 1 C
(o13C � -75o/oo) and enrichment of C02 in 13C.
Residual C02 due to progressive, but incomplete,
reduction by H2 into methane attains o13C values
up to about +21o/oo (Deuser, 1979). Therefore it
appears that o13C values of C02 in the Me zone
vary between about -25o/oo and +21o/oo (cf. Whiticar
et a/., 1986). Regardless of the dominating Me
pathway, the earliest formed methane is isotopically
2
more enriched in 1 C. High rates of C02 production
by reaction (4), which cause no change in the pH of
the pore waters, lead initially to the dissolution
rather than precipitation of carbonates. Carbonates
precipitated in this zone have 'intermediate' o13C
values (mostly between -22 and +2o/oo). Conversely,
carbonates that have very positive carbon isotopic
values are relatively rare (cf. Clayton, 1994).
As a result of the anoxic, low sulphate concentra­
tions in the Me zone, carbonates expected to form
include siderite and ferroan dolomite/ankerite
(Gautier & Claypool, 1984). The precipitation of
these carbonates occurs in sediments rich in reac­
tive detrital iron (Coleman, 1985), as follows:
2Fe20 + 7CH20 4FeC0 + 3CH4 + H20 (7)
3 3
=
The solubility of methane in pore waters is limited
and depends on the pressure, temperature and
salinity. Excess methane dissipates upwards and is
6 S. Morad
oxidized anaerobically in the BSR zone and aerobi­
cally in the suboxic zones as follows:
2 1 993). These workers have argued that there is
CH4 + S04 - HS- + HC0 - + H20 (8)
3
=
CH4 + 202 H20 + HC0 - + H+ (9)
3
=
2
These two reactions contribute 1 C to the pore
waters in the sulphate reduction and, particularly,
the suboxic zones. Methane seepages on the seafloor
are accompanied by the formation of authigenic
2
calcite and aragonite that are highly enriched in 1 C
(Hovland et a!. , 1 987).
Within the zone of methane oxidation, rates of
sulphate reduction may be seasonally and spatially
variable. Iron carbonates form in the BSR zone due
2
to reduction of Fe3+ to Fe + by sulphate-reducing
bacteria (Coleman et a!., 1 993). Alternating zones
of dolomite and siderite (Morad, unpublished data)
occur due to fluctuations in the positions of the
transition from FeR to BSR and from BSR to Me
zones. Alternating bands of siderite (o13C � -6o/oo)
and ankerite (o13C �- I Io/oo in Jurassic marine
sandstones from the Barents Sea have probably
been formed due to this FeR to BSR or BSR to Me
fluctuation mechanism (Morad et a!. , 1 996). Fluc­
tuations in the geochemical zones are brought about
due to the episodic oxygenation of anoxic basins or
changes in the rate of sedimentation and flux of
organic matter.
In some cases, deep-sea carbonates have o180
values that cannot be explained even if the bottom
water temperature is assumed to be o·c (Wada et
al., 1 982). Such an anomalous 180-enrichment of
carbonates (o180PoB up to +7.9o/oo) has been ar­
gued to be related to the destabilization of gas
hydrates (Matsumoto, 1 989).
The Me zone may extend from the surface to
burial depths corresponding to a temperature in­
crease to about 7 5 ·c , where biological activity is
decreased or largely inhibited. However, formation
waters at temperatures> 8o·c with o13C values as
high as + 5%o have been reported by Carothers &
Kharaka (I 980), suggesting that methanogenesis
may occur at higher temperatures.
Carbonates from thermal decarboxylation
of organic matter
As bacterial activity diminishes due to an increase
in temperature, the diagenetic reactions in which
organic matter plays an important part will be
thermally controlled to temperatures perhaps as
high as � 2so·c (Carothers & Kharaka, 1 978,
1 980; Surdam e t al., 1 984; Giordano & Kharaka,
sufficient evidence indicating that carboxylic acids,
as well as C02 and H20, are produced in the early
stages of the thermocatalytic degradation of ali­
phatic acids incorporated in kerogen before hydro­
carbon generation. At temperatures between 80 and
1 20"C relatively high concentrations (up to
I 0 000 mg/ 1) of carboxylic acids, particularly ace­
tate, are detected in oil-field brines (Hanor &
Workman, I 986; Kharaka et al. 1 986; MacGowan
& Surdam, 1 990). Over this temperature range the
pH of the carbonate system is externally buffered by
carboxylic acid anions (Surdam et al. , 1 984). Hence
the decarboxylation of organic matter and conse­
quent increase in Pco, would enhance the precipi­
tation rather than dissolution of carbonate cements.
External pH buffering and enhanced carbonate
precipitation may also occur due to silicate reac­
tions (e.g. the dissolution and albitization of detrital
feldspar, chloritization of mica) in the diagenetic
system (Smith & Ehrenberg, 1 989; Hutcheon &
Abercrombie, 1 990). The o13C values of the carbon
derived from organic matter is � - 1 5o/oo. The o 13C
of carbonate cements in this zone is usually consid­
erably influenced by the redistribution of earlier
formed carbonates, but is �- I Oo/oo.
At temperatures greater than � 1 0o·c , thermal
degradation of carboxylic acids produces methane
and carbon dioxide (Surdam et al. , 1 9 84). As the
carboxylic acid anions are consumed due to increas­
ing temperature, the carbonate system becomes
internally buffered, and thus the pH may decrease
due to increased Pco, in the system, leading to
carbonate dissolution and the enhancement of sec­
ondary porosity (Surdam et al. , 1 984). Factors
influencing the thermal destruction rate of organic
acids include coupled sulphate reduction and hy­
drocarbon oxidation, and the mineralogy of host
sediments (Bell, I 99 1 ); the presence of hematite
causes rapid rates of acetic acid decomposition.
Over the temperature interval 1 20- 1 60·c the
carboxylic acid anions completely decarboxylate
and the alkalinity is dominated by the carbonate
system. Consequently, any increase in Pco,- will
cause further dissolution.
A variety of carbonate cements occurs in the de­
carboxylation zone depending on the mineralogy of
the host sediments and earlier formed carbonates,
as well as incursion by deep-seated thermobaric
waters. Sediments containing abundant reactive,
detrital Fe-minerals result in the formation of
 Geochemical evolution of carbonate cements
ferroan calcite and ankerite (e.g. Kantorowicz,
1 98 5 ).
FACIES-RELATED DISTRIBUTION OF
CARBONATE CEMENTS
Like other diagenetic minerals in siliciclastic se­
quences, eogenetic carbonate cements may display
a strong relationship with depositional facies in
continental and marine settings.
Continental calcite and dolomite
Calcretes and dolocretes are the dominant forms of
carbonate cements in continental and nearshore
sediments, which develop in warm to hot, arid to
semi-arid regions, with low, seasonal rainfall and
high evaporation (Goudie, 1 98 3). However, cal­
cretes composed of low-Mg calcite may develop in
wet, cold areas and in dry Arctic soils by freezing
(Swett, 1 9 74; Bunting & Christensen, 1 9 80;
Drozdowski, 1 9 80). Strong et a!. ( 1 992) found that
in cold, wet areas calcrete formation is enhanced by
the presence of abundant carbonate clasts and a
high degree of biological activity beneath forest
covers. The stable isotopic composition of these
carbonates is a powerful tool for inferring palaeo­
environmental and palaeo-ecological variables such
as climate, vegetation type and atmospheric levels
of C0 (Cerling, 1 984, 1 99 1 ; Cerling & Hay, 1 986;
2
Cerling et a!., 1 989; Mack et a!. , 1 99 1 ; Mora et a!.,
1 99 1 ; Driese & Mora, 1 993).
Carbonate precipitation in the vadose zone of hot
arid to semi-arid regions is enhanced by a decrease
in Pco, and PH,o due to increasing temperature and
evaporation. Conversely, carbonate leaching is en­
hanced by a humid climate, which prevents the
evaporative concentration of dissolved Ca2+ and
Mg2+. Loss of water through uptake by plants was
argued by Klappa ( 1 980) to be a likely mechanism
for the precipitation of carbonates around roots.
Carbonate precipitation around roots (rhizocre­
tions) may also be enhanced by microbial activities
(Krumbein, 1 968) and an increase in alkalinity due
to the decay of dead plants.
Sources of Ca and Mg for calcrete and dolocrete
are uncertain, but are often believed to be wind­
blown dust. Calcium and Mg may also be derived
from pyroclastic material (Bestland & Retallack,
1 993) and oceanic aerosols (Quade et a!., 1 995).
These sources are also relevant to phreatic carbon­
ates. In some cases the groundwater may bring ions
7
from carbonate rock terranes to siliciclastic se­
quences. Additional sources include Ca dissolved in
rainwater ( � 6-7 ppm; Goudie, 1 973), Ca­
plagioclase, Ca in tissues of certain plants, and
carbonate bioclasts (e.g. land snails).
Dissolution of carbonate grains may occur as a
consequence of: (i) a build up of Pco, in the vadose
zone due to the extensive respiration of plants and
micro-organisms; (ii) an increase in the concentra­
tion of organic (humic-fulvic) acids due to secre­
tion by, or decay of, plants; and (iii) mixing between
waters with chemically different compositions, par­
ticularly in terms of Pco, (e.g. vadose and phreatic
waters), which is referred to as mixing corrosion
(Wigley & Plummer, 1 976).
Calcretes and dolocretes occur as concretions
and laterally extensive cements in floodplain and
nearshore sediments (Figs 2 & 3). The carbonate­
cemented zones reach thicknesses exceeding I 0 m
and dimensions of over I 0 km x I 00 km (Arakel &
McConchie, 1 982; Arakel, 1 986). Calcretes and
dolocretes may also develop in fluvial channel
sandstones (Tandon & Narayan, 1 98 1 ; Arakel et a!.,
1 990; Arakel, 1 99 1 ). These are dominantly phreatic
carbonates formed by the dissolution and re­
precipitation of carbonate intraclasts derived from
the erosion of floodplain pedogenic calcretes. The
high permeability of channel sandstones also en­
hances the dissolution and kaolinization of detrital
silicates, particularly in semi-arid regions with ac­
tive groundwater systems (Fig. 2).
Dolocretes are common in fine-grained distal
fluvial facies, whereas calcretes dominate in coarse­
grained, proximal facies (Fig. 3). The precipitation
of dolomite is enhanced by an increase in the
Mg/Ca ratio of flowing groundwaters due to the
precipitation of calcite in proximal sediments and
to evaporative ionic concentration. Dolomite pre­
cipitation in lacustrine environments is believed to
occur from mixed groundwater and lake brines,
which sink into the sediments during periods of
intensive evaporation and density increase (Colson
& Cojan, 1 996; Spot! & Wright, 1 992). Calcretes
and dolocretes composed of alternating bands of
calcite and ferroan to non-ferroan dolomite are
believed to reflect precipitation from mixing be­
tween fresh phreatic waters and more saline, vadose
waters (Watts, 1 980; Morad et a!., this volume;
Saigal et a!., in preparation). Dolomite precipitates
when the pore waters are enriched in Mg2+ due to
its evaporative concentration, whereas calcite pre­
cipitates from fresh waters during rainy periods.
Shallow marine sands and gravel rich in carbonate
8 S. Morad

Fig. 2 Distribution of eogenetic

carbonates and clay minerals in a
meandering fluvial system under
semi-arid climatic conditions.

Fig. 3. Variations in relative importance of the different geochemical zones of diagenesis (see text) and the diagenetic
minerals formed in a profile covering proximal to distal continental arid to semi-arid environments, as well as
subaquatic marine environments. Mn- and Fe-oxides should be encountered in the oxic zone of these different settings.
 Geochemical evolution of carbonate cements
bioclasts may also be incurred and cemented by
meteoric waters. In some cases the cement is
low-Mg calcite which occurs as concretions oriented
parallel to the flow pathways of groundwater
(Johnson, 1 989; McBride et a!., 1 994).
The concretions may contain cracks that result
from repeated wetting and drying events. These
cracks are filled by clays and silt in areas which are
episodically flooded, or filled by phreatic carbonate
cement composed of coarsely crystalline calcite,
dolomite or alternating bands of calcite/dolomite,
or rarely fibrous radiaxial calcite (Saigal et a!., in
preparation). Criteria for the identification of va­
dose cements include: (i) pendant or meniscus
texture; (ii) carbonate precipitation in close relation
to rootlets (rhizocretions); (iii) displacive and grain­
shattering carbonate cements (Braithwaite, 1 989;
Saigal & Walton, 1 988); and (iv) patchy lumines­
cence due to episodic cementation related to tem­
poral filling of the pores with water.
Calcretes and, particularly dolocretes in hot, arid
climates are commonly associated with Mg-clays
(sepiolite and palygorskite), silcrete and gypcrete
(Watts, 1 9 80; El-Sayed et a!., 1 99 1 ; Spot! & Wright,
1 992; Colson & Cojan, 1 996). However, authigenic
silica is preferentially associated with calcretes and
dolocrete developed on chemically unstable volca­
nic bedrocks (Hay & Wiggins, 1 980). Conversely,
carbonates formed under semi-arid conditions con­
tain both smectite and kaolinite (Fig. 2) (Morad et
a!., this volume). Dolocretes are often closely asso­
ciated with ultramafic bedrocks, which result in an
2 2
increase in the Mg + /Ca + ratio of the groundwa­
ters (Watts, 1 9 80; Maizels, 1 9 87; Bums & Matter,
1 995). In some occurrences there is a close link of
dolocrete formation with dolostone bedrock, such
as in Miocene palaeosols from Spain (Alonso Zarza
et a!., 1 992). Dolomite cement is also common in
sandstones that are closely associated with evapor­
ite deposits in coastal and inland sabkha settings
(Strong & Milowdowski, 1 987; Shew, 1 99 1 ; James,
1 992; Morad et al., 1 995). In these settings, dolo­
mite precipitation is enhanced by an increase in the
Mg/Ca ratio of pore waters due to the evaporation of
marine or mixed marine/meteoric waters (Patterson
& Kinsman, I 982) and the precipitation of calcite
and calcium sulphate cements (Kinsman, 1 969).
Marine calcite and dolomite
Eogenetic calcite cement dominates in shallow
marine siliciclastic sediments, and accompanies
sulphate reduction and methane oxidation (Kantor-
9
owicz et a!., 1 987; Wilkinson, 1 99 1 ). Dolomite
occurs in relatively small amounts, mainly in the
sulphate reduction zone as overgrowths on detrital
dolomite and by the diagenetic replacement of cal­
cite and aragonite precursors. The main sources of
ions for carbonate cements are sea water, biogenic
carbonates and carbonate intraclasts. Sea water Ca,
Mg and HC0 - are introduced into the pore waters
3
by diffusion, or advection by storms and tidal cur­
rents. Chemical gradients are established due to the
onset of carbonate precipitation as a consequence of
the oxidation of local concentrations of organic
matter, and hence an increase in alkalinity. Berner
( 1 968) demonstrated experimentally that the bacte­
rial decomposition of fish caused an increase in pH
of the solution and consequently the precipitation
2
of Ca + as a mixture of calcium fatty acids salts or
soaps. Berner ( 1 968) suggested that some ancient
calcite concretions, especially those enclosing the
skeletons of soft-bodied organisms, may have ini­
tially formed as calcium soaps which later con­
verted to CaC0 . No evidence of this process has
3
yet been provided for natural settings.
The rapid (tens of years) carbonate cementation
(high-Mg calcite and aragonite) of sand deposits
which occurs in Recent tropical and subtropical
marine coastal settings and results in the formation
of tight beach rocks (Krumbein, 1 979; Amieux et
a!., 1 989; Strasser et a!., 1 9 89; Guo & Friedman,
1 990) was probably also common in the geological
past. Carbonate precipitation occurs in the marine
vadose zone within intertidal and low supratidal
sediments, most probably due to evaporation and
C02 degassing (Hanor, 1 978) and photosynthesis
by algae (Holail & Rashed, 1 992). There are no
well-established criteria with which to recognize
ancient beach rocks, as they are often subjected to
recrystallization and dolomitization ( Ingvald,
1 995). However, dolomitized beach rocks usually
preserve two characteristic features: (i) the presence
of carbonate fringes around well rounded, unre­
placed framework grains, and (ii) the microcrystal­
line habit of the intergranular carbonate (see Fig. 4).
Cement fabrics (see Fig. 4) typically comprise
rims of numerous scalenohedral crystals or syntax­
ial overgrowths around carbonate bioclasts and
intraclasts which grade into micritic or blocky
crystals towards the pore centre (Spadafora et a/.,
this volume). The earliest formed rims and over­
growths are often non-luminescent due to a lack of
Mn, indicating an oxic marine origin. Sands
enriched in detrital carbonates and bioclasts are
rapidly cemented by fringing calcite while on the
10 S. Morad
seafloor. This leads to stabilization of the arenite
framework against porosity destruction by compac­
tion during subsequent burial. Early cementation is
kinetically enhanced by nucleation on carbonate
substrates. Upon burial, the progressive addition of
coarse blocky or mosaic calcite over early calcite
(Wilkinson, 1 99 1 ; Carvalho et a!., 1 995) and minor
dolomite cements may lead to the formation of
extensively cemented sandstones. These eogenetic
cements are strata-bound, nodular or laterally con­
tinuous from hundreds of metres to several kilome­
tres (Kantorowicz et a!., 1 987; Prosser et a!., 1 993).
Shallow marine sandstones often are enriched in
biogenic carbonates which act as nuclei for calcite
precipitation and as a cement source during burial
(Bj0rkum & Walderhaug, 1 990). The deposition of
such sandstones occurs in wave- and storm­
dominated, shallow marine environments, and to a
smaller extent in muddy, fair weather sediments,
tidal channels and tidal point bars. Shell-dominated
layers also form by reworking into the slope apron
(Hendry et a!., 1 996) and as a consequence of short­
term mortality due to catastrophic events such as an
increase in water-column turbidity and a decrease in
dissolved oxygen concentration. In warm, oxygen­
ated marine pore waters the bioclasts themselves
usually do not dissolve because they are originally
formed in equilibrium with sea water. However, the
dissolution of metastable aragonite and high-Mg cal­
cite may begin in the suboxic and bacterial sulphate­
reduction zones (Morse & Mackenzie, 1 990).
In nodular cemented sandstones, the areas left
uncemented often reveal evidence of later burial
diagenetic modifications, such as compaction and
quartz cementation (Morad et a!., 1 995). Burial
cements are believed to be sourced from meteoric
or dissolution of detrital carbonates and bioclasts
(cf. Wilkinson, 1 99 1 ). As the sandstone framework
is expected to be stabilized due to early cementa­
tion, the burial dissolution of bioclasts may be
recognized by oversized pores and mouldic pores
filled with cement.
Although abundant skeletal bioclastic fragments
play an important part in the development of
calcite-cemented sandstones, they should not be
considered as the only source of such cements.
Evidence for this is the common presence of calcite­
cemented sandstones in Precambrian sequences.
Additional evidence is the absence of bioclastic
carbonates in Jurassic sandstones with strata-bound
calcite cements (Bj0rkum & Walderhaug, 1 993;
Prosser et a!., 1 993). This suggests that other
sources such as sea water and carbonate mud
intraclasts are at least as important as bioclasts.
Highly reactive volcaniclastic sediments may also
enhance carbonate cementation at shallow depths
below the seafloor. Alteration of these sediments
may cause the establishment of Ca2+, Mg2+ and
HC03- diffusion gradients between pore waters and
overlying seawater (Morad & De Ros, 1 994). The
domination of calcite over other carbonates in
volcaniclastic sediments (De Ros et a!., 1 996) is
unclear, but may be related to the preferential
incorporation of Fe2+ and Mg2+ in trioctahedral
smectite, and to a diagenetically open system with
respect to the overlying seawater. The mechanisms
bringing marine pore waters into supersaturation
with respect to calcite in volcaniclastic sediments
are poorly understood.
Another potential mechanism responsible for the
formation of laterally continuous, strata-bound
calcite-cemented sandstones is the episodic up­
welling of anoxic seawater (see Kempe, 1 990; Grot­
zinger & Knoll, 1 995). The upwelling of such high
alkalinity waters to shelf and coastal areas may
occur subsequent to periods of sea water stratifica­
tion accompanying sea-level rise.
Subsurface, carbonate-cemented sandstone beds
can be recognized from geophysical well logs and
cores. Moreover, concretionary cemented sand­
stones are differentiated from continuously ce­
mented horizons based on these methods. The
latter sandstones show as tight intervals on sonic,
density and neutron logs. Scattered small concre­
tions give a less distinct response on density and
neutron logs because of their limited lateral extent
and show resistivity readings that vary around the
borehole. Unlike eogenetic strata-bound cementa­
tion, continuous mesogenetic carbonate-cemented
sandstone horizons are structurally controlled and
cut across stratification when precipitation is re­
lated to water flow along faults.
Calcite cement in ancient marine sediments is
consistently a low-Mg variety (Magaritz et a!., 1 979;
Spadafora et a!., this volume), which is either a
primary precipitate or results from the stabilization
of metastable high-Mg calcite and aragonite precur­
sors. There is evidence indicating that inorganic
carbonates precipitated from sea water have varied
between low-Mg calcite during periods of global
warming (greenhouse mode due to an increase in
atmospheric Pco ) and high-Mg calcite and arago­
nite during periods of global cooling (Sandberg,
1 983). Increased atmospheric Pco, has been related
to periods of high plate tectonic activity, which
leads to the release of more C0 derived from the
2
 Geochemical evolution of carbonate cements ll

2
metamorphism of calcareous sediments at subduc­ tremely 1 C-rich, isopachous Mg-rich calcite has
tion zones (Wilkinson et al., 1985). Calcite stabili­ also been reported from other modern non-tropical
zation during these periods is further enhanced by shallow marine terrigenous sediments, including the
lower Mg/Ca ratios in sea water due to the interac­ northeast USA shelf(Hathway & Degens, 1969), the
tion with ejected oceanic crust at mid-oceanic Mississippi River delta (Roberts & Whelan, 1975)
ridges. Thus low-Mg calcite fringes in some ancient and the Kattegat Sea (J0rgensen, 1976, 1979).
marine sediments, such as the extensively studied Simulations of marine-meteoric mixing (e.g.
Jurassic sandstones of the North Sea (Girard, this Plummer, 1975; Wigley & Plummer, 1976) pre­
volume), are likely to be primary. dicted calcite oversaturation in waters with 20-70%
This issue can be extended further to include a sea water. However, the saturation degree of the
discussion on the variation in abundance of dolo­ mixed waters varies depending on the initial calcite
mite in marine sandstones during geological times. saturation index, Pc02 and temperature. Neverthe­
This variation in the calcite/dolomite ratio, with the less, predictive models constructed by Frank &
greater abundance of dolomite in old sedimentary Lohmann (1995) for low-Mg calcite precipitation in
rocks, is probably not only the result of burial
diagenesis, but also due to changes in palaeo­
oceanographic conditions (Given & Wilkinson,
1987). Possible factors include the following(Purser
et a!., 1994): (i) climate-tropical to subtropical
climate favours the precipitation of dolomite
Sea level1
(Tucker & Wright, 1990); and (ii) global sea-level
change-sea-level rise leads to the incursion of
nearshore areas by sea water, which, upon mixing
with meteoric waters and evaporation, enhances the
local precipitation of dolomite due to an increase in
Sea level2
the Mg/Ca ratio of pore waters.

Mixed marine-meteoric water carbonates

phreatic calcite
The degree of mixing between marine and meteoric vadose calclle
waters, and hence the mineralogy, texture and
pattern of carbonate cementation in coastal sand­
1.2
stones, are strongly influenced by sea-level fluctua­
tion (Fig. 4). The precipitation of eogenetic calcite
Meteoric water
and dolomite in nearshore sandstones occurs as vadose and/or phreatic
alternating bands formed by precipitation from meteoric caicite
mixed marine-meteoric waters (Morad et a!.,
3 4
1992). Evidence from present day settings suggests
that the influence of marine mixing with fresh
groundwater, and hence dolomite formation, may
extend landward for distances of 25-30 km (Ma­
Mixed to meteoric
garitz et a!., 1981).
Carbonate precipitation from mixed waters is h!IJh-Mg calcite/aragonite
fnnges pores
enhanced by an increase in alkalinity due to the 5
oxidation of organic matter and methane (Lunde­
gard, 1994). Gas pockets are common in Holocene
sediments rich in organic matter (e.g. McMaster,
1984). Nelson & Lawrence (1984) and Simpson &
Hutcheon (1995) reported the formation of Ho­
locene, high-Mg calcite nodules (1513C � -49o/oo to
-7o/oo) in hybrid, bioclastic deposits of the modem
Fig. 4 Influence of sea-level drop on the composition
Fraser River delta(� 49.N) due to methane oxida­ and texture of carbonate cements in sandstones situated
tion close to the seafloor. Early diagenetic, ex- in shallow marine, coastal and nearshore settings.
12 S. Morad
carbonate sediments from mixed waters suggest
that the zone of oversaturation with respect to
calcite can expand to encompass the full range of
mixing. If high Mg/Ca ratios are maintained, arago­
nite rather than calcite may rarely precipitate from
mixed waters (Kimbell & Humphrey, 1994).
Siderite
Siderite precipitates from reducing, non-sulphidic
pore waters that evolve in the suboxic and micro­
bial methanogenesis zones of all depositional envi­
ronments. These geochemical conditions occur in
organic-rich sediments containing appreciable
amounts of reactive iron minerals and in which the
pore waters are so/--poor meteoric or brackish
(Postma, 1982).
Siderite is most common in continental and
coastal sediments due to the much lower contents of
dissolved sulphate in meteoric and brackish waters
than in sea water. In these environments, small
amounts of iron sulphide are formed, which allows
an increase in Fe2+ concentration in pore waters,
and hence promotes siderite formation. Siderite is
abundant in fine-grained, organic-rich marsh and
swamp sediments associated with deltaic and
coastal sediments. Siderite slightly enriched in Ca
and Mg formed in Holocene intertidal marsh and
sandflat sediments from both marine and mixed
marine-meteoric pore waters (Pye et al., 1990;
Moore et al., 1992). In these sedimentary facies,
siderite is closely associated with pyrite and Fe­
dolomite/ankerite. In fluvial sediments, siderite
preferably forms in fine-grained floodplain and
crevasse splay or in oxbow lake and pond sedi­
ments. The presence of plant remains in semi-arid
to semi-humid regions enhances its formation
(Fig. 3). Authigenic siderite spherules and thread­
like morphologies related to the replacement of
detrital mica (Morad et al., this volume) are com­
mon in pedogenic profiles (Besly & Fielding, 1989;
Kantorowicz, 1990; Browne & Kingston, 1993).
According to Mozley ( 1989a), the elemental com­
position of siderite is controlled by the chemistry of
depositional waters, with meteoric siderites being
more enriched in Mn, but depleted in Ca and
particularly Mg compared with siderite in marine
sediments. However, Morad et al. (this volume)
found that eogenetic siderites formed in a continen­
tal setting are highly enriched in Ca and Mg.
Additionally, high-Mg siderites are typically formed
at increased temperatures (Morad et al., 1994).
Thus, to apply the findings of Mozley ( 1989a), it is
important to determine precisely the diagenetic
regime of siderite formation.
Unlike calcite and dolomite, siderite rarely forms
as an extensive pore-filling cement, but rather as
discrete fine crystals, spherules and nodules scat­
tered in the host sediments. Nevertheless, Baker et
al. ( 1996) found that early diagenetic siderite con­
cretions (0.5-2 mm) form up to 30% of Triassic
sandstones and mudstones from eastern Australia.
Laterally continuous siderite-cemented offshore
shelf sandstone sheets ( 15 em thick) occur in Upper
Cretaceous sequences from Canada (McKay et al. ,
1995).
Rhodochrosite
Rhodochrosite occurs mainly in fine-grained ma­
rine and brackish water sedimentary basins, such as
the Baltic Sea (Lynn & Bonatti, 1965; Suess, 1979;
Pedersen & Price, 1982; Minoura, 199 1). In deep­
sea sediments, Ca- or Fe-rich rhodochrosite occurs
as scattered crystals, microspherules and as nodules
within host pelagic sediments (Coleman et al.,
1982; Wada et al. , 1982; Matsumoto, 1992; Chow
et al. , 1996). However, Bruhn ( 1993) observed
fine-crystalline rhodochrosite nodules in fine­
grained sandstones and siltstones of Lower Tertiary,
submarine turbidites from Brazil.
Magnesite
Eogenetic magnesite cement in sandstones is rela­
tively rare because its formation requires pore
waters to be enriched in Mg2+ and depleted in
Ca2+, SO/- and Cl-. These conditions may occur
in arid climates in which marine pore waters evap­
orate and become successively saturated with re­
spect to calcium carbonates, calcium sulphates and
halite, such as in sabkha settings (Kinsman, 1969;
Morad et al., 1995). Continental brines enriched in
Mg2+ are also suitable for the formation of eoge­
netic magnesite due to the low sulphate and chlo­
ride ion concentrations. Most recent magnesite
cements form in the fine-grained sediments of
alkaline/saline lakes (Last, 1992; Warren, 1990)
and, less commonly, in freshwater lacustrine sedi­
ments (Zachmann, 1989).
Magnesite precipitates at depths of a few decime­
tres below the sediment-water interface, such as in
the ephemeral salt pans of Recent playa lakes in
north-east Spain, where precipitation is enhanced
 Geochemical evolution of carbonate cements
by increases in carbonate alkalinity due to bacterial
activity (Pueyo Mur & Ingles Urpinell, 1 987). In the
Permian Rotliegend reservoir sandstones from
southern North Sea, intergranular, eogenetic mag­
nesite occurs in interdune sabkha as well as dune
and fluvial facies (Purvis, 1 992). Eogenetic magne­
site occurs as nodules and layers in Permian playa
lake mudstones and as intergranular cements in
alluvial fan sandstones from Austria (Spot! &
Bums, 1 994). The precipitation of this magnesite
has been attributed to high-Mg brines derived from
the weathering of Devonian dolostones and associ­
ated massive magnesite deposits in the catchment
area (Spot! & Burns, 1 994).
Marine eogenetic magnesite is also known to pre­
cipitate in deep-sea sediments. Matsumoto ( 1 992)
described rhombic, microcrystalline (2- 1 5 J.Lm) Ca­
Mn-Fe rich magnesite and Fe-Mn rich lansfordite
(hydrous Mg-carbonate) in Miocene to Pliocene
mudstones from ODP Site 799 in the Japan Sea. He
concluded that on progressive burial and increase
in temperature (� 435 mbsf, T � 43 oq, the meta­
stable lansfordite is transformed into magnesite.
DISSOLUTION OF
CARBONATE CEMENTS:
MECHANISMS AND CONSEQUENCES
When carbonate cements are subjected to physico­
chemical conditions that vary considerably from
those under which they formed, they may dissolve
and re-precipitate at various scales. Carbonate dis­
solution and the creation of secondary porosity may
occur during eodiagenesis or telodiagenesis or in
response to progressive burial. Eogenetic secondary
pores may survive subsequent burial and compac­
tion in sandstones that have been subjected to early
overpressuring or hydrocarbon emplacement, or if
dissolution is incomplete, and leave evenly distrib­
uted remnants of carbonate cement.
The scales of carbonate redistribution, and thus
reservoir quality enhancement, are difficult to con­
strain. Several workers have argued that the reser­
voir properties of sandstones are greatly enhanced
due to large-scale carbonate dissolution (L0n0y et
a!., 1 986; Schmidt & McDonald, 1 979). As the un­
dersaturated waters have to circulate through large
volumes of permeable sediment to cause economi­
cally important carbonate cement dissolution, it is
expected that such secondary porosity develops in
partially rather than pervasively cemented sand-
13
stones. Mesogenetic waters probably do not migrate
along a wide front, but are instead focused. Hence
aggressive waters that cause cement dissolution in
a sandstone unit are usually derived from deeper
levels.
Over the past 20 years a major debate has centred
on the mesogenetic dissolution of carbonate ce­
ments. One side of the argument suggests that
carbonate dissolution is caused by acidic waters and
C0 derived on thermal maturation of organic
2
matter in mudstones (Schmidt & McDonald, I 979;
Morton & Land, 1 987). On the other side, mass
balance calculations suggest that the amounts of
organic matter may be insufficient to provide nec­
essary C02 that could produce the observed carbon­
ate dissolution and secondary porosity seen in most
sandstones (Lundegard & Land, 1 986). Moreover,
acidic waters may be neutralized within the mud­
stones due to interactions with carbonate bioclasts
and silicate minerals before reaching adjacent sand­
stones (Giles & Marshall, 1 9 86). Carbonate cement
dissolution can also be accomplished by means of
carboxylic acids and carboxylic acid anions formed
by redox reactions during the hydrocarbon invasion
of hematite-bearing sandstones (Surdam et a/.,
1 993).
Alternative mechanisms that account for the
mesogenetic dissolution of carbonate cements in
sandstones include: (i) the cooling of ascending hot
waters aided by the retrograde solubility of carbon­
ates (Giles & de Boer, 1 990; Wood & Hewett,
1 9 84); and (ii) the mixing of two waters (Runnells,
1 969). The resulting saturation state of carbonate
cements due to mixing depends on: Pco,, tempera­
ture, ionic strength (salinity), degree of carbonate
saturation and pH of the end-member waters before
mixing (Thraikill, 1 968; Plummer, 1 97 5 ; Wigley &
Plummer, 1 976; James & Choquette, 1 990).
Dissolution of carbonate cements in the shallow
subsurface realm is attributed to the infiltration of
meteoric waters, which are weak carbonic acids, or
to mixing corrosion. The overall leaching capacity
of meteoric waters is strongly controlled by: (i) the
amounts of dissolved C02 available in the soil
profile; (ii) the type and extent of organic-inorganic
reactions that produce or consume protons; (iii) the
permeability and depositional geometry of the
sandstones; and (iv) the hydraulic heads. The disso­
lution capacity is expected to be more significant in
permeable, laterally extensive sandstones in basins
with a high hydraulic head. However, meteoric
waters are unlikely to cause deep burial mineral
14 S. Morad
dissolution because they probably attain equilib­
rium with carbonates and silicates in the soil profile
and in the relatively shallow subsurface. The more
reactive the mineral contents in these regimes, the
shallower is the dissolution capacity of meteoric
waters. Morad et a!. (this volume) concluded that in
the Triassic Lunde Formation, North Sea, meteoric
waters dissolved carbonate cements and framework
silicates within a few tens of metres below the
Kimmerian unconformity surface. In areas where
the Lunde Formation was buried at depths> 350 m
below this surface, meteoric waters mainly caused
the dissolution of carbonate cements.
Criteria for the recognition of secondary porosity
due to the dissolution of carbonate cements in sand­
stones include (Schmidt & McDonald, 1 979) the
presence of: (i) oversized pores formed by the disso­
lution of grain-replacive carbonate cements-over­
sized pores may, however, result from the dissolu­
tion of carbonate bioclasts and intraclasts; (ii)
partially dissolved carbonate cements with etched
rather than euhedral crystal outlines; and (iii) grain­
replacive carbonates surrounded by open pores.
Secondary cement dissolution porosity that mimics
or enhances primary intergranular porosity is
difficult to recognize. Nevertheless, dissolved car­
bonate cements may leave framework grains with
corroded margins that can be best recognized by
scanning electron microscopy (Burley & Kantorow­
icz, 1 986). Dissolution of calcite cement is more
pervasive than the less soluble dolomite, ankerite
and siderite.
RECRYSTALLIZATION
AND REPLACEMENT OF
CARBONATE CEMENTS
In addition to dissolution, the destabilization of car­
bonate cements may result in recrystallization and
replacement by other carbonates. Microcrystalline
calcite and dolomite are sensitive to recrystallization
at various burial depths. The recrystallization of
dolomite has been reviewed by Mazzullo ( 1 992).
Burial recrystallization of micritic/microsparitic ce­
ments in sandstones may result in the formation of
poikilotopic calcite (Saigal & Bj0rlykke, 1 9 8 7).
However, poikilotopic calcite is also a common
primary cement in calcretes (e.g. Knox, 1 977; Tan­
don & Narayan, 1 9 8 1 ). Recrystallized calcite and
dolomite are recognized as patchily distributed,
coarsened crystals. In contrast, precipitational vari-
ations in crystal size of drusy carbonates show
trends of increasing crystal size from pore walls to
pore centre. Siderite and ankerite are less soluble
and thus less sensitive to recrystallization than
calcite and dolomite (Matsumoto & Iijima, 1 98 1 ;
Mozley & Bums, 1 992). Spot! & Bums ( 1 994)
argued that magnesite is resistant to deep burial
recrystallization, but might undergo recrystalliza­
tion by interaction with meteoric waters at low
temperatures.
Recrystallization may influence the crystal struc­
tural, elemental and isotopic compositions of the
carbonate in question (Gregg et a!., 1 992; Chafetz &
Rush, 1 994; Malone et a!., 1 994; Kupecz & Land,
1 994). Carbonate cements formed by recrystalliza­
tion are characterized by lower 8180 values than the
microcrystalline precursor cements. This suggests
the involvement of meteoric waters or increased
burial temperatures. Therefore recrystallization
must be considered when 8180 is used for studies
on palaeoclimate, the timing of cementation and
palaeo-water composition. Unlike 8180, the carbon
and strontium isotopic compositions of carbonates
may be preserved during recrystallization, particu­
larly in low permeability rocks (Dutton & Land,
1 98 5 ; Siegel et a!., 1 98 7 ; Cerling, 1 99 1 ; Driese &
Mora, 1 993; Kupecz & Land, 1 994).
In addition to recrystallization, the replacement of
one carbonate cement by another is common during
burial diagenesis. Eogenetic calcite cement may be
replaced partially to completely by ferroan dolomite/
ankerite during mesodiagenesis (Boles, 1 978). The
dolomitization of calcite cement, which is wide­
spread in limestones, is less frequently reported for
sandstones (Hudson & Andrews, 1 9 87; Lawrence,
1 99 1 ; Morad et a!., 1 995). Complete replacement of
calcite cement by dolomite and ankerite is difficult to
recognize. However, its recognition may be possible
by the presence of mimetically replaced bioclasts
(Richter & Fuchtbauer, 1 9 78; Morad et a!., 1 996)
and by the similarity of the dolomite and ankerite
fabric to the eogenetic calcite. Replacement of sider­
ite cement and intraclasts by ankerite occurs in res­
ervoir sandstones from offshore Norway (Morad et
a!., 1 996). Upon uplift and invasion by meteoric
waters, dolomite/ankerite may be dissolved or re­
placed by calcite ± hematite (Morad et a!., 1 995). ·
Calcitization of dolomite cements may also occur in
the eogenetic regime due to subtle modifications in
pore water chemistry caused by variations in inten­
sity of rainfall and/or sea- or lake-level fluctuations
(e.g. Colson & Cojan, 1 996) (Fig. 4).
 Geochemical evolution of carbonate cements IS

EQUILIBRIUM RELATIONSHIPS of ankerite becomes narrower, giving way to siderite

AMONG DIAGENETIC CARBONATES
Studies on the stability of diagenetic minerals in
relation to temperature and formation water chem­
istry provide important insights into the overall
mineralogical and chemical evolution of the host
sediments (Boles, 1 982; Kaiser, 1 984; Morad et
a!., 1 990, 1 994). The precipation conditions and
equilibrium relationships of carbonate are com­
plex issues and controlled by several inter-related
parameters, such as pore water chemistry (ionic
activities, pH, alkalinity, dissolved organic com­
pounds), kinetics and temperature. The tempera­
ture-dependent equilibrium relationships among
calcite-ankerite-siderite, calcite-dolomite-mag­
nesite, siderite-magnesite, and dolomite-ankerite
have been calculated as functions of aMgl+/Uca'+,
<lFe2 +/<lc32+ and <lFe2+/UMg2 + (Fig. S).
Considering the stability relationships among
calcite-ankerite-magnesite, it is interesting to note
that as the temperature increases the stability field
and calcite (Fig. Sa). Ankerite has been reported by
several workers to be a common deep burial, meso­
genetic carbonate cement (Boles, 1 9 78; Kantorow­
icz, 1 98 S ; Sharp et a!., 1 9 8 8), but also forms as a
near-surface cement (Morad et a!., 1 996; Mozley &
Hoernle, 1 990). However, calcite may post-date
ankerite (e.g. Girard, this volume) if suitable
geochemical conditions and temperatures prevail.
The stability relationships in Fig. Sa suggest that as
the temperature decreases, lower aFeH/Uca'' ratios
in pore waters are required to stabilize ankerite at
the expense of calcite.
Ferroan carbonate formed during mesodiagenesis
may be siderite rather than ankerite if the formation
waters have a sufficiently high Fe/Ca activity ratio.
For instance, formation waters in Triassic reservoir
sandstones in southern Tunisia ( T -:::::. 8 0 " C) have log
(aFe2+ /Uc32+) values between -3 and -2.S, and thus
fall within the stability field of siderite, which agrees
well with the petrographic observations of Morad et
a!. ( 1 994). Probably due to kinetic reasons, siderite

·3.2
siderite
�ro
+
-3.4
� -3.5
N

0
�
�" ·3.6 +
N

u.
ro
�" -4.0
c; -3.8 c;
.Q .Q
·4.5
-4.0

·4.2
-1.8
;r- ·2.0
C>
::;
ro ·2.2
:t"
� ·2.4
c;
.Q ·2.6
·2.8
·3 d
·3.0 -4 L...
� .. �...L...o ....�
... ..o.L�.L.......
... �.J. ........
.L.... ...:l
0 25 50 75 100 125 150 175 200 0 25 50 75 100 125 150 175 200
T (OC) T (OC)
0 North Sea formation waters
D. Pore waters from deep-sea sediments

Fig. 5. Equilibrium diagrams for common diagenetic carbonate cements as a function of ionic activities and
temperature constructed using the thermodynamic computer programme CHEMSAGE. The North Sea formation water
data are from Egeberg & Aagaard ( 1 989) and deep-sea pore water data are from Egeberg ( 1 990).
16 S. Morad
may not precipitate despite the saturation of pore
waters with respect to it (Emerson & Wildmer,
1978). The activity ratios of Fe/Ca calculated for
pore waters from the Fraser River Delta (Simpson
& Hutcheon, 1995; � -2.6 and -0.2) and from the
Amazon fan (ODP Leg 155; Flood et a/., 1995;
> 1.0) fall within the stability field of siderite in
Fig. Sa. However, no siderite has been detected in
these sediments. Low log (a.Fe, . fa.ca2 + ) ratios, and
hence the plot of pore waters in the stability field of
calcite, is often related to the precipitation of
Fe-sulphides.
The stability relationship and perhaps even the
extent of solid solution between siderite and magne­
site depend on the temperature and the Fe/Mg
activity ratio. As the temperature increases, the sta­
bility field of magnesite increases, which means that
higher Fe/Mg activity ratios are required to stabilize
siderite at the expense of magnesite (Fig. 5b). For­
mation waters from Triassic Tunisian reservoirs
( T � 80 " C; Morad et a/., 1994) are characterized by
log (a.FeH/a.Mg, . ) ratios of -3.9 to -2.4 and hence fall
within the stability field of siderite, which is far more
dominant than magnesite (Morad et a/., 1994). Al­
ternating zones of magnesian siderite and ferroan
magnesite in these Triassic sandstones formed at 55-
60 ' C (Morad et a/., 1994). Magnesium-rich siderites
with low Ca and Mn contents have also been formed
at increased temperatures (� 70-90 ' C) in other
sedimentary basins (Macaulay et a/., 1993; Mozley
& Hoernle, 1990; Rezaee & Rojahn-Schulz, this
volume). When several generations of siderite occur
in a sedimentary sequence, it appears that the later
generations are more enriched in Mg (e.g. Mozley,
1989b). Iron-rich magnesite cements in Permian
mudstones and sandstones from Austria have been
reported by Spot! & Burns ( 1994).
Eogenetic ferroan magnesite (FeC03 � 1.5-25.5
mol%) also forms in mudstones and sandstones of
deep-sea sediments (Matsumoto & Matsuda, 1987;
Matsumoto, 1992). Unlike magnesite formed at in­
creased temperatures (Morad et a/., 1994; Spot! &
Burns, 1994), these deep-sea magnesites contain
substantial amounts of Ca (6.5- 17.0 mol%) and Mn
(0.5-20.5 mol%). Eogenetic magnesian siderites in
marine sediments contain appreciable amounts of
Ca (McKay et a/., 1995; Mozley, l 989a; see also
Browne & Kingston, 1993; Morad et a/., this vol­
ume). Deep-sea siderites are enriched in Mn (Chow
et a!., 1996). Conversely, near pure or slightly to
moderately Mn-rich (�2- l0 mol%) siderites form
during the eodiagenesis of continental sediments
(Mozley, l 989a; Browne & Kingston, 1993; Baker et
a/., 1996).
The stability relationship between dolomite and
ankerite depends on the temperature and activity
2
ratio of Fe 2+/Mg + (Fig. 5c). Formation waters
from Norwegian North Sea reservoirs (Egeberg &
Aagaard, 1989) have log (a.FeH/a.Mg, . ) of � -3 to -2
and fall within the stability field of dolomite and
ankerite (Fig. 5c). Both of these minerals are widely
reported as mesogenetic cements in these sediments
(Saigal & Bje�rlykke, 1987; Morad et a/., 1990).
The stability relationships between calcite, dolo­
mite and magnesite depend on the temperature and
activity ratio of Mg 2 + /Ca 2 + (Fig. 5d). Lower
Mg/Ca activity ratios are required to induce the
dolomitization of calcite and to stabilize magnesite
at the expense of dolomite (Fig. 5d) (Usdowski,
1994). Formation waters from the Norwegian
North Sea reservoirs have an average log (a.Mg2 + /
Uc3H) � - 1.0 t o 0.0 and thus fall within the stability
field of dolomite. Nevertheless, both calcite and
dolomite are common cements in these rocks,
indicating that dolomitization is a kinetically con­
trolled reaction. Further evidence of this is revealed
from Recent sediments, such as the Fraser River
delta in Canada (Simpson & Hutcheon, 1995) (log
(a.MgH/Uc3H) � -2.2 to +1.0), where the pore wa­
ters are saturated with respect to dolomite, but it is
calcite rather than dolomite that precipitates. Cal­
cite rather than dolomite forms below the deep-sea
floor, yet the pore waters plot at shallow, near sea
bottom temperatures in the stability field of dolo­
mite and shift with an increase in depth towards the
stability field of calcite (Fig. 5d). This shift is due to
a diffusion-controlled, downhole decrease in Mg/Ca
activity ratio caused by the incorporation of Mg in
Mg-silicate that results from the alteration of volca­
nic material, a process which is coupled with the
release of calcium (McDuff & Gieskes, 1976).
PATTERNS OF FLUID FLO W :
CLUES T O THE ORIGIN
AND MECHANISMS OF
MESOGENETIC CARBONATE
CEMENTATION
There is ample evidence of active, large-scale fluid
flow in the subsurface, which should be considered
in diagenetic modelling (Sullivan et a/., 1990; Glu­
yas & Coleman, 1992; Gaupp et a/., 1993). Direct
evidence of fluid flow is manifested by hot springs,
 Geochemical evolution of carbonate cements
geyser fields, seafloor vents and seepages, and a rise
in groundwater level during and after earthquakes
(Sibson, 1990). The most important regimes of fluid
flow in sedimentary basins (Fig. 6) are related to
compaction by sediment loading, tectonic compres­
sion, deep meteoric infiltration in areas of tectonic
uplift, thermo-chemical convection due to density
gradients around salt diapirs and convection due to
the presence of thermal gradients, such as in the
vicinity of rising magmas.
Fracturing, folding and thrusting greatly influ-
Fig. 6. Patterns of fluid flow
envisaged for three common types
of sedimentary basins.
17
ence the style and extent of fluid flow in sedimen­
tary basins. Faults, however, may either act as high
permeability conduits and thus enhance fluid flow
(Knipe, 1993) or as seals that result in compartmen­
talization, and thus the restriction of water flow
(Harding & Tuminas, 1989; Hindle, 1989). Tec­
tonic stresses cause rapid, pulse-like changes in fluid
flow (Sibson et a/., 1975; Muir Wood, 1993). Fluid
flow along fracture systems is episodic and occurs
by seismic pumping and seismic valving (Sibson,
i 98 1 ). Seismic pumping occurs due to pressure
18 S. Morad
gradients, whereas flow by seismic valving occurs as
a result of dilation and fault failure induced by high
pore pressures in the vicinity of overpressured
�ones. The release of overpressure may be accom­
panied by hydrofracturing and fluid migration
along pressure gradients (Sullivan et a!., 1 990;
Caritat & Baker, 1 992; Schulz-Rojahn, 1 993).
Fracturing and fluid flow along pressure gradients
may result in mesogenetic carbonate cementation in
intergranular pores of sandstones and along frac­
tures according to any of the following mechanisms.
I Decrease in Pco, induced when fluids migrate to
high permeability, underpressured lithologies, such
as at interface between mudstones and sandstones,
or along fault zones that are connected to under­
pressured zones. The precipitation of calcite can
thus be envisaged as follows:
2 calcite cement in sandstone sequences (Morad et
Ca + + 2HC03- CaC03 + C02 + H20
=
In a manner similar to carbonate precipitation in
fractures, wellbore-scale precipitation and forma­
tion damage occur due to pressure release in
hydrocarbon-producing wells (Fisher & Boles,
1 987).
2 Addition of C02 may induce carbonate precipi­
tation when the pH is externally buffered. Migra­
tion of C02 occurs along pressure gradients either
in gaseous form driven by buoyancy, or dissolved in
water by diffusion or advection. C02 in sedimen­
tary basins forms by inorganic reactions and by
organic matter maturation. Reservoirs containing
large volumes of C02 may be formed by the
metamorphism of calcareous sequences due to the
emplacement of igneous intrusions (Studlick et a!.,
1 990). C02 can also be produced as a consequence
of the pervasive dissolution of carbonate cements
and carbonate rocks (Lundegard & Land, 1 9 86).
3 Increase in HC03 concentrations due to the
degradation of oil by incurred meteoric waters. This
is evidenced by carbonate cementation along the
oil-water surface, such as in Tertiary, turbiditic
reservoir sandstones from northern North Sea
(Watson et a!., 1 995).
Although the presence of cements along fractures
is indicative of water flow, precipitation does not
necessarily occur by advection, but rather by ionic
diffusion from the host sediments. Advective ce­
mentation requires the circulation of huge water
volumes. For each pore volume of cement, 1 04 to
1 05 water volumes are required (Bathurst, 1 97 5 ;
Wood, 1 986; Sharp e t a!., 1 988). The distinction
between cements formed by diffusive and advective
material flux is difficult, but certainly important in
mass transfer studies. In contrast with diffusion,
advection may indicate the derivation of external
waters that have been subjected to temporal varia­
tions in chemical composition. This would result in
complex chemical zonations within the carbonate
crystals.
Mesogenetic carbonate cements are derived inter­
nally from within the sandstones and externally
from interbedded and juxtaposed beds as well as
from waters migrated from deeper parts of the
basins along fractures. The dissolution and repre­
cipitation of eogenetic carbonate cements and bio­
clasts are among the important internal sources.
Albitization of Ca-plagioclase has also been consid­
ered as an internal source of calcium (Schulz et al.,
1 989), but probably accounts for a small portion of
a!., 1 990). External sources include interbedded
and tectonically lower or juxtaposed lithologies
such as mudstones, carbonate rocks and evaporites
(e.g. Purvis, 1 992; Gaupp et a!., 1 993). Evidence
used in support of external sources includes a
greater abundance of carbonate cements at the
boundaries with adjacent mudstones (e.g. Carvalho
et a!., 1 99 5 ; Moraes & Surdam, 1 993). However,
Sullivan & McBride ( 1 99 1 ) found no relationship
between carbonate cement distribution in sand­
stones and the mudstones of the Gulf Coast Ter­
tiary. Moreover, in the absence of pH buffering
agents, waters charged with high Pco, derived from
mudstones may indeed induce carbonate dissolu­
tion rather than precipitation. Ca-charged dolomi­
tizing waters derived from deeply buried carbonate
rocks migrate upwards and contribute to the calcite
cementation of sandstones (Morad et al., 1 994).
Burial carbonate cementation occurs subsequent
to considerable compaction, leading to a successive
decrease of both intergranular volume(IGV) and of
o 180 of the carbonate. However, in some basins,
carbonate cementation may occur by ascending hot
basinal brines to shallow depths (Sullivan et al.,
1 990). Such cements occur in weakly compacted
sediments and are characterized by low 8180 values
and fluid inclusions with high homogenization tem­
peratures. This mechanism imposes difficulties in
recognizing these cements from those formed by
recrystallization at increased temperatures, as both
mechanisms preserve a high, pre-cement porosity.
A few workers (Giroir et a!., 1 989; Souza et a!.,
1 995) argued that the early emplacement of calcite
cement in sandstones of rift basins may take place
 Geochemical evolution of carbonate cements
from hot convected waters driven by the high
geothermal gradients related to the oceanic open­
ing. The role of hot water circulation due to the
emplacement of diabase on the fracturing and
diagenesis of sandstones has been proposed by
Girard et a!. ( 1 988).
DIRECTIONS FOR
FUTURE RESEARCH
Although a considerable advance has been made in
our understanding of clastic diagenesis and of car­
bonate cementation in particular, factors control­
ling the cementation of ancient shallow marine
sandstones which lack present day analogues, such
as eogenetic, strata-bound calcite-cemented, marine
sandstones are unclear. What are the sources of
calcite cement in these sandstones when carbonate
bioclasts are totally absent? Has the global and
regional change in ocean chemistry and pattern of
circulation any impact on cementation of sand on
the sea floor?
The numerical modelling of patterns, extent and
mechanism of water flow in the subsurface and their
influence on the mineralogy, geochemistry and dis­
tribution of carbonate cements should be an area of
further future research. Questions that need to be
answered include the following. Are the basinal
brines in equilibrium with the conductive thermal
field of the basin, and are they static, actively
flowing or moving only sluggishly? Are water move­
ments induced essentially by extrinsic tectonic and
thermal factors? What are the sources of deep,
mesogenetic cements? What are the importance and
scales of advection versus diffusion in their forma­
tion? Are metamorphic, magmatic, and perhaps
even mantle-derived waters, involved in the diage­
netic evolution of formation waters and host sedi­
ments? Can carbonate cement redistribution and its
influence on the properties of deep reservoirs be
quantified?
Volcanic events have been frequent throughout
most of geological history, yet there are few studies
documenting their importance in sandstone diagen­
esis and carbonate cementation in particular. Is this
related to difficulties in recognizing volcaniclastic
sediments due to their rapid and extensive diage­
netic alteration?
Future research in clastic diagenesis would ben­
efit from an interdisciplinary approach with respect
to other water-related disciplines, such as igneous/
19
hydrothermal, metamorphic, structural, ore and
hydrogeology. Our view of diagenetic evolution of
sandstones is currently strongly biased towards the
rapidly subsiding basins of the Gulf Coast of USA
and the North Sea in north-west Europe. Studies
should include a wider diversity of basinal settings
to approach a more realistic picture of clastic
diagenesis. Finally, the sharp line between scientists
dealing with the diagenesis of mudstones and car­
bonate rocks should be removed, and instead we
should learn from what have been achieved by
them, for instance about coastal and nearshore pore
water chemistry and diagenesis, and about the
factors .and mechanisms of massive dolomitization
of limestones.
ACKNOWLEDGEMENTS
I thank I.S. Al-Aasm, L.D. De Ros, W. Dickinson,
Q. Fisher, C. Macaulay, J. Hendry and C. Spot! for
constructive reviews of the manuscript. I am grate­
ful to the Swedish Natural Science Research Coun­
cil (NFR) for supporting my research activities.
REFERENCES
ALONSO ZARZA, A.M., WRIGHT, V.P., CALVO, J.P. & CARCIA
DEL CURA, M.A. ( 1 992) Soil-landscape and climatic
relationships in the Middle Miocene of the Madrid
Basin. Sedimentology, 39, 1 7-35.
AMIEUX, P., BERNIER, P., DALONGEVILLE, R. & ME D WECKI,
V. ( 1 989) Cathodoluminescence of carbonate-cemented
Holocene beachrock from the Togo coastline (West
Africa): an approach to early diagenesis. Sediment.
Geol. , 65, 2 6 1 -272.
ARAKEL, A.V. ( 1 986) Evolution of calcrete in paleodrain­
ages of Lake Napperby area, central Australia. Paleo­
geogr. Paleoclimatol. Paleoecol., 54, 2 8 3-303.
ARAKEL, A.V. ( 1 99 1 ) Evolution of Quaternary duricrusts
in Karinga Creek drainage system, central Australian
groundwater discharge zone. Ausl. J. Earth Sci., 38,
3 34-347.
ARAKEL, A.V. & McCoNCHIE, D. ( 1 982) Classification and
genesis of calcrete and gypsite lithofacies in palaeo­
drainage systems of inland Australia and their relation­
ship to carnotite mineralization. J. sediment. Petrol., 52,
1 1 49- 1 1 70.
ARAKEL, A.V., JACOBSON, G. & LAYONS, W.B. ( 1 990)
Sediment-water interaction as a control on geochemical
evolution of playa lake systems in the Australian arid
interior. Hydrobiologia, 197, 1 - 1 2.
BAKER, J.C., KASSAN, J. & HAMILTON, P.J. ( 1 996) Early
diagenetic siderite as an indicator of depositional envi­
ronment in the Triassic Rewan Group, southern Bowen
Basin, eastern Australia. Sedimentology, 43 , 77-88.
20 S. Morad
BAKER, P.A. & KASTNER, M. ( 1 9 8 1 ) Constraints on the for­
mation of sedimentary dolomite. Science, 213, 2 1 4- 2 1 6 .
BALTZER, F., KENIG, F., BOICHARD, R . , PLAZIAT, J.-C. &
PuRSER, B. H. ( 1994) Organic matter distribution, water
circulation and dolomitization beneath the Abu Dhabi
Sabkha (United Arab Emirates). In: Dolomites (Eds
Purser, B., Tucker, M. & Zenger, D). Spec. Pubis int.
Ass. Sediment., 21, 409-427. Blackwell Science, Ox­
ford.
BATHURST, R.G.C. ( 1 9 7 5 ) Carbonate Sediments and their
Diagenesis, 2nd edn. Elsevier, Amsterdam, 658 pp.
BELL, J.L. ( 1 99 1 ) Acetate Decomposition in Hydrothermal
Solutions. Pennsylvania State Uni versity, Pennsylvania.
BERNER, R.A. ( 1968) Calcium carbonate concretions
formed by the decomposition of organic matter. Sci­
ence, 159, 195-197.
BERNER, R.A. ( 1 98 1 ) A new geochemical classification of
sedimentary environments. J. sediment. Petrol., 51 ,
359-365.
BERNER, R.A. ( 1 984) Sedimentary pyrite formation: an
update. Geochim. Cosmochim. Ac/a, 48, 605-6 1 5 .
BERNER, R.A., ScoTT, M.R. & THOMLINSON, C . ( 1 9 70)
Carbonate alkalinity in the pore waters of anoxic ma­
rine sediments. Limnol. Oceanogr., 15, 544-549.
BESLY, B.M. & FIELDING, C.R. ( I 989) Paleosols in West­
phalian coal-bearing and redbed sequences, central and
northern England. Paleogeogr. Paleoclimatol. Paleo­
ecol., 70, 303-330.
BESTL-\ND, E. A. & RETALLACK, G.J. ( 1993) Volcanically
influenced calcareous paleosols from the Miocene Kia­
hera Formation, Rusinga Island, Kenya. J. geol. Soc.
London, ISO, 293-310.
BJ0RKUM, P.A. & WALDERHAUG, 0. ( 1 990) Geometrical
arrangement of calcite cementation within shallow ma­
rine sandstones. Eanh Sci. Rev., 29, 1 4 5- 1 6 1 .
BJ0RKUM, P.A. & WALDERHAUG, 0. ( 1993) Isotopic COm­
position of a calcite-cemented layer in the Lower Juras­
sic Bridport Sands, southern England: implications for
formation of laterally extensive calcite-cemented layers.
J sediment. Petrol. , 63, 678-682.
BOLES, J.R. ( 1978) Active ankerite cementation in the
subsurface Eocene of southwest Texas. Contrib. Min­
eral. Petrol., 68, 1 3-22.
BOLES, J.R. ( 1 982) Active albitization of plagioclase, Gulf
Coast Tertiary. Am. J. Sci., 282, 1 65-1 80.
BRAITHWAITE, C.J.R. ( 1989) Displacivc calcite and grain
breakage in sandstones. J. sediment. Petrol., 59, 258-
266.
BRowNE, G. H. & K.lNGSTON , D.M. ( 1 993) Early diagenetic
spherulitic siderites from Pennsylvanian paleosols in
the Boss Point Formation, Maritime Canada. Sedimen:
cology, 40, 467-474.
BRUHN, C.H.L ( 1993) High-resolution stratigraphy, reser­
voir geometry and facies characterization of Cretaceous
and Tertiary wrbidites from rift and passive margin
Brazilian basins. PhD thesis, McMaster University,
Hamilton.
BUNTING, B.T. & CHRISTENSEN, L. ( 1 980) Micromorphol­
ogy of calcareous crusts from the Canadian High Arctic.
Ceo!. Foren. Stockholm Forhandl. . 100, 3 6 1 -36 7.
BURLEY, S.D. ( 1993) Models of burial diagenesis for deep
exploration plays in Jurassic fault traps of the Central
and Northern North Sea. In: Petroleum Geology of
Northwest Europe: Proceedings of the 4th Conferen�·e
(Ed. Parker, J.R.), pp. 1 3 5 3 - 1 3 7 5 . Geological Society,
Bath.
BURLEY, S.D. & KANTOROWIC'Z, J.D. ( 1 986) Thin section
and SEM textural criteria for the recognition of cement­
dissolution porosity in sandstones. Sedimentology, 33.
587-604.
BURNS, S.J. & MATTER, A. ( 1 995) Geochemistry of carbon­
ate cements in surficial alluvial conglomerates and their
paleoclimatic implications, Sultanate of Oman. J. sedi­
ment. Res., A65. 1 70-l 77.
BURNS, S.J., BAKER, P.A. & SHOWERS, W.J. { 1 988) The
factors controlling the formation and chemistry of
dolomite in organic-rich sediments: Miocene Drakes
Bay Formation, California. In: Sedimentology and
Geochemistry of Dolostones (Eds Shukla, V. & Baker.
P.A.). Spec. Pubis int. Ass. Sediment. 43, 4 1 -52. Black­
well Science, Oxford.
CARITAT, P.O. & BAKER, J.C. ( 1 992) Overpressure release,
cross-formational porewater flow and diagenesis. In:
Water-Rock Interaction (Eds Kharaka, Y.K. & Maest,
A.S.). Proceedings of the 7th International Symposium
on Water-Rock Interaction. Rouerdam. The Nether·
lands, Vol. 2. Moderate and High Temperawre Environ­
ments, pp. 1 1 6 1 - 1 1 64.
CAROTHERS, W.W. & KHARAKA, Y.K. ( 1 978) Aliphatic acid
anions in oil-field waters-implications for origin of
natural gas. Bull. Am. Ass. petrol. Ceo/., 62, 2441-24 5 3 .
CAROTHERS, W .W . & KHARAKA, Y.K. ( 1 9 8 0 ) Stable carbon
isotopes of HC03- in oil-field waters-implication for
the origin of C02• Geochim. Cosmochim. Acta, 44.
323-332.
CARVALHO, M.V.F., DE Ros, L.F. & GoMES, N.S. ( 1 995)
Carbonate cementation patterns and diagenetic reser­
voir facies in the Campos Basin Cretaceous turbidites,
offshore eastern Brazil. Mar. Petrol. Geol., 12, 741-758.
CERLING, T. ( 1991) Carbon dioxide in the atmosphere:
evidence from Cenozoic and Mesozoic paleosols. Am. J.
Sci., 291, 3 77-400.
CERLING, T.E. ( 1 984) The stable isotopic composition of
modern soil carbonate and its relationship to climate.
Earth planet. Sci. Leu., 71, 229-240.
CERLING, T.E. & HAY, R.L. ( 1 986) An isotopic study of
paleosol carbonates from Olduvai Gorge. Quat. Res.,
25, 63-78.
CERLING, T.E., QUADE, J., WANG, Y. & BOWMAN, J.R.
( 1989) Carbon isotopes in soils and paleosols as ecology
and paleoecology indicators. Nature, 341, 1 38 - 1 3 9 .
CHAFETZ, H.S. & RUSH, P.F. ( 1994) Diagenetically altered
sabkha-type Pleistocene dolomite from the Arabian
Gulf. Sedimentology, 4 1 , 409-42 1 .
CHOW, N., MORAD, S. & AL-AASM, I.S. ( 1 996) Origin of
authigenic carbonates in Eocene to Quaternary sedi­
ments from the Arctic and Norwegian-Greenland Sea.
In: Proceedings ofthe Ocean Drilling Program, Scient(jic
Results (Eds Myhre, A., Thiede, J. et at.). Ocean Drilling
Program, College Station, TX.
CLAYPOOL, G.E. & KAPLAN, I.R. ( 1 974) The origin and
distribution of methane in marine sediments. In: Natu­
ral Gases in Marine Sediments (Ed. Kaplan, I.R.)
pp. 97-139.
CLAYTON, C.J. ( 1 994) Microbial and organic processes. In:
Quantitative Diagenesis: Recent Developments and Ap-
 Geochemical evolution of carbonate cements
plications to Reservoir Geology (Eds Parker, A. & Sell­
wood, B.W.). NATO ASI Series C: Mathematical and
Physical Sciences, 453, 1 25-1 60. Kluwer, Dordrecht.
COLEMAN, M.L. ( 1 985) Geochemist ry of diagenetic non­
silicate minerals: kinetic considerations. Philos Trans.
R. Soc. London, Ser. A 3 1 5, 39-56.
COLEMAN, M.L. & RAISWELL, R. ( 1 993) Microbial mineral­
ization of organic matter: mechanisms of self­
organization and inferred rates of precipitation of
diagenetic minerals. Philos. Trans. R. Soc. London. Ser.
A. 344, 69-87.
COLEMAN, M.L., FLEET, A.J. & DONSON, P. ( 1 982) Prelim­
inary studies of manganese-rich carbonate nodules from
Leg 68, Site 503, eastern equatorial Pacific. In: Initial
Reports of the Deep Sea Drilling Project (Eds Prell, W.L.
& Gardner, J.V.), No. 68, pp. 48 1 -489. US Govern­
ment Printing Office, Washington, DC.
COLEMAN, M.L., HEDRICK, D.B., LOVELY, D.R., WHITE,
D.C. & PYE, K. ( 1 993) Reduction ofFe(III) in sediments
by sulphate-reducing bacteria. Nature, 36 1 , 436-438.
COLSON, I. & COJAN, I. ( 1 996) Groundwater dolocretes in a
lake-marginal environment: an alternative model for do­
locrete formation in continental settings (Danian of the
Provence Basin, France). Sedimentology, 43, I 7 5- 1 88.
CuRTIS, C.D. ( 1 967) Diagenetic iron minerals in some
British Carboniferous sediments. Geochim. Cosmo­
chim. Acta, 3 1 , 2 1 09-2 1 23.
CuRTIS, C.D. ( 1 987) Mineralogical consequences of or­
ganic matter degradation in sediments: inorganic/
organic diagenesis. In: Marine Clastic Sedimentology
-Concepts and Case Studies (Eds Leggett, J.K. & Zuffa,
G.G.), pp. 1 08- 1 23 . Graham & Trotman, London.
DE Ros, L.F., MORAD, S. & AL-AASM, I.S. ( 1 997) Diagen­
esis of siliciclastic and volcaniclastic sediments in the
Cretaceous and Miocene sequences of the NE African
margin. Sediment. Geol., 1 1 2, 1 3 7- 1 56.
DEUSER, W.G. ( 1 979) Extreme 1 3 C/ 1 2C variations in Qua­
ternary dolomites from the continental shelf. Earth
planet. Sci. Lett., 8, 1 1 8- 1 24.
DRIESE, S.G. & MORA, C.I. ( 1 993) Physico-chemical envi­
ronment of pedogenic carbonate formation in Devonian
vertic palaeosols, central Applachians, USA. Sedimen­
tology, 40, 1 99-2 1 6.
DROZDOWSKI, E. ( 1 980) Cementation of clastic sediments
by calcium carbonates under stagnant ice sheets. In:
INQUA Till Norway (Ed. Sollid, J.L.), 34, 97-98.
DUTTON, S.P. & LAND, L.S. ( 1 985) Meteoric burial diagen­
esis of Pennsylvanian arkosic sandstones, southwestern
Anadarko Basin, Texas. Bull. Am. Ass. petrol. Geol. , 69,
22-3 8.
EGEBERG P.K. ( 1 990) Unusual composition of pore waters
found in the Izu-Bonin fore-arc sedimentary basin.
Nature, 344, 2 1 5-2 1 8.
EGEBERG, P.K. & AAGAARD, P. ( 1 989) Origin and evolution
of formation water from oil fields on the Norwegian
shelf. Appl. Geochem., 4, 1 3 1 - 1 42 .
EL-SAYED, M.I., FAIRCHILD, I.J. & SPIRO, B . ( 1 99 1 ) Kuwaiti
dolocrete: petrology, geochemistry and groundwater
origin. Sediment. Geol. 73 , 59-75.
EMERSON, S. & WIDMER, G. ( 1 978) Early diagenesis in
anaerobic lake sediments. II. Thermodynamic and ki­
netic factors controlling the formation of iron phos­
phate. Geochim. Cosmochim. Acta, 42, 1 307- 1 3 1 6.
21
FISHER, J.B. & BOLES, J.R. ( 1 987) Water-rock interaction
in Tertiary sandstones, San Joaquin Basin, California,
U.S.A.: Diagenetic controls on water composition.
Chern. Geol. , 82, 83- 1 0 1 .
FLOOD, R.D., PIPER, D.J.W. & PARTY, S.S. ( 1 995) Sites
930-946. In: Proceedings of the Ocean Drilling Pro­
gram, Initial Reports, No. ! 5 5. Ocean Drilling Program,
College Station, TX, 1 23 3 pp.
FRANK, T.D. & LOHMANN, K.C. ( 1 995) Early cementation
during marine-meteoric fluid mixing: Mississippian
Lake Valley Formation, New Mexico. J. sediment.
Petrol. , A65, 263-273.
FROELICH, P.N., KLINKHAMMER, G.P., BENDER, M.L. et a/.
( 1 979) Early oxidation of organic matter in pelagic
sediments of eastern equatorial Atlantic: suboxic di­
agenesis. Geochim. Cosmochim. Acta, 43, 1 075- 1 090.
GALIMOV, E.M. ( 1 985) The Biological Fractionation of
Isotopes. Academic Press, London, 2 6 1 pp.
GARRISON, R.E., KASTNER, M. & ZENGER, D.H. ( 1 984)
Dolomites of the Monterey Formation and Other
Organic-rich Units. Soc. Econ. Paleontol. Mineral. Pa­
cific Sect. 4 1 , 2 1 5 pp.
GAUPP, R., MATTER, A., PLATT, J., RAMSEYER, K. &
WALZEBUCK, J. ( 1 993) Diagenesis and fluid evolution of
deeply buried Permian (Rotliegend) gas reservoirs,
northwest Germany. Bull. Am. Ass. petrol. Geol. , 77,
1 1 1 1 - 1 1 28.
GAUTIER, D.L. & CLAYPOOL, G.E. ( 1 984) Interpretation of
Early Diagenesis in Ancient Sediments. Society of Eco­
nomic Paleontologists and Mineralogists, Tulsa, Short
Course 1 7, 6-78.
GILES, M.R. & DE BOER, R.B. ( 1 990) Origin and signifi­
cance of redistributional secondary porosity. Mar.
Petrol. Geol. , 6, 378-397.
GILES, M.R. & MARSHALL, J.D. ( 1 986) Constraints on the
development of secondary porosity in the subsurface:
re-evaluation of processes. Mar. Petrol. Geol., 3, 243-
255.
GIORDANO, T.H. & KHARAKA, Y.K. ( 1 993) Organic ligand
distribution and speciation in sedimentary basin brines,
diagenetic fluids and related ore solutions. In: Geoj/uids:
Origin, Migration and Evolution of Fluids in Sedimen­
tary Basins (Ed. Parnell, J.) Spec. Pub!. geol. Soc.
London, 78 , 1 75-202.
GIRARD, J.-P., DEYNOUX, M. & NAHON, D. ( 1 988) Diagen­
esis of the Upper Proterozoic siliciclastic sediments of
the Taoudeni Basin (West Africa) and relation to dia­
base emplacement. J. sediment. Petrol., 59, 2 3 3-248.
GIROIR, G., MERINO, E. & NAHON, D. ( 1 989) Diagenesis of
Cretaceous sandstone reservoirs of the south Gabon rift
basin, West Africa: mineralogy, mass transfer, and
thermal evolution. J. sediment. Petrol., 59, 482-493.
GIVEN, R.K. & WILKJNSON, B.H. ( 1 987) Dolomite abun­
dance and stratigraphic age: constraints on rates and
mechanisms of Phanerozoic dolostone formation. J.
sediment. Petrol., 57, 1 068- 1 078.
GLUYAS, J. & COLEMAN, M. ( 1 992) Material flux and
porosity changes during sediment diagenesis. Nature,
356, 52-54.
GouDIE, A.S. ( 1 983) Calcrete. In: Chemical Sediments and
Geomorphology: Precipitates and Residues in the Near­
Surface Environment (Ed. Goudie, A.S. & Pye, K.), pp.
9 3-1 3 1 . Academic Press, San Diego.
22 S. Morad
GREGG, J.M., HOWARD, S.A. & MAZZULLO, S.J. ( 1 992)
Early diagenetic recrystallization of Holocene ( < 3000
years old) peritidal dolomites, Ambergris, Belize. Sedi­
mentology, 39, 1 43- 1 60.
GROTZINGER, J.P. & KNOLL, A.H. ( 1 995) Anomalous car­
bonate precipitation: is the Precambrian the key to the
Permian? Pa/aios, 10, 578-596.
Guo, B. & FRIEDMAN, G.H. ( 1 990) Petrophysical charac­
teristics of Holocene beachrock. Carbonates Evaporites,
5, 85-92.
HANOR, J.S. ( 1 978) Precipitation of beachrock cements:
mixing of marine and meteoric waters vs. C02-
degassing. J. sediment. Petrol., 48, 489-50 1 .
HANOR, J.S. & WORKMAN, A.L. ( 1 986) Distribution of
dissolved volatile fatty acids in some Louisiana oil-field
brines. App/. Geochem., I, 37-46.
HARDING, T.P. & TUMINAS, A. C. ( 1 989) Structural inter­
pretation of hydrocarbon traps sealed by basement
normal blocks and at stable flank of foredeep basins and
rift basins. Bull. Am. Ass. petrol. Geol. , 73, 8 1 2-840.
HATHWAY, J.C. & DEGENS, E.T. ( 1 969) Methane-derived
marine carbonates of Pleistocene age. Science, 165,
690-692.
HAY, R.L. & WIGGINS, B. ( 1 980) Pellets, ooids, sepiolite
and silica in three calcretes of the southwestern United
States. Sedimentology, 27, 5 5 9-576.
HENDRY, J.P., TREWIN, N.H. & FALLICK, T.E. ( 1 996)
Low-Mg calcite marine cement in Cretaceous turbidites:
origin, spatial distribution and relationship to seawater.
Sedimentology, 43, 877-900.
HINDLE, A.D. ( 1 989) Downthrown traps of the NW Witch
Ground Graben, UK North Sea. J. petrol. Geo/., 1 2,
405-4 1 8.
HOLAIL, H. & RASHED, M. ( 1 992) Stable isotopic composi­
tion of carbonate-cemented recent beachrock along the
Mediterranean and the Red Sea coasts of Egypt. Mar.
Geo/., 106, 1 4 1 - 1 48.
HOVLAND, M., TALBOT, M.R., Q VALE, H., 0LAUSSEN, S. &
AASBERG, L. ( 1 987) Methane-related carbonate cements
in pockmarks of the North Sea. J. sediment. Petrol., 57,
8 8 1 -892.
HUDSON, J.D. & ANDREWS, J.E. ( 1 987) The diagenesis of
the Great Stuarine Group, Middle Jurassic, Inner He­
brides, Scotland. In: Diagenesis of Sedimentary Se­
quences (Ed. Marshall, J.D.), Spec. Pub!. geol. Soc.
London, 36, 259-276.
HUTCHEON, I. & ABERCROMBIE, H. ( 1 990) Carbon dioxide
in clastic rocks and silicate hydrolysis. Geology, 18,
5 4 1 -544.
INGVALD, E. ( 1 995) Carbonate diagenesis of the Permian
Ajligidos Formation, Reconcavo Basin, Bahia, NE Bra­
zil: evidencefrom geochemistry and isotopes. MSc thesis,
Uppsala University, Sweden.
JAMES, N.P. & CHOQUETTE, P.W. ( 1 990) Limestones-the
meteoric diagenetic environment. In: Diagenesis (Eds
Mcllreath, I.A. & Morrow, D.W.). Geosci. Can. Reprint
Ser., 4, 3 5-73.
JAMES, W.C. ( 1 992) Sandstone diagenesis in mixed
siliciclastic-carbonate sequences: Quadrant and Ten­
sleep formations (Pennsylvanian), northern Rocky
Mountains. J. sediment. Petrol. , 62, 8 1 0-824.
JoHNSON, M.R. ( 1 989) Paleogeographic significance of
oriented calcareous concretions in the Triassic Katberg
Formation, South Africa. J. sediment. Petrol. , 59, I 008-
1 0 1 0.
JORGENSEN, N.O. ( 1 976) Recent high magnesian calcite/
aragonite cementation of beach and submarine sedi­
ments from Denmark. J. sediment. Petrol. , 46, 940-95 1 .
JORGENSEN, N.O. ( 1 979) Magnesium incorporation in
Recent marine calcite cement from Denmark. J. sedi­
ment. Petrol. , 49, 945-950.
KAISER, W.R. ( 1 984) Predicting reservoir quality and
diagenetic history in the Frio Formation (Oligocene) of
Texas. In: Clastic Diagenesis (Ed. McDonald, D.A. &
Surdam, R.C.). Mem. Am. Ass. petrol. Geol., Tulsa, 37,
1 95-2 1 6.
KANTOROWICZ, J.D. ( 1 985) The origin of authigenic anker­
ite from the Ninian Field, UK North Sea. Nature, 315,
2 1 4-2 1 6.
KANTOROWICZ, J.D. ( 1 990) Lateral and vertical variations
in pedogenesis and other early diagenetic phenomena,
Middle Jurassic Ravenscar Group, Yorkshire. Proc.
Yorks. Geol. Soc. 48, 6 1 -74.
KANTOROWICZ, J.D., BRYANT, I. D. & DAWANS, J.M. ( 1 987)
Controls on the geometry and distribution of carbonate
cements in Jurassic sandstones: Bridport sands, south­
em England and Viking Group, Troll Field, Norway. In:
Diagenesis of Sedimentary Sequences (Ed. Marshall,
J.D.), Spec. Pub!. geol. Soc. London, 36, 1 03- 1 1 8.
KASTNER, M., ELDERFIELD, H., MARTIN, J.B., SUESS, E.,
KVENVOLDEN, K.A. & GARRISON, R.E. ( 1 990) Diagenesis
and interstitial-water chemistry at the Peruvian conti­
nental margin-major constituents and strontium iso­
topes. In: Proceedings of the Ocean Drilling Program,
Scientific Results (Ed. Suess, E., von Huene, R. et a/.).
1 1 2, 4 1 3-440.
- Ocean Drilling Program, College Station,
TX.
KEMPE, S. ( I 990) Alkalinity: the link between anaerobic
basins and shallow water carbonates? Naturwissen­
schafien, 77, 426-427.
KHARAKA, Y.K., LEROY, M.L., CAROTHERS, W.W. & GOER­
LITZ, D.F. ( 1 986) Role of organic species dissolved in
formation waters from sedimentary basins in mineral
diagenesis. In: Roles of Organic Maller in Sediment
Diagenesis (Ed. Gautier, D.L.), Spec. Pub!. Soc. econ.
Paleont. Miner., Tulsa, 38, 1 1 1 - 1 22.
KIMBELL, T.N. & HUMPHREY, J.D. ( 1 994) Geochemistry
and crystal morphology of aragonite cements of mixing­
zone origin, Barbados, West Indies. J. sediment. Res.,
A64, 604-6 1 4.
KINSMAN, D.J.J. ( 1 969) Modes of formation, sedimentary
association, and diagnostic features of shallow-water
and supratidal evaporites. Bull. Am. Ass. petrol. Geo/. ,
53, 830-840.
KAPPA, C. F. ( 1 980) Rhizoliths in terrestrial carbonates:
classification, recognition, genesis and significance. Sed­
imentology, 27, 6 1 3-629.
KNIPE, R.J. ( 1 993) The influence of fault zone processes
and diagenesis on fluid flow. In: Diagenesis and Basin
Development (Eds Horbury, A.D. & Robinson, A.G.).
Am. Ass. petrol. Geol., Tulsa, Stud. Geol., 36, 1 35-1 5 1 .
KNox, G.J. ( 1 977) Caliche profile formation, Saldanha
Bay (South Africa). Sedimentology, 24, 657-674.
KRUMBEIN, W.E. ( 1 968) Geomicrobiology and geochemis­
try of the "nari lime crust" (Israel). In: Recent Develop­
ments in Carbonate Sedimentology in Central Europe
 Geochemical evolution of carbonate cements
(Eds Muller, G . & Friedman, G.M.), pp. 1 38- 1 47.
Springer Verlag, Berlin.
KRUMBEIN, W.E. ( 1 979) Phytolithotrophic and chemoor­
ganotrophic activity of bacterias and algae related to
beachrock. Geomicrobiol. J., 1, 1 39-203.
KuPECZ, J.A. & LAND, L.S. ( 1 994) Progressive recrystalli­
zation and stabilization of early-stage dolomite: Lower
Ordovician Ellenburger Group, west Texas. In: Dolo­
mites (Eds Purser, B., Tucker, M. & Zenger, D.), Spec.
Pub!. int. Ass. Sediment. 2 1, 2 55-2 79. Blackwell Scien­
tific Publications, Oxford.
LAST, W.M. ( 1 992) Petrology of modem carbonate hard­
grounds from East Basin Lake, a saline maar lake, south
Australia. Sediment. Geol. , 8 1 , 2 1 5-229.
LAWRENCE, M.J.F. ( 1 99 1 ) Early diagenetic dolomite con­
cretions in the late Cretaceous Herring Formation,
eastern Marlborough, New Zealand. Sediment. Geol. ,
75, 1 25- 1 40.
LUNDEGARD, P.D. ( 1 994) Mixing zone origin of 1 3 C­
depleted calcite cement: Osberg Formation sandstones
(Middle Jurassic), Veslefrikk Field, Norway. Geochim.
Cosmochim. Acta, 58, 266 1 -2675.
LUNDEGARD, P.D. & LAND, L.S. ( 1 986) Carbon dioxide
and organic acids: their role in porosity enhancement
and cementation, Paleogene of the Texas Gulf Coast. In:
. Roles of Organic Mauer in Sediment Diagenesis (Ed.
Gautier, D.L.), Spec. Pub!. Soc. econ. Paleont. Miner.,
Tulsa, 38, 1 29- 1 46.
LYNN, D.C. & BONATTI, E. ( 1 965) Mobility of manganese
in diagenesis of deep-sea sediments. Mar. Geol. , 3,
547-574.
L0N0Y, A., AKSELSEN, J. & R0NNING, K. ( 1 986) Diagenesis
of deeply buried sandstone reservoirs, Hild Field, north­
ern North Sea. Clay Miner. , 2 1 , 497-5 1 1 .
McARTHUR, J.M., BENMORE, R.A., COLEMAN, M.L., SOLD!,
C., YEH, H.-W. & O'BRIEN, G.W. ( 1 986) Stable isotopic
characterization of francolite formation. Earth planet.
Sci. Leu., 77, 20-34.
MACAULAY, C.J., HASZELDINE, R.S. & FALLICK, A.E. ( 1 993)
Distribution, chemistry, isotopic composition and ori­
gin of diagenetic carbonates: Magnus Sandstone, North
Sea. J. sediment. Petrol. , 63, 3 3-43.
McBRIDE, E. F., PICARD, M.D. & FOLK, R.L. ( 1 994) Oriented
concretions, Ionian coast, Italy: evidence of groundwater
flow direction. J. sediment. Petrol. , A64, 53 5-540.
McDUFF, R.E. & GIESKES, J.M. ( 1 976) Calcium and
magnesium profiles in DSDP interstitial waters: diffu­
sion or reaction? Earth planet. Sci. Leu., 33, 1 - 1 0.
MAcGowAN, D. B. & SuRDAM, R.C. ( 1 990) Carboxylic acid
anions in formation waters, San Joaquin Basin and
Lousiana Gulf Coast, USA-implications for clastic
diagenesis. Appl. Geochem . , 5, 687-70 1 .
MACK, G.H., COLE, D.R., GIORDANO, T.H., SCHAAL, W.C.
& BARCELOS, J.H. ( 1 99 1 ) Paleoclimatic controls on
stable oxygen and carbon isotopes in caliche of the Abo
Formation (Permian), south-central New Mexico,
U.S.A. J. sediment. Petrol. , 6 1 , 458-4 72.
McKAY, J.L., LoNGSTAFFE, F.J. & PuNT, A.G. ( 1 995) Early
diagenesis and its relationship to depositional environ­
ment and relative sea-level fluctuations (Upper Creta­
ceous Marshybank Formation, Alberta and British
Columbia). Sedimentology, 42, 1 6 1 - 1 90.
McMASTER, R. ( 1 984) Holocene stratigraphy and deposi-
23
tiona! history of the Narragansett Bay System, Rhode
Island, U.S.A. Sedimentology, 3 1 , 777-792.
MAGARITZ, M., GAVISH, E., BALKER, N. & KAFRI, U. ( 1 979)
Carbon and oxygen isotope composition-indicators of
cementation environment in Recent, Holocene, and
Pleistocene sediments along the coast of Israel. J.
sediment. Petrol. , 49, 40 1 -4 1 2.
MAGARITZ, M., KAUFMAN, A. & YAALON, D.H. ( 1 9 8 1 )
Calcium carbonate nodules i n soils: 1 80/160 and 1 3 C/
1 2 C ratios and 1 4C contents. Geoderma, 25, 1 57- 1 72.
MAIZELS, J.K. ( 1 987) Plio-Pleistocene raised river channel
systems of the western Shargiya (Wahiba), Oman. In:
Desert Sediments: Ancient and Modern (Eds Frostick,
L.E. & Reid, I.), Spec. Pub!. geol. Soc. London, 35,
3 1 -50.
MALONE, M.J., BAKER, P.A. & BuRNS, S.J. ( 1 994) Recrys­
tallization of dolomite: evidence from the Monterey
Formation (Miocene), California. Sedimentology, 4 1 ,
1 223- 1 239.
MATSUMOTO, R. ( 1 989) Isotopically heavy oxygen­
containing siderite derived from the decomposition of
methane hydrate. Geology, 1 7 , 707-7 1 0.
MATSUMOTO, R. ( 1 992) Diagenetic dolomite, calcite,
rhodochrosite, magnesite, and landsfordite from Site
799, Japan Sea-implications for depositional environ­
ments and the diagenesis of organic-rich sediments. In:
Proceedings of the Ocean Drilling Program, Scientific
Results (Eds Pisciotto, K.A., Ingle, J.C. et a/.). 1 27/1 28,
75-98.
MATSUMOTO, R. & IIJIMA, A. ( 1 98 1 ) Origin and diagenetic
evolution ofCa-Fe-Mg carbonates in some coalfields of
Japan. Sedimentology, 28, 239-259.
MATSUMOTO, R. & MATSUDA, H. ( 1 98 7) Occurrence, chem­
istry and isotopic composition of carbonate concretions
in the Miocene to Pliocene siliceous sediments of
Aomori, northeastern Japan. J. Fac. Sci. , Univ. Tokyo,
2 1 , 35 1 -377.
MAZZULLO, S.J. (I 992) Geochemical and neomorphic
alteration of dolomite: a review. Carbonates Evaporites,
7, 2 1 -37.
MINOURA, K. ( 1 9 9 1 ) Origin of manganese carbonates in
Jurassic red shales, central Japan. Sedimentology, 38 ,
1 37- 1 52.
MOORE, S.E., FERRELL, R.E. & AHARON, P. ( 1 992) Diage­
netic siderite and other ferroan carbonates in a modem
subsiding marsh sequence. J. sediment. Petrol. , 62,
357-366.
MORA, C.I., DRIESE, S.G. & SEAGER, P.G. ( 1 9 9 1 ) Carbon
dioxide in Paleozoic atmosphere: evidence from
C-isotopic compositions of pedogenic carbonate. Geol­
ogy, 1 9, 1 0 1 7- 1 020.
MORAD, S. & DE Ros, L.F. ( 1 994) Geochemistry and
diagenesis of stratabound calcite cement layers within
the Rannoch Formation of the Brent Group, Murchison
Field, North Viking Graben (northern North Seal­
comment. Sediment. Geol. , 93 , 1 35- 1 4 1 .
MORAD, S., BERGAN, M., KNARUD, R. & NYSTUEN, J.P.
( 1 990) Albitization of detrital plagioclase in Triassic
reservoir sandstones from the Snorre Field, Norwegian
North Sea. J. sediment. Petrol. , 60, 4 1 1 -425.
MORAD, S., MARFIL, R., AL-AASM, I.S. & GOMEZ-GRAS, D.
( 1 992) The role of mixing-zone dolomitization in sand­
stone cementation: evidence from the Triassic Bunt-
24 S.
sandstein, the Iberian Range, Spain. Sediment. Geol. ,
80, 5 3�65.
MoRAD, S., B EN IsMAIL, H., D E Ros, L.F., AL-AASM, I.S. &
SERRHINI, N.-E. ( 1 994) Diagenesis and formation water
chemistry of Triassic reservoir sandstones from south­
em Tunisia. Sedimentology, 41, 1 25 3 - 1 272.
MORAD, S., AL-AASM, l.S., LONGSTAFFE, F.J., MARFIL, R.,
DE Ros, L.F., JoHANSEN, H. & MARZO, M. ( 1 995)
Diagenesis of a mixed siliciclastic/evaporitic sequence
of the Middle Muschelkalk (Middle Triassic), the Cata­
lan Coastal Range, NE Spain. Sedimentology, 42, 749-
768.
MoRAD, S., D E Ros, L.F. & AL-AASM, I.S. ( 1 996) Origin of
low o 1 80, pre-compactional ferroan carbonates in the
marine St0 Formation (Middle Jurassic), offshore NW
Norway. Mar. Petrol. Geol. , 1 2 , 263-276.
MORAES, M.A.S. & SuRDAM, R.C. ( 1 993) Diagenetic heter­
ogeneity and reservoir quality: fluvial, deltaic, and
turbiditic sandstone reservoirs, potiguar and rec6ncavo
rift basins, Brazil. Bull. Am. Ass. petrol. Geol. , 77,
1 1 42- 1 1 58.
MORSE, J.W. & MACKENZIE, F.T. ( 1 990) Geochemistry of
Sedimentary Carbonates. Elsevier, Amsterdam, 707 pp.
MORTON, R.A. & LAND, L.S. ( 1 987) Regional variations in
formation water chemistry, Frio Formation (Oligocene),
Texas Gulf Coast. Bull. Am. Assoc. petrol. Geol. , 7 1 ,
1 9 1 -206.
MozLEY, P.S. ( 1 989a) Relation between depositional envi­
ronment and the elemental composition of early diage­
netic siderite. Geology, 17, 704-706.
MOZLEY, P.S. ( 1 989b) Complex compositional zonation in
concretionary siderite: implications for geochemical
studies. J. sediment. Petrol. , 59, 8 1 5-8 1 8.
MOZLEY, P.S. & BURNS, S.J. ( 1 992) Oxygen and carbon
isotopic composition of marine carbonate concretions,
an overview. J. sediment. Petrol. , 63, 73-83.
MOZLEY, P.S. & HOERNLE, K. ( 1 990) Geochemistry of
carbonate cements in the Sag River and Shublik Forma­
tions (Triassic/Jurassic), North Slope, Alaska: implica­
tions for the geochemical evolution of formation waters.
Sedimentology, 37 , 8 1 7-836.
MUIR WooD, R. ( 1 993) Earthquakes, strain-cycling and
the mobilization of fluids. In: Geofluids: Origin, Migra­
tion and Evolution ofFluids in Sedimentary Basins (Ed.
Parnell, J.), Spec. Publ. geol. Soc. London, 78, 85-98.
NELSON, C.S. & LAWRENCE, M.F. ( 1 984) Methane-derived
high-Mg calcite submarine cement in Holocene nodules
from the Fraser Delta, British Columbia, Canada. Sedi­
mentology, 3 1 , 645-654.
PATTERSON, R.J. & KINSMAN, D.J.J. ( 1 982) Formation of
diagenetic dolomite in coastal sabkha along Arabian
(Persian) Gulf. Bull. Am. Ass. petrol. Geol. , 66, 28-43.
PEDERSEN, T.F. & PRICE, N.B. ( 1 982) The geochemistry of
manganese carbonate in Panama Basin sediments.
Geochim. Cosmochim. Acta, 46, 59-68.
PLUMMER, LN. ( 1 975) Mixing of sea water with calcium
carbonate ground water. Geol. Soc. Am. Mem., Boul­
der, 142, 2 1 9-236.
PosTMA, D. ( 1 982) Pyrite and siderite formation in brack­
ish and freshwater swamp sediments. Am. J. Sci., 282,
1 1 5 1 - 1 1 85.
PROSSER, D.J., DAWS, J.A., FALLICK, A.E. & WILLIAMS,
B.P.J. ( 1 993) Geochemistry and diagenesis of stra-
Morad
tabound calcite cement layers within the Rannoch
Formation of the Brent Group, Murchison Field, North
Viking Graben (northern North Sea). Sediment. Geol. ,
87, 1 39-1 64.
PUEYO MuR, J.J. & INGLES URPINELL, M. ( 1 987) Magnesite
formation in recent playa lakes, Los Monegros, Spain.
In: Diagenesis of Sedimentary sequences (Ed. Marshall,
J.D.), Spec. Publ. geol. Soc. London, 36, 1 1 9- 1 22.
PURSER, B.H., TUCKER, M.E. & ZENGER, D.H. ( 1 994)
Problems, progress and future research concerning do­
lomites and dolomitization. In: Dolomites (Eds Purser,
B.H., Tucker, M.E. & Zenger, D.H.), Spec. Publ. int.
Ass. Sediment., 2 1 , 3-20. Blackwell Science, Oxford.
PURVIS, K. ( 1 992) Lower Permian Rotliegend sandstones,
southern North Sea: a case study of sandstone diagene­
sis in evaporite-associated sequences. Sediment. Geol. ,
77, 1 5 5- 1 7 1 .
PYE, K., DICKSON, A.D., SCHIAVON, N., COLEMAN, M.L. &
Cox, M. ( 1 990) Formation of siderite-Mg-calcite-iron
sulfide concretions in intertidal marsh and sandflat
sediments, north Norfolk, England. Sedimentology, 37 ,
325-343.
Q uADE, J., CHIVAS, A.R. & McCuLLOCH, M.T. ( 1 995)
Strontium and carbon isotope tracers and the origin of
soil carbonate in South Australia and Victoria. Palaeo­
geogr. Palaeoclimatol. Palaeoecol., 1 13, 1 03- 1 1 7.
RICHTER, O.K. & FOCHTBAUER, H. ( 1 978) Ferroan calcite
replacement indicates former magnesian calcite skele­
tons. Sedimentology, 25, 843-860.
ROBERTS, H.H. & WHELAN, T. ( 1 975) Methane-derived
carbonate cements in barrier and beach sands of a
subtropical delta complex. Geochim. Cosmochim. Acta,
39, 1 085- 1 089.
RuNNELLS, D.O. ( 1 969) Diagenesis of chemical sediments
and the mixing of natural water. J. sediment. Petrol. , 39,
1 1 88- 1 20 1 .
SAIGAL, G.C. & BJ0RLYKKE, K. ( 1 987) Carbonate cements
in clastic reservoir rocks from offshore Norway- rela­
tionships between isotopic composition, textural devel­
opment and burial depth. In: Diagenesis ofSedimentary
Sequences (Ed. Marshall, J.D.), Spec. Publ. geol. Soc.
London, 36, 3 1 3-324.
SAIGAL, G.S. & WALTON, E.K. ( 1 988) On the occurrence of
displacive calcite in Lower Old Red Sandstone of
Carnoustie, Eastern Scotland. J. sediment. Petrol. , 58,
1 3 1 - 1 35 .
SAIGAL, G.S., MORAD, S . & NYSTUEN, J.P. Paleosols o f the
Snorre Field, Norwegian North Sea, part II. Origin of
carbonate nodules. In preparation.
SANDBERG, P.A. ( 1 983) An oscillating trend in Phanero­
zoic non-skeletal carbonate mineralogy. Nature, 305,
1 9-22.
ScHMIDT, V. & McDoNALD, D.A. ( 1 979) The role of
secondary porosity in the course of sandstone diagene­
sis. In: Aspects of Diagenesis (Eds Scholle, P.A. &
Schluger, P.R.), Spec. Publ. Soc. econ. Paleont. Miner.,
Tulsa, 29, 1 75-207.
SCHULTZ, J.L., BOLES, J.R. & TILTON, G.R. ( 1 989) Tracking
calcium in the San Joaquin basin, California: a stron­
tium isotopic study of carbonate cements at North
Coles Levee. Geochim. Cosmochim. Acta, 53, 1 99 1 -
1 999.
ScHULZ-ROJAHN, J.P. ( 1 993) Calcite-cemented zones in the
 Geochemical evolution of carbonate cements
Eromanga Basin: clues t o petroleum migration and
entrapment. APEA J., 1993, 63-76.
SHARP, J.M., GALLOWAY, W.E., LAND, L.S., et a/. ( 1 988)
Diagenetic processes in northwest Gulf of Mexico sedi­
ments. In: Diagenesis, II (Eds Chilingarian, G.V. &
Wolf, K.H.). pp. 43- 1 1 4. Developments in Sedimentol­
ogy, 43. Elsevier, Amsterdam.
SHEW, R.D. ( 1 9 9 1 ) Upward-shoaling sequence of mixed
siliciclastics and carbonates from the Jurassic Smack­
over Formation of Central Mississippi. In: Mixed
Carbonate-Siliciclastic Sequences (Eds Lomando, A.J.
& Harris, P.M.), 15, 1 3 5- 1 67.
SHOLKOVITZ, E. ( 1 973) Interstitial water chemistry of the
Santa Barbara Basin sediments. Geochim. Cosmochim.
Acta, 37, 2043-2073.
SIBSON, R.H. ( 1 9 8 1 ) Fluid flow accompanying faulting:
field evidence and models. In: Earthquake Prediction:
an International Review (Eds Simpson, D.W. & Rich­
ards, P.G.), American Geophysical Union, Maurice
Ewing Series, 4, 593-603.
SIBSON, R.H. ( 1 990) Conditions of fault-valve behaviour.
In: Rheology and Tectonics (Eds Knipe, J.R. & Rutter,
E.H.), Spec. Pub!. geol. Soc. London, 54, 1 5-28.
SIBSON, R.H., MooRE, J.M. & RANKIN, A.H. ( 1 975) Seismic
pumping-a hydrothermal fluid transport mechanism.
J. geo/. Soc. London, 1 3 1 , 65 3-659.
SIEGEL, D.l., CHAMBERLAIN, S.C. & DOSSERT, W.P. ( 1 987)
The isotopic and chemical evolution of mineralization
in septarian concretions: evidence for episodic paleohy­
drologic methanogenesis. Geo/. Soc. Am. Bull. , 99,
385-394.
SIMPSON, G. & H UTCHEON, I. ( 1 995) Pore-water chemistry
and diagenesis of the modern Fraser River Delta. J.
sediment. Petrol. , A65, 648-655.
SLAUGHTER, M. & HILL, R.J. ( 1 9 9 1 ) The influence of
organic matter in organogenic dolomitization. J. sedi­
ment. Petrol. , 6 1 , 296-303.
SMITH, J.T. & EHRENBERG, S.N. ( 1 989) Correlation of
carbon dioxide abundances with temperature in clastic
hydrocarbon reservoirs: relationships to inorganic
chemical equilibrium. Mar. Petrol. Geo/. , 6, 1 2 9- 1 35.
SouzA, R.S., D E Ros, L.F. & MORAD, S. ( 1 995) Dolomite
diagenesis and porosity preservation in lithic reservoirs:
Carmop6lis Member, Sergipe-Alagoas Basin, northeast­
ern Brazil. Bull. Am. Ass. petrol. Geo/. , 79, 725-748.
SP6TL, C. & BURNS, S.J. ( 1 994) Magnesite diagenesis in red
beds: a case study from the Permian of the Northern
Calcareous Alps (Tyrol, Austria). Sedimentology, 4 1 ,
543-565.
SP6TL, C. & WRIGHT, V.P. ( 1 992) Groundwater dolocretes
from the Upper Triassic of the Paris Basin, France: a
case study of an arid, continental, diagenetic facies.
Sedimentology, 39, 1 1 1 9- 1 1 36.
STRASSER, A., DAVAUD, E. & Jmoui, Y. ( 1 989) Carbonate
cements in Holocene beachrock: example from Bahiret
el Biban, southeastern Tunisia. Sediment. Geo/. , 62,
89- 1 00.
STRONG, G.E. & MILODOWSKI, A. E. ( 1 987) Aspects of the
diagenesis of the Sherwood Sandstones of the Wessex
Basin and their influence on reservoir characteristics.
In: Diagenesis of Sedimentary Sequences (Ed. Marshall,
J.D.), Spec. Pub!. geol. Soc. London, 36, 325-337.
STRONG, G.E., GILES, J.R.A. & WRIGHT, V.P. ( 1 992) A
25
Holocene calcrete from North Yorkshire, England: im­
plications for interpreting paleoclimates using calcretes.
Sedimentology, 39, 33 3-347.
STUDLICK, J.R.J., SHEW, R.D., BAYSE, G.L. & RAY, J.R.
( 1 990) A giant carbon dioxide accumulation in the
Norphlet Formation, Pisgash Anticline, Mississippi. In:
Sandstone Petroleum Reservoirs (Eds Barwis, J.H.,
McPherson, J.G. & Studlick, J.R.J.), pp. 1 8 1 -203.
Springer Verlag, Berlin.
SUCHECKI, R.K., HUBERT, J.F. & WET, C.C.B.D. ( 1 988)
Isotopic imprint of climate and hydrogeochemistry on
terrestrial strata of the Triassic-Jurassic Hartford and
Fundy Rift Basins. J. sediment. Petrol., 58, 80 1 -8 1 1 .
SuESS, E. ( 1 979) Mineral phases formed in anoxic sedi­
ments by microbial decomposition of organic matter.
Geochim. Cosmochim. Acta, 43, 3 39-3 52.
SuLLIVAN, K.B. & McBRIDE, E.F. ( 1 99 1 ) Diagenesis of
sandstones at shale contacts and diagenetic heterogene­
ity, Frio Formation, Texas. Bull. Am. Ass. petrol. Geol. ,
75, 1 2 1 - 1 38.
SULLIVAN, M.D., HASZELDINE, R.S. & FALLICK, A.E. ( 1 990)
Linear coupling of carbon and strontium isotopes in
Rotliegend Sandstone, North Sea: evidence for cross­
formational fluid flow. Geology, 18, 1 2 1 5- 1 2 1 8.
SURDAM, R.C., BOESE, S.W. & CROSSEY, L.J. ( 1 984) The
chemistry of secondary porosity. In: Clastic Diagenesis
(Eds Surdam, R.C. & McDonald, D.A.), Mem. Am. Ass.
petrol. Geol., Tulsa, 37, 1 27- 1 49.
SuRDAM, R.C., JIAO, Z.S. & MAcGowAN, D.B. ( 1 993)
Redox reactions involving hydrocarbons and mineral
oxidants: a mechanism for significant porosity enhance­
ment in sandstones. Bull. Am. Ass. petrol. Geo/. , 77,
1 509- 1 5 1 8.
SwETT, K. ( 1 974) Calcrete crusts in an Arctic permafrost
environment. Am. J. Sci., 274, 1 05 9- 1 063.
TANDON, S.K & NARAYAN, D. ( 1 9 8 1 ) Calcrete conglomer­
ate, casehardened conglomerate and constone: a com­
parative account of pedogenic and non-pedogenic
carbonates from the continental Siwalik Group, Punjab,
India. Sedimentology, 28, 3 5 3-367.
THRAIKILL, J. ( 1 968) Chemical and hydrologic factors in
the excavation of limestone caves. Geo/. Soc. Am. Bull. ,
79, 1 9-26.
TuCKER, M.E. & WRIGHT, V.P. ( 1 990) Carbonate Sedimen­
tology. Blackwell Scientific Publications, Oxford, 482
pp.
UsDOWSKJ, E. ( 1 994) Synthesis of dolomite and geochem­
ical implications. In: Dolomites (Eds Purser, B., Tucker,
M. & Zenger, D.), Spec. Pub!. int. Ass. Sediment., 2 1 ,
345-360. Blackwell Science, Oxford.
WADA, H., NIITSUMA, N., NAGASAWA, K. & OKADA, H.
( 1 982) Deep sea carbonate nodules from the Middle
America Trench area off Mexico, Deep Sea Drilling
Project Leg 66. In: Initial Reports of the Deep Sea
Drilling Project (Ed. Moore, C.E.A.), No. 66, pp. 453�
474. US Government Printing Office, Washington, DC.
WARREN, J.K. ( 1 990) Sedimentology and mineralogy of
dolomitic Coorong Lakes, south Australia. J. sediment.
Petrol. , 60, 843-858.
WATSON, R.S., TREWIN, N. & FALLICK, A. ( 1 995) The
formation of carbonate cements in the Forth and Bal­
moral Fields, northern North Sea: a case for biodegra­
dation, carbonate cementation and oil leakage during
26 S. Morad
early burial. In: Characterization ofDeep Marine Clastic
Systems (Eds Hartley, A.J. & Prosser, D.J.), Spec. Pub!.
geol. Soc. London, 94, 1 77-200.
WATTS, N.L. ( 1 980) Quaternary pedogenic calcretes from
the Kalahari (southern Africa): mineralogy, genesis and
diagenesis. Sedimentology, 27, 66 1 -686.
WHITCAR, M.J., FABER, E. & SCHOELL, M. ( 1 986) Biogenic
methane formation in marine and freshwater environ­
ments: C02 reduction vs. acetate fermentation-isotopic
evidence. Geochim. Cosmochim. Acta, 50, 693-709.
WIGLEY, T.M.L. & PLUMMER, L.N. ( 1 976) Mixing of
carbonate waters. Geochim. Cosmochim. Acta, 40, 989-
995.
WILKJNSON, B.H., OWEN, R.M. & CAROLL, A.R. ( 1 985)
Submarine hydrothermal weathering, global eustacy,
and carbonate polymorphism in Phanerozoic marine
oolites. J. sediment. Petrol. , 55, 1 7 1 - 1 83.
WILKJNSON, M. ( 1 99 1 ) The concretions of the Bearreraig
Sandstone Formation: geometry and geochemistry. Sedi­
mentology, 38, 899-9 1 2.
WooD, J.R. ( 1 986) Thermal mass transfer in systems
containing quartz and calcite. In: Roles of Organic Mat­
ter in Sediment Diagenesis (Ed. Gautier, D.L.), Spec.
Pub!. Soc. econ. Paleont. Miner., Tulsa, 38, 1 69- 1 80.
WooD, J.R. & HEWETT, T.A. ( 1 984) Reservoir diagenesis
and convective fluid flow. In: Clastic Diagenesis (Eds
McDonald, D.A. & Surdam, R.C.), Mem. Am. Ass.
petrol. Geol., Tulsa, 37 , 99- 1 1 0.
ZACHMANN, D.W. ( 1 989) Mg-carbonate deposits in fresh­
water environment. In: Magnesite: Geology, Mineral­
ogy, Geochemistry and Formation of Mg-Carbonates,
(Ed. Meller, P.), Monograph Series on Mineral De­
posits, 28, 6 1 -94.
Spec. Pubis int. Ass. Sediment. ( 1 998) 26, 27-5 1

Origin and spatial distribution of early vadose and

phreatic calcite cements in the Zia Formation,
Albuquerque Basin, New Mexico, USA

J.R. B E C K N E R a n d P . S . M OZLEY
Department of Earth and Environmental Science, New Mexico Institute of Mining
and Technology, Socorro, NM 87801, USA

A BSTRACT

The Miocene Zia Formation consists of sands and muds depos ted i in fluvial, aeolian and playa lake
environments. Although much of the formation is poorly consolidated, resistant zones of calcite
cementation are common. These range in size from isolated nodules to tabular cemented zones several
metres thick t hat extend for over 2 km laterally .The calcite cemented zones are h ghly i complex,
exhibiting a wide range of macroscop ci and m croscopic
i textures and geometries .Af er t cons dering
i a
combination of microscopic, macroscopic and geochemical characteristics, we have inferred the
environment of precipitation (i e.. pedogenic, vadose non-pedogenic, phreat ci ) of the pr ncipal
i types of
cementat oni Nodules
. and rhizocretions with micrit ci fabrics and alveolar structures are inferred to be
vadose carbonates. Ovoid or elongate concretions, characterized by blocky spar cements and
preservation of primary sedimentary structures, are inferred to be phreatic carbonates .Most cemented
units in the Zia Formation reflect characteristics of both phreatic and vadose zone cementation (e g..
3
preservation of sedimentary structures plus rhizocret ons i and alveolar microtextures ). 81 C values for
18
vadose cement tend to be heavier and 8 0 values tend to be similar or slightly lighter than phreatic
cements .813 C and 818 0 values for units with mixed features tend to have intermediate values. Most
cementation types that exhibit a m xture
i of features may reflect past fluctuations of the water table,
where vadose cements were moved into the phreatic zone. V adose zone cementation occurred
princ pally
i in association with soil development, whereas phreatic zone cementation occurred
preferent ally
i in zones of high primary permeability .In many cases early vadose cements provided
nucleation sites for later phreatic cementation .Tabular units in the Zia Formation are of en t laterally
extensive, decreasing potential reservoir/a q u ifer quality by forming significant barriers to vertical f uid
l
flow .These barriers could result in compartmentalizat on i of the reservoir/a q u ifer, and extensively
reduce production if wells were screened on only one side of a cemented layer .

INTRODUCTION

Understanding fluid flow in aquifers and hydrocar­
bon reservoirs requires an understanding of hetero­
geneities in porosity and permeability in the
material. A number of workers have examined the
influence of primary depositional controls on aqui­
fer heterogeneity (e.g. Weber, 1982; Anderson,
1989, 1990; Davis et a!., 1993). To date, however,
few studies have examined the influence of dia­
genetic alterations on porosity and permeability
heterogeneities. To predict the subsurface distribu­
tion of diagenetic alterations that influence flow, it
is necessary to understand the diagenetic processes
involved and to determine the controls on the
spatial distribution of diagenetic alterations. In this
paper we examine controls on the origin and spatial
distribution of early calcite cements in the Miocene
Zia Formation of New Mexico, in which calcite­
cemented low-permeability zones can extend for
several kilometres laterally.
Unlike marine sediments, where early diagenesis
typically occurs entirely within the phreatic (satu­
rated) zone, early diagenetic alterations in terres­
trial sediments occur in both vadose (unsaturated)
and phreatic zones. Furthermore, in terrestrial

Carbonate Cementation in Sandstones Edited by Sadoon Morad 27

© 1998 The International Association of Sedimentologists
ISBN: 978-0-632-04777-2
28 J.R. Beckner and P.S. Mozley
sediments significant alterations can occur during
pedogenesis. Thus, a fundamental problem in any
study of early terrestrial diagenesis is identifying
vadose versus phreatic alterations. Although there
are many studies that have investigated pedogenic
carbonate formation, few have focused on non­
pedogenic cementation. Fewer still have addressed
the problem of differentiating among different types
of cements. In the Zia Formation we have been able
to infer the environments of cement formation and
the relationships between these environments and
the subsequent spatial distribution of cementation.
Our principal conclusion is that cementation in the
phreatic zone occurred preferentially in zones of
high primary permeability, whereas vadose cemen­
tation occurred principally in association with soil
development. Furthermore, pedogenic carbonates
apparently served as nucleation sites for later
phreatic cementation, leading to complex zones of
mixed pedogenic and phreatic cements.
TERMINOLOGY
Because the terminology for early carbonate ce­
ments is complex and somewhat ambiguous, it is
necessary to define the terms used in this study.
Carbonate cements are subdivided into three prin­
cipal types:
1 Pedogenic carbonate is carbonate that precipi­
tated in an active soil (i.e. the precipitation was
related to pedogenic processes such as weathering,
evapotranspiration, biological activity, etc.). Most
detailed studies of early (i.e. before significant
burial diagenesis) calcite cementation in semi-arid
and arid settings have been on pedogenic carbon­
ates (Gile et a!., 1 966; Reeves, 1976; Esteban &
Klappa, 1 983; Klappa, 1983; Rabenhorst et a!.,
1984; Machette, 1985; Monger et a!., 199 1 ; Mack et
a!., 1993).
2 Vadose non-pedogenic carbonate is carbonate that
precipitated in the vadose zone but is not related to
pedogenesis. Vadose non-pedogenic carbonates
have been reported in the literature (e.g. Carlisle,
1983; Goudie, 1 983; Semeniuk & Searle, 1 985;
Wright & Tucker, 1 991 ) , although few criteria were
described to distinguish them from pedogenic ce­
ments.
3 Phreatic carbonate is carbonate that precipitated
by non-pedogenic processes in the phreatic zone.
Terrestrial phreatic carbonates have been described
by many workers (Mann & Horwitz, 1 979; Arakel &
McConchie, 1 982; Carlisle, 1 983; Netterberg,
1969; Arakel et a!., 1 989; Wright & Tucker, 199 1 ;
Spot! & Wright, 1 992; Burns & Matter, 1995;
Mozley & Davis, 1 996).
The terms calcrete and caliche are frequently used
to describe some of the cement types mentioned
above. They are most often used to describe a
variety of cryptocrystalline calcium carbonate de­
posits resulting from pedogenic processes, that
eventually forms indurated masses (Gile et a!.,
1966; Read, 1974; Reeves, 1976; Semeniuk &
Meager, 198 1 ; Esteban & Klappa, 1983; Klappa,
1 9 83; Netterberg & Caiger, 1 983; Machette, 1985;
Milnes, 1 992). The term calcrete has also been used
to describe a wide variety of calcium carbonate
deposits resulting from groundwater processes
(Netterberg, 1 969; Mann & Horwitz, 1 979; Seme­
niuk & Meager 1 981; Arakel & McConchie, 1982;
Carlisle, 1 983; Semeniuk & Searle, 1985; Jacobson
et a!., 1 98 8; Arakel et a!., 1 989; Wright & Tucker,
1 99 1 ; Spot! & Wright, 1 992). These terms will be
avoided, as they have been used in a variety of ways
by different workers.
GEOLOGICAL SETTING
The Zia Formation is the basal rift filling unit of the
Santa Fe Group in the Albuquerque basin, part of
the 1000 km long Rio Grande rift of Colorado and
New Mexico (Lozinsky, 1994). The 10-2 1 Ma Zia
Formation is exposed in a 55 km long arc extending
from the Rio Puerco in the west to 24 km north of
Albuquerque, New Mexico (Gawne, 198 1 ) . The
upper part of the Zia Formation ( 1 6- 1 0 Ma; Ted­
ford, 1982) was deposited during the period of most
active rifting (Chapin & Cather, 1 994). The study
site is on the western margin of the Albuquerque
Basin, about 20 km from Albuquerque, on the King
Ranch (Fig. I ). The Zia Formation in this area is
typified by exposed, resistant, well cemented hori­
zons bounding poorly consolidated sediments. It can
be divided into sand-dominated, aeolian (Piedra
Parada) and fluvial-aeolian (Chamisa Mesa) mem­
bers; a mud-dominated, flu vial member (Canada
Pillares Member); and the sand-dominated aeolian­
fluvial member (Unnamed Member; Gawne, 198 1 ;
Tedford, 1982) (Figs 2 and 3). The lower contact of
the Zia Formation is unconformable with the
Eocene Galisteo Formation and the Crevasse Can­
yon Formation of the Cretaceous Mesaverde Group
(Gawne, 1 98 1 ; Tedford, 1982). The upper contact
 Calcite cements in the Zia Formation 29

1070 1060

360

New Mexico

35 0
0
Albuquerque
Basin

Explanation

D Albuquerque Basi�

'-- Basin boundary

� Fault-hachures on
r'-downthrown side;
N dashed where
inferred or buried.

Fig. 1 . Map of the Albu q

uer q
ue 0 20mi
Basin showing the King Ranch study 1..__..
1 "'
'--11
0 20km
area. Modified from Lozinsky
( 19 94).

is the Sand Hill fault, a major normal fault that
offsets the Zia Formation and units of the Upper
Santa Fe Group by about 600 m (Mozley & Good­
win, 1995a (Fig. 3)).
Facies associations (Miall, 1990) were defined
from a detailed analysis of lithofacies in the study
area (Table I ; Fig. 4). The classification used for
fluvial sediments is from Miall ( 1990) and Davis et
a/. ( 199 3). The terms facies and facies/lithofacies
association are also used to define aeolian sedi­
ments and sedimentary characteristics (Kocurek,
1981; Kocurek & Dott, 1 9 81; Porter, 1987; Chan,
1989). The symbols used for flu vial and aeolian
facies associations (e.g. CH, OF, EC, ES in Table I
and Fig. 4) were developed for this study. Palaeosol
formation is a function of surface exposure time
and landscape stability. Palaeosols are important to
an understanding of depositional environments and
ancient flood basin accretion rates (Leeder, 1975;
Allen, 1 9 86; Atkinson, 1986; Kraus & Bown, 1986;
Davies et al., 1993). Because of this they will be
considered separately from crevasse splay and over­
bank deposits.
METHODS
Sections of the Zia Formation were measured along
four transects to examine lateral and vertical varia­
tions in lithology and cementation (Fig. 3). Key
30 J.R. Beckner and P.S. Mozley

Lithology
Sample Locations
72895-6 72895-2 72895-3
72895-4 72895-5 72896-6a
72895-6b 72895-7
6295-4
81995-20
81995-19
81994-18
81994-17
81994-16
8594-16
72194-3
81994-15
81994-14 8594-14
8594-13 81994-9
81994-8
81994-7
81994-6
81994-5
81994-4
72195-2 EXPLANT!ON
81994-3 8594-12
81994-2 Muds
8594-11 81994-1
t===���===��� 72195-1
122394-1
122394-2
�
l:::::::l
Silty Sand
122394-3
122394-4 r:.:7:l
122394-5
122394-6 L::d
122394-7 Sand
122394-8 �
122394-9 1395-1
122394-10 1395-2 �
. . . . . . .

1395-3
. .
Unconformable
. . . . . . .
· . · . · . · . · . · . · . · . · . · . · . · . · . · . · . · . · . ·

.
. . . . . . . .
Contact
.
• • 0 • • 0 0 • • • • • • • • • •
. . . . .
• • • • • • • • • • 0 • • • 0 • •
. . . . 1395-4
. .
.
. . . .
1395-5 Faulted Contact
1395-6
1395•7

1111!111111��
1395-8
1395-9
1395-10
1395-11 8594-1 0
81895-1
r,=,=,.=,=,.=�=,=,-=,=;,=,.=,=,.=�=,=,-=,=,l\\\\' m�s�2
8594-8
8394-7 8594-7
81895-3
. . . . .
• • • • • 0 0 • • 0 0 • • • 0 • •

8594-6
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.8394-6
.
.
.
.
.
.
.
.
.
. . . .
. . . . . .
. .
.

0 • • • 0 0 8594-5
• • • • • • • • • • •
. . . . . .
8594-4A.
. .
.
.
.
.
.
.
.
.
.
.
.
.
.8594-4
.
.
.
.
.
.
.
. . . . . . .

8394-5
· . · . · . · . · . · . · . · . · . · . · . · . · . · . · . · . · .
. . . . . . . .
8594-2
. . . .

• • • • • • 8394-4
• • • • • • 0 .
· . · . · . · . · . · .
8394-3 . .
· . · . · . · . ·
.
. .

.
.
.
.
.
.
.
.
.
.
.
.
.8594-1A
.
.
. . . . . . . . . .
. . . . . . .

Fig. 2. Generalized strat graphical

i column of the Zia F ormation showing ages, lithologies, depositional en vironments
and locations of the samples. This stratigraphical column was constr u
cted from the fo u r detailed col u
m ns whose
locations are shown in Fig .3. Terminology and ages from Tedford ( 1 982) and L ozins ky ( 1 988) .

beds were traced laterally throughout the study area the four measured sections for petrographical and
to evaluate the continuity of cementation and vari­ geochemical analysis (Fig. 2). Laterally continuous
ations in bedding thickness and cement morpholo­ units were sampled in more than one area to
gies. Cemented units were classified by outcrop examine variations in petrographic and geochemi­
morphology, surface textures and sedimentary cal characteristics. Thin sections were made from
structures. Seventy-six samples were collected along most of the samples, which were impregnated with
 Calcite cements in the Zia Formation
000000000000000000000000000000
000000000000000000000000000000
Parada
Member
11;:11 Galisteo
liiil Formation
� Crevasse
� Canyon
Formation
� Fault
Fig. 3. Geological m
ap of the study area showing locations of stratigraphical colu m
blue-dyed epoxy before thin section preparation to
identify porosity. These thin sections were analysed
for authigenic textures and mineralogy using a
standard petrographic microscope under plain light,
crossed-polarized light and cathodoluminescence.
Mean grain size, sorting and roundness data were
collected from outcrops and thin sections using
visual comparators (grain size: Amstrat Inc.; sort­
ing: Pettijohn et al., 1987; roundness: Powers,
1953). The cathodoluminescence was performed on
a microscope equipped with a MAAS/Nuclide
model ELM-3 Luminoscope. A Chittick apparatus
(modified from Dreimanis, 1962) was used to de­
termine the total percentage of calcite, and to test
for the presence of other carbonates. The analytical
precision based on I 0 samples is better than 3%. On
selected samples a JEOL-733 Superprobe, equipped
31
COLUMN4
ns. Modified fro mGawne ( 1 98 1 ) .
with a high-resolution back-scattered electron de­
tector, X-ray mapping features and image analysis
software, was used to determine elemental compo­
sition and zoning in cements. Sample operating
conditions were 20 nA sample current and 1-10 Jlm
beam diameter. Carbonate standards were used and
sample totals are I 00 ± 2% for all values. Finally, a
Finnigan MAT Delta E isotope ratio mass spec­
trometer was used to analyse carbon and oxygen
isotope values for each sample. Carbon and oxygen
values were measured from C02 gas liberated from
whole rock samples using 100% phosphoric acid.
Data are reported in parts per million (o/oo), relative
to PDB for oxygen and carbon. The analytical preci­
sion, determined from six standards, is better than
0.1 o/oo for both carbon and oxygen.
Table 1. S
u mmary o flithological in ormation
f or
f aci
f se associations; t reminology modifi ed rom
f Miall ( 1 990) and D
avis et a/. ( 1 993) w
N

Faci se association Litho aci

f se pr se n
et G o
e m tery Grain siz /esorting C m
e n
e tation typ se

CH Trough cross-b d ed de sand ( S t) Tabular to l n

e ticular Fin eto coars ,e Typ el and typ e3
Chann le+ l v
e ee planar laminat d e sand ( S p) 0.2-3 m thick mod reat ley sort d
e (phr aetic ) tabular units
low angl ecross b- d ed de sand ( SI) 1 0 m to >2 km in lat real sand/sandston e
horizontally laminat d e sand ( S h) ext n
et
rippl ecross-laminat d e sand ( S r)
massiv esand ( S m)
massiv ,e crud ley b ded d e silts and muds (Fm )
fin ley laminat de to rippl de silts and muds (F I)
laminat ed silt, sand, and clay (Fsc )
cs Trough cross-b ded de sand ( S
t) Tabular, massiv ,e Poorly sort d
e sands and Thin sandston esh ee
ts
Cr v
e ass esplay d p
e osits planar laminat de sand ( Sp) l n
e ticular, to thin silty sands ar eusually w lel
low angl ecross-b d
ed de sand ( Sl) w de g -eshap de sands, c m
e net d
e
horizontally laminat ed sand ( S
h) 0.2-4 m thick
rippl ecross-laminat d
e sand ( Sr) 0.5 m to 0.5 km in �
massiv esand ( Sm ) lat real ext n
et (l:l
t:l:l
OF Massiv e, crud ley b ded de silts and muds (Fm ) Tabular to thin and Muds and silts Poorly c me n
et de, �
?;-
Ov rebank fin se fin ley laminat de to rippl d
e silts and muds (Fl ) lobat e isolat d
e nodul se, platy, �
laminat ed silt, sand, and clay (Fsc ) 0. 1 - 3 m thick and rod concr teions ""
"'
silts and clays w/rhizocr teions (Fr ) 0.5 m to 0.2 km in 1=:>
�
lat real ext n
et l=:l..
p Pala o e sols on sand (Ps ) Tabular to Muds, silts, v ery fin eto Nodular, platy and rod :-tl
Yl
Pala o
esol horizons pala oe sols on silts and clays (Psc ) discontinuous and m de ium silty and clay ye concr teions, and typ e2
silt and clays w/rhizocr teions (Fr ) patchy sand/sandston se and typ e3 (vados e)
massiv esand w/rhizocr teions ( Smr ) 0. 1 - l m thick tabular units
�
N
0. 1 t o > 1 km lat real �
ext net
EC Trough cross-b ded de sand ( S t e) Tabular, l enticular and Fin eto low er coars ,e Scatt re d
e ovoid to
Cross-stratifi d
e a o
e lian planar laminat de sand ( Sp e) w de g -eshap d
e mod reat ley to w lel leongat econcr teions and
dun ebodi se low angl ecross-b d
ed de sand ( S l e) 1 -3 m thick sort d
e sand/sandston e small typ e1 and typ e3
horizontally laminat de sand ( S h e) > 1 km lat real ext n
et phr a etic tabular units
rippl ecross-laminat d
e sand ( Sr e)
massiv esand ( Sm e)
E S L ow angl ecross-b d
ed de sand ( S
l e) Tabular Fin eto m d e ium Coars re lay res of te
n
A o
e lian sandsh ee
t horizontally laminat d
e sand ( Sh e) 1 -2 m thick mod erat ley to poorly orm
f w lel c em n et d
e typ e
d p
e osits rippl ecross-laminat d
e sand ( S
r e) > 1 km lat r
eal ex
t n
et sort d
e sand/sandston e 1 and typ e3 (phr a etic )
ID L ow angl ecross-b d
ed de sand ( S
l e) Tabular, l ne ticular and All siz se, g n
e really S
m all typ e1 and typ e3
I
nt erdun ed p
e osits horizontally laminat d
e sand ( Sh e) discontinuous poorly sort d e or bimodal (phr aetic ) tabular units
rippl ecross-laminat d
e sand ( S
r e) 0. 1 -0.5 m thick
massiv esand ( SM ) 1 0 m to > 1 km lat real
ex
t net
 Calcite cements in the Zia Formation 33

Sand Dominated Fluvial

Environment

r:7':77l Crevasse Splay

l:i:::2d Deposits (CS)

D
Overbank
Fines (OF)

• Paleosol
Horizon (P)

D
Cross-stratified Eolian
Dune Deposits (EC)

E3 Eolian Sheetsand
� Deposits (ES)
� Interdune
� Deposits (ID)

Fig. 4. Sch me atic d p

e ositional
g o
e m ter y o fth eZia Formation.
Faci se ass ociations ar e rom
f Tabl eI.

SANDSTONE PETROGRA PHY (Piedra Parada Member) to lithic arkoses (Chamisa

Most of the Zia Formation in the King Ranch area
can be classified as lithic arkoses (Fig. 5). The Zia
Formation can be further subdivided into two
distinct domains on the QFL diagram: one contains
the lower Zia Formation (Piedra Parada, Chamisa
Mesa and Canada Pillares Members; the other
contains the Unnamed Member. The lower Zia
Formation changes from a feldspathic litharenite
Mesa, Canada Pillares Members). The Unnamed
Member exhibits scattered compositions, but is
differentiated from the lower members by greater
amounts of feldspar (Fig. 5).
Volcanic rock fragments of intermediate compo­
sition are generally the most abundant lithic frag­
ments, averaging 70-90% of all rock fragments
(Fig. 5). Chert is the most common sedimentary
rock fragment, although some units contain abun-
34 J.R. Beckner and P.S. Mozley
Su
b a ko
r se
• Unna medM em b er
IJ Ca n
a d a Pilla resM em b er
• Cha m i as M esa M em b er
6 Pi edra Pa ar da M ember
dant detrital carbonate (Fig. 5). These carbonate
fragments resemble pedogenic carbonates and may
be caused by erosion of the underlying pedogenic
units.
Most volcanic lithic fragments are fresh and well
rounded; however, chemical alteration has removed
unstable phenocrysts such as hornblende and plagio­
clase from some volcanic grains, leaving euhedral
voids. More irregular voids indicate dissolution of
aphanitic/glassy groundmasses. Potassium feldspars
vary from fresh to deeply altered to clays. Most of the
plagioclase is unaltered, and dissolution along cleav­
age planes is more common than alteration to calcite
or clays.
TYPES OF CALCITE CEMENTATION
Calcite cementation in the Zia Formation is com­
plex, exhibiting a wide range of macroscopic and
microscopic morphologies. Four principal types of
isolated concretions, and three principal types of
laterally extensive tabular units, were identified. A
summary and description of facies associations,
lithofacies types, lithologic data and cementation
types is given in Table I . Descriptive data and
interpretations for each cementation type are pro­
vided in Table 2. Details of spatial distribution and
Iitholacies/Iithologic associations of these cementa­
tion types are shown in Figs 6 and 7.
Fig. 5. T renary plot o f
s
asnd/ and s
ton eco m po istion by
mem b re o f th eZia F or m ation .
S
a mpl ethat plot sa sa lithar enite
contain sa larg ea m ount o fd terital
carbonat e. Cla ss ification rof mF ol k
( 1 974 ) .
Concretions
Nodules
Nodules can be subdivided into two types. The first
consists of small (0. 1 -5 em diameter) subspherical
to irregular forms (Fig. 8A) and is common in
reddened clays and clay-rich silty sands in overbank
fine (OF), and palaeosol (P) horizons (Table I ;
Fig. 6). Some of this first type of nodule exhibit two
stages of concentric zonation, distinguished by a
colour change from grey or greenish grey in the
middle to pink on the outside. Dense micrite forms
the usual matrix, and crystallaria (with some cir­
cumgranular forms) are common (Fig. 8B). The
second type of nodule is roughly the same size and
shape, but is characterized by oval grooved and
tubular surface pitting (Fig. 8C). This type is more
common in the silts and silty sands (crevasse splay
(CS), overbank fines (OF), and palaeosols (P)) of the
upper Unnamed Member. It has a micritic matrix,
circumgranular cracks, micrite-spar, and some alve­
olar textures as well (Fig. 8D). A micrite-spar mi­
crotexture is where grains or groups of grains are
coated with micritic cements, and the areas in
between are filled with spar (16-50 �m diameter).
Ovoid and elongate concretions
These concretions range from small ( 1 -4 em diam-
Table 2. Summ a
ry of d secri p
t iv ed a
t an
a d int re p
r te a
t ions of c m
e n
et a
tion ty pe
s in th eZi aForm a
tion

Environm n et of
Cem n
e t taion ty pe Host lithology Outcro pmor phology Surf a
c et x
e tur se Microt x
etur se p
r cei p
it taion

Nodul a r Cl a
ys , cl y
a-rich silty s a
nd 0. 1 -5 em di am te re ovoid Smooth to p i tt d
e , tub ed Micritic f b a ric , m n e iscus V a
dos e
concr teions to irr geul a
r sh ape
s n
a d groov d
e c em ents , circumgr a nul ar
cr a
cking , cryst lal rai a,
a
lv o
el art x e tur se , gr ain
dissolution
Ovoid to leong ta e Fin eto co a
rs es a
nd l-4 e m di am te re ovoid to Smooth to w a
rty Poikiloto p
ic to blocky s pa
r Phr ea
tic
concr teions > l 0 m leong a
t esh ape s
Pl a
ty concr teions Cl y
as , cl a
y-rich silty s n
ad 5-50 e m caross p l ta se th ta Smooth to p i tt d
e , tub d
e Micritic f a bric , m ne iscus V a
dos e
s eem to follow r elict n
a d groov d
e c m
e n e ts , circumgr a nul ar Q
t x
e tur se cr caking , lav o el a
r t xetur es , �
gr ai n dissolution �
'"'
"'
Rod concr teions Cl a
ys , silty s and , n
ad s a
nd 0. 1 -5 em di a m te re , Mostly smooth , but Micritic f abric , V d
a os e �
"'
3-50 e m long ; singl eor som teim se pi tt d
e , tub d
e circumgr a nul ar cr a cking , �
br a nching , th ni downw ards n
a d groov d
e a
lv o
el a
r t xetur se , gr a
in �
dissolution ;:;·
Ty pel t a
bul a
r Fin eto co a
rs es and G n e re laly > l 0 m l ta re a l Smooth to w a
rty surf a
c e Poikiloto p
i c to blocky s pa
r Phr ea
t ic s.
"'
c m
e n
et de unit xet ne t , with p r se rev d
e N
s de im n e t ar
y structur se ; iS'
sh a r plow re , n ad g n e re ally
sh a r pu ppe r bound rai se �
....
�
Ty pe2 t b
a ul a
r V rey fin eto m de ium- G n e re laly l 0-200 m l ta re la Smooth to p i tt d
e , tub d
e Micritic f a bric , m ne iscus V a
dos e �
c m
e n
et de unit gr a
in de cl y
a y
e to silty s n
ad xet n
e t ; m sasiv e, mottl d e, n
a d groov d
e c me ne ts , circumgr a nul ar (3
w a vy- p l tay , br ceci a
t de, cr acking , r a di al s par, �
t eepee, l a m in ar f eat ur se ; lav o
el ar n ad f ne setr al
sh ar pu ppe r a n d diffus e t xetur se , gr a
i n dissolution
low er bound rai se
Ty pe3 ( ph r ea
t ic ) Fin eto m d
e ium-gr a
in d
e G n e re laly > l 0 m l a t er la Smooth to w a rty , Blocky s pa r na d s parry Phr ea
tic >>
t a
bul a
r c em ne t ed s a
nd xet ent , with p r se rev de som teim se v rey irr geul ra flo a
ting gr ain t x
e tur se v a
dos e
unit s de im n et ary structur se p l us with p i ts , tub se , a
nd
rod sh ape s groov se
Ty pe3 (v a dos )e V rey fin eto m d e ium- M sasiv e, mottl d
e , with Smooth to p i tt d
e , tub ed Micritic f a bric , m ne iscus V ados e>>
t b
a ul a
r c em ent d
e gr a
in de cl a
y ye to silty s a
nd nodul se , rods , a
nd p l a
t se ; a
n d groov d
e c em ne ts, circumgr a nul ar p
h r ea
t ic
unit som es de im n et a
ry cr caking , lav oel ar nad
structur se pr se rev de micrit -es pa r t xetur se

w
U>
36 J.R. Beckner and P.S. Mozley

z
0
(/) GRAIN SIZE 0
w !z z
a:
::;"' I="' ��=
"'
EXPLANATION
1-
w
::; ���d
1"1iflllclay
u
0
"' ��
<

LITHOLOGY
sand
--------
j< J -------

- - ....-
106 --------
--- -
-
------- --
�:...
----
OF)
- _
- - -
: silty sand
------ -
I:J - - ---
-------
-------

:: shaly sand
-
Scattered ovoid concretions
t::: �
k:::<:d sandy silt
104
. . .. .
..
: . ··:.··:. ··:.
..
........ . :.
....
...
:··...
:··... ··: .
....
:·· .. .. ··:. ··
··:.. CS) V in lenticular sandbodies, in
silty sand.
:.··: ··:. ··:..:·· . :·· :. ·· ··:.··. ··
·· :. -
::: ::::::: silt
: ··:··: ·:· :··": :·· :·· · :· :··"
Thinly bedded type-3 p ( hreatic)
J J
unit interbedded with uncemented
silty mud
!'
t:::::1
muds
102
..
.....
.. . ..
.
:·: .�
. .. .....
.. . . .... .
...:.:r-
. .. .
· ·:· · � I CH silty sand and clay, giving outcrop
a platy appearance.

.
�
. .·
�:::::::�
... ...
��
�: 1\ Thinly bedded type ! units
·... · .·. · . .
..
· · ··
.
. · · · · ...
· · ··
\interbedded with uncemented
-

t�ii
1
silty sand and clay lenses.
100
-F)
(O
r--- Some detrital carbonate.
PHYSICAL STRUCTURES
t-
'<.J trough cross-stratificaton
(CS/

llt�
/// high angle planar bedding OF)
Scattered nodules and rod
=low angle tabular bedding
98
concretions in a mudstone
-=..planar cross stratification r---- r- matrix, interbedded with thin
a.,, ·rip-up clasts
type-3 (vadose and phreatic)
oooornicritic nodules tabular lenses.
OF)
ICHNOFOSSILS
.u, rhizocretions
burrows
96

:������:���=�i�: r--- Thinly bedded type-3(phreatic)

tabular unit.Coarser laminae
.r:
------ooco.
C
( S/
OF)
1/ are better cemented.giving
unit a platy appearance.
94 This unit has detrital

DEGREE OF CEMENTATION tJ����H����������:r; carbonate.

: =:= ��·
r----
CH Massive to brecciated and
• very well cemented

• well cemented 92
.
.

.·..
:
�.u:.:.:.: � �
.
.. u·.
·

.
·.:. .
·.· .
·....
: ::
- t7'P\�
�
mottled type- 2 tabular unit with
abundant branching rod
concretions.
moderately cemented EC)
· · . ...
· . ·. . ... ·. ·
• ...
......///'.',',
..

poorly cemented (I ) Scattered .2 to .3m concretions

· · · · · ·
D D
trace to no cement t-- along crossbedding planes.
D ....
......
..
... . .
....
..'<.J....
.
EC) Some are elongate, othere are
....
..
.. .
...... .. .. . irregular type-! tabular units.
90
..

(I )

88
���������������i I JI I..._�___
.
·· ·· ·· ·· ·· ·· · ·· ·· ·· ·
· · · ·· · · · · · ·
I
D
(ES)
Moderately cemented type-!
tabular units. These units are
coarser and better sorted than
those above and below them.
· · · · · · ·

Fig. 6. Str a
tigr ap
h i ca
l o
c lumn showing d te a
i ls of th er le taionshi p
s b tew ee
n gr a
i n siz e, lithology , sorting a
nd d g er eeof
ce
m n e t taion in s d
e im n e ts from th eC h a m is aM se a a n d Can ad aPill a r se M m e b res. Not e o c rr le taion b tew ee n oc ars re
n
a d b tet re sort d
e h c an n le (C H) a sso ic taions n a d good ce m ne t taion. Nodul a r, pl a
ty n a d rod-sh ape d oc n rc teions ra e
a
sso ic ta de with rc ve as s es p l y
a (C S ) , pal oe sol (P ), int redun ef aci se (!D) n a d ov reb na k fin e(OF) s d e im ne ts. S catt re d
e
ovoid to leong a t eoc n rc teions domin a t ein th e rcoss-str taifi de ae oli n
a f aci se (EC). Not eth a t th e o c ras re portions o f
ae
oli na s n a dsh ee
ts (ES ) a r epr fe re n
e ti laly ce m n et de.

eter) ovoid and oblate forms, to elongate cemented mented masses (generally <I 0 m long) in which the
masses (Fig. 9A-C). Ovoid concretions are found long axis of the concretion is oriented subparallel to
isolated or coalesced in botyroidal masses (0.2-1 m bedding. Elongate concretions occur alone or in
diameter). Elongate concretions are elongate ce- irregularly shaped groups with similar orientations.
 Calcite cements in the Zia Formation 37

Stratigraphic E Lithology
e
and SPATIAL DISTRIBUTION OF CEMENTED ZONES
Units � �;;���:�t �----�T�A�B�U�L�A�R�U�N7.I�T�S�--.---�C�O�N�C�RE�T�I�O�N�S.-�
340 thin units of elongate concretions and
nodules and rhizocretions
320
scattered nodules and rhizocretions

JOO �
· Sand ····
Dominate nd�·§
. Fluvial ·.-:·
�
'-' ·· =================== � sc�tt�er�e�d�e�lo�n�
a� on
co�n�c�re�ti�
ga�t�e�
scattered nodules and rhizocretions
s
� � t �
280
sea ere e on a e concretiOns
thin (.1-.3m thick) type-3 vadose Type-I tabular
260 tabular units with scattered nodules units (phreatic)
and rhizocretions
thin (.1-.4m thick) type-3 vadose and Type-3 tabular
phreatic tabular units units {phreatic)
PZZZZZZZJ
220
Sand
scattered 2-10m long lenticular Type-3 tabular
type-1 and type-3 phreatic tabular units (vadose)
200
. Fluvtal :.
' ..
:::: ::::;;.::;;·:::;;::;m: units
180
�E�Ii·a��
� F I�vl al: = �
·.:.. �����������;:;�� ���� �
····= t
scattered nodules, rhizocretions and
t� pe::::_·2�a�
y� nd
� typ'.!
':'::e-3:.'
. ta ---1
la�r_'_u'."n�its
.�b�u'." ·� Type-2 tabular
units (vadose)
160 1 =====�---------- ., scattered ovoid, elongate and small
(1-lOm) tabular units (type-1 and
Hili

·:Eolian.· f=== type-3 (phreatic))

140

CANADA scattered ovoid and (3-5 em thick)

PILLARES
120
§ Mud � cemented sand lenses and scattered
nodules a.:::.nd rh=izoc.:.:re :.:i:o.ns
: t.: --
MEMBER -:.: n omina t ed·>·
- Fluvial �F=
== =========== ===== =\- ;.:. :.: :.:::.: :::.::
- ::.C:..: ": :' : "�-:- ---j
�-::=-�7- - �-�-�-�$
- ;;;:;:;;:
CHA�'IISA
l\·1ESA
100 scattered thin (3-10em thick)
MEMBER , ,,•Horo , ---:;:;;::;:;;:::;:�;:" cementedsand lenses, and scattered
�u,.,,,.,_,.,,.,,,..
� .,., 1
80 �:
..,..,.,...

"'""'""""
'"'"""
• ••'•••
M'"'"'' .. nodules and rhizocretions
scatttered ovoid and elongate
60
concretions following crossbedding.
·Eolian:·
common 3-5m tabular bodies (type-1
40 and type-3 (phreatic)).
rare 2-3 m tabular bodies (type 1 and
20 type 3(phreatic))
�FI; ;-;T,;J7:-c:J +----------------'1 common 1.5 to 5 m lentici.Jlar units
f : Fluvial·: km--- ------. 1 (typc-3(phreatic))

Fig. 7. Sp a
ti la distribution of c eme
nt taion t
yp se in th eZi aFor ma
t ion. Not eth e a
ssoci a
t ion of l ta re laly xet n
e siv e
phr eat ic units with s n
a d lithologi se. La
t re laly xet n
e siv ev a
dos eunits a
r enot sasoci a
t d
e with n
ay p a
rticul a
rd p
e osition a
l
n
e viron me nt.

Most ovoid to elongate concretions are covered
with millimetre-sized wart-like structures (hereafter
referred to as warts). Where it is possible to tell,
ovoid to elongate concretions seem to form in
coarser units with better sorting than beds either
above or below them. This relationship is not well
demonstrated in the cross-stratified dune (EC), or
aeolian sand sheet (ES) deposits (Table 1 ; Fig. 6),
but is more evident in channel (CH) sand bodies
(Table 1; Fig. 6). Calcite cements in ovoid concre­
tions are poikilotopic to blocky spar (15 11m-1.0 mm
diameter). Although most ovoid to elongate concre­
tions are not zoned, some exhibit internal zonation,
of which two types can be recognized. The first
consists of two concentric zones, differentiated by
only a colour change from grey (inner zone) to pink
(outer zone). The second type consists of six or
more concentric layers of radial spar cement (layers
vary from 0.5 to 3 mm in thickness; Fig. 9D).
Platy concretions
Platy concretions are small (5-50 em diameter),
flat, irregularly shaped masses that most often occur
in groups or masses with a consistent planar orien­
tation, and are often subparallel to bedding. This
type of concretion is usually associated with cre­
vasse splay deposits (CS), overbank fines (OF),
palaeosol horizons (P), and interdune (ID) deposits
(Table 1 ; Fig. 6). The surfaces of platy concretions
commonly have 1 -3 em diameter pits, tubes or
grooves (Fig. 1OA), although smooth surfaces are
38 JR. Beckner and P. S. Mozley

Fig. 8. (A ) I
rregul a
r mi rciti cnodules in a lc y
a-ri hc silt from the Ch am is aMes aMember. Divisions on s ca le a re in
ecntimetres. (B ) Photomi rcogr aph of anodule showing mi rciti cm a trix , rcyst a
ll a
ri a a
n d icr u
c mgr a
nul ar rc acking ( c).
(C) Nodule with pitted, tubul a r (T ), na d grooved (G ) surf ac e textures. Tubul ar stru tcures rae filled with s pa r ca l icte in
the ecntre of the nodule. Divisions on s ca le a re in ecntimetres. (D ) Photomi rcogr ap h of p revious nodule showing a
mi rciti cm atrix, nad a o
cm p lex mixture of laveol a r textures (A ) a
n d icr uc mgr anul ar rc ac
king ( a
rrows ). Alveol a r
textures a re more rounded th an icr u c mgr a nul a r rc acking, an d not asso ic taed with fr a
m ework gr ains.

also found. Platy concretions with pitted, tubed or
grooved surfaces usually have a micritic matrix,
with micrite-spar and alveolar textures (Fig. I OB),
whereas those with smooth upper and lower sur­
faces usually have a microspar (7-15 Jlm diameter)
matrix.
Calcite cements associated with these concretions
are dominantly micritic and exhibit circumgranular
cracking, alveolar, micrite-spar and meniscus mi­
crotextures. Radial spar microtexture is present
locally, characterized by bladed radial spar formed
around a micritic nucleus (Fig. I OD).

Rod concretions Tabular cemented units

Tube and rod concretions are small (0. 1 -5 em
diameter, 3-50 em long) horizontal to vertical
masses that occur both individually and in groups.
They are associated with overbank fines (OF),
palaeosol horizons (P), aeolian sand sheet deposits
(ES) and interdune (ID) deposits (Table I; Fig. 6).
They often branch, and most thin downwards (Fig.
I OC). Some rod concretions have pitted, tubular
and grooved surface textures; most are smooth.
Tabular cemented units are 0.2-3 m thick tabular
bodies that commonly extend for hundreds of
metres or more laterally. They can be divided into
three types: those with original sedimentary struc­
tures preserved (type I ); those in which sedimentary
structures are not preserved, with tube, groove and
pitted surface textures (type 2); and those in which
some mix of both type I and type 2 characteristics
are present (type 3).
 Calcite cements in the Zia Formation 39

Fig. 9. (A ) Isol a t d
e na d grou p s of ovoid concr teions from th ePi d e r aP ra d
a aM me b re. Divisions on th esc a l ea r ein
c n
e tim ter se. (B ) Elong a t econcr teions th at s eem to b econstruct de from ovoid concr teions. Th se eshow conc n e tric
int rn
e la zon a tion. (C) Elong a t econcr teions from th eu ppe r part of th eUnn a m de M m e b re. Not eth econsist n e cy of th e
ori net a
tion. Sc a l e� I 0 e m. (D ) C onc entric ovoid concr teion from th eu ppe r pa
rt of th eUnn a m de M m e b re. Divisions
on th esc a l e ra ein c ne tim ter se.

to elongate concretions are commonly found on

Type 1 (sedimentary structures preserved)
Sedimentary structures such as trough and planar
cross-bedding are common features of type I tabu­
lar cemented units (Fig. l l A). These cemented
units are coarser grained and better sorted than
units immediately below and above (Fig. 6). Lower
contacts are most often sharp and locally erosive.
Upper contacts are usually sharp. Bed outlines can
be lenticular, wavy and irregular, depending on the
original sedimentary structures preserved. These
units are usually associated with channel associa­
tions (CH), and coarser, better sorted units in
crevasse splay deposits (Table 1; Fig. 6). They vary
from 0.2 to 3 m in thickness and can be of great
lateral extent (> 1 km) (Fig. 7). Calcite cementation
textures are mainly blocky spar. Coalesced ovoid
the tops of these units: they are usually less than
1 m in thickness and can extend for tens of metres
laterally.
Type 2 (no sedimentary structures preserved)
Type 2 tabular units lack original sedimentary
structures and are often associated with reddened
clays and clayey sands from overbank fine (OF),
palaeosol (P) and interdune (ID) deposits (Table 1 ;
Fig. 6). Micritic calcite i s the main cement, and
micrite-spar textures, grain dissolution, alveolar
structures, circumgranular cracking and meniscus
cement are common. Type 2 tabular units are
subdivided by outcrop morphology into massive,
platy, wavy bedded, fractured and laminar types.
40 J.R. Beckner and P.S. Mozley

Fig. I 0. (A ) Pl tay concr teion from th emiddl eof th eUnn am de M m e b re. Notic eth emillim ter -esiz de tub e( T ) a nd
groov e(G) structur se. Divisions on th esc a l e ra ein c n
e tim ter se. (B) Photomicrogr ap h of a
lv oel ar t xetur se (A ) from a
p
l a
ty concr teion in th eUnn a m d e M m e b re. (C ) Rod-sh ape d concr teions from n a aeoli n
a s n
a dston ein th eCh a m is a
M se aM m e b re. Not eth ta s v
e re a
l of th erods br a nch a n d thin downw a rds. Divisions on th esc la e ra ein c n e tim ter se.
(D) R a di al s pa
r (microcodium) microt x etur e.

The most common type 2 morphology is charac­
terized by massive bedding, with abundant branch­
ing or isolated rod structures and pitted, tubular
and grooved surface textures (Fig. l l B). Lower
contacts are usually gradational. This morphology
is generally 0.3-1 m thick, and occasionally can be
of great lateral extent (>I km) (Fig. 7).
Some outcrops are thin ( l 0-20 em), platy or
wavy bedded, with pitted, tubular and grooved
surfaces (0.5-3 em diameter). These thin bedded
units are generally less than l 0 m in lateral extent.
Other outcrops are characterized by millimetre­
sized calcite-filled fractures that are in places irreg­
ular, unoriented and fenestral, and sometimes re­
semble small folds (Fig. l l C). Original sedimentary
structures are generally not preserved. These units
may also be associated with tubular, rod and platy
concretions. These outcrops exhibit alveolar and
fenestral microtextures, and displacement laminae
in thin section.
Some outcrops have an irregular wavy laminar
(3- 1 0 em thickness) morphology. Individual lami­
nae vary from l to 2 mm in thickness. Units usually
have sharp upper and lower contacts. These forms
exhibit abundant alveolar and fenestral microtex­
tures (Fig. 1 1 D).
Type 3 (tabular units with mixedfeatures)
The above descriptions are pure end-member ce­
mentation types. However, most tabular cemented
units in the Zia Formation show a mixture of
characteristics of these end-members. Units that are
closest in appearance to the type l end-member
have excellent preservation of sedimentary struc­
tures, with rare pit and tube structures (type 3)
 Calcite cements in the Zia Formation 41

Fig. 11. (A) Ty p

e I t b
aul a
r unit from the middle of the Unn a
med Member. Note the good p
reserv taion of sediment ary
structures. Units on the sc ale a
re in decimetres. (B) Ty p e2 t b
a ul a
r unit from the Piedr aP ra da aMember. Note b a sence
of sediment ary structures. Units on the sc a l e rae in decimetres. (C ) Tee pee structure from the middle of the Unn amed
Member. Units on the sc a le are in centimetres. (D) Photomicrogr ap h of fenestr a
l/l a
m in a
r microtextures common in
t a
bul a
r units with l a
m in a
r, brecci ated an d tee p ee outcro pmor p hologies.

(Fig. 12A). The most common, thickest and most
laterally extensive units (>2 km) are those that are
close in appearance to type I tabular units (Fig. 7).
Mixed feature cements near the type 2 end-member
are associated with more poorly sorted, finer
grained layers and pit, tube and rod structures, with
some evidence of the original sedimentary struc­
tures (type 3) (Fig. 128).
Type 3 units also show a mixture of cement
textures, including floating grain and micrite-spar
types. Floating grain microtextures are usually char­
acterized by grains surrounded by drusy to iso­
pachous sparry cements, with the remaining void
spaces filled with micrite or microspar (Fig. 1 2C).
This type of cement is most commonly found in
units near the type I end-member. The micrite-spar
microtexture is most common in mixed units near
the type 2 end-member (Fig. 12D). In these units
the spar is generally equal to or more abundant than
the micrite cements.
CATHODOLUMINESCENCE AND
ELEMENTAL COM POSITION
Authigenic calcite varies from bright orange to
non-luminescent, whereas detrital carbonate is a
dull orange. Poikilotopic and blocky spar associated
with ovoid and elongate concretions and type I
tabular units is typically a dull orange to non­
luminescent. Some type I tabular units, and most
type 3 (phreatic), show some zonation (bright
orange to dull orange-red and non-luminescent). In
most cases this is not visible under plane polarized
light. Micritic cements are either a dull orange-red
or non-luminescent. Spar-filled alveolar and fenes­
tral textures associated with these micrites are only
luminescent along the edges. Oscillatory zoning
(regular and irregular) in this spar occurs rarely.
Although zonation is visible under cathodolumines­
cence, it is not visible using back-scattered electron
imaging. Microprobe analysis shows that, regardless
42 J.R. Beckner and P.S. Mozley

Fig. 12. (A ) Ty pe3 ( ph r ea t ic ) t b

a ul a r unit. Although th re eis good p r se rev taion of s d e i ment ary structur se , atub eth ta
br anch se downw a rds is shown by th e a rrow. Th eh amme r is app roxi ma t ley 1 8 emlong. (B) Ty pe3 (v a dos e) t b
aul a r
unit. Th e a rrows p o int to r leict s d e i me n t ray structur se. Divisions on th esc a l e ra ein d cei met r se. (C) S par- m icrit e
microt x etur efro m aty pe3 ( p h r eat ic ) unit. Fr ame work gr a i ns a
r eco ta de by dis pl caiv eiso pa chous s pa r, na d th es pa c e
b tew een is fill d
e with micrit e. (D ) Micrit e- s par microt xetur efro m aty pe3 (v a dos e) unit . Gr ai ns na d grou p s of gr ai ns
ra eco ta d
e with micrit ,e a nd th es pa c eb tew ee n is fill d
e with s par.

of microtexture, the cements are very near the regardless of stratigraphical position (Fig. 14) .

calcite end-member composition (Fig. 13). Magne­
sium is the main impurity, and even this is less than
I mol o/o. Cements from the Sand Hill fault at the
top of the section show slightly more magnesium
than Zia Formation samples (Mozley & Goodwin,
1995a) (Fig. 13).
ISOTOPE GEOCHEMISTRY
The isotopic composition of the various calcite
types does not vary greatly. Carbon isotope values
(8 1 3 C) range from -3.0 to -5.5o/oo PDB, whereas
oxygen isotope values (8 1 8 0) range from -7.3 to
-13.6o/oo PDB (Fig. 14). 8 1 3 C values for nodular,
platy, rod-shaped concretions and type 2 tabular
units are generally heavier than those of other types
There is also a weak upward stratigraphical trend of
increased 8 1 3 C values in the Unnamed Member for
type 1 and type 3 tabular units. 8 1 8 0 values for the
lower part of the Zia Formation show no definite
trend with stratigraphical position, but there is an
increase in 8 1 8 0 values in type 1 and type 3 tabular
units higher in the section within the Unnamed
Member (Fig. 14). The highest value for Zia Forma­
tion cements (-7.3o/oo PDB) approaches the average
value of the fault cements (-7.1 o/oo PDB) (Mozley &
Goodwin, 1995b). Samples collected along a 500 m
lateral traverse of a single cemented horizon that
intersects the fault exhibit no consistent variation in
8 1 8 0 with distance from the fault. The sample
closest to the fault (0.5 m) has the closest value to
the fault cements (-7.3o/oo PDB). Type 2 tabular units
and nodular, platy and rod-shaped concretions are
 Calcite cements in the Zia Formation
0 Zia cements (spar)
Fig. 13. Tern a ry di g
ar am showing com p osition of
micrite, s pa r nad S na d Hill f u
a lt cements from the study
rae a
. The sc ale of the p lot is a
t 99 mol% Ca C 0 3 . D ta a
for f ua lt cements from Mozley & Goodwin ( 1 99 Sb ).
generally more enriched in 1 3 C and depleted in 1 8 0
than those associated with type I and type 3 tubular
units and ovoid and elongate concretions (Fig. 1 5).
D ISCUSSION
Environments of cement formation
We have inferred the environments of cement
formation in the Zia Formation by comparing
microscopic and macroscopic characteristics with
those of cements of known origin described in the
literature. In this section we discuss known charac­
teristics of vadose and phreatic cements, and use
this as the basis for identification of cementation
environments in the Zia Formation.
Characteristics of vadose cementation
Despite the complexities and variations in surficial
environments of precipitation, vadose zone ce­
ments in arid environments have a number of
distinctive characteristics.
1 A dense micritic fabric, crystallaria, circumgran­
ular cracking and alveolar textures have been fre­
quently associated with pedogenic cementation
(Esteban & Klappa, 1 9 83; Wright, 1 990; Wright &
Tucker, 1 99 1 ; Mora et a!., 1 993). Microcodium has
also been associated with pedogenic cementation.
Microcodium, which exhibits a radial spar micro-
43
texture, is associated with either root filaments, or
casts of fruiting or resting stages of soil fungi
(Klappa, 1 978, 1 979; Esteban & Klappa, 1 9 83;
Goudie, 1 9 83; Wright, 1 990; Monger et a!., 1 99 1 ;
Wright & Tucker, 1 99 1 ; Mora et a!., 1 993).
2 Permeability in the vadose zone tends to be higher
in finer sediments because flow occurs preferentially
along grain surfaces rather than the centre of large
pores. Finer sediments have more surfaces on which
vadose flow can occur (Palmquist & Johnson, 1 962;
Hillel, 1 9 80; Jury et a!., 1 99 1 ; Mozley & Davis,
1 996). If cementation is limited by the supply of
Ca 2 + and/or HC0 3- to the precipitation site, vadose
cements should occur preferentially in the finer
sediments (Mozley & Davis, 1 996).
3 Vadose cements are commonly associated with
soil zonation and the alteration of parent material
during soil development, resulting in reddened
clays and clay-rich sands in which there is little or
no preservation of original sedimentary structures
(Retallack, 1 990; Mack et a!., 1 993; Mora et a!.,
1 993).
4 Vadose cementation is intimately associated with
rhizocretions, which record the orientation and
position of former root systems as root casts or
moulds (Klappa, 1 980b; Esteban & Klappa, 1 9 83;
Goudie, 1 983; Retallack, 1 9 8 8 , 1 990; Wright &
Tucker, 1 99 1 ; Gardner et a!., 1 992; Milnes, 1 992).
5 Vadose cementation is sometimes associated with
distorted or disrupted bedding, such as brecciation
and teepee structures. Brecciation can result from
cracking and drying during dewatering, or cracking
and dissolution when well indurated carbonate lay­
ers are disturbed by growing roots (Klappa, 1 9 80a;
Esteban & Klappa, 1 9 83). Growing roots also play a
role in the formation of some teepee structures, when
expansion along a single layer forces sediment up­
wards (Klappa, 1 980a). Tepee structures can also
arise from expansive calcite and/or evaporite min­
eral growth (Klappa, 1 980a; Warren, 1 9 82; Goudie,
1 983).
6 Cementation in the vadose zone can also result in
irregular, wavy, laminar cement morphologies.
Laminar cemented zones with abundant root traces
and alveolar and fenestral microtextures are
thought to result from root mats forming in the
zone of capillary rise (Cohen, 1 98 2 ; Semeniuk &
Searle, 1 985; Wright et a!. , 1 98 8). Laminar units
high in the vadose zone may have etched upper
surfaces due to exposure (Semeniuk & Meager,
1 98 1 ), or have fewer and more vertically oriented
rhizocretions (Cohen, 1 982).
44 J.R. Beckner and P.S. Mozley

-2
CARBON

-3
� •
Q 0 .• 0 0 ...
0 0 0 D D
�
;:o
-4
D 00

- c}] OJD D
�0. -� �
o oBIJ.p OJ
D D
l
•
-S
•
• •
-6

OXYGEN t
-7
�
0 �
-8 • 0 �

0
-9

� -10 •
0 0
r:9
�
0
'21 • oo ��
�
<-0
-I I
� 0&' o •
:� o
�
0 0 • 0
-12 0 0 0 •

-13
Fig. 14. Plot of c a
rbon an d oxygen
isoto p e v laues versus str taigr ap h ic
- 14
so 1 00 I SO 200 2SO 300 3SO osition. o 1 3C v laues exhibit
p
0
signific a n t sc a
tter even within a
Stratigraphic Position (m)
single horizon (see 1 80- 1 90 m ).
Phre taic c rabon v a lues , however ,
• Vadose (nodul ra , pl tay, and rod concretions) do p l ot consistently below v d a ose
v laues , reg a rdless of str atigr ap h ic
o Mixed:type-3 (vadose)tabular units o sition. o 1 80 v laues show more
p
• Phreatic (ovoid and elong tae concretions, and type- It b
a ul ra units) sc tater th a n c rabon v laues, a nd
D Mixed: type-3 (phreatic ) t b
aul ra units incre a se in the u pp er two-thirds of
� F u
alt cements the Unn a med Member.

precipitation, and the unfilled voids could subse­

Vadose cementation in the Zia Formation
Nodules, platy concretions, rod-shaped concretions
and type 2 tubular units all have micritic matrices,
alveolar microtextures, circumgranular cracks and
cross-cutting fractures. Cementation is associated
with finer-grained layers in reddened clays and clay­
rich sands in which there is very little or no
preservation of original sedimentary structures. The
radial spar microtexture associated with some con­
cretions resembles microcodium. All of this implies
that these cementation types are vadose. Micrite­
spar cement textures could have initially formed in
the vadose zone as pendant and meniscus envelopes
around grains or groups of grains (Jacka, 1974;
Reeves, 1976; Warren, 1983). These initial vadose
cements would provide sites for further calcite
quently be filled with sparry calcite in the phreatic
zone (Jacka, 1970; Funk, 1979).
Platy concretions have been described by several
workers as resulting from initial disruption of relict
bedding; similar rod concretions have been de­
scribed as rhizocretions (Kappa, 1980b; Esteban &
Kappa, 1983; Retallack, 1988, 1990).
Irregular, unoriented and fenestral millimetre­
sized calcite-filled fractures found in some units are
interpreted as brecciation structures. Structures that
resemble small folds are probably teepee structures
because they are associated with rhizocretions and
alveolar microtextures. There is not enough clay in
these units to cause expansion, although expansive
calcite growth cannot be ruled out. The laminar
cemented units in the Zia Formation have abun-
 Calcite cements in the Zia Formation 45

-2 cements should be associated with coarser, more

• Vadose permeable sediments (Lynch, I 996; Mozley &
o Mixed (vadose)
Davis, 1 996).
• Phreatic 3 Pedogenesis usually destroys original structures,
-3 Vadose D Mixed (phreatic)
•
so the preservation of original sedimentary struc­
� •
Q 0 • •
•
D
D tures such as cross-bedding is evidence of a non­
e • • • 0
0 pedogenic origin, and has been attributed by others
u o oe 0
Do [p
Oo
0 0
to phreatic-groundwater cementation (Wright &
B,-4 0 Mixed Tucker, 1 991; Spot! & Wright, 1 992; Mora et al.,
oo 0 �q_::ipo o 0
0
1 993).
J
�------------- • 4 Phreatic cementation is very rarely associated
with rhizocretions (Wright & Tucker, l 991; Sp6tl
-5 •
•

Phreatic & Wright, 1 992; Mora et a!., 1 993). The lack of

• •
rhizocretions indicates that cementation occurred
below the zone in which plants had their roots, in
-6+-----�----�---r--� the phreatic zone.
-12 -I I -10 -9 -8 -7
&ISO (PDB)
Phreatic cementation in the Zia Formation
Fig. 15. Plot of li 1 3C versus 8180, with individu a
l p
o ints
identified by ce m ent taion ty p e s. V da ose ty pes include Ovoid and elongate concretions and type l tabular
nodule, p l tay na d rod concretions, a s well as ty p e2
units appear to have formed principally in the
t b
a ul a
r units. Phre a t ic ty p es include ovoid to elong tae
concretions sa well a s ty p e I t b a ul a
r units. In gener a l phreatic zone, because they have poikilotopic and
v d
a ose ce ments h a ve he a v ier c rabon v laues a nd lighter blocky spar cements, are associated with coarser,
oxygen v laues th a n p h re at ic ce m ents. Phre a t ic n
a d ty pe better sorted units, show preservation of original
3( p hre taic) units th a t p lot with oxygen v laues gre a ter sedimentary structures, and are not associated with
th na - 10 a re fro mthe u pp er pa rt of the Unn am ed
Me m ber. rhizocretions. In the Zia, preferential cementation
of coarser, better sorted layers operates on the scale
of both thin section and outcrop, something also
dant root traces and alveolar textures, and thus are noticed by Lynch ( 1 996). Elongate concretions have
interpreted as root mats associated with the zone of been noted by other workers and attributed to
capillary rise. groundwater flow in the phreatic zone (McBride et
a!., 1 994, 1 995; Mozley & Davis, 1 996). Orienta­
tions of these elongate concretions tend to be
Characteristics of phreatic cementation
uniform within a single outcrop, often on the scale
Cementation in a terrestrial phreatic environment of several kilometres, as would be unexpected in
also has several distinctive characteristics. vadose-zone cementation (Mozley & Davis, 1 996).
1 Calcite precipitation under phreatic conditions
can continue uninterrupted by an air-water inter­
Mixed vadose and phreatic cementation
face (Morse & Mackenzie, 1990). Thus, isopachous
or drusy, poikilotopic and blocky spar cements are Most cemented tabular units in the Zia Formation
most often associated with precipitation in the are difficult to classify as strictly pedogenic, vadose
phreatic zone (Jacka, 1 9 70; Folk, 1 9 7 4; Retallack, non-pedogenic or phreatic carbonates. Cementation
1 990; Bums & Matter, 1 995). Sparry cements can in these units forms a continuum between vadose
also form in the vadose zone as calcans or crystic and phreatic end-members. The most common type
nodules, but they are associated with soil zonation, of mixed unit is near the phreatic end-member. In
highly dense micritic cements and nodules (Weider these units, vadose influence is indicated by the rare
& Yaalon, 1982). Because these cements are not occurrence of rhizocretions in outcrop. Vadose in­
associated with such features they are unlikely to fluence on cements may also be indicated by the
represent calcans or crystic nodules. presence of sparry, floating grain microtextures.
2 As previously discussed, if cementation is limited These appear to be the result of initial vadose cemen­
by the supply of Ca2+ and/or HC0 3 - to the precip­ tation (grain-coating micrite), followed by circum­
itation site, and supply is limited by flow, phreatic granular cracking and then expansive phreatic ce-
46 J.R. Beckner and P.S. Mozley
mentation (spar) caused by burial below the water
table (see Fig. 12C). Expansive calcite growth is a
common feature of some phreatic carbonates
(Wright & Tucker, 199 1 ; Mora et a!., 199 3 ), and
there is no evidence of grain or cement dissolution as
described by Tandon & Friend ( 1 989).
Cements near the vadose end-member are asso­
ciated with typical vadose features; however, these
features are less apparent than in type 2 tabular
units, and sparry void filling cements are sometimes
more abundant than micrite. As stated previously,
micrite-spar cement textures could have initially
formed in the vadose zone as pendant and meniscus
envelopes around grains or groups of grains (Jacka,
1 974; Reeves, 1 976; Warren, 1 983). Upon burial,
these initial vadose cements would provide sites for
further calcite precipitation and the unfilled voids
could subsequently be filled with sparry calcite in
the phreatic zone (Jacka, 1 970; Funk, 1 979). The
vadose contribution to cementation may have been
overlooked in the past because of this overprinting.
Cathodoluminescence and elemental composition
Phreatic cements in the Zia Formation consist of
large crystals of almost pure calcite which show no
zoning in cathodoluminescence. This suggests that
the cements formed in a relatively short time,
during which pore water chemistry was relatively
constant (Bums & Matter, 1995).
Although vadose cements in the Zia Formation
are typically non-luminescent, occasionally mul­
tiple complex irregular zonations do occur. Wright
& Peeters ( 1 989) suggested that such zonations
result from multiple stages of crystal growth, com­
plex crystal dissolution, and reprecipitation. For the
most part floating grain textures in the Zia Forma­
tion seem to be the result of expansive calcite
growth, and not grain dissolution as described by
Tandon & Friend (1989).
Cements from type 3 tabular (vadose) units are
similar to vadose cements in luminescent character­
istics. Cements from type 3 tabular (phreatic) units
sometimes exhibit regular zoning in cathodolumi­
nescence. Generally no zonation is present in calcite
crystals under plain light, suggesting that crystal
growth may not be multigenerational.
Isotope geochemistry
The environment of precipitation has a direct effect
on carbon and oxygen isotope values for vadose and
phreatic zone carbonates. Units interpreted to be of
vadose origin have generally higher o 1 3 C values and
similar or slightly lower o 1 8 0 values than cemented
units inferred to have formed dominantly in the
phreatic zone (Fig. 1 5). Type 3 units (mixed fea­
tures) typically have o 1 3 C and o 1 80 values between
those of phreatic and vadose units (Fig. 1 5). Further
complications arise because o 1 3 C values for mostly
phreatic mixed units resemble vadose values,
whereas their o 1 8 0 values resemble the phreatic
values.
H igher o 1 3 C values for vadose cement have been
attributed by other workers to either greater diffu­
sion of heavy atmospheric carbon, or a larger
relative percentage of isotopically heavier c4 plant
biomass (Talma & Netterberg, 1 9 83; Mora et a!.,
1 993). Lower o 1 8 0 values for vadose cement have
several possible explanations.
1 The main mechanism for the precipitation of
calcite in the vadose zone was transpiration­
induced drying, and not evaporation (transpiration
does not fractionate oxygen, whereas evaporation
does (Quade et a!., 1 989; Cerling & Quade, 1 993).
Evaporation removes the lighter oxygen (by frac­
tionation), making the o 1 80 values in the vadose
cement heavier.
2 Waters that recharged the aquifer had undergone
water-rock interaction, mixing oxygen values from
meteoric waters with those derived from dissolu­
tion of 1 8 0-enriched minerals in the rock. Dissolu­
tion of framework grains, particularlY,_ volcanic rock
fragments and feldspar, is common in the Zia
Formation.
3 Winter rainfall is isotopically lighter, so that main
recharge events that penetrated into the phreatic
zone may have occurred during the summer when
isotopic values are heavier (Quade et a!., 1989;
Cerling & Quade, 1 993; Wang et a!., 1 993).
Stratigraphical variation in isotopic compositions
may also mask the relationship between isotopic
values and the environment of precipitation.
Changes in the local vegetation, precipitation rates
or seasonal temperatures can affect isotope values
(Mora et a!., 1 993; Wang et a!., 1993). The similar­
ities in oxygen values for both phreatic and vadose
units implies that they were precipitated from fluids
with a similar origin, in this case meteoric water.
Detailed age data for the Zia Formation are not
currently available, and so correlation of Zia For­
mation isotope changes with changes elsewhere is
not possible. The isotopic signature of the vadose
cement may also be contaminated by later phreatic
 Calcite cements in the Zia Formation
cementation (or vice versa), especially in type 3
units. A similar complex variation in isotope com­
position resulting from the mixing of vadose and
phreatic (hydromorphic) cementation is observed
in other fluvial settings (Slate et al., 1996). Also,
bulk samples were analysed, and so possible isoto­
pic differences between spar and micrite cements
were not observed. Clearly, further data need to be
collected on Zia Formation isotopes before any
definite conclusions can be made.
TIMING OF CEMENTATION
Because most of the vadose cements appear to be
pedogenic, they must have formed shortly after
deposition of the host sediments (while the sedi­
ments were still exposed to surficial weathering).
The exact timing of phreatic cementation is more
difficult to determine. Evidence from some type 3
tabular units indicates that at least some of the
phreatic cementation also occurred very early. The
most common surface texture for type 3 tabular
units is root moulds (pits, tubes and grooves) (see
Figs 8C and l OA), indicating that the cement must
have formed around the root while it was still
physically present. Because phreatic cements gener­
ally do not fill the root moulds, cementation must
have occurred before the oxidation of the root. In
an oxidizing, arid, alluvial environment organic
root material will not last long after burial, therefore
the phreatic cementation probably occurred very
early.
CONTROLS ON THE
SPATIAL DISTRIBUTION O F
CEMENTATION : IM PLICATIONS FOR
GROUNDWATER AND
PETROLEUM RESOURCES
The dominant types of cementation in the Zia
Formation are pedogenic and phreatic. By defini­
tion, the spatial distribution of pedogenic carbonate
is a function of the spatial distribution of palaeo­
sols, which is a function of facies architecture and
the length of time a particular surface was exposed.
Most pedogenic carbonate in the Zia Formation is
poorly developed, discontinuous and associated
with finer-grained sediments in overbank fines
(OF), crevasse splay (CS), and interdune (ID) facies
associations (see Table I ; Fig. 6). Unlike discontin-
47
uous vadose cements, the distribution of extensive,
well developed pedogenic units in the Zia Forma­
tion is controlled primarily by the duration of
surface exposure and landscape stability.
Phreatic cementation is typically associated with
coarser and better sorted facies associations such as
fluvial channel deposits (CH), cross-stratified dune
deposits (EC), aeolian sheetsand (ES) deposits and
some interdune deposits (ID) (see Table 1; Figs 4
and 6). This indicates that phreatic cements formed
preferentially in initially highly permeable portions
of the Zia Formation, presumably because of the
initial high groundwater flow rates in such zones (i.e.
permeable zones would have an abundant supply of
dissolved Ca2+ and/or HC0 3 -). The distribution of
phreatic cementation can also be explained by
groundwater flow effects. Where fluvial channel sand
deposits are surrounded by silty sands, silts and
shales, flow (and thus cementation) is focused into
thinner, more isolated sands (Lynch, 1996). In tex­
turally more homogeneous sediments, flow is not
focused and cementation is less extensive (Lynch,
1996), a feature we see in the aeolian sediments of
the Zia Formation. Where there is no evidence of
textural control on phreatic cementation, vadose cal­
cite is present and thus could have acted as a nucleus
for later phreatic-zone precipitation.
Although our study is based upon outcrop sam­
ples, and consequently does not directly relate to
groundwater or hydrocarbon production problems,
the Zia Formation in the subsurface is an important
local aquifer, and similar alluvial units form signifi­
cant aquifers and hydrocarbon reservoirs elsewhere.
Thus the cementation relationships observed in the
Zia Formation are of more than local interest. Cal­
cite cementation in the Zia Formation has adversely
affected potential reservoir/aquifer quality in two
main ways:
1 Phreatic-zone cementation occurred preferen­
tially in units that had the highest primary perme­
abilities (i.e. coarser-grained and better-sorted lay­
ers). Thus extensive calcite cementation has resulted
in a permeability inversion, in which zones of high
primary permeability are now low-permeability
zones.
2 Type 3 and some type I tabular units are often
laterally extensive, in some cases extending for over
2 km (see Fig. 7). These units would form signifi­
cant barriers to vertical fluid flow, perhaps resulting
in compartmentalization of the reservoir/aquifer.
Such compartmentalization can result in dramati­
cally reduced production if wells are screened on
48 J.R. Beckner and P.S. Mozley
only one side of the cemented layer (Kantorowicz et
a!., 1 987).
CONCLUSIONS
Vadose cements in the Zia Formation are character­
ized by the presence of rhizocretions and associated
microtextures (alveolar, fenestral, circumgranular
cracking), and by a lack of primary sedimentary
structures. Phreatic cements in the Zia Formation
are characterized by poikilotopic and blocky spar
cements, the preservation of original sedimentary
structures, and the absence of rhizocretions and as­
sociated microtextures. They occur as isolated or
groups of ovoid or elongate concretions, and as lat­
erally extensive tabular bodies. Type 3 (mixed) units
in the Zia Formation reflect characteristics of both
phreatic and vadose zone cementation (e.g. preser­
vation of sedimentary structures plus rhizocretions
and alveolar microtextures). Type 3 (mixed) units
may reflect movements of the water table, such that
vadose cements are moved into the phreatic zone, or
vice versa.
o 1 3 C values for vadose cements tend to be heavier
and o 1 8 0 values tend to be similar to or slightly
lighter than those of phreatic cements. Type 3 units
also have mixed isotope values, with o 1 3 C and 8 1 8 0
values between the end-member vadose and phreatic
values.
Cementation in the phreatic zone occurred pref­
erentially in zones of high primary permeability,
whereas vadose cementation occurred principally in
association with soil development. Pedogenic car­
bonates may have served as nucleation sites for
later phreatic cementation, leading to complex
zones of mixed pedogenic and phreatic cements.
Calcite cementation in the Zia Formation has
greatly reduced potential reservoir/aquifer quality.
Most permeable units are extensively cemented
with phreatic calcite. Many tabular units are often
laterally extensive, forming significant barriers to
vertical fluid flow and conceivably resulting m
compartmentalization of the reservoir/aquifer.
ACKNO WLED GEMENTS
Mike Spilde assisted in the microprobe analysis of
the cements. Bill DeMarco developed many of the
photographs. Dr David Johnson provided the use of
his microscope and camera. Dr Andrew Campbell
provided the use of his stable isotope laboratory,
participated in numerous discussions and reviewed
preliminary versions of the manuscript. Drs Laurel
Goodwin, David Love and Bruce Harrison partici­
pated in numerous discussions, accompanied me in
the field, and reviewed preliminary versions of the
manuscript. The manuscript also greatly benefited
from the comments and suggestions of Drs Steven
Burns, Sadoon Morad, Antonio Garcia and V. P.
Wright. Special thanks are due to the King and
Parker families for allowing access to the study area.
Partial funding for this study was provided by the
Office of Graduate Studies at New Mexico Tech,
and the New Mexico Geological Society. In addi­
tion, acknowledgement is made to the Donors of
the Petroleum Research Fund, administered by the
American Chemical Society, for the partial support
of this research.
REFERENCES
ALLEN , J.R.L . ( 1 986) Pedogeni c acl rcetes in the Old Red
Sandstone fa ices (L ate Silurian-Early C arboniferous ) of
the Anglo-Welsh area , Southern Britain. In: Paleosols,
Their Recognition and Interpretation (Ed. Wright, V.P. )
pp. 58-86. Prin ecton University Press, Prin ecton , NJ.
ANDERSON, M.P. ( 1 989) Hydrogeologi cfa ices models to
delineate large-s acle s p atial trends in gla cial and gla ico­
fluvial sediments. Geol. Soc. Am. Bull. 1 0 1 , 50 1 -5 1 1 .
ANDERSON, M.P. ( 1 990) Aquifer heterogeneity-A geolog­
i acl p ers p
e tcive. In: Fifth Canadian/American Confer­
ence on Hydrogeology, Proceedings (Ed. Ba h c u, S. ),
pp. 3-22. National Well Water Asso ication , Dublin ,
Ohio.
ARAKEL, A.V. & McCoNCHIE, D. ( 1 982) C lassifi action and
genesis of a cl rcete and gy p site lithofa cies in p aleodrain­
age systems of inland Australia and their relationshi pto
acrnotite mineralization. J. sediment. Petrol. , 52, 1 1 49-
1 1 70.
ARAKEL, A.V., JACOBSON, G. , SALEHI, M. & HILL , C .M.
( 1 989) Sili icfi action of acl rcete in p alaeodrainage basins
of the Australian arid zone. A ust. J. Earth Sci. , 36,
73-89.
ATKINSON, C . D. ( 1 986) Te tconi c o c ntrol on alluvial sedi­
mentation as revealed by an an icent actena of the
C a p ella F ormation (Eo ecne ) of Northern S p a in. In:
Paleosols, Their Recognition and Interpretation (Wright ,
V.P. ) , pp . 1 39- 1 79. Prin ecton University Press, Prin ­c
eton , NJ.
BuRNS, S.J. & MATTER, A. ( 1 995) Geo c hemistry of acrbon­
ate ecments in surfi cial alluvial o c nglomerates and their
p
aleo lcimati cim p l i actions, Sultanate of Oman. J. sedi­
ment. Res. , A65( 1 ), 1 70- 1 77.
CAR LI SLE , E. ( 1 983) C on ecntration of uranium and vana­
dium in a cl rcetes and gy pc retes. In: Residual Deposits:
Surface Related Weathering Processes and Materials
(Ed. Wilson ,R.C.I. ) ,S p
e .c Pub!. Geol. So c. London , 1 1 ,
1 85- 1 95.
 Calcite cements in the Zia Formation
CERLING, T.E. & Q uADE, J. ( 1 993) Stable carbon and
oxygen isotopes in soil carbonates. In: Climate Change
in Continental Isotopic Records, (Eds Swart, P., Lohman,
K.C. & MacKenzie, J.A.) pp. 2 1 7-23 1 . Geophys.
Monogr., Am. Geophys. Union, 78.
CHAN, M.A. ( 1 989) Erg margin of the Permian White Rim
Sandstone, SE Utah. Sedimentology 36, 2 3 5-2 5 1 .
CHAPIN, C.E. & CATHER, S.M. ( 1 994) Tectonic setting of
the axial basins of the northern and central Rio Grande
rift. In: Basins of the Rio Grande Rift: Structure, Strati­
graphy, and Tectonic Setting (Eds Keller, G.R. & Cather,
S.M.), Geol. Soc. Am. Spec. Paper, 291 , 5-2 5.
CoHEN, AS. ( 1 982) Paleoenvironments of root casts from
the Koobi Fora Formation, Kenya. J. sediment. Petrol. ,
52, 40 1 -4 1 4.
DAVIS, J.M., LOHMANN, R.C., PHILLIPS, F.M., W I LSON , J.L.
& LovE, D.W. ( 1 993) Architecture of the Sierra Lad­
rones Formation, central New Mexico: depositional
controls on the permeability correlation structure. Geol.
Soc. Am. Bull. , 105, 998- 1 007.
DREIMANIS, A. ( 1 962) Quantitative gravometric determi­
nation of calcite and dolomite by using a Chittick
apparatus. J. sediment. Petrol. , 32(3), 520-529.
ESTEBAN, M. & KLAPPA, C.F. ( 1 983) Subaerial exposure
environment. In: Carbonate Depositional Environments
(Eds Scholle, P.A., Bebount, D.G. & Moore, C.H.).
Mem. Am. Ass. petrol. Geol. , Tulsa, 33, 1 -54.
FOLK, R.L. ( 1 97 4) The natural history of crystalline cal­
cium carbonate: effect of magnesium content and salin­
ity. J. sediment. Petrol. , 44, 40- 5 3 .
FuNK, J . M . ( 1 979) Distribution o f carbonate cements in
Quaternary alluvial-fan deposits, Birch Creek Valley,
East-Central Idaho. Diagenetic model. Bull. Am. Ass.
petrol. Ceo/. , 63(3), 454.
GARDNER, L.E., DIFFENDAL, R.F. & WI LLI AMS , D.F. ( 1 992)
Stable isotope composition of calcareous paleosols and
ground-water cements from the Ogallala Group (Neo­
gene), western Nebraska. Contributions to Geology, Uni­
versity of Wyoming, 29(2), 97- 1 09.
GAWNE, C.E. ( 1 98 1 ) Sedimentology and stratigraphy of the
Miocene Zia Sand of New Mexico: summary. Ceo/. Soc.
Am. Bull. , 99( 1 ), 999- 1 007.
G!LE, L.H., PETERSON, F.F. & GROSSMAN, R.B. ( 1 966)
Morphological and genetic sequences of carbonate ac­
cumulation in desert soils. Soil Sci. , 101(5), 347-360.
GouDIE, A.S. ( 1 98 3 ) Calcrete. In: Chemical Sediments and
Geomorphology: Precipitates and Residua in the Near­
Surface Environment (Eds Goudie, AS. & Pye, K.),
pp. 93- 1 3 1 . Academic Press, London.
HILLEL, D. ( 1 980) Fundamentals ofSoil Physics. Academic
Press, San Diego.
JACKA, A.D. ( 1 970) Principles of cementation and porosity
-occlusion in Upper Cretaceous sandstones, Rocky
Mountain Region. In: Wyoming Geological Association
Guidebook, Twenty-second Annual Field Conference,
pp. 265-285.
JACKA, A.D. ( 1 974) Differential cementation of Pleis­
tocene carbonate fanglomerate, Guadalupe Mountains.
J. sediment. Petrol. , 44( 1 ), 85-92.
JACOBSON, G., ARAKEL, A.V. & YIJIAN, C. ( 1 988) The
central Australian groundwater discharge zone: evolu­
tion of associated calcrete and gypcrete deposits. A ust. J.
Earth Sci. , 35, 549-565.
49
JURY, W.A., GARDNER, W.R. & GARDNER, W.H. ( 1 99 1 )
Soil Physics, 5th edn. John Wiley, New York.
KANTOROWICZ, J.D., BRYANT, J.D. & DAWANS, J .M. ( 1 98 7)
Controls on the geometry and distribution of carbonate
cements in Jurassic sandstones: Bridport Sands, south­
ern England and Viking Group, Troll Field, Norway. In:
Diagenesis of Sedimentary Sequences (Ed. Marshall,
J.D.), pp. I 03- 1 1 8. Blackwell Scientific Publications,
Oxford.
KLAPPA, C.F. ( 1 978) Biolithogenesis of Microcodium:
elucidation. Sedimentology, 25, 489-522.
KLAPPA, C.F. ( 1 979) Calcified filaments in Quaternary
calcretes: organo-mineral interactions in the subaerial
vadose environment. J. sediment. Petrol. , 49, 95 5-968.
KLAPPA, C. F. ( 1 980a) Brecciation textures and teepee
structures in Quaternary calcrete (caliche) profiles from
eastern Spain: the plant factor in their formation. Ceo!.
J. , 15(2), 8 1 -89.
KLAPPA, C.F. ( 1 980b) Rhizoliths in terrestrial carbonates:
classification, recognition, genesis and significance. Sed­
imentology, 27, 6 1 3-629.
KLAPPA, C.F. ( 1 983) A process-response model for the
formation of pedogenic calcretes. In: Residual Deposits:
Surface Related Weathering Processes and Materials.
(Ed. Wilson, R.C.I.). Spec. Publ. Geol. Soc. London, 1 1 ,
22 1 -2 3 3 .
KocuREK, G . ( 1 98 1 ) Significance o f interdune deposits
and bounding surface in aeolian dune sands. Sedimen­
tology, 28, 75 3-780.
KOCUREK, G. & DOTT, R.H. ( 1 98 1 ) Distinctions and use of
stratification types in the interpretation of eolian sand.
J. sediment. Petrol. , 51(2), 5 79-59 5 .
KRAUS, M.J. & BowN, T.M. ( 1 986) Paleosols and time
resolution in alluvial stratigraphy. In: Paleosols, Their
Recognition and Interpretation (Ed. Wright, V.P.), pp.
1 80-207. Princeton University Press, Princeton, NJ.
LEEDER, M.R. ( 1 97 5) Pedogenic carbonates and flood
sediment accretion rates: a quantitative model for allu­
vial arid-zone lithofacies. Ceo!. Mag. , 1 1 2(3), 25 7-270.
LozrNSKY, R.P. ( 1 994) Cenozoic stratigraphy, sandstone
petrology, and depositional history of the Albuquerque
Basin, central New Mexico. In: Basins ofthe Rio Grande
Rift: Structure, Stratigraphy, and Tectonic Setting (Eds
Keller, G.R. & Cather, S.M.), Geol. Soc. Am. Spec.
Paper, 291 , 7 3-8 1 .
LYNCH, F.L. ( 1 996) Mineral/water interaction, fluid flow,
and Frio Sandstone diagenesis: evidence from the rocks.
Bull. Am. Ass. petrol. Ceo/. , 80(4), 486-504.
MACHETTE, M.N. ( 1 985) Calcic soils of the southwestern
United States. Geol. Soc. Am. Spec. Paper, 203, 1 -2 1 .
MACK, G.H., JAMES, W.C. & MoNG ER H.C. ( 1 993) Classi­
,
fication of paleosols. Ceo!. Soc. Am. Bull. , 105, 1 29-
1 36.
MANN, A.W. & HORWITZ, R.D. ( 1 979) Groundwater cal­
crete deposits in Australia; some observations from
Western Australia. J. Geol. Soc. Aust., 26, 293-303.
McBRIDE, E.F., PICARD, M.D. & FOLK, R.L. ( 1 994) Ori­
ented concretions, Ionian coast, Italy, evidence of
groundwater flow direction. J. sediment. Petrol. , A64(3),
53 5-540.
McBRIDE, E.F., MILLIKEN, K.L., CAVAZZA, W. et a/. ( 1 995)
Heterogeneous distribution of calcite cement at the
outcrop scale in Tertiary sandstones, Northern Apenni-
50 J.R. Beckner and P.S. Mozley
nes, Italy. Bull. Am. Ass. petrol. Geol. , 79(7), 1 044-
1 06 3 .
MIALL, A.D. ( 1 99Q) Principles ofSedimentary Basin A nal­
ysis, 2nd edn. Springer-Verlag, New York.
MILNES, A.R. ( 1 992) Calcrete. In: Weathering, Soils, and
Paleosols (Eds Martini, J.P. & Chesworth, W.), pp. 309-
347. Elsevier, Amsterdam.
MONGER, H. C., DAUGHERTY, L.A. & GILE, L.H. ( 1 99 1 ) A
microscopic examination of pedogenic calcite in an
aridisol of southern New Mexico. In: Occurrence, Char­
acteristics, and Genesis of Carbonate, Gypsum, and
Silica Accumulations in Soils (Ed. Nett Leton, W.O.),
Soil Sci. Soc. Am. Spec. Pub!., 26, 3 7-59.
MORA, C.!., FASTOVSKY, D.E. & DRIESE, S.G. ( 1 993)
Geochemistry and Stable Isotopes of Paleosols. Univer­
sity of Tennessee Department of Geological Sciences,
Studies in Geology, 23, 65 pp.
MORSE, J.W. & MACKENZIE, F.T. ( 1 990) Geochemistry of
Sedimentary Carbonates. Developments in Sedimentol­
ogy, 48. Elsevier, Amsterdam.
MOZLEY, P. & DAVIS, J.M. ( 1 996) Relationship between
oriented calcite concretions and permeability correla­
tion structure in an alluvial aquifer, Sierra Ladrones
Formation, New Mexico. J. sediment. Res., A66, 1 1 - 1 6.
MOZLEY, P. & GOODWIN, L. ( 1 995a) Patterns of cementa­
tion along a Cenozoic normal fault: a record of paleo­
flow orientations. Geology, 23, 5 3 9-542.
MoZLEY, P. & GOODWIN, L. ( 1 995b) Patterns of cementa­
tion along the Sand Hill Fault, Albuquerque Basin, NM:
implications for paleoflow orientation and mechanisms
of fault-related cementation. In: Characterization of
Hydrogeologic Units in the Northern Albuquerque Basin.
(Eds Haneberg, W.C. & Hawley, J.W.). New Mexico
Bureau of Mines and Mineral Resources Open-File
Report, 402-C, 3- 1 1 -3-23 .
NETTERBERG, F ( 1 969) The interpretation o f some basic
calcrete types: South Africa. Arch. Bull. , 24, 1 1 7- 1 22 .
NETTERBERG, F . & CAIGER, J . H. ( 1 983) A geotechnical
classification of calcretes and other pedocretes. In:
Residual Deposits: Surface Related Weathering Pro­
cesses and Materials (Ed. Wilson, R.C.I.). Spec. Pub!.
Geol. Soc. London, 1 1 , 2 3 5-243.
PALMQUIST, W.N. & JOHNSON, A.!. ( 1 962) Vadose flow in
layered and nonlayered materials. US Geol. Surv. Prof.
Paper, 450-c, C l 42-C l 43 .
PETTIJOHN, F.J., POTTER, P.E. & SIEVER, R . ( 1 987) Sand
and Sandstone, 2nd edn. Springer-Verlag, New York.
PoRTER, M.L. ( 1 987) Sedimentology of an ancient erg
margin: the Lower Jurassic Aztec Sandstone, southern
Nevada and southern California. Sedimentology, 34,
6 6 1 -680.
PowERS, M.C. ( 1 9 5 3 ) A new roundness scale for sedimen­
tary particles. J. Sediment. Petrol. , 28, I 08- 1 1 0.
QuADE, J., CERLING, T.E. & BowMAN, J.R. ( 1 989) System­
atic variations in the carbon and oxygen isotopic com­
position of pedogenic carbonate along elevation
transects in the southern Great Basin, United States.
Geol. Soc. Am. Bull. , 101, 464-475.
RABENHORST, M.C., WILDING, L.P. & WEST, L.T. ( 1 984)
Identification of pedogenic carbonates using stable car­
bon isotope and microfabric analyses. Soil Sci. Soc. Am.
J. , 48, 1 25 - 1 32.
READ, J.F. ( 1 974) Calcrete deposits and Quaternary sedi-
ments, Edel Province, Shark Bay, Western Australia. In:
Evolution and Diagenesis of Quaternary Carbonate Se­
quences, Shark Bay, Western A ustralia (Ed. Logan,
B.W.), Mem. Am. Ass. petrol. Geol., 22, 250-282.
REEVES, C.C. ( 1 976) Caliche: Origin, Classification, Mor­
phology and Uses. Estacado Books, Lubbock, Texas.
RETALLACK, G.J. ( 1 988) Field recognition of paleosols.
Geol. Soc. Am. Spec. Paper, 216, 1 -20.
RETALLACK, G.J. ( 1 990) Soils of the Past. Unwin-Hyman,
· Boston.
SEMENIUK, V. & MEAGHER, T.D. ( 1 9 8 1 ) Calcrete in Qua­
ternary coastal dunes in southwestern Australia: a
capillary-rise phenomenon associated with plants. J.
sediment. Petrol. , 51( 1 ), 47-68.
SEMENIUK, V. & SEARLE, D.J. ( 1 98 5 ) Distribution of
calcrete in Holocene coastal sands in relationship to
climate, Southwestern Australia. J. sediment. Petrol. ,
55( 1 ), 86-95.
SLATE, J.L., SMITH, G.A., YANG, w . & CERLING, T.E. ( 1 996)
Carbonate-paleosol genesis in the Plio-Pleistocene St
David Formation, Southeastern Arizona. J. sediment.
Res. , 66( 1 ), 85-94.
SPOTL, C. & WRIGHT, V.P. ( 1 992) Groundwater dolocretes
from the Upper Triassic of the Paris Basin, France: a
case study of an arid, continental diagenetic facies.
Sedimentology, 39, 1 1 1 9- 1 1 36.
TALMA, A.S. & NETTERBERG, F. ( 1 983) Stable isotope
abundances in calcretes. In: Residual Deposits (Ed.
Wilson, R.C.I.), Spec. Pub!. Geol. Soc. London, 1 1 ,
22 1 -2 3 3 .
TANDON, S.K. & FRIEND, P.F. ( 1 989) Near-surface shrink­
age and carbonate replacement processes, Arran Com­
stone Formation, Scotland. Sedimentology, 36 , 1 1 1 3-
1 1 26.
TEDFORD, R.H. ( 1 982) Neogene stratigraphy of the north­
western Albuquerque basin. In: New Mexico Geological
Guidebook, 3 3rd Field Conference, Albuquerque
County II, pp. 273-278.
WANG, Y., CERLING, T.E., QUADE, J. & BOWMAN, J.R.
( 1 993) Stable isotopes of paleosols and fossil teeth as
paleoecology and paleoclimate indicators: an example
from the St David Formation, Arizona. In: Climate
Change in Continental Isotopic Records (Eds Swart, P.,
Lohman, K.C. & MacKenzie, J.A.), Geophys. Monogr.,
Am. Geophys. Union, 78, 2 4 1 -248.
WARREN, J.K. ( 1 982) The hydrological significance of
Holocene teepees, stromatolites, and boxwork lime­
stones in coastal salinas in south Australia. J. sediment.
Geol. , 52(4), 1 1 7 1 - 1 20 1 .
WARREN, J.K. ( 1 983) Pedogenic calcrete as it occurs in
Quaternary calcareous dunes in coastal South Australia.
J. sediment. Petrol. , 53(3), 7 8 7-796.
WEBER, K.J. ( 1 982) Influence of common sedimentary
structures on fluid flow in reservoir models. J. Petrol.
Tech., 34, 665-672.
WEIDER, M. & YAALON, D.H. ( 1 982) Micromorphological
fabrics and developmental stages of carbonate nodular
forms related to soil characteristics. Geoderma, 28,
203-220.
WRIGHT, V.P. & PEETERS, C. ( 1 989) Origins of some early
Carboniferous calcrete fabrics revealed by cathodolumi­
nescence: implications for interpreting the sites of cal­
crete formation. Sediment. Geol. , 65, 345-3 5 3 .
 Calcite cements in the Zia Formation 51

WRIGHT, V.P., PLATT, N.H. & WIMBLEDON, W.A. ( 1 988)
Biogenic laminar calcretes: evidence of calcified root­
mat horizons in paleosols. Sedimentology, 35, 603-620.
WRIGHT,V.P. ( 1 990) A micromorphological classification
of fossil and recent calcic and petrocalcic microstruc­
tures. In: Soil Micromorphology: a Basic and Applied
Science (Ed. Douglas, L.A.), pp. 40 1 -407. Develop­
ments in Soil Science, 1 9 . Elsevier, Amsterdam.
WRIGHT, V.P. & TUCKER, M.E. ( 1 9 9 1 ) Introduction. In:
Calcretes (Eds Wright, V.P. & Tucker, M.E.) Int. Ass.
Sediment., Reprint Series, 2, 1 -22.
Spec. Pubis int. Ass. Sediment. ( 1 998) 26, 5 3-85

Carbonate diagenesis and porosity evolution

in sheet-flood sandstones: evidence from the
Middle and Lower Lunde Members (Triassic) in the Snorre Field,
Norwegian North Sea

S. MORAD,* L . F. DE RO S,*' J . P. N Y S T U E Nt a n d M. BER GAN t

*Sedimentary Geology Research Group,
Institute of Earth Sciences, Uppsala University, S-752 36 Uppsala, Sweden,
e-mail sadoon. morad@geo. uu.se; ifderos@ifufrgs.br; and
tSaga Petroleum, Kjorboveien 16, PO Box 490, N-1301 Sandvika, Norway
e-mail johan-petter. nystuen@saga.com; morten.bergan@saga.com

ABSTRACT

Facies-related cementation and porosity reduction occurred in Triassic sheet-flood sandstones of the
Lower and Middle Members of the Lunde Formation (the Snorre Field, Norwegian North Sea). These
processes took place during near-surface eodiagenesis and syncompaction to mesodiagenesis. The early
cements are dominated by calcite, dolomite and siderite, which were precipitated at ""3o·c and
postdate some of the authigenic kaolinite. The precipitation temperatures of syncompactional to
mesogenetic calcite and dolomite/ankerite are calculated to be 30-70·c and 40-9o·c, respectively.
Meteoric telodiagenesis during the Kimmerian uplift (late Jurassic-early Cretaceous) resulted in the
dissolution of carbonate cement and of framework silicates, as well as in the precipitation of kaolinite.
This telogenetic kaolinite was formed in sandstones located a few tens of metres below the
unconformity.

INTRODUCTION

Carbonate cements are often among the dominant
components of diagenesis and hence are of decisive
importance in determining the reservoir quality of
sandstone sequences. Despite this, the timing, the
geochemical conditions of precipitation and disso­
lution, as well as the source and fate of these
cements are not fully understood. In continental
and near-shore sediments, cements commonly pre­
cipitate as calcretes and dolocretes in the vadose
and phreatic zones, and attain a variety of mineral­
ogical, textural and distribution patterns as well as
elemental and isotopic compositions. These ce­
ments form lenses and layers of densely cemented
1 Present address: Universidade Federal do Rio Grande do
Sui, Instituto de Geociencias, Departamento de Mineral­
ogia e Petrologia, Av. Bento Goncalves, 9500, CEP
9 1 5 0 1 -970 Porto Alegre, RS, Brazil.
alluvial deposits u p to 1 0 k m wide and 100 km
long, such as in Quaternary sediments from Austra­
lia (Arakel, 1986, 199 1 ; Arakel & Wakelin-King,
199 1) and Tertiary sediments of Kuwait (El-Sayed
et al., 199 1 ). Therefore, phreatic calcretes and
dolocretes can profoundly influence fluid flow, in­
cluding petroleum migration and production.
Stable isotopic compositions of eogenetic carbon­
ate cements largely reflect biological activity, the
detrital composition of the host sediment, latitude,
climatic conditions, depositional facies and palaeo­
hydrology. However, eogenetic carbonates are usu­
ally sensitive to the different physicochemical
conditions that may prevail during burial diagene­
sis, which might lead to their dissolution, recrystal­
lization or replacement by other carbonates. The
two latter processes may overprint, and hence
complicate the interpretation of, original carbon

Carbonate Cementation in Sandstones Edited by Sadoon Morad 53

© 1998 The International Association of Sedimentologists
ISBN: 978-0-632-04777-2
54 S. Morad e t a/.
and oxygen isotopic signatures and, above all, limit
their use in unravelling the conditions of eogenetic
carbonate cementation.
Despite the complexity involved in the interpre­
tation of carbonate cementation owing to the pres­
ence of multiple generations and subsequent
alteration, diagenetic carbonates can provide valu­
able information about the geochemical evolution
of pore waters involved. This evaluation can be
achieved by detailed petrographic and geochemical
studies of these cements, coupled with a proper
understanding of the depositional environments
and facies, as well as of the burial history of the
sandstone sequences. It is consequently possible to
unravel and perhaps even predict the factors con­
trolling the composition and distribution patterns
..
•
of carbonate cements in sedimentary basins. This
study is an attempt in this direction, and is made
using Triassic sandstones of the Lower and Middle
Lunde Members in the Snorre Field, northern
North Sea (Fig. 1). We also aim to elucidate the
reasons for the overall reduced sandstone reservoir
quality of these two members compared with those
of the U pper Lunde Member. At Snorre, the latter is
an important reservoir containing 369 x 106 Sm3.
In this paper the terms dolocrete and calcrete are
used to indicate sediments extensively cemented by
displacive dolomite and calcite under conditions
ranging from the soil horizon to shallow phreatic.
Conversely, we use the expression carbonate ce­
ment in sandstones where is no evidence of displa­
cive, near-surface precipitation. Kaolin is used as a
Fig. I. The North Sea area and the
location of the Snorre Field on the
Tampen Spur structural high in the
northernmost North Sea. Major
structural elements are of late
Jurassic-Early Cretaceous age.
 Sheet-flood sandstones in the Snorre Field 55

common name for the polymorphs kaolinite and
dickite. The terms eodiagenesis, mesodiagenesis
and telodiagenesis are used sensu Schmidt & Mc­
Donald ( 1979).
GEOLOGICAL SETTING
Structure, stratigraphy and palaeoclimate
The Snorre Field is located in the northern North
Sea area within the Tampen Spur, a structural high
formed during Late Jurassic-Early Cretaceous rift­
ing (Fig. 1 ). The Tampen Spur, consisting of several
rotated fault blocks, constitutes the northeastern­
most structural continuation of the Shetland Plat­
form, forming the northwestern border region of the
Viking Graben. The Snorre Field reservoir is a
structural trap within one of these rotated fault
blocks (Fig. 2). The Snorre fault block became
progressively uplifted and rotated during the
Kimmerian tectonic phase from the Callovian to
Oxfordian time, culminating with an erosion of
1200-1500 m of uppermost Triassic and Jurassic

� Lower Lunde Reservoir

(Prospect)
- Middle Lunde Reservoir

E=:::J �tJ;,����)
nde reservoir

�·········: Upper Lunde reservoir

'
········· (L02·L05)

c=::J Statljord reservoir

I
3414
34f7

Fig. 2. Map of the Snorre Field

showing positions of wells 34/4-1
and 34/7-A-3H, sampled for the
present study, and cross-section
lines A-A' and B-B' shown in
Fig. 6. t151211101.ERS.M6e.EOm
56 S. Mor ad et a/.

strata in the northeastern part of the block, suc­ During the Early Triassic the rift basin in the
ceeded by submergence and renewed burial from northern North Sea area was bordered by major
the Early Cretaceous (Fig. 3) (Dahl & Solli, 1993). north-south-running basin margin faults to the
The Snorre and neighbouring fields were fi lled east, towards the Norwegian mainland, and to the
during the late Cretaceous and Cainozoic. The west towards the Shetland Platform region. The
O xfordian-Ryazanian Draupne Shale (Kim­ approximately 200 km broad basin was segmented
meridge Clay equivalent) is the primary hydro­ into a series of half-grabens along north-south­
carbon source rock, sealed by late Jurassic and trending intrabasinal faults during the syn-rift
Cretaceous-Tertiary shales (Horstad et a/., 199 5). phase. This phase was succeeded by the develop­
The alluvial reservoir rocks in the Tampen Spur ment of a wide, thermally subsiding post-rift basin
area were deposited within a Permian-early Trias­ from the Anisian-Ladinian to Middle J urassic
sic rift basin that comprised most of the North Sea (Badley et a/. 1988; Steel & R yseth, 1990; Steel,
area (e.g. Ziegler, 1988; Glennie, 1990, 1995). 1993; Nystuen & Fait, 1995). Morton (1993) also

Priabonai
Bartonai n
Lutetai n Balder_ � - -_
....:. ______
Ypresai n
Thane tian Lista/Sele -
Danian
-

Maastrichtian Jr
o salfar _

Campanian
= = =-= _
� -
Sa nt
o n
Coniacian - -.
T uronian
an
i
H� ;';; ,;; ;;!' �;- - '--J-��� = = : iB do ok s
=
� ::-::-�

t C-"
f ------- --' e--
no'--"ma'-'n'-=ia--
n __--l�·��dbytSo la
- -
A l b ian - �
la!Mime
L

�
A ptian

-�Tuxen/AsgArd
Barremian
hH.;oa70 � u tec:,; riv
':' ia"'-
'
n -l =- =- = -
Valangn i ian --- _ �
Rvazanian
Volgian -
f--v:;
t<.l""m
l m ="'r =- ----l � · ·��
e'"' o"all ' a-' n
Oxfor dian - - ch

�.
Callovian m
o
f--'B::a:::.tho::.: :.:.:nian ::::..__ __---j . N s .
_ � � _ .' . . .. l_ )
·. .. . ·
. ·
_
;..
" ·

Bajocai n .

Toarcian
Pli ensbach ai n
Sinemurai n
Rhaetian

�=·�=�=�=-�\
n
·-
::-:-:-:
1--� �a-� n-----1:· ����- � = === .. . Fig. 3. General stratigraphy of the
-4· . . . ... -. :- . -: . . :- . �. . .
.- - -- .
. .
1-- ----+--:-La - -::- din,-
. ian - -
Tampen Spur area, revised from
·Lomvi .· · · · · ·
.

· · · · · · · ·
Campbell & Ormaasen ( 1987) with
Anisian
1---:-----+-- - - - - - - -_ u _ _+le � - -
isf : · ::- .. - ---=-=- =-=
· lithostratigraphical nomenclature
0 l ene k ai n. lnd an from Vollset & Dore (I 984) and
? ? ?
Isaksen & Tonstad ( 1989) and
time scale from Gradstein et a/.
(I 995).
 Sheet-flood sandstones in the Snorre Field 57

suggested syn-rift faulting during the Rhaetian­
Norian to Lower Sinemurian in the North Sea area.
Although the crustal extension was moderate to
negligible during this post-rift phase, tectonic reac­
tivation or passive differential movements have
occurred along marginal and a few intrabasinal
faults (Steel & Ryseth, 1990; Nystuen & Fiilt, 1995).
The Lunde Formation was deposited during the
early post-rift phase, in Olenekian Scythian to Late
Rhaetian time, a period of about 30 million years
(Figs 3 and 4). The formation belongs to the Hegre
Group and, in the Snorre Field area, is about
1200 m thick where its maximum thickness is
preserved (Fig. 4). Together with the overlying,
overall continental Statfjord Formation, the Hegre
Group constitutes a composite stack of megase­
quences reflecting variation in the basin's accom­
modation rate and sediment influx, controlled
primarily by the rate of tectonic subsidence of the
basinal area and tectonic uplift of the hinterland, as
well as eustasy (Nystuen et al., 1989; Steel &
Ryseth, 1990; Steel, 1993) .
The palaeolatitude o f the northern North Sea
area during the Middle and Late Triassic was about
20-26 °N, and the climate generally warm and arid.
However, over about 30 million years climatic
variations are assumed to have occurred, brough t
about by plate tectonic movements which took the
region from lower to higher latitudes, by orbital
forcing of the climate (Milankovitch periodicities)
and by regional variations between hinterland highs
and lowland basinal areas. Climatic variations re­
lated to topographical differences are thought to
have caused regional variations in the amount of
annual precipitation and runoff pattern.
Depositional environments
During deposition of the Lunde Formation, the
palaeodrainage direction in the Tampen Spur area

Chronostratigraphy Group Formation/

member

Pliensbachian Dunlin Amundsen

(60 ·130m)

Statfjord
(100- 200m)

Upper
(850m)

214
Norian

Middle
(150m)
Lunde

Hegre

Lower
(200m)

Fig. 4. Stratigraphic column of

Triassic-Lower Jurassic on the Lomvi (1OOm)
Tampen Spur showing general
lithostratigraphy of the Lunde
Formation (modified from Teist (1000m)
Nystuen & Fait, 1995). Time scale
is according to Gradstein et a/. lnduan
( 1995).
58 S. Morad e t al.
was northerly towards an ultimate base level of an
epeiric Borealic seaway between the present Nor­
way and Greenland. The area was a wide alluvial
plain or terminal basin with clastic material coming
from hinterland highs to the east and west (Steel &
Ryseth, 1990; Nystuen & Fai t, 1995).
The lower member of the Lunde Formation in
the Snorre Field (type area) is 155-230 m thick, as
revealed by four wells in which the unit is totally
penetrated. Three other wells have partially pene­
trated this member. Only 1 2 m altogether have been
cored in the 34/4-1 well, from the uppermost and
middle parts of the unit. The lower member consists
of interchanging beds of mudstone, siltstone and
very fine- to fine-grained sandstone. The sandstone
beds increase and thicken upwards. The Lower
Lunde Member was formed by progradation of an
alluvial wedge from the basin's margin towards the
axial basinal area (Steel, 1993). The mudstones are
reddish brown to greenish grey and were probably
deposited on a distal alluvial plain. The sandstones
are interpreted to represent sheet-flood deposits and
fluvial-channel facies (Nystuen & Fait, 1995). Well­
to-well log correlation of individual depositional
packages of strata is difficult within the lower
member. This property suggests the dominance of
ribbon-shaped fluvial channel sandstones embed­
ded in overbank fine-grained sediments, interlay­
ered with more laterally continuous sheet-flood
deposits. The proportion of sandstones with poros­
ity and permeability high enough to be defined as
reservoir rocks (net reservoir) relative to the total
(gross volume) amount of rock (the N /G ratio) is
relatively high. A low degree of carbonate cementa­
tion of the sandstones in this unit accounts for this
relatively high N/G value.
The Middle Lunde Member ( 100-150 m thick) is
characterized by fine-grained beds of high lateral
continuity and a low N /G ratio, although the total
sandstone content is high. The change from the
lower member occurs vertically within a transi­
tional zone of about 10-20 m. The member consists
of greyish mudstones, siltstones and sandstones
organized in depositional units that can be corre­
lated on electric well logs for several kilometres
within the field. The cored sections show succes­
sions of parallel-laminated and current-ripple lami­
nated, very fine- to fine-grained sandstone beds
interchanging with siltstone and mudstone units.
The sandstone beds, up to about 3 m thick, are
blocky, slightly graded or inversely graded. The
beds may form composite, slightly upward-fining
bedsets up to about 5 m thick. These sandstones,
together with the mudstones, are organized in mo­
tifs that are upward fining, upward coarsening or
upward coarsening to fining. These motifs are de­
fined as the allostratigraphical reservoir units M l­
M4. Rip-up mud intraclasts up to 2-3 em long and
smaller clastic clay aggregates are common in the
sandstone beds, particularly in the lower part.
Sandstone- and mudstone-siltstone facies are
slightly bioturbated, mostly as single vertical to hor­
izontal burrows; some finely laminated mudstone
units are non-bioturbated. Desiccation cracks, filled
with very fine sand, are common. Desiccation has
also given rise to disturbed primary lamination.
Massively carbonate-cemented beds and concre­
tions of 2-3 em occur in both the sandy and the
muddy sediments. Root structures are rare. The
overall high resistivity and low N /G ratio of the
middle member reflects the extensive carbonate
cementation in sandstones.
The overall depositional environment of the Mid­
dle Lunde Member represents a distal alluvial plain
or terminal basin (Nystuen et al., 1989; Steel &
R yseth, 1990; Nystuen & Fait, 1995). Depressed
areas were flooded by ephemeral sheet floods, leav­
ing blankets of sand, silt and mud of high lateral
extent. During certain periods these depressions
could tum into shallow temporal lakes in which
laminated, current-ripple laminated and wave­
ripple laminated mud and silt aggraded. Drying up
of these shallow lakes gave rise to frequent desicca­
tion cracks, distorted lamination and mud fl akes.
The episodic flooding promoted the infiltration of
suspended clay particles into the sand blankets.
The Upper Lunde Member (�850 m thick) marks
another change in the depositional facies and style
of these U pper Triassic continental beds. The lower
boundary of this member is usually assigned to the
base of the first marked fluvial-channel sandstone.
The lower part of the upper member consists of
braided stream channel sandstones and units of
reddish-brown floodplain mudstones characterized
by palaeosols with calcrete concretions. The upper
part of the member also comprises middle-sinuous
stream deposits, interchanging with rather mature
reddish-brown, calcrete-rich palaeosols (Nystuen &
Fait, 1995).
Burial history
The burial history of the Lower and Middle Lunde
Members in the northeastern part of the rotated
 Sheet-flood sandstones in the Snorre Field 59

WELL 34/4-1

17 Nordland Gp.

1000

Balder Fm.
2000 Sele Fm.
Shetland Gp Maasrr.

L.Maaslr.

Campanian

Santonian
Coniac
U. Turonian
Cromer Knoll Gp.
Time(Ma)
U. Lunde

L.& m Lunde

Lomvi

Fig. 5. Subsidence curve for strata in well 34/4- 1. The latest Jurassic-earliest Cretaceous uplift caused the erosion of
about 1300 m of sediments above the Middle Lunde Member.

Snorre fault block is complex (Fig. 5). It includes an
initial period of subsidence down to a depth of
about 1500-1600 m until the Middle Jurassic
(Bathonian-Callovian), followed by a period of
uplift an d erosion during the Late Jurassic and
Early Cretaceous. The uplift culminated with a
removal of 1200-1500 m of sediments, followed by
a second period of subsidence with an onset in
Valanginian-Hauterivian. A phase of very rapid
subsidence took place during Late Cretaceous
(Campanian-Maastrichtian), and another phase of
high subsidence rate started in Pliocene-Pleistocene
an d is still going on.
The Late Jurassic to Early Cretaceous Kimme­
rian uplift and tilting of the Snorre fault block
caused the formation of an erosional surface that
cuts into the underlying succession of strata to
various stratigraphical depths. Thus, the top of the
cored middle Lunde section in well 34/4-1 is lo­
cated 24 m beneath this subaerial un conformity,
whereas the top of the cored middle Lunde interval
in well 34/7-A-3H is in a more southerly position
and located 358 m below the unconformity (Fig. 6).
SAMPLES AND ANALYTICAL
METHODS
Two wells, 34/4-1 and 34/7-A-3H (Fig. 2), were
selected based on the availability of coring of the
Middle and Lower Lunde Members. The cored
intervals from these two wells do not overlap
stratigraphically. Samples examined from well 34/
7 -A-3H are exclusively sandstones that represent
the uppermost part of the Middle Lunde. Samples
from well 34/4-1 comprise mudstones and sand­
stones, and represen t the upper part of the Lower
Lunde Member and middle and lower intervals of
the Middle Lunde Member.
One hundred an d fifty-five thin sections were
prepared from core samples impregnated with blue
epoxy resin, stained with alizarin red and potassium
60 S. Morad et a/.

Strat. halogen lamp. These examinations were performed

Depth
m
A A' in order to detect zonations and different genera­
(f) tions of calcite and dolomite cements.
:::J
0
UJ 34/4-1 Studies of crystal habits and paragenetic relatidn­
u
;5
UJ
ships were performed on 13 small gold-coated
-2400 a:
u chips, using a J EOL JSM-T330 scanning electron
microscope (SEM) with an acceleration voltage of
z
< I 0 kV. The < 2 J..Lm fraction was separated from 84
u iii
< samples by standard sedimentation methods and
I
(f) a:
(f)
examined by a Philips X-ray diffractometer
-3000 "'
- z
equipped with Cu(Ku) radiation and a nickel filter.
a: "'
a: The chemical composition of minerals was deter­
1- @ Legend:
mined in 29 polished carbon-coated thin sections
1km
>---,----<
0
I
cored
using a Cameca Camebax BX50 microprobe

-3600
Jwom 1m
8 B'
-2400
-3000
-3600
Fig. 6. Cross-section of the rotated 'Snorre fault block'
showing structural position of the wells cored in the
Lower and Middle Members of the Lunde Formation.
The depth of erosion beneath the base Cretaceous
unconformity (BCU) increases towards the crest line of
the fault block in the northeast. The positions of
chronostratigraphical boundaries are uncertain. See
positions of cross-section lines on Fig. 2.
equipped with three spectrometers and a back­
scattered electron detector (BSE). Operating condi-
tions were 20 kV acceleration voltage, 8 nA (for
carbonates and clay minerals) to 12 nA (for feld­
spars) measured beam current, and a 1 -10 J..L m
beam dia meter (depending on the extent of homo­
geneous areas). Standards and count times were:
wollastonite (Ca, 10 s), orthoclase (K, 5 s), albite
(Na, Si, 5 and I 0 s, respectively), corundum (AI,
20 s), MgO (Mg, 10 s), MnTi03 (Mn, 10 s) and
hematite (Fe, 1 0 s). Precision of analysis was better
than 0. 1 molo/o.
For the purpose of carbon and oxygen isotope
analysis of the carbonates, 26 samples were reacted
with I OOo/o phosphoric acid at 25 oc for calcite for
1 h and at 50oC for 24 h for dolomite. The C02
released from siderite was collected after 6 days at
50oC. Samples containing more than one carbonate
phase were analysed after sequential chemical sep­
aration treatments (Al-Aasm et a!., 1990). The
evolved gas for each carbonate fraction was anal­
ysed using a SIRA-1 2 mass spectrometer. The phos­
phoric acid fractionation factors used were 1 .0 1025
for calcite (Friedman & O'Neil, 1 977), 1.0 I 060 for
dolomite and 1.0 1045 4 for siderite (Rosenbaum &
Sheppard, 1 986). Carbon and oxygen isotope data
are presented in the normal 8 notation relative to
PDB (Craig, 1957) and SMOW (Craig, 196 1 ). Preci­

ferrocyanide for carbonates, and examined with
standard petrographic microscopes. The modal
compositions of 34 representative samples were
obtained by counting 300 points in each thin
section.
Twenty-two polished thin sections were exam­
ined with a Technosyn cathodo-luminoscope at an
acceleration voltage of 12-1 5 kV and a beam cur­
rent intensity of 0 . 42-0 . 43 rnA, and under blue UV
light in an Olympus BX60 microscope with I 00 W
sion ( 1 cr) was monitored through daily analysis of
the NBS-20 calcite standard and was better than
±0.05o/oo for both 8 1 3C and 8 1 8 0. Because of the
common presence of zonation and of more than one
generation of the same carbonate mineral in the
same sample, the isotopic data should be consid­
ered as average values. However, after careful
cathodoluminescence (CL) , UV and BSE examina­
tion we have attempted to select samples containing
one dominant cement type and generation.
 Sheet-flood sandstones in the Snorre Field
7
The 8 Sr/8 6 Sr ratios of calcite and dolomite in 12
samples were determined after washing the samples
with distilled water to remove the pore salts that
result from drying. The calcite samples were then
reacted with dilute acetic acid and the dolomite
samples with 0.1 HC':i, and analysed using an auto­
mated Finnigan 2 6 1 mass spectrometer equipped
with nine Faraday collectors. All analyses were
performed in the static multicollector mode using
rhenium filaments. Correction for isotope fraction­
ation during the analysis was made by normaliza­
tion to 8 6 Sr/88Sr 0 . 1 1 94. The mean standard error
=
of mass spectrometer performance was ± 0.00003
for NBS-98 7 .
Fluid inclusions i n authigenic carbonates were
examined in 14 double-polished 100 mm thick
sections using a Linkham THpOO stage calibrated
for the temperature range between -I 00 and 400oC.
COMPOSITION, PROVENANCE AND
DIAGENETIC MODIFICATION OF
THE FRAMEWORK GRAINS
According to the classification of McBride (1963),
the Lower and Middle Lunde sapdstones are arkoses
and subarkoses in which the quartz grains are do­
minantly monocrystalline (av. Omono 16.5 vol%;
= (Fig. 7A). If albitized feldspars (av. 5 . 7 vol%) are
Q poly 7 . 1 vol%), and K-feldspar (av. 7 . 3 vol%)
= considered as framework constituents and included
dominates over plagioclase (av. 2. 1 vol%). Electron
microprobe (EMP) analyses of detrital K-feldspars
revealed small amounts of albite solid solution
(Ab < 5 mol%) The K-feldspars contain variable
amounts of albite ex-solution lamellae, and are thus
perthitic. EMP analysis of detrital plagioclase grains
revealed small extents of anorthite (An < I 0 mol%)
and orthoclase (Or < I mol%) solid solutions. The
detrital feldspars have been substantially elimi­
nated by a variety of diagenetic processes, including
dissolution, kaolinization, albitization and replace­
ment by carbonate cements. Of these processes only
albitization preserves the detrital grain shape al­
most perfectly, and thus careful consideration of
their recognition criteria is necessary (Morad et al. ,
1990).
The original framework composition of the sand­
stones, which was exclusively arkosic (av.
Q50.6 F46 .7L2.7), typical of basement-uplift prove­
nance from a dominantly plutonic and high-grade
metamorphic source area (Dickinson et al., 1983),
was changed by diagenesis into a dominantly sub­
arkosic composition with average Q72.4F24.2L3.4
61
A
QUARTZ
B
50 10
ROCK
FRAGMENTS
Fig. 7. Present and original detrital composition of
representative Lunde sandstones plotted on McBride's
( 1 963) diagram: A, considering albitized feldspars as
diagenetic constituents (not in F pole)-observe the shift
from the arkose (original composition: white dots) to the
subarkose field (present composition: black dots); B,
plotting albitized feldspars in F pole-a less substantial
field dislocation is seen from original (white squares) to
present composition (black squares).
in the F pole of the sandstone classification triangle,
then the present average composition is still arkosic
(Q 6 3F34L3), but with considerably lower detrital
feldspar content than the original composition
(Fig. 7B), owing to feldspar dissolution, kaoliniza­
tion and replacement by carbonate cements. The
difference in framework composition between sand­
stone samples from the two wells studied is small,
a lthough the stratigraphical position and facies vary
between sandstones in the two wells.
The provenance of the Lunde Formation in the
Tampen Spur area is considered to be primarily
crystalline high-grade rocks of Precambrian age at
the British side of the northern North Sea basin
(Nystuen & Fait, 199 5 ). This is indicated by the
overall palaeogeography, the arkosic composition of
the sandstones, lithic fragments of granite and
gneiss, and of Precambrian N d/Sm provenance ages
(Mearns et al., 1989). However, U pper Proterozoic
to U pper Palaeozoic rocks, unroofed from the
Precambrian basement, may well a lso have acted as
source rocks.
Finely crystalline lithic fragments are scarce (av.
62 S. Morad et al.
1.3 vol%), being dominated by micaceous low-rank
metamorphic and altered volcanic rocks. Detrital
micas occur in small amounts (av. �I volo/o), mainly
in the fine-grained sheet-flood sandstones, and are
often extensively kaolinized or chloritized. Mud
intraclasts (av. 0. 1 volo/o) reworked from floodplain
deposits are concentrated in some sheet-flood sand­
stone beds. Finely crystalline dolomitic, sideritic
and calcitic intraclasts (av. 0. 7 vol%) derived from
reworking of early diagenetic carbonate deposits are
common. Heavy minerals (av. 0 . 5 vol%) include
scattered or locally concentrated altered Fe-Ti ox­
ides, rutile grains, and smaller amounts of zircon,
monazite, epidote, tourmaline, apatite and garnet.
PETROGRAPHY AND CHEMISTRY
OF DIAGENETIC MINERALS
Calcite
Calcite is the main cement in the sandstones,
averaging �15% and forming up to 70% of calcretes
(Table 1 ). Calcite cement in the calcretes occurs as
bright yellow fluorescing (Plate 1A, facing p. 6 2 ),
red to orange luminescing (Plate 1B), microcrystal­
line (� 10 J.Lm) mosaic or blocky aggregates (�30-
120 J.Lm) that replace and displace the host sedi­
ments. The cement may display a drusiform texture
with an increase in crystal size from the rim to the
pore centre (Fig. 8A). Within large pores, such as
vugs, burrows, root casts and shrinkage cracks,
calcite occurs as coarse blocky or, rarely, divergent
radiaxial-like crystals, engulfing kaolin (Fig. 8B) and
dolomite rims. Shrinkage cracks have developed in
near-surface, subaerial pedogenic environments
owing to repeated wetting and drying. Both in the
calcretes and in the sandstones, coarse blocky to
poikilotopic (up to 2 mm), calcite crystals engulf and
partially replace (and hence postdate) detrital clays,
kaolin, dolomite and siderite (Fig. 8C). Calcite also
corrodes the framework grains and pervasively re­
places the feldspars (Fig. 8C). The coarse crystalline
calcite is non-fluorescing or shows complex zona­
tion, with dull green and non-fluorescing zones
(Plate lA). Coarsely crystalline, intergranular pore­
filling calcite shows overall orange luminescence,
with thin, red-luminescing zones (Plate I C).
The framework grains in sandstones cemented by
microcrystalline and coarse to poikilotopic calcite
display a loose grain packing even when their
replacement by carbonates is considered. Evidence
of displacement includes expansion of micas along
their cleavage planes and of quartz and feldspar
grains along fractures.
Coarsely crystalline calcite has replaced and was
replaced by dolomite and ankerite, as evidenced by
the presence of corroded intercrystalline bound­
aries, suggesting a recursive precipitation of both
minerals (Fig. 80). Indeed, calcite and dolomite
occur in some cases as alternating bands (Fig.
8D,E). Poikilotopic calcite only partially fills the
intergranular pores of medium- to coarse-grained
sandstones as patchy (up to 3 mm), heterogeneously
distributed cement. Patches devoid of early carbon­
ate cements are highly compacted and cemented by
quartz overgrowths, which therefore postdate these
carbonates. In laminated sediments this calcite is
segregated along the coarser-grained laminae. These
partial pore-filling cements show euhedral crystal
terminations, indicating that the intergranular po­
rosity in these sandstones is primary and not
formed by calcite dissolution.
Poikilotopic calcite cement reveals evidence of
substantial dissolution and creation of secondary
porosity. In well 34/4-1 calcite dissolution was
accompanied by the precipitation of kaolinite. Evi­
dence for calcite dissolution includes: (i) the scat­
tered, patchy corroded remnants; (ii) similarity of
corroded shapes of framework grains in areas where
calcite is no longer present to those in areas ce­
mented by calcite; and (iii) the presence of replacive
calcite cement within the framework grains but not
in adjacent pores. Sandstones subjected to partial
cement dissolution contain undeformed ductile
grains such as micas.
In addition to cement, microcrystalline calcite
occurs as laminae in lacustrine sediments that often
reveal evidence of disruption, presumably due to
desiccation shrinkage during periodic exposure.
Fibrous calcite similar to radia.xial cements which
occur in ancient limestones commonly occurs as
vugular void fillings in these microcrystalline calcite
laminae. Similar microcrystalline calcite is in some
cases interlaminated with grey bioturbated mud­
stones.
Calcites are Mg, Fe and Mn poor (av. �0. 7 mol%)
(Fig. 9; Table 1), yet in a few instances concentra­
tion of these elements may reach up to 3. 5 , 2.4 and
1.6 mol%, respectively (Table 1 ). The dark­
luminescing zones have low Mn content compared
with the yellow-orange luminescing zones, which
contain up to �2 . 5 mol% MnC03• The crystals are
usually chemically homogeneous and unzoned, but
Table I. Chemical (mol%), stable (%o) and Sr isotopic compositions of diagenetic carbonates from the Lunde Formation, Snorre Field

Sample, constituent MgC03 CaC03 MnC03 FeC03 813C PDB 8180 PDB 8 18sMOW 87 Sr/s6sr Observations

Calcite
34/4- 1 , 25 1 6.45 av. 0.0 99.3 0. 1 0.7 -2.5 - 1 0.3 20.3 5% coarse/poikilotopic calcite cement
range 0.0-0.0 99. 1 -99.4 0. 0-0. 1 0.5-0.8
34/4- 1 , 2 5 5 7.87 av. 0.0 98.7 0.0 1 .2 1 .6 -4.8 26.0 0. 7 1 1 1 2 7 42% coarse replacive calcite and
range 0.0-0.0 98. 1 -99.3 0.0-0. 1 0.6- 1 .8 drusiform pore-fill
34/4- 1 , 25 59.35 av. 0.4 99.5 0.0 0.0 1 .6 -5.0 25.8 calcrete: =5 5 % mosaic calcite pore-fill,
range 0.4-0.5 99.4-99.6 0.0-0.0 0.0-0. 1 replacing dolomite
34/4- 1 , 25 62. 1 av. 0.0 99.9 0.0 0. 1 - 1 .3 -6.6 24.2 0.7 1 1 65 5 = 2 5 % coarse mosaic replacive calcite
range 0.0-0.0 99.9- 1 00.0 0.0-0.0 0. 1 -0. 1
34/4- 1 , 2 5 7 8.3 av. 0.3 98.5 0. 3 0.9 - 1 .9 -8.9 2 1 .8 1 3% calcite coarse pore-filling and
range 0.0-0.7 97.9-99.1 0. 1 -0.5 0.8- 1 . 1 replacing dolomite
34/4- 1 ,2 5 96.3 av. 0. 1 98.2 0.3 1 .3 -3.0 -9.9 20. 7 8% blocky replacive calcite �
""
�
range 0.0-0.3 97.4-98.9 0.0-0. 5 0.6-2.2
34/4- 1 , 2 5 9 7.8 5 av. 1 .5 97.4 0.5 0.6 24% coarse replacive calcite
s,
<:;)
range 0. 1 -3 . 5 96.2-98 . 3 0. 1 - 1 . 3 0.2- 1 . 2 <:;)
34/4- 1 ,2599. 1 av. 0.0 96.6 0.3 1 .0 -2.8 - 1 0.7 1 9.9 0.7 1 1 493 I 0% poikilotopic replacive calcite �
"'
range 0.0-0.2 97.2-99.4 0.2-0.7 0.2-2 . 4 $:)
;:s
34/4- 1 ,2604.5 av. 0.3 98. 1 0.5 1 .0 -2.2 - 1 0.8 1 9. 6 0.7 1 1 5 74 -20% poikilotopic calcite
range 0.0-0.6 9 8 . 1 -9 8 . 1 0.2-0.8 0.4- 1 . 6 t;.
B
34/4- 1 , 2667.6 av. 0. 2 97.9 0.9 0.9 -3.2 - 1 0.2 20.4 0.7 1 1 205 Calcrete: 43% nodular and poikilotopic ;:s
range 0.0-0 . 5 97.2-99. 1 0.0- 1 .6 0. 1 -2 . 5 replacive calcite �
34/4- 1 , 2669.95 av. 0.0 98.4 0.5 1 .0 -3.9 - 1 0.6 20.0 30% replacive poikilotopic calcite s·
range 0.0-0.0 9 8 .0-98.9 0.2-0.8 0. 1 - 1 . 8 s.
""
34/4- 1 , 2670.55 av. 0. 1 97.6 0.9 1 .4 -3.8 - 1 2.0 1 8.5 0. 7 1 1 62 3 I 0 % replacive poikilotopic calcite
range 0.0-0. 7 96. 1 -98.5 0.6- 1 .3 0.5-2.4 �
<:;)
3 4/4- 1 , 267 1 .8 av. 1 .0 98.3 0. 1 0.5 -2.4 -9.0 2 1 .6 0. 7 1 1 1 99 Calcrete: 44% microcrystalline rims and .....
.....
range 0.0- 1 . 6 98.3-98.4 0.0-0.3 0.0- 1 .4 coarse zoned calcite ""

34/7-A-3H,2 7 3 8 . 9 av. 0.3 98.5 0.8 0.3 -5.6 -1 1 .9 1 8 .6 5% replacive intergranular coarse calcite �
""
range 0.0-0.7 9 8 . 1 -99.0 0.6- 1 . 1 0.2-0.6
i.S::
34/7-A-3H, 2760.7 av. 0. 1 98.9 0.5 0.3 -5.3 - 1 1 .8 1 8.8 Replacive blocky/patchy calcite
range 0.0-0. 3 98.3-99.6 0.3-0.6 0. 1 -0.6
34/7-A-3H, 2767.7 av. 0.6 98.0 0.9 0.6 -5.2 - 1 2.0 1 8.6 1 2% blocky/patchy calcite replacing
range 0.0-0.8 97. 1 -99.1 0.6- 1 . 2 0.3-0.9 pseudomatrix
34/7-A-3H, 2770.4 av. 0.0 98.8 0. 7 0.4 -5 . 1 - 1 0. 1 20. 5 9% coarse calcite blocky/radiaxial
range 0.0-0. 1 9 6 . 3-99.1 0.3- 1 .0 0.0-0.8 filling vugs
34/7-A-3H, 2774.0 av. 0.4 98. 1 1.1 0.3 -4.9 - 1 0.7 1 9.9 0. 7 1 1 1 84 2 8% replacive poikilotopic calcite
range 0. 1 -0.7 9 7 .7-99.0 0.8- 1 .4 0.0-0.5
34/7-A-3H, 2779.2 av. 0.6 96.0 1 .0 0.3 -4.2 - 1 1 .5 1 9. 1 0.7 1 1 3 1 0 20% replacive poikilotopic calcite
range 0.5-0.6 9 8 .0-98.0 0.9- 1 . 1 0.3-0. 3
34/7-A-3H,2 7 62.7 av. 0.0 96.5 1.1 0.4 -3.9 - 1 2 .4 1 8. 1 1 6% replacive poikilotopic calcite
range 0.0-0.0 98.5-98.5 1 . 1 -1 . 1 0.4-0.4
34/7-A-3H, 2783.0 av. 0.0 96.9 0.7 0.3 I% blocky calcite in cracks with kaolin
range 0.0-0.0 98.8-99.0 0. 7-0.7 0.2-0.4 0..
34/7-A-3H, 2786.8 av. 0.3 98.4 0.6 0.6 -4. 7 - 1 1 .2 1 9 .4 20% coarse, replacive and crack-filling w

range 0.0-0.6 9 7 . 6-99.8 0. 1 -0.9 0.0- 1 .6 calcite

 0\
.,.

Table 1. (Continued)
3 87Sr/86 Sr
Sample, constituent MgC03 CaC03 MnC03 FeC03 o1 Cpos 0180pos 018sMOW Observations

Dolomite/ankerite
34/4- 1 , 25 1 6.45 av. 26.3 57.8 0.4 1 5 .4 -3.2 - 1 1 .9 1 8. 7 1 2% blocky, zoned Fe-dolomite/ankerite
range 23.0-29 . 6 5 7 . 1 -58.9 0.3-0.5 1 2 . 8- 1 9. 2
34/4- 1 , 2522.6 av. 39.8 57.1 0.0 2.9 = 1 5% small dolomite rhombs in clay
range 39.0-40.5 5 6 . 5- 5 7 . 8 0.0-0.0 1 . 5-4.3 matrix
3 4/4- 1 , 2526.55 av. 28.1 58.2 0.2 1 3.5 - 1 .7 -9.4 2 1 .2 23% blocky Fe-dolomite rhombs
range 24.1 -44.0 5 5 . 5-80.0 0.0-0.5 0.5- 1 7 . 8
3 4/4- 1 , 2 5 3 8 . 3 5 av. 26.6 59.3 0. 5 1 3.5 -1.1 -9.9 20. 7 = 5% microcystalline blocky Fe-dolomite
range 25.2-27 . 5 5 8 .4-60. 1 0.5-0.5 1 1 .8- 1 5 . 8
3 4/4- 1 , 2 5 5 7 . 8 7 av. 34.6 57.7 0. 1 7.2 3.9 - 1 .5 29.4 30% zoned Fe-dolomite-ankerite
range 2 1 . 1 -44. 1 54.5-6 1 .4 0.0-0.5 0.2- 1 9.5
3 4/4- 1 , 2 5 5 9.35 av. 39.2 58.7 0.2 1 .5 3.1 -5.2 25.5 Calcrete/dolocrete: =20% dolomite in
range 3 7 . 9-40.7 5 7 .0-60. 1 0.0-0.6 1 . 1 - 1 .6 mica and rims before calcite
34/4- 1 , 2562. 1 av. 29.8 56.6 0.4 1 3. 1 0.3 -7.4 23.3 -I 0% zoned and unzoned
Yl
range 1 0.6- 3 9 . 5 5 3 . 1 -6 1 .5 0.0- 1 .0 2.0-35.8 Fe-dolomite/ankerite rims
34/4- 1 , 2 5 7 9.05 av.
range
28.5
24.3-40.3
59.6
5 5 .4-6 1 . 3
0.0
0.0-0.0
1 1 .8
3.6-1 4.6
-0.4 -3.8 2 7.0 Palaeosol: -20% dolomite replacing
kaolin and clay cutans
�
i3
34/4- 1 , 2 5 7 8 . 3 av. 36.5 57.1 0.4 5.9 0.7 -4.6 26.2 0. 7 1 1 448 Dolocrete: 29% microcrystalline dolomite l'l..
range 1 8 .6-4 1 . 9 5 5 .4-58.9 0.2-0.8 1 .6-22.2 rims and crusts �
34/4- 1 , 2596.3 av. 23. 1 59.4 0.3 1 7 .2 -3.5 - 1 0. 1 20.6 9% Fe-dolomite rhombs in clay matrix !::>
:--
range 22.0-23.3 57.7-60.5 0.2-0.5 1 5 . 7 - 1 8.9
34/4- 1 , 2666.0 av. 1 9. 5 57.1 0.6 22.7 -3.8 - 1 1 .6 1 8.9 5% large Fe-dolomite/ankerite rhombs
range 1 4.9-2 3 . 6 5 5 . 5-59.0 0.0- 1 . 4 1 9.3-28. 1
3 4/4- 1 , 2667.6 av. 21.1 59.0 0.4 1 9 .4 -3.7 - 1 0.6 20.0 I% zoned ankerite-dolomite rhombs
range 1 5.9-25.8 5 5 . 2-60.4 0.3-0.6 1 4.3-28.4
3 4/4- 1 ,266 8.6 av. 1 9 .0 5 7.6 0.7 22.7 -3.8 - 1 0.9 1 9. 7 2% intergranular Fe-dolomite/ankerite
range 1 4. 8-22.2 5 5 . 2-59.0 0.3- 1 . 8 1 8.5-2 8 . 3 rhombs
3 4/4- 1 , 2668.95 av. 1 8 .4 57.9 0. 3 23.2 -4.8 - 1 1 .9 1 8 .6 20% Fe-dolomite/ankerite, part replaced
range 1 4. 8-22.4 5 5 . 1 -59.9 0. 1 -0.6 1 8 .3-29.2 by calcite
34/4- 1 , 2670.65 av. 1 5 .6 56.6 0.5 27.3 3% coarse ankerite, engulfing siderite
range 1 5 .5- 1 5 . 7 56.4-5 6 . 8 0. 3-0.6 27.2-27 . 5
34/7-A-3H, 27607.7 av. 43.7 54.8 0.0 1.3 Dolomite mostly i n expanded mica
range 43.7-43.7 54.8-54.8 0.0-0.0 1 . 3- 1 . 3
34/7-A-3H, 2 7 70.4 av. 4 1 .2 54.2 0.2 4.0 -2.4 -4.3 26.5 0. 7 1 1 269 Dolomite: 43% displacive dolomite rims
range 38.9-43.4 50.9-58.2 0.0-0.4 2.2-5 . 1
34/7-A-3H, 2783.0 av. 40.6 55.0 0.8 3.3 -0.9 -3.8 2 7.0 0. 7 1 1 1 34 Dolocrete: 52% rims of zoned dolomite
range 3 5 . 8-45 . 5 52.9-58.0 0.2- 1 . 9 0.2-7.8
Table I. (Continued)

Sample, constituent MgC03 CaC03 MnC03 FeC03

3
&1 C PDB 018 0 PDB 018SMOW 8 7 Sr/ s6sr Observations

Siderite
3 4/4- 1 , 25 1 6.45 av. 1 2 .2 4.4 0.9 82.5 6% Mg-siderite engulfed by ankerite
range 1 2 .2- 1 2 .2 4. 3-4. 3 0.9-0.9 8 2 . 5-82.5
34/4- 1 , 2 5 22.5 av. 1 2.8 3.4 0.2 83.7 - 1 .7 -8.9 2 1 .8 3% siderite dissolved rhombs in clay �
range 1 1 . 1 - 1 4.2 2. 7-4.6 0.0-0.5 82.6-84.3 matrix �
34/4- 1 , 2526.65 av. 1 3 .6 5.3 0. 1 81.1 - 1 .2 -9.4 2 1 .2 1 0% siderite partially dissolved rhombs �
0
range 1 3 .2- 1 3.9 4.7-6. 1 0.0-0.2 80.4-82.0 0
34/4- 1 , 2596.3 av. 1 5 .2 3.6 0.3 80. 8 -2.3 -9.6 2 1 .0 1 3% siderite in mica and intergranular $::)..
"'
range 1 3 . 7 - 1 8.4 2. 1 -5 . 7 0. 1 -0. 7 7 8 . 7-82.8 rhombs !::>
�
34/4- 1 , 2 5 9 7 . 8 5 av. 19.1 3.4 0.3 77.3 4% small siderite rhombs in mica and
calcite
�
range 1 8.8- 1 9 . 3 2. 3-4.4 0. 1 -0.5 76.2-78.4 0
34/4- 1 , 2599. 1 av. 1 4.0 4.4 0.4 8 1 .2 8% partially dissolved siderite rhombs �

range 1 4.0- 1 4.0 4.4-4.4 0.4-0.4 8 1 .2-8 1 .2 �

34/4- 1 , 2604.5 av. 1 4.0 3.5 0.2 82.3 -3.0 - 1 0.2 20. 5 -4% partially dissolved siderite rhombs ::l"
range 1 3.9- 1 4.4 2. 8-4.0 0. 1 -0.3 8 1 .7-8 3 . 0 s.
1 4. 5 3. 1 0.5 82.0 2% microcrystalline siderite in mica and "'
34/4- 1 , 2666.0 av.
range 1 2 . 7 - 1 7.0 2.3-3.9 0. 1 -0. 8 80.2- 8 3 . 8 large, part dissolved rhombs �
0
34/4- 1 , 2667.6 av. 1 3.7 2.3 1 .4 82.5 -4.2 -8.4 22.3 3% large siderite rhombs, with Ti oxides ....
range 1 0.9-20. 1 1 . 2-5 . 5 0.8-2.3 7 3 . 5-8 5 . 7 �
34/4- 1 , 2 6 6 8 . 6 av. 1 0. 8 2.3 0.4 86.6 -4. 7 - 1 1 .9 1 8.6 2% microcrystalline siderite in micas and �
"'
range 7 . 8- 1 4. 2 0.6-4.3 0.0-0.6 82.2-9 1 .2 as large rhombs
�
34/4- 1 , 2668.95 av. 1 2 .2 1.7 0.5 85.5 - 1 0. 1 - 1 6.9 1 3. 5 3% microcrystalline siderite rhombs,
range 7.3- 1 6.9 1 . 4-2.0 0.3-0.8 80.8-90.9 intergranular and in micas
34/4- 1 , 2670. 5 5 av. 1 4. 1 2.7 0.6 82.6 -3.9 -8. 1 22.6 1 4% siderite in mica and in kaolinized
range 1 1 . 2- 1 6 .2 1 . 3-3.8 0.4- 1 .0 80.4-8 6.5 pseudomatrix
34/4- 1 , 267 1 .8 av. 1 2. 5 5.8 1.1 80. 6 -3.8 -8.4 22.2 3% siderite rhombs, within dissolved
range 1 1 .9- 1 3 .2 5 . 3-8.2 1 .0- 1 . 1 80.4-80.8 feldspar

a-­
u.
Fig. 8. (A) Photomicrograph of displaced 'floating' quartz grains in a sandstone cemented by calcrete with drusiform
rims followed by blocky pore-filling calcite (p); crossed polars; (B) photomicrograph of coarse blocky calcite engulfing
and replacing kaolinite (k) within a large vug; crossed polars; (C) backscattered electron (BSE) image of a sandstone
with poikilotopic calcite cement replacing pore-filling kaolinite (k), dissolved feldspars and bright siderite remnants;
(D) BSE image of complex oscillatory precipitation of calcite and dolomite; rhombohedral dolomite (d l ) is replaced by
calcite (c l ), which is covered by a laminated microcrystalline dolomite rim (d2) with some calcite intercalations
(arrow), which is overgrown by ankerite (ak), followed by coarse pore-filling calcite (c2); (E) BSE image of a calcrete
and dolocrete with microcrystalline calcite replacing dolomite (ca), covered in large pores by collomorphic laminated
microcrystalline dolomite (d), followed by coarse pore-filling calcite (cb); (F) BSE image of a dolocrete with
microcrystalline dolomite (medium grey) as displacive rims and expanding mica flakes, followed by coarse pore-filling
calcite (bright).
 Sheet-flood sandstones in the Snorre Field
Fig. 9 . Chemical composition o f diagenetic carbonates
from the Lower and Middle Members of Lunde
Formation.
a few show irregular, weak outwards Fe increase.
The o 1 80p08 values of cal cite range from -12. 4o/oo
tO -4. 8o/oo, 0 1 3Cpos values from -5.6o/oo to + J . 6o/oo,
and 87Sr/8 6 Sr ratios between 0. 7 1 1127 and
0 . 7 1 1655 (Table 1).
Fluid inclusions are relatively rare in the poikilo­
topic cal cite and mainly single phased, which re­
mained so after freezing, indicating entrapment at
temperatures :o;; s o o c (Goldstein & R eynol ds,
199 4). A total of eight two-phase inclusions with
very small gas bubbles (i.e. high liquid/gas ratios)
yielded a very narrow range of homogenization
temperatures of 62-68 °C. These values were not
corrected for pressure. The precise melting temper­
atures of the first and last ice crystals were not
possible to obtain. The inclusions are rounded in
shape, :o;;6 Jlm in diameter, and display no fluores­
cence under UV light.
Dolomite and ankerite
Dolomite and ankerite together are second in abun­
dance (av. 8 volo/o) after calcite, and occur both as
cement in sandstones and in dolocretes and cal­
cretes (up to ;;;. so volo/o). In the dolocretes, dolomite
forms rims on detrital grains an d extensive inter­
granular cements composed of small euhedral to
subhedral rhombs (<5-20 Jlm), which are covered
and engulfed by coarse, blocky pore-filling calcite
(Fig. 8F). These dolomite crystals have displaced
the framework grains and caused the expansion of
mica flakes to several times their original size
67
(Fig. 8F). Some dolocretes contain crusts of micro­
crystalline dolomite with collomorphic textures
composed of alternating bands of dolomite and
calcite (Fig. 8 E).
Strongly ferroan dolomite does not occur as
discrete crystals in the dolocretes, but only as
overgrowths on non-ferroan dolomite (Fig. 8D).
Conversely, in sandstones dolomite and ankerite
occur as coarser, euhedral blocky crystals (20-
200 Jlm; Fig. I OA) randomly scattered in intergran­
ular pores (Fig. I OB) and covering, engulfing and
partially replacing clay pseudomatrix (Fig. I OC),
clay coatings (Fig. I OD) and, to a l esser extent, the
framework grains. Dolomite and ankerite displace
mica flakes along th e cleavages (Fig. 8F) an d engulf
kaolin and siderite crystals (Fig. I OC). The non­
ferroan dolomites reveal evidence of intracrystal­
line dissolution, whereas the ferroan dolomite and
ankerite that occur as discrete crystals and as
overgrowths on non-ferroan dolomite are not dis­
solved. The microcrystalline dolomite displays a
red to reddish -brown luminescence owing to low
Mn/Fe ratios (Fig. 8D,E). Strongly ferroan dolomite
and ankerite are non-luminescing. Microcrystalline
dolomite in dolocretes shows a homogeneously
bright yellow UV fl uorescen ce (Plate 1 F), whereas
the coarser blocky dolomite and ankerite show no
fluorescence.
The EMP analyses revealed that dolomite and
ankerite vary considerably in chemical composition,
particularly in terms of Fe content (FeC03 0 . 3-
=
36 mol%) (Table I ; Fig. 9 ). The microcrystalline do­
lomite rims an d crusts in dolocrete, as well as the
small rhombs replacing clays and occurring within
mica in the sandstones, are relatively Fe poor to
moderately ferroan (0. 3-7 . 8 % FeC03) (Table I ;
Fig. 9). However, both the small rhombs and the
large blocky crystals in sandstones are zoned, with Fe
increasing outwards to an ankeritic composition
(Fig. I OE). Furthermore, the dolomites are slightly
to moderately calcian (�53-62 mol% CaC03)
(Table I ) and have relatively low Mn content (0.2-
1.9 mol%).
Iron is replacing magnesium, which is revealed by
their strong negative correlation (Fig. 1 1 ) (r2 =
-0.97). Ca is also n egatively correlated with Fe in
ankerite an d ferroan dolomite, with > I 0 mol%
FeC03 (Fig. 12a) (r2 -0. 76). Dolomites with
=
< I 0 mol% FeC03 show no correlation between Ca
and Fe (Fig. 12a) (r2 -0. 16). These features indi­
=
cate that in the crystal structure of ankerite and
ferroan dolomite there is an increase in alternating
68 S. Morad et al.

Fig. 10. (A) Scanning electron micrograph of euhedral rhombohedral dolomite crystals associated with finely
crystalline kaolin in a sandstone; (B) BSE image of scattered dolomite rhombs (bright) in a sandstone with
intergranular kaolin and kaolinization of feldspar (lower centre) and mica (centre, with expanded edges); (C) BSE
image of dolomite rhombs (medium grey) engulfing and replacing kaolinized pseudomatrix (dark grey) and siderite
crystals (bright); (D) scanning electron micrograph of small dolomite rhombs on top of clay-coated grains; (E) BSE
image of zoned dolomite cement with external zones of ankerite composition (bright); (F) scanning electron
micrograph of poorly shaped, flattened m icrocrystalline siderite.
·
 Sheet-flood sandstones in the Snorre Field 69

50 (a) 62

40
D dolomite
• Fe-dolomite/ankerite
• siderite
60

58
···
> ·

#- I
(fl. 3 0 (')
0 56
• ••
I
"'

\
u
0 <t1 I&
() u
•
� 20 ....
54
• •

•
'
52
... 0
10
• 50
0 10 15 20 25 30 35 40

0 0 Dolomite
FeC03 %

0 20 40 60 80 1 00 • Fe-dolomile/ankerrte

FeC03 %
(b) 62

Fig. 1 1 . MgC03 mol% versus FeC03 mol% plot of

...
· 0

,z.-"r(
60
dolomite, ferroan dolomite, ankerite (>I 0% FeC03) and
siderites, showing a highly negative correlation. · o rSl

#-
58
�
(') I 't · • 0
layers of Ca atoms with decreasing FeC03 content, 0 56 DQJ 0
u 0 o DO @
and vice versa. This may reflect an increase in <t1 •
u Ooo 'O
dolomite crystal-structural ordering with increase in 54 •
0
il • 0 [j!J
Fe content. A weak negative correlation between Ca
52
and Mg (Fig. 12) (r2 -0. 52) for dolomites with
=

0
< l 0 mol% FeC03 indicates that the stoichiometric
50 molo/o CaC03 in dolomite structure is ap­ 10 15 20 25 30 35 40 45 50
proached as the amount of Mg is increased. Because MgC03 %
of the strong negative correlation between Fe and
Mg in the ankerite and ferroan dolomite with Fig. 12. Compositional plots of CaC03 mol% against
(a) FeC03 and (b) MgC03 mol% in dolomite, ferroan
> 10 FeC0 3 (Fig. l l ) (r2 -0.97), Ca and Mg con­
dolomite and ankerite (>I 0% FeC03). See discussion in
=

tent in these carbonates conversely show a positive the text.

correlation (Fig. l 2b) (r2 + 0.66).
=

The o 1 80PDB values of dolomite range from

- l l .9o/oo to - l . 5 o/oo, and the o 1 3CPDB values range
from -4. 8o/oo to +3. 9 o/oo (Table 1). The 87Sr/86Sr
ratio of dolomite in three dolocrete samples yielded
values between 0. 7 111343 and 0. 7 1 1448 (Table l ).
The fluid inclusions in the microcrystalline and
coarse blocky Fe-dolomite and ankerite cement are
extremely rare and too small (< 2 J..L m) to be used for
reliable determination of homogenization tempera­
tures.
Siderite
Siderite is most abundant (up to ::::: 14%) in fine­
grained sandstones rich in mica and clay pseudo­
matrix. It occurs as small subhedral or flattened
rhombs (< 3-15 J..L m) (Fig. l OF) that replaced the
detrital clays, and expanded as well as replaced the
mica flakes (Fig. l 3A). In coarser-grained sand-
stones, siderite also occurs as relatively large euhe­
dral rhombs (� 120 J..L m) around and within
dissolved and kaolinized feldspars and within
Fe-Ti oxides that have been dissolved and replaced
by euhedral anatase (Fig. l 3B). Both the finely and
coarsely crystalline siderites are partially dissolved,
preferentially in the crystal cores (Fig. 1 3C). These
intracrystalline dissolution pores are partially filled
by authigenic chlorite. Siderite is engulfed by other
carbonate cements (e.g. Fig. lOC), but in a few
sandstones microcrystalline siderite occurs as rims
around dolomite rhombs.
The siderites are moderately to strongly enriched
in magnesium (7 . 3-20.9 molo/o MgC03; av. 13. 7 %)
(Table l ; Fig. l l ), and display no crystal-chemical
zonation. The o 1 8 PDB values of siderite range from
- l 6 . 9 o/oo to -8 . l o/oo, and the 0 1 3Cp08 values from
-I 0. 1 o/oo to - l .2o/oo (Table I ).
Fig. 13. (A) BSE image of a biotite which has been expanded and partially replaced by microcrystalline siderite;
(B) BSE image of coarse, partially dissolved siderite crystals (s) surrounding an anatase rim (at) after a dissolved
detrital Fe-Ti mineral (partially filled by dolomite and siderite); the interstitial spaces are filled by feldspar- and
pseudomatrix-replacing kaolinite (dark) and calcite (ca) with bright pyrite framboids; (C) BSE image showing a
dissolved siderite crystal-the dissolution void is partially filled by chlorite; (D) BSE image of a sandstone with
extensive kaolin replacement of feldspars, pseudomatrix and mica; there are partially dissolved feldspars and
Ti-minerals (bright crown); (E) scanning electron micrograph of kaolinite vermicules made by thin, irregular-edged
platelets which replaced clay pseudomatrix; (F) scanning electron micrograph of dickitized kaolinite vermicules-thick,
euhedral dickite crystals grew between and replaced almost totally thin kaolinite platelets (arrows).
 Sheet-flood sandstones in the Snorre Field
Clay minerals
The diagenetic clay minerals are mainly kaolin and
chlorite. Kaolin replaces feldspar, mica and
pseudomatrix, and fills intergranular pores in the
sandstones (Fig. 13D) as well as vugs, burrows and
root moulds in the calcretes and dolocretes
(Fig. 8B). SEM examination revealed that both
kaolinite and dickite are present. Kaolinite occurs
as thin, irregular-edged platelets that are stacked in
vermicular aggregates with delicate texture, indicat­
ing an in situ authigenic origin (Fig. 13E). Kaolin­
ized micas display the typical expanded texture
(Figs 1 OB and 13D); kao1inization is more intensive
along terminations of the mica flakes. Compared
with kaolinized feldspars and pore-filling kaolinite,
kaolinite vermicules which have replaced pseudo­
matrix reveal less intercrystalline microporosity
(Fig. 13E) and contain abundant microcrystalline
remnants of precursor smectitic clays as well as iron
and titanium oxides (Fig. 1OC).
Dickite is distinguished from kaolinite by SEM
and XRD (X-ray diffraction) analysis. Dickite oc­
curs as euhedral monoclinic blocky crystals and has
typical XRD reflections at 4. 13 A, 3. 79 A, 2. 5 0 A
and 2.33 A in randomly oriented samples. Dickite
occurs together with pervasively etched remnants of
kaolinite, from which it inherited the vermicular
habit (Fig. 13F). Dickite crystals are much thicker
(:::::; :;;. 1-8 Jlm) than kaolinite (< 1 Jlm), and show no
etching. Morad et al. (1994) concluded that these
textural features are indicative of kaolinite transfor­
mation into dickite via small-scale dissolution­
reprecipitation. Kaolinite is engulfed (and thus
postdated) by calcite, dolomite and siderite, both in
the sandstones and in the calcretes and dolocretes.
Kaolinite totally engulfed by these carbonates is
well preserved or only slightly dickitized. Dickite is
covered by, and hence predates, authigenic chlorite.
Both kaolinite and dickite are engulfed by quartz
overgrowths and coarsely crystalline calcite.
Chlorite replaces kaolinite, clay pseu domatrix,
infiltrated clays, micas and heavy minerals. It oc­
curs as rims composed of platelets oriented perpen­
dicularly to grain surfaces. The rims were formed by
replacing infiltrated smectitic clay coatings, which
were originally oriented tangentially to grain sur­
faces (Fig. 14A) (see Moraes & De Ros, 1990).
These infiltrated clays were presumably introduced
into the vadose zone of alluvial continental sedi­
ments under semi-arid conditions by episodic
floods (Walker et a/., 1978; Moraes & De Ros,
71
1990). Typically, mechanically infiltrated clays are
originally detrital smectites formed under semi-arid
weathering conditions (see Keller, 1970; Walker et
a/., 1978). This is evidenced by the dominance of
smectitic clays in the mudstone samples and in the
mud intraclasts. Infiltrated coatings and derived
chloritized rims are conspicuous, particularly in
medium-grained sheet-flood sandstones (up to
2. 7 volo/o).
Transformation of smectitic coatings into chlo­
rites has occurred through an intermediate stage
honeycombed aggregates of mixed layers of
chlorite/smectite (CIS). Chloritization is incom­
plete, leaving remnants of smectite and CIS clays
beneath the chlorite platelets (Fig. 14A). Chlorite
has also pseudomorphically replaced biotite fl akes
and vermicular kaolinite aggregates, with preserva­
tion of the original stacked habit (Fig. 14B).
Illite is a minor diagenetic constituent occurring
as fibres closely associated with chloritized pseudo­
matrix and infiltrated coatings. The presence of
honeycombed mixed-layer illite/smectite (liS)
might indicate that illitization occurred via this
intermediate stage.
Quartz and feldspars
Quartz cement forms on average 2.6 %. , but is
abundant (up to 9 . 7 %) in sandstones poor in detrital
and authigenic clays. Quartz occurs both as over­
growths on detrital quartz and as prismatic out­
growths in the presence of relatively thick infiltrated
clay coatings or authigenic clay rims (Fig. 14C).
Quartz cements are covered by, but also cover and
engulf, diagenetic carbonates, kaolin and chlorite
(Fig. 14D), suggesting a recurrent precipitation dur­
ing burial diagenesis.
Detrital plagioclase and K-feldspar grains are
albitized, distinguished by the typical petrographic
and chemical features characterized for the Upper
Lunde Member by Morad et a/. ( 1990). Albitized
detrital feldspars contain dissolution voids and are
untwinned or show irregular blocky to tabular
extinction. Detrital plagioclases are far less calcian
(An < 10 mol%) than those analysed in the Upper
Lunde (An � 28 mol%) by Morad et a/. ( 1990). The
albitized K-feldspar grains are composed of a larger
number of smaller lath-like albite crystals, arranged
parallel to each other in two directions that presum­
ably reflect traces of cleavage planes.
On average, diagenetic albite replaced 6.2 bulk
rock-volume o/o, corresponding to :::::; 2/5 of the detri-
72 S. Morad et a!.

Fig. 14. (A) Scanning electron micrograph of chlorite platelets which grew perpendicularly on coatings of infiltrated
smectitic clays (background); (B) scanning electron micrograph of chlorite which replaced pseudomorphically and
pervasively kaolinite vermicules; small remnants of corroded siderite crystals (s) which were probably involved in the
reaction; (C) scanning electron micrograph of discontinuous quartz overgrowths and prismatic outgrowths on top of
clay-coated grains; (D) scanning electron micrograph of late quartz outgrowths which engulf chlorite; (E) BSE image of
an anatase rim (at) around dissolved Fe-Ti grain filled by zoned dolomite (dl); showing intergranular kaolin (dark),
siderite (s) as well as finely crystalline and framboidal pyrite (py); (F) BSE image of coarse barite (white) engulfing and
replacing kaolinite (dark) within a vugular pore rimmed by microcrystalline dolomite (dl).
 Sheet-flood sandstones in the Snorre Field
tal feldspars which survived early dissolution an d­
kaolinization. Plagioclase grains were more affected
by albitization (3. 1 relative to 2. 1 o/o remaining
detrital plagioclase) than detrital K-feldspar (2.6
relative to 7 . 3% remaining K-feldspar). However,
the greater abundance of K-feldspar may be due to
preferential elimination of detrital plagioclase by
other earlier diagenetic processes, such as dissolu­
tion, kaolinization and replacement by carbonates.
Albite (Ab �100 mol%) also occurs as small (< 1-
30 Jlm) discrete crystals associated with chloritic
clays, and is engulfed by late quartz cements. The
detrital K-feldspar grains show overgrowths
(,;2. 3%) with ragged or sawtooth-like outline. In
some cases the overgrowths occur around mouldic
pores that resulted from the post-overgrowth disso­
lution of detrital K-feldspar cores. This is probably
related to the near end-member composition of the
overgrowths, which renders them more resistant to
dissolution and albitization than the detrital core.
Other diagenetic constituents
Hematite occurs sparsely in the fine-grained flood­
plain sediments, as tiny pigments that are either
evenly distributed in the sediment or closely associ­
ated with infiltrated clay coatings around frame­
work grains, and as alteration products of detrital
Fe-bearing minerals such as Fe-Ti oxides.
Diagenetic Ti-oxides are more abundant than
iron oxides in sandstones (up to 2.6 %). They occur
as local aggregates of bipyramidal anatase crystals,
apparently formed by the complete alteration of
detrital Fe-Ti oxides, wh ich are commonly associ­
ated with siderite and ankerite (Figs 13B an d 1 4E)
(see Morad, 1988) or are scattered in chloritized
pseudomatrix and biotite.
Pyrite averages 0.2 volo/o, and only in a few
samples forms up to 1.3 volo/o. It shows two occur­
rence habits: (i) fine crystals (< 2 Jlm) or framboids
scattered in kaolinized or chloritized detrital clays
and micas, or engulfed by coarse carbonate cements
(Fig. 15E); and (ii) coarsely crystalline (up to
�200 Jlm across), intergranular replacive cement.
Barite occurs as scarce, large crystals (up to
2 mm) filling vugs and cracks an d engulfing as well
as replacing kaolinite and carbonate cements in
dolocretes an d calcretes (Fig. 14F). In the sand­
stones, barite occurs as a few poikilotopic and small
crystals which cover, and thus postdate, chlorite
rims aroun d framework grains.
Some epidote, monazite an d zircon grains show
73
reddish-brown envelopes of solid bitumen which
were polymerized from oil by the radioactive emis­
sion of the grains. Their presence within present­
day water zones probably indicates that either the
original oil column was thicker than at present, or
that emplacement of oil was gradual, from the base
to the top of the structure.
DISCUSSIO N
Paragenetic sequence and overall
diagenetic evolution
The relative timing of the main diagenetic processes
in Lower and Middle Lunde Members is presented
schematically in Fig. 15. Because of the complex
diagenetic patterns and burial histories (see Fig. 5),
a precise timing cannot be achieved for all the
diagenetic effects observed. N evertheless, the volu­
metrically important diagenetic processes occurred
under early, n ear-surface conditions. Products of
the first burial phase in the Jurassic, of the tela­
diagenesis during the Kimmerian uplift (Late
Jurassic-Early Cretaceous) and of the second burial
diagenesis phase (Middle Cretaceous to Recent) are
volumetrically less significant than during eodiagen­
esis.
Eodiagenesis and telodiagenesis
At near-surface conditions the eogenetic and teloge­
netic processes and products are strongly controlled
by several interrelated parameters. These include
the chemical composition of meteoric waters, cli­
mate, hydrological setting, rate of deposition versus
erosion, as well as detrital composition, permeabil­
ity, biological activities and organic-matter content
in the sediments and soil horizo11s. What follows is
a discussion of the role of eodiagenesis and tela­
diagenesis on the overall diagenetic evolution of the
Lower and Middle Lunde Members.
Silicates
The climatic conditions and episodic flooding that
characterized the depositional setting of the Lunde
enhanced the infiltration of suspended clay particles
and the formation of coatings around framework
grains in the sandstones. Clay minerals formed by
weathering processes in the hinterland under arid to
semi-arid climatic conditions would be expected to
74 S. Morad et a!.

240 200 1 50 1 00 so 0
a p proxi m a t e
time (Ma)

stage � odiagenesi� mesodiag. 1 l l

telo dia g . mesodiagenesis 2

1-
infiltrated coatings

I I I I
iron oxides -
I I I I
- _, _,
r- r--i
dissolution

I
-
kaolinite
I I r-
siderite - 1- I
I
dolom ite
__,. - I
I
I -I - - -
I
a n k erite

,_
I I
calcite 1

K-fe l d s p a r
I
- I I
quartz
I - II- -
overgrowth$

- I __;tg=-
mechanical compaction \ -I --
chemical compaction
I I - I , _ - -
d i c kit e
I I I ?-
c h l o rite
I I I --
albite I I I _ _,

titanium oxides I - I I �? , _ Fig. 15. Simplified paragenetic

barit e I I -- I I , _
sequence of the main diagenetic
processes in the Lower and Middle
I I I I Lunde sandstones.

be dominated by smectite, as were the infiltrated 34/4-1 (358 m and 24 m, respectively}, pervasive
clays, clay pseudomatrix and mudstones. The near­ calcite and, to a lesser extent, dolomite dissolution
surface eogenetic interaction of meteoric waters and creation of secondary porosity was not accom­
with detrital minerals resulted in the formation of panied by kaolinite formation. The meteoric waters
kaolinite at the expense of detrital feldspars and were thus aggressive towards detrital silicates imme­
micas. Eogenetic kaolinite was conceivably formed diately below the unconformity, but apparently re­
during periods (either seasonal or several years' mained undersaturated only in relation to the car­
variability of climate) of increased rainfall, which bonates at greater depths. Meteoric water incursion
were followed by dry conditions that enhanced the indicates the presence of considerable hydraulic
formation of calcretes and dolocretes. Evidence of head, as well as exposure of the Lunde above sea
kaolinite formation during eodiagenesis includes level.
the engulfment of kaolinite and dissolved and Grains and cement dissolution and kaolinization
kaolinized feldspar by calcretes and dolocretes. were conceivably enhanced by the humid cli matic
Our results indicate that kaolinite distribution in conditions that prevailed in the area during the Late
the Lunde Formation is not strictly controlled by the Jurassic to Early Cretaceous. In contrast to eoge­
Kimmerian uplift and erosion. This is due partly to netic kaolinite, telogenetic kaolinite replaces com­
the formation of kaolinite during eodiagenesis and pactionally deformed micas and clay pseudomatrix.
partly to the strong relationship between kaolinite Evidence indicating that this dissolution was tela­
abundance and detrital composition of the sand­ genetic and not related to the second mesogenetic
stones, particularly the original amounts of feldspars phase (Middle Cretaceous to Recent) (see Figs 5
and mud intraclasts. Pervasive kaolinite formation, and 15) includes the presence of later undissolved
coupled with dissolution of calcite and dolomite ce­ euhedral calcite and dolomite cements that post­
ments, has been substantial in well 34/4- 1. In well date q uartz overgrowths, chlorite rims and feldspar
34/7 -A-3H sandstones, the top of which was buried albitization. The presence of randomly scattered
deeper below the unconformity than that of well patches of carbonate cement left by telogenetic
 Sheet-flood sandstones in the Snorre Field
dissolution promoted the preservation o f a loose
sandstone framework and of telogenetic secondary
porosity during the second burial phase. Inhibition
of compaction in these sandstones is evidenced by
the presence of undeformed ductile grains such as
micas.
Kaolinite is dominantly replacing feldspar, mud
intraclasts and pseudomatrix. Replacement of the
pseudomatrix indicates that kaolinite formation
occurred, at least partially, by telodiagenesis during
the Kimmerian uplift, subsequent to compaction
caused by the first burial phase (J urassic) (Figs 5
and 1 5).
Carbonates
The high intergranular volume (IGV) in carbonate­
cemented Lunde sediments indicates an early,
pre-compactional timing. Siderite was among the
first carbonates to precipitate, after feldspar disso­
lution and kaolinization and calcite and dolomite
cementation. According to Mozley ( 1 989), the
relatively high Mg content (av 1 3. 7 mol%) in the
siderites should indicate precipitation from marine­
influenced pore waters. However, no marine­
related sedimentary facies were detected in the
sequence, and it is believed that the sea might have
been up to hundreds of kilometres away from the
Lunde depositional sites during the Ladinian to
Norian (Steel & Ryseth, 1 990; N ystuen & Fait,
1 99 5). Therefore, the elevated Mg content in
siderite is considered primarily to reflect high aM8> +
related to alteration of the detrital magnesian
minerals, such as biotite, heavy minerals and smec­
titic mud intraclasts, by infiltrated meteoric pore
waters. Indeed, siderite is associated with dissolved
and kaolinized micas and clays, which perhaps
indicates that even iron, as well as suitable pH
values, were provided by these altered silicates
(see Boles & Johnson, 1 984; Morad, 1 990). Such
siderite is more enriched in Mg than that in the
open pores, which supports our hypothesis.
Cementation by Fe-poor calcite and dolomite
occurred recurrently during eodiagenesis, as indi­
cated by the mutual partial replacement and by the
presence of alternating rims of both minerals. Dis­
tinction between vadose and phreatic cementation
is not easy (see Purvis & Wright, 1 99 1 ; Spot! &
Wright, 1 992). However, the samples lack typical
vadose features, such as meniscus and pendant
cements, rhizocretions and glaebules (Esteban &
Klappa, 1 983; Arakel & McConchie, 1 982). It is
75
therefore believed that cementation was accom­
plished in the phreatic zone. Additional evidence
for this postulation is the coarse crystalline texture
and the presence of crystal-chemical zonation in the
carbonate cements. Moreover, the microcrystalline
carbonate cements display a homogeneous lumines­
cence (see Plate 1 B,D) which reflects periodically
homogeneous pore-water compositions more typi­
cal of the phreatic zone. Conversely, vadose cal­
cretes and dolocretes are expected to have patchy
variations in luminescence as a result of periodic
influx of waters into the sediments. The dominance
of phreatic over vadose cementation may be due to
extensive alluvial reworking and poor vegetation.
Calcite cement in calcretes shows lower Mn
and/or Fe contents than do the pre-compactional,
poikilotopic calcite cements, indicating formation
under generally more oxidizing conditions. How­
ever, the presence of small-scale CL zonations in
eogenetic, vug-filling calcite is related to fluctua­
tions in aMn>+ in the pore waters, which probably
took place in the sub-oxic phreatic zone. Microcrys­
talline dolomite in the dolocretes is characterized
by dark red luminescence and low Mn and Fe
contents, suggesting likewise more oxic conditions
than those of the Fe-rich dolomites in the sand­
stones. The bright fl uorescence of microcrystalline
calcite and dolomite cements in calcretes and dolo­
cretes is attributed to adsorbed organic matter from
microbial remnants (see Dravis & Yurewicz, 1 98 5).
The influence of microorganisms, such as bacteria,
lichens and algae, in calcrete and dolocrete precip­
itation is indicated by the preservation of bacterial
and algal cell remnants and calcified filaments in
these deposits (see Phillips et a!., 1 98 7 ; J ones, 1 988;
Folk, 1 99 3).
The alternating bands of calcite and dolomite in
the calcretes and dolocretes resemble those formed
by mixing between marine and ]lleteoric waters
(Ward & Halley, 1 98 5 ; Machel & Mountj oy, 1 986;
Humphrey & Radjef, 1 99 1 ; Morad et a!., 1 992).
However, there is no facies evidence of marine
influence on the studied sequence, and diagenesis
was thus fully meteoric. Therefore, the alternating
bands are attributed to episodic fl uctuations in the
amounts of rainfall and dilution of the pore waters,
and shifting between dolomite and calcite equilib­
rium fields. Dolomite was formed during dry peri­
ods of increase in the Mg/Ca ratio of pore waters
caused by water-sediment interaction (e.g. alter­
ation of biotite and mud intraclasts), coupled with
evaporation. This is supported by the higher Sr
76 S. Morad et al.
contents in dolomite (up to :::; 7 00 ppm) compared
with calcite (up to :::; 2 70 ppm). Similar ranges of
87Sr/8 6Sr ratios in dolomite and calcite, however,
indicate a similar source of strontium. Watts ( 1 980)
observed similar alternating bands of calcite and
dolomite in pedogenic calcretes from the Kalahari
Desert, which he attributed to mixing between fresh
phreatic waters and more saline, vadose waters.
Dramatic fl uctuations in the near-surface geochem­
ical environment due to climatic changes would
explain the close succession and sometimes alterna­
tion of kaolinite, siderite, dolomite and calcite.
The sources of eogenetic calcite and dolomite
cements in alluvial sediments of the Lunde and
similar successions elsewhere are often not immedi­
ately clear. This is particularly true when the strata
are not associated with carbonatic bedrocks or
bioclasts, and there is no evidence for the presence
of extraformational carbonate rock fragments.
When no such carbonate sources are visible, cement
may be derived from rainwater, airborne carbonate
dust, and from the breakdown of calcian silicates
and Ca-bearing plants (e.g. Goudie, 1 983; R eeves,
1 976). In the U pper Lunde sandstones much of the
detrital plagioclase, which is moderately calcian (An
.;;; 2 8 mol%; Morad et al., 1 990), was dissolved and
kaolinized during eodiagenesis, and was thus a
likely source of calcium ions. Magnesium as well as
calcium was also derived from the kaolinization of
mud intraclasts. The kaolinization of detrital biotite
was an additional source of magnesium as well as
iron. Moreover, microcrystalline carbonate intrac­
lasts derived from the erosion and redeposition of
palaeosol sections are common in the studied rocks.
These intraclasts were probably important sources
for the syncompactional to mesogenetic carbonate
cements. Increases in ionic concentrations of
groundwaters, and enhanced carbonate precipita­
tion, may have subsequently occurred by evapora­
tion under the overall semi-arid climatic conditions
(see White et al., 1 963).
Mesodiagenesis: role o f eogenetic minerals
and temperature
The mesogenetic reactions in the studied rocks were
largely controlled by increases in temperature and
by the patterns of eogenetic and telogenetic modifi­
cation. Syncompactional to early mesogenetic mod­
ifications were accomplished during two burial
phases (see Figs 5 and 1 5), yet assignment of at least
some of the diagenetic events to a specific burial
phase might be difficult. Carbonates formed during
these modifications include: (i) calcite precipitated
as euhedral blocky crystals and overgrowths on
eogenetic cements; and (ii) ferroan dolomite an d
ankerite cements precipitated as thin zones around
early non-ferroan dolomite and siderite, and as
discrete, zoned blocky crystals in sandstones. The
blocky calcite was affected by telogenetic dissolu­
tion, and was thus mainly formed during the first
burial phase (Jurassic) (see Figs 5 and 1 5), whereas
calcite overgrowths, Fe-dolomite and ankerite dis­
play no signs of dissolution, which indicates that
they were formed during the second burial phase
(Middle Cretaceous to R ecent) .
The important mesogenetic silicates formed as a
consequence of considerable increases in tempera­
ture during the second burial phase (Fig. 1 5) in­
clude dickite, albite, chlorite and q uartz. Dickite is
formed almost exclusively by the replacement of
eogenetic kaolinite, a process that occurs at :::; 80-
1 30 " C (Ehrenberg et al. , 1 993; McAulay et al.,
1 993; Morad et al. , 1 994). Albitization of plagio­
clase occurred apparently simultaneously with dic­
kite formation . This process can result in the forma­
tion of minor amounts of kaolin and calcite owing to
the presence of calcium and excess aluminium in the
detrital plagioclase, compared with authigenic albite
(Morad et al., 1 990). We are, however, unable to
distinguish precisely these particular calcite and
kaolin byproducts from the abundant earlier-formed
kaolinite an d calcite cements. Nevertheless, kaolin
booklets are commonly closely associated with albi­
tized plagioclase, and some of the early calcite
cements display minor overgrowths that might
be formed by mesogenetic calcite addition as a
consequence of plagioclase albitization. As the
provenance of Lunde sediments has not changed
considerably with time, the low amounts and lesser
extent of anorthite solid solution of plagioclase
compared with those of the U pper Lunde (Morad et
al., 1 990) is attributed to a more pervasive albitiza­
tion and elimination of particularly the calcian
plagioclases in the Lower an d Middle Lunde
Members.
Q uartz overgrowths are absent to minor in sand­
stones totally cemented by calcretes and dolocretes,
but quite common in sandstones cemented by
syncompactional carbonate cements, suggesting
that part of the overgrowths formed during early
compaction. Mesogenetic q uartz occurs as over­
growths and outgrowths that engulf earlier-formed
minerals, including dickite, albite and chlorite.
 Sheet-flood sandstones in the Snorre Field 77

Eogenetic feldspar dissolution and kaolinization are ORIGINAL POROSrTY DESTROYED BY CEMENTATIOI'< ( % )

potential sources for the early-burial quartz over­

40 ...J
growths. Determination of the source for quartz <(
(.)
35
z
outgrowths and overgrowths is beyond the scope of <(
I
this study. � (.)
w �
;;:; 30 ::;: z
The marked differences in diagenetic mineralogy ::;: >-
"' 0
::0 0 f5
between sediments of the two wells studied indicate 6 25 w <(
> a.
variations in depositional facies and perhaps dif­
ferent diagenetic evolution pathways. In well 34/7 -
a:
:'5
::0
z
20 � ::;:
0
(.)
w ...J
0 <(
(.)
A-3H sandstones, chloritization occurs in the <(
a: 15 � :;;:
CJ (/) w
uppermost Middle Lunde Member and continues a: 0 I
w a: (.)
into the Upper Lunde sandstones. Chlorite covers, � 10
1i: 0
...J z
<(
and hence postdates, albite and dickite. The smaller �
amounts of diagenetic kaolinite in well 34/7 -A-3H l'i
a:
are attributed to less significant telogenetic dissolu­ 0
0
tion of silicates because of the presence of Middle 10 15 20 25 30 35 40
Lunde sandstones at a greater depth below the CEMENT (%)

Kimmerian unconformity than the sandstones of

well 34/4-1 (358 and 24 m, respectively).
Although both cores display floodplain mud­
stones and sheet-flood sandstones, the Middle
Lunde samples in 34/7 -A-3H are dominated by
fine- to medium-grained sheet-flood sandstones.
The elevated initial porosity and permeability of
these sandstones compared with the mudstones,
siltstones and very fine to fine-grained sandstones
has perhaps allowed larger amounts of mechanically
infiltrated clays, which are preserved as smectitic
coatings and/or transformed into chloritic or CIS
and liS rims in the sandstones.
Porosity evolution: compaction versus cementation
and reservoir implications
The overall large intergranular volume (IGV; av.
34.5%) and low packing values (average packing
proximity index, Pp of Kahn, 1956; 25. 8%) indi­
= related to chlorite rims covering the grains and
cate that cementation occurred early and limited
the compaction during subsequent burial. Near­
f
surface cementation is evident iom grain displace­
ment, the presence of undeformed ductile grains
such as micas within the cement, and the occur­
rence of intraclasts containing carbonate cements
similar to those in the sandstones. The plot of IGV
versus cement vol% in sandstones with IGV < 40%
reveals that cementation was a much more impor­
tant agent of porosity destruction than compaction
(Fig. 16).
The low values of petrographic macroporosity
(av. 1 1 . ( %), and petrophysical porosity (av. 17. 4%;
range 0.05-28. 4%) and permeability (av. 42.6 mD;
range < 0.01-672 mD) are due not only to compac-
Fig. 16. Plot of intergranular volume (%) versus cement
(%) for Lower and Middle Lunde sandstones with an
intergranular volume .;; 40% (see Houseknecht, 1 98 7).
tion and carbonate cementation, but also to diage­
netic clay mineral. The large amounts of authigenic
kaolin and chlorite have substantially reduced the
permeability, but to a lesser extent the porosity,
considering the abundant intercrystalline micro­
porosity of these clays (see Hurst & Nadeau, 1995).
The eogenetic and telogenetic dissolution of feld­
spars and mud intraclasts resulted in the redistribu­
tion, rather than enhancement, of porosity (see
Giles & deBoer, 1990), as it was accompanied by
the formation of abundant kaolinite in the inter­
granular pores adjacent to the dissolved feldspars, a
process which has resulted in a net loss of perme­
ability. Further deterioration of permeability is
occluding pore throats. Secondary porosity is best
developed in sandstones as a result of the teloge­
netic dissolution of carbonate cements, which was
not accompanied by the precipitation of intergran­
u lar kaolinite.
Sandstones with potentially better porosity pres­
ervation are characterized by: (i) coarser grain size
and better sorting; (ii) lower tendency to host
extensive eogenetic carbonate cement than the finer
sediments, which are more represented by well
34/4-1 samples; and (iii) chlorite rims evolved from
the infiltrated clay coatings, which are more abun­
dant in coarse-grained sands which inhibited pre­
cipitation of pore-occluding quartz and carbonate
cements.
78
Diagenetic conditions of carbonate cementation:
constraints from isotopes and fluid inclusions
The oxygen and carbon isotopic compositions
(Table 1 ; Fig. 1 7), the loose grain packing and the
large amounts and displacive growth h abits of the
carbonate cements indicate a phreatic, precompac­
tional origin. Small amounts of ferroan dolomite,
ankerite and calcite were, however, formed during
mesodiagenesis.
Oxygen isotopes
In order to use o 1 80 in discussing the physicochem­
ical conditions under which carbonate precipitation
occurred, it is important to understand the o 1 80
constraints of the pore waters involved. A precise
knowledge of the 8 1 80 compositions of these pore
waters is difficult to achieve. N evertheless, as the
Middle Lunde sediments were situated at a palaeo­
latitude between �20 and 26"N, the 0 1 80sMOW Of
pore waters during eodiagenesis can be inferred
from the global meteoric water map to be � -6%o to
-4%o (av. -5o/oo). The absence of authigenic evapor­
itic minerals, Mg-rich clay minerals (e.g. sepiolite)
and silcrete (see Thiry & Milnes, 1 99 1 ; Spot! &
Wright, 1 992), coupled with the presence of abun­
dant eogenetic kaolinite, suggests that no great
fluctuations in the 0 1 80water values have occurred
due to evaporation. The present-day formation
waters in the Lunde sediments of the Snorre Field
0
D
• calcite
D dolomite
<> siderite D DO D I
-5 D
r = 0.84974
• i="
m D
0
�
c.. -10
0
"'
<.0
-15
<>
-20
-1 2 -9 -6 -3 0 3
o13C PDB
Fig. l7. o1 3CPDB versus 01 80PDB plot of representative
diagenetic carbonates from Lower and Middle Members
of Lunde Formation.
S. Morad et a!.
have an average o 1 80sMow value of c. -3%o (Ege­
berg & Aagaard, 1 989), indicating a relatively mod­
erate, burial-diagenetic enrichment in 1 80 relative
to the depositional meteoric waters.
If we consider that precipitation of eogenetic
carbonates occurred from unmodified meteoric wa­
ters, it is possible to deduce the precipitation tem­
perature. However, 8 1 80 values of the siderite
(-8. 1 o/oo to - 1 6. 9%o) (Table 1 ) are very low consid­
ering its postulated near-surface origin, and would
indicate unreasonably high precipitation tempera­
tures of �45- 1 20 " C (Fig. 1 8). The higher tempera­
tures exceed even the maximum burial temperature
achieved by the sediments, which is l OO " C. More­
over, siderite is a mineral known not to undergo
recrystallization and isotopic re-equilibration (Mat­
sumoto & Iij ima, 1 9 8 1 ; Curtis & Coleman, 1 986;
Faure et a!., 1 99 5 ; Morad et a!., 1 996). Thus,
assuming a near-surface eogenetic origin (�25 " C),
which is indicated by petrographic examination,
siderite precipitation would have occurred from
pore waters with anomalously low o 1 8sMow values
(- 1 8%o to -9%o) (Fig. 1 8). This in turn contradicts
the assumed 8 1 80 compositions of meteoric pore
water (-6%o to -4o/oo) based on palaeolatitude.
Morad and De Ros ( 1 994) and Morad et a!. ( 1 996)
have proposed that anomalously low 8 1 80 values in
early precompactional carbonates might result from
120
9' 80
40
-20 -15 -10 -5 0
o 1 8 Q(SMOW)
Fig. 18. Range of temperature and isotopic composition
6
of the pore fluids constrained for the precipitation of the
analysed siderites (0' 80PDB - 1 6. 9%o to -8. 1 %o). Field
=
marked for 01 80water between -5 and -3%o SMOW; bar
shows the 01 80watcr that would be required to precipitate
siderite at 25 "C (fractionation equation after
Rosenbaum & Sheppard, 1 986).
 Sheet-flood sandstones in the Snorre Field
low-temperature interaction between pore waters
and chemically unstable vol caniclastic sediments.
No significant amounts of such sediments were,
however, detected in the Lower and Middle Lunde
Members. Another possible alternative mechanism
for the strong enrichment of siderite in 1 6 0 would
be microbial oxygen fractionation by preferential
bacterial metabolism of light oxygen in organic
matter and the production of 1 6 0-enriched HC03-
(Duan et al., 1 995).
The � 1 80 val ues of precompactional calcite ce­
ments (- 1 0.2%o to -4.8o/oo) (Table I ) indicate pre­
cipitation at temperatures of � I 0-30 "C, from
meteoric pore waters with � 1 80 of -5o/oo (Fig. 1 9).
The range obtained may reflect either true changes
in mean annual air temperature or variations in
� 1 80 of pore waters due to fluctuations in dry and
wet periods (see Cerling, 1 9 84). Heavy rainfall, for
example, may have lower � 1 80 values than mean
annual rainfall (Vogel & Van U rk, 1 975).
Syncompactional, mesogenetic euhedral calcite
has � 1 80 values between - 1 2 .4o/oo and -6.6o/oo
(Table I ) and higher Mn contents than eogenetic
cal cite (r2 + 0.76) (Fig. 2 1 b). Assuming that pre­
= eogenetic calcite and dolomite are in line with the
cipitation occurred from moderately evolved mete­
oric pore waters with an average � 1 80 composition
similar to that of the present-day average formation
water (-3o/oo), these late calcites must have precipi­
tated at temperatures of �30-7 o · c (Fig. 1 9). The
maximum precipitation temperature of mesoge-
P' 40
i='
20
0 ·15
-10 -5 0 5
1) 1 BQ( SMOW)
Fig. 19. Range of temperature and isotopic composition
of the pore fluids constrained for the precipitation of the
analysed calcites {ll' 80pos - 1 2.4%o to -4.8%o). Field
= Field marked for 01 80watcr between -5 and -3%o SMOW
marked for 01 80water between -5 and -3%o SMOW
(fractionation equation after Friedman & O'Neil, 1 977).
79
netic calcite was, however, probably higher than
7 o · c, considering that these burial cements are
contaminated by eogenetic calcite.
The fluid-inclusion microthermometry supports
the above postulations that precompacti,anal calcite
precipitated at � �5o·c, whereas syncompactional
to early mesogenetic calcite yielded ho mogeniza­
tion temperatures between 62 and 6 8 · c. The latter
val ues are close to those measured for poikilotopic
calcite by Saigal & Bje�rlykke ( 1 98 7 ) in other North
Sea sandstones. The narrow range of homogeniza­
tion temperatures and their agreement w ith petro­
graphic and oxygen isotopic data indicate that the
fluid inclusions are primary. The absence of fluores­
cence in the inclusions under UV light indicates the
absence of hydrocarbon, and that cementation oc­
curred prior to the emplacement of oil in the Lunde
sandstones of the Snorre Field.
Dolocretes have � 1 80p08 values between -4.3o/oo
and - 1 . 5o/oo (Table 1 ), which indicate precipitation
temperatures of 1 5-30" C from meteoric waters with
assumed average � 1 80sMow val ue of -5o/oo (Fig. 20).
The similar ranges of precipitation temperatures for
petrographic observations, which reveal a con­
temporaneous, alternating precipitation of these
carbonate cements. Mesogenetic dolomite and
ankerite (8 ' 80PDB = - 1 1 . 9o/oo tO -4.6o/oo (Table 1 )
were formed at temperatures o f �40-9o · c, assum-
P' 50
i='
30
10
·10 -5 0
o 1 BQ (SMOW)
Fig. 20. Range of temperature and isotopic composition
of the pore fluids constrained for the precipitation of the
analysed dolomites/ankerite (ll' 80 - 1 1 . 9%o to - 1 . 5%o).
=
(fractionation equation after Rosenbaum & Sheppard,
1 986).
80 S. Morad et a!.

ing precipitation from waters with the average o 1 80
value of -3%o of present-day formation waters
(Fig. 20). The average Fe content in dolomite has
increased with increases in temperature, which is
confirmed by its strong negative correlation with
o 1 8 0 values (r2 -0.88) (Fig. 2 1 a). N either calcite
= Therefore, sandstones have both high IGV and high
nor siderite displays such a relationship. The Mn
content in dolomite (but not siderite) shows a much
weaker correlation with o 1 80 v alues (r2 -0.32)
(Fig. 2 1 ) than for calcite.
A plot of IGV v ersus o 1 80 values normally
displays a positive correlation, reflecting a succes­
sive decrease in available pore space with increasing
burial depth and temperature (e.g. Boles & Ram­
seyer, 1 98 7). The correlation between these two
parameters in the analysed sequence is, however,
esis. In sandstones containing small amounts of
patchy eogenetic cements the uncemented areas
were subjected to normal compaction and elimina­
tion of available pore space during burial, whereas
the cemented areas remained loosely packed.
o 1 80 values if early cementation was extensive, or
high o 1 80 and low IGV values if early cementation
= was only partial and patchy.
Strontium isotopes
The Sr isotopic ratios of the analysed calcites and
dolomites display a relatively narrow range and are
typical of meteoric water diagenesis, being more
enriched in 87Sr (87Sr/8 6Sr 0. 7 1 1 1 27-0. 7 1 1 65 5 )
=

weak (Fig. 22). For a narrow range of o 1 8 0 values than Triassic seawater (87Sr/8 6 Sr � 0 . 7078) (Burke
(� - 1 2%o to -8%o) there is a wide range of IGV et al., 1 983). The similarity in the range of Sr
values between �25% and 50%. This is attributed to isotope ratios between calcite and dolomite ex­
variations in the amount of early carbonate cements cludes the involvement of marine waters in the
and to partial cement dissolution during telodiagen- precipitation of dolomite.
The addition of 87Sr to meteoric pore waters
occurred from the diagenetic alteration of clays,
feldspars and micas. The only slightly higher Sr
(a) 1 00
isotopic ratio of the mesogenetic calcite suggests
• • that Sr isotopic composition is buffered by the
80 .. ..
redistribution of eogenetic carbonates, which were
the main source for the mesogenetic cements. It
rf!. 60
"' further suggests that these calcites pre-date Sr sup­
0
u ply by pervasive mesogenetic feldspar albitization
� 40
(see Schultz et al., 1 989).

20
��D D D D
55
0

D dolomite
50 • siderite II 0 0
1 .5
(b) • �
0

Q)
0 calcite
E 45
0 0 • :::J
0 0 lD
0 0 >
�
"'
0
DO
D �:::J 40 QJ
0
0 0 DO
0 m •
'<? O S • 00 c o o• 0
� • D
o aoo
D D � 35 0
• ot
O «l
D D
e>
Q)
0 0 00 c 0
30 � 10
• B!J O •
-0.5 25
-20 -15 -1 0 -5 0 -20 -15 -10 -5 0
i) 1 8Q PDB %o o 1 80 PDB %o

Fig. 21. Plot of o 1 80p08 versus (a) FeC03% and (b) Fig. 22. Plot of intergranular volume (%) versus <5 1 80 of
MnC03% of dolomite/ankerite, siderite and calcite dolomite/ankerite, siderite and calcite. See text for
cements. See comments in the text. discussion.
 Sheet-flood sandstones in the Snorre Field
Carbon isotopes
The carbon isotopic compositions of calcretes and
dolocretes usually reflect the abundance and type of
vegetation in depositional sites, the degree of isoto­
pic equilibration with atmospheric C02, and the
mechanism of organic matter oxidation, which is
strongly controlled by the dissolved oxygen content
in the pore waters. The 8 1 3C values of the analysed
carbonates show a slight to moderate variability
(mainly � -5%o to +4%o) and are indicative of
carbon derivation from the alteration of organic
matter. The highest values are attributed to carbon
derivatio_n within the microbial methanogenesis
zone, which is not surprising considering the abun­
dance of ferroan carbonates, which are indicative of
anoxic conditions. However, similar values can be
obtained where precipitation occurred in isotopic
equilibrium with atmospheric C02 (8 1 3C -7%o)=
during degassing at a temperature of l O "C (Sa­
lomons et al., 1 97 8). The 8 1 80 value of this partic­
ular dolomite indicates precipitation at � 1 5 ·c.
The sources of negative 8 1 3C values is uncertain,
but could be related to derivation of carbon from the
oxidation of organic matter in the soil profiles.
Organic matter in Triassic soils is expected to be
dominated by C3 plants, which contribute dissolved
carbon with a 8 1 3C signature of �- 1 2%o (Cerling,
1 984). As it is generally agreed that there were no C4
plants during the Triassic, which could contribute
dissolved C with a 8 1 3C signature of -4%o to +4%o
(Cerling, 1 984), 8 1 3C values higher than - 1 2%o in
calcrete and dolocrete may be attributed to substan­
tial involvement of atmospheric C02 (dissolved C
with 8 1 3C of � + 2o/oo). Carbon derived from micro­
bial methanogenesis might display a wide and rather
continuous range of8 1 3C values (�-2 5%o to + 1 6%o)
(Whiticar et al., 1 986; Clayton, 1 994).
2
The positive correlation (r = + 0.85) between
3
8 1 C and 8 1 8 0 values (Fig. 22) of the eogenetic
Lunde carbonates may be related to relatively rapid
near-surface precipitation caused by evaporation
and C02 degassing (e.g. Salomons et al., 1 978;
Schlesinger, 1 98 5 ; Salomons & Mook, 1 986; Spot! &
Wright, 1 992), which increases the enrichment of
1 3C and 1 80 isotopes. However, it is believed that at
depths of a decimetre the evaporation rate is sub­
stantially reduced, and there is thus very little oppor­
tunity for significant 1 8 0 enrichment in soil water
before the next rainfall causes sufficient infiltration
to obliterate this effect (Hellwig, 1 973).
Nevertheless, groundwater calcretes and dolo-
81
cretes are expected to be less influenced by evapo­
rative isotopic enrichment, and should thus have
lower 8 1 80 values than vadose carbonates (see
Talma & Netterberg, 1 98 3). Conversely, in conjunc­
tion with petrographic evidence, the 8 1 8 0 values of
some Lunde calcites and dolomites reflect the ele­
vated precipitation temperatures. As these carbon­
ates have depleted carbon isotopic values it is likely
that there was an addition of some light carbon
from the thermal decarboxylation of organic matter
(8 1 3C � - 1 5%) (Irwin et al., 1 97 7 ) as burial depth
and temperature increased. The lower 8 1 3C value
obtained ( - 1 0. 1 o/oo) is in a siderite sample with an
anomalously low 8 1 80 value (- 1 6.9%o) (Table 1 ;
Fig. 22). The reason for this is unclear, but could
reflect the extent and pattern of microbial isotopic
2
fractionation rather than input of 1 C from the
decarboxylation zone, for the reasons discussed
above.
CONCLUSIONS
The siliciclastic sheet-flood sediments of the Lower
and Middle Members of the Lunde Formation In
the Snorre oilfield, Norwegian North Sea, have been
subjected to pervasive eogenetic kaolinization of
feldspar, mud intraclasts and micas. The great
extent of eogenetic kaolinization may have been
accomplished by episodic heavy rainfall alternating
with extended dry periods. Additional kaolinization
is related to the Kimmerian uplift and exposure.
Sandstones which subcropped a few tens of metres
below the Kimmerian unconformity were affected
by extensive kaolinization and carbonate dissolu­
tion, whereas sandstones that remained deeper than
300 m experienced mainly carbonate dissolution.
Sediments were also extensively cemented by
near-surface and pre-compactionql eogenetic sider­
ite (8 1 80 = - 1 6. 9%o to -8. 1 %o), calcite (8 1 8 0 =
- 1 0.2o/oo to -4. 8%o) and dolomite (8 1 80 = -4.3%o to
- 1 . 5o/oo). The anomalously low 8 1 8 0 values of sider-
ite are attributed to microbial oxygen isotopic
fractionation, as it would otherwise indicate unrea­
sonably high precipitation temperatures (up to
1 2o · q. The eogenetic calcites and dolomites pre­
cipitated from such meteoric waters at tempera­
tures of 1 0-3o · c.
Mesogenetic carbonates include slightly Mn-Fe­
rich calcite W 8 0 - 1 2.4%o to -6.6%o) as well as
=
dolomite and ankerite W 80 = - 1 1 .9%o to -4.6%o)
that were precipitated at 30-70"C and 40-9o·c,
82 S. Morad et al.
respectively. This was followed by albitization of
K-feldspars and preferentially of plagioclase, chlor­
itization of kaolinite, smectitic mud intraclasts,
pseudomatrix, infiltrated clays and biotite.
The 8 1 3C values (� -5%o to + 4%o) of the eoge­
n etic and mesogenetic carbonate cements suggest
derivation of carbon from microbial methanogene­
sis, oxidation of plant remains, and a possible
contribution from atmospheric C02. The radio­
genic 87Sr/8 6 Sr ratios (0. 7 1 1 127-0 . 7 1 1655) are typ­
ical of meteoric water interaction with detrital
silicates, and vary relatively slightly between the
eogenetic and mesogenetic calcites. This suggests
that mesogenetic calcite was formed by redistribu­
tion of eogenetic carbonates prior to 87Sr-producing
silicate reactions, such as albitization of feldspar.
ACKNOWLEDGEMENTS
We thank Saga Petroleum a.s. and the other Snorre
Field partners (Esso, Enterprise, Elf, Amerada
Hess, Statoil, Norsk Hydro, Idemitsu and Dem­
inex) for permission to publish this paper. We are
grateful to H. Harryson for assistance with the
microprobe analyses, L. Ravdal and G.B. R0ed
(Saga) for draughting the figures, S. Hvoslef (Saga)
for the b urial diagrams, H. Walderhaug (University
of Bergen) for calculations of palaeolatitudinal po­
sitions, and C. Back and B. Gios for the photo­
graphic work. Comments by reviewers R. Gaupp
and M. Ramm helped to improve the manuscript.
L. F.D.R. acknowledges the support from Brazilian
National R esearch Council (CNPq; grant 200465/
92. 9 -GL). S.M. thanks the Swedish Natural Science
Research Council (NFR) for financial support.
REFERENCES
AL-AASM, I.S., TAYLOR, B.E. & SouTH, B. ( 1 990) Stable
isotope analysis of multiple carbonate samples using
selective acid extraction. Chern. Geol. , 80, 1 1 9- 1 2 5 .
ARAKEL, A.V. ( 1 986) Evolution of calcrete in paleodrain­
ages of Lake Napperby area, central Australia. Paleo­
geogr. Paleoclimatol. Palaeoecol. 54, 283-303.
ARAKEL, A.V. ( 1 99 1 ) Evolution of Quaternary duricrusts
in Karinga Creek drainage system, central Australian
groundwater discharge zone. Aust. J. Earth Sci. , 38,
3 34-347.
ARAKEL, A.V. & McCoNCHIE, D. ( 1 982) Classification and
genesis of calcrete and gypsite lithofacies in palaeo­
drainage systems of inland Australia and their relation-
ship to carnotite mineralization. J. sediment. Petrol. , 52,
1 1 49- 1 1 70.
ARAKEL, A.V. & WAKELIN-KING, G.A. ( 1 99 1 ) Chemical
sedimentation and facies relationships in the Karinga
Creek palaeo-drainage system, Amadeus Basin, North­
ern Australia. Bull. Bur. Miner. Res. , 236, 5 8 1 -589.
BADLEY, M.E., PRICE, J.D., RAMBECH DAHL, C. &
AGDESTEIN, T. ( 1 98 8 ) The structural evolution of the
northern Viking Graben and its bearing upon exten­
sional models of basin formation. J. geol. Soc., Land. ,
145, 45 5-472.
BoLES, J.R. & JOHNSON, S.K. ( 1 984) Influence of mica
surfaces on pore water pH. Chern. Geol. , 43, 303-3 1 7 .
BOLES, J.R. & RAMSEYER, K. ( 1 987) Diagenetic carbonate
in Miocene sandstone reservoir, San Joaquin Basin,
California. Bull. Am. Ass. Petrol. Geol. , 7 1 , 1 475- 1 487.
BURKE, W.H., DENISON, R.E., HETHERINGTON, E.A., KOEP­
NICK, R.B., NELSON, N.F. & OTTO, J.B. ( 1 983) Variation
of seawater 87 Sr/86 Sr throughout Phanerozoic time.
Geology, 10 , 5 1 6-5 1 9.
CAMPBELL, C.J. & 0RMAASEN, E. ( 1 9 87) The discovery of oil
and gas in Norway: an historical synopsis. In: Geology of
the Norwegian Oil and Gas Fields (Ed. Spencer,
A.M.E.A.), pp. 1 -3 7 . Graham & Trotman, London.
CERLING, T.E. ( 1 984) The stable isotopic composition of
modern soil carbonate and its relationship to climate.
Earth Planet. Sci. Lett. , 71, 229-240.
CLAYTON, C.J. ( 1 994) Microbial and organic processes. In:
Quantitative Diagenesis: Recent Developments and Ap­
plications to Reservoir Geology (Eds Parkers, A. &
Sellwood, B.W.). NATO ASI Series C: Mathematical
and Physical Sciences, 453, 1 25- 1 60. Kluwer Aca­
demic, Dordrecht.
CRAIG, H. ( 1 9 57) Isotopic standards for carbon and oxygen
correction factors for mass spectrometric analysis of
carbon dioxide. Geochim. Cosmochim. Acta, 1 2, 1 3 3-
1 49.
CRAIG, H. ( 1 96 1 ) Standards for reporting concentrations
of deuterium and oxygen- 1 8 in natural waters. Science,
133, 1 8 3 3 - 1 934.
CuRTIS, C. D. & CoLEMAN, M.L. ( 1 986) Controls on the
precipitation of early diagenetic calcite, dolomite and
siderite concretions in complex depositional sequences.
In: Roles of Organic Matter in Sediment Diagenesis (Ed.
Gautier, D.L.), Spec. Pub!. Soc. Econ. Paleont. Miner.,
Tulsa, 38, 23-33.
DAHL, N. & SoLu, T. ( 1 993) The structural evolution of
the Snorre Field and surrounding 'areas. In: Petroleum
Geology of Northwest Europe. Proceedings of the 4th
Conference, Vol. 2 (Ed. Parker, J.R.), pp. 1 1 59- 1 1 66.
London.
DICKINSON, W.R., BEARD, L.S., BANKENRIDGE, G.R. et a/.
( 1 983) Provenance of North American Phanerozoic
sandstones in relation to tectonic setting. Geol. Soc. Am.
Bull., 94, 222-23 5 .
DRAVIS, J.J. & YUREWICZ, D.A. ( 1 9 85) Enhanced carbonate
petrography using fluorescence microscopy. J. sediment.
Petrol. , 55, 795-804.
DUAN, W.M., PYE, K., COLEMAN, M., WHITE, D.C.,
RINGELBERG, D. & HEDRICK, D. ( 1 995) Formation of
unique sideritic concretions in modern salt marsh sedi­
ment, Norfolk, UK. EUG Abstract Volume. European
 Sheet-flood sandstones in the Snorre Field
Union of Geosciences Meeting, Strasbourg, France.
EGEBERG, P.K. & AAGAARD, P. ( 1 989) Origin and evolution
of formation water from oil fields on the Norwegian
shelf. Appl. Geochem., 4, 1 3 1 - 1 42 .
EHRENBERG, S.N., AAGAARD, P . , WILSON, M.J., FRASER,
A.R. & DuTHIE, D.M.L. ( 1 993) Depth-dependent trans­
formation of kaolinite to dickite in sandstones of the
Norwegian Continental shelf. Clay Miner., 28, 3 2 5-
352.
EL-SAYED, M . I . , FAIRCHILD, I.J. & SPIRO, B. ( 1 99 1 ) Kuwaiti
dolocrete: petrology, geochemistry and groundwater
origin. Sediment. Geol. , 73 , 59-7 5 .
ESTEBAN, M. & KLAPPA, C.F. ( 1 98 3 ) Subaerial exposure
environment. In: Carbonate Depositional Environments
(Eds Scholle, P.A., Bebout, D. G. & Moore, C. H.). Mem.
Am. Ass. Petrol. Geol., Tulsa, 33, 1 -54.
FAURE, K., HARRIS, C. & WILLIS, J.P. ( 1 995) A profound
meteoric water influence on genesis in the Permian
Waterberg Coalfield, South Africa: evidence from stable
isotopes. J. sediment. Res., A65, 605-6 1 3 .
FoLK, R.L. ( 1 993) SEM imaging of bacteria and nanno­
bacteria in carbonate sediments and rocks. J. sediment.
Petrol. , 63, 990-999.
FRIEDMAN, I. & O'NEIL, J.R. ( 1 977) Compilation of Stable
Isotopic Fractionation Factors of Geochemical Interest.
US Geol. Surv. Prof. Pap., 440-KK, 1 2 pp.
GILES, M.R. & DEBOER, R.B. ( 1 990) Origin and signifi­
cance of redistributional secondary porosity. Mar.
Petrol. Geol. , 6, 378-397.
GLENNIE, K.W. (I 990) Outline of North Sea history and
structural framework. In: Introduction to the Petroleum
Geology of the North Sea, 3rd edn (Ed. Glennie, K.W.),
pp. 3 4-77. Blackwell Scientific Publications, Oxford.
GLENNIE, K.W. ( 1 995) Permian and Triassic rifting in
northwest Europe. In: Permian and Triassic Rifting in
Northwest Europe (Ed. Bold, S.A.R.). Spec. Pub!. geol.
Soc. London, 91, 1 -5 .
GOLDSTEIN, R.H. & REYNOLDS, T.J. ( 1 994) Systematics of
Fluid Inclusions in Diagenetic Minerals. Society of Eco­
nomic Paleontologists and Mineralogists (Society for
Sedimentary Geology), Short Course, 31, 1 9 8 pp.
GouDIE, A.S. ( 1 9 83) Calcrete. In: Chemical Sediments and
Geomorphology: Precipitates and Residua in the Near­
Surface Environment (Eds Goudie, A.S. & Pye, K.),
pp. 9 3 - 1 3 1 . Academic Press, Orlando, Florida.
GRADSTEIN, F.M., AGTERBERG, F.P., 0GG, J.G. et a/. ( 1 995)
A Triassic, Jurassic and Cretaceous time scale. In:
Geochronology, Time Scales and Global Stratigraphic
Correlation (Eds Berggren, W.A., Kent, D.V., Aubry
M.-P. & Hardenbol, J.). Spec. Pub!. Soc. Econ. Paleont.
Miner., Tulsa, 54, 95- 1 26.
HELLWIG, D.R. ( 1 9 73) Evaporation of water from sand. J.
Hydro/. , 18, 3 1 7-327.
HoRSTAD, 1., LARTER, S.R. & MILLS, N. ( 1 995) Migration of
hydrocarbons in the Tampen Spur area, Norwegian
North Sea: a reservoir geochemical evaluation. In: The
Geochemist of Reservoirs (Eds Cubitt, J.M. & England,
W.A.). Spec. Pub!. geol. Soc. London, 86, 1 59- 1 8 3 .
HousEKNECHT, D.W. ( 1 987) Assessing the relative impor­
tance of compaction processes and cementation to
reduction of porosity in sandstones. Bull. Am. Ass.
Petrol. Geol. , 7 1 , 63 3-642.
83
HUMPHREY, J.D. & RADJEF, E.M. ( 1 99 1 ) Dolomite stoichi­
ometric variability resulting from changing aquifer con­
ditions, Barbados, West Indies. Sediment. Geol. , 7 1 ,
1 29- 1 36.
HURST, A. & NADEAU, P.H. ( 1 995) Clay microporosity in
reservoir sandstones: an application of quantitative
electron microscopy in petrophysical evaluation. Bull.
Am. Ass. Petrol. Geol. , 79, 563-573.
IRWIN, H., CURTIS, C. & COLEMAN, M. ( 1 97 7) Isotopic
evidence for source of diagenetic carbonates formed
during burial of organic-rich sediments. Nature, 269,
209-2 1 3 .
ISAKSEN, 0. & TONSTAD, 0. (Eds) ( 1 9 89) A revised Creta­
ceous and Tertiary lithostratigraphic nomenclature for
the Norwegian North Sea. Norwegian Petrol. Directorate
Bull. 5, 59 pp.
JoNES, B. ( 1 98 8 ) The influence of plants and microorgan­
isms on diagenesis in caliche: example from the Pleis­
tocene lronshore Formation on Cayman Brae, British
West Indies. Bull. Can. Petrol. Geol. , 36, 1 9 1 -20 1 .
KAHN, J.S. ( 1 956) The analysis and distribution of the
properties of packing in sand-size sediments: I . On the
measurement of packing in sandstones. J. Geol. , 64,
3 8 5-395.
KELLER, W.O. ( 1 970) Environmental aspects of clay min­
erals. J. sediment. Petrol. , 40, 7 8 8-8 1 3 .
MACHEL, H.-G. & MOUNTJOY, E.W. ( 1 9 86) Chemistry and
environments of dolomitization-a reappraisal. Earth
Sci. Rev. , 23, 1 75-222.
MATSUMOTO, R. & IIJIMA, A. ( 1 9 8 1 ) Origin and diagenetic
evolution of Ca-Fe-Mg carbonates in some coalfields of
Japan. Sedimentology, 28, 2 3 9-2 59.
McAULAY, G.E., BuRLEY, S.D. & JoHNES, L.H. ( 1 993)
Silicate mineral authigenesis in the Hutton and NW
Hutton fields: implications for sub-surface porosity
development. In: Petroleum Geology of Northwest Eu­
rope: Proceedings of the 4th Conference (Ed. Parker,
J.R.), pp. 1 3 7 7- 1 394. London.
McBRIDE, E. F. ( 1 963) A classification of common sand­
stones. J. sediment. Petrol., 33, 664-669.
MEARNS, E.W., KNARUD, R.N., STANLEY, K.O. & STOCK­
BRIDGE, C.P. ( 1 989) Samarium-neodymium isotope
stratigraphy of the Lunde and Stat fjord Formations of
Snorre Oil Field, northern North Sea. J. geol. Soc.
London, 146, 2 1 7-228.
MORAD, S. ( 1 988) Diagenesis of titaniferous minerals in
Jurassic sandstones from the Norwegian Sea. Sediment.
Geol. , 57, 1 7-40.
MORAD, S. ( 1 990) Mica alteration reactions in Jurassic
reservoir sandstones from the Haltenbanken area, off­
shore Norway. Clays Clay Miner. , 38, 584-590.
MORAD, S. & DE Ros, L.F. ( 1 994) Geochemistry and
diagenesis of stratabound calcite cement layers within
the Rannach Formation of the Brent Group, Murchison
Field, North Viking Graben (northern North Seal­
comment. Sediment. Geol. , 93 , 1 3 5- 1 4 1 .
MORAD, S., BERGAN, M., KNARUD, R. & NYSTUEN, J.P.
( 1 990) Albitization of detrital plagioclase in Triassic
reservoir sandstones from the Snorre Field, Norwegian
North Sea. J. sediment. Petrol. , 60, 4 1 1 -425.
MORAD, S., MARFIL, R., AL-AASM, I.S. & GOMEZ-GRAS, D.
( 1 992) The role of mixing-zone dolomitization in sand-
84 S. Morad et a/.
stone cementation: evidence from the Triassic Bunt­
sandstein, the Iberian Range, Spain. Sediment. Geol. ,
80, 5 3-65.
MORAD, S., BEN ISMAIL, H., DE Ros, L.F., AL-AASM, l.S. &
SERRHINI, N.-E. ( 1 994) Diagenesis and formation water
chemistry of Triassic reservoir sandstones from south­
em Tunisia. Sedimentology, 4 1 , 1 2 5 3 - 1 272.
MoRAD, S., DE Ros, L.F. & AL-AASM, I.S. ( 1 996) Origin of
low 01 80, precompactional ferroan carbonates in the
marine St0 Formation (Middle Jurassic), offshore NW
Norway. Mar. Petrol. Geol. , 13, 263-276.
MoRAES M.A.S. & DE Ros, L.F. ( 1 990) Infiltrated clays in
fluvial Jurassic sandstones of Rec6ncavo Basin, north­
eastern Brazil. J. sediment. Petrol. , 60, 809-8 1 9 .
MORTON, N. ( 1 993) Potential reservoir and source rocks in
relation to Upper Triassic to Middle Jurassic sequence
stratigraphy, Atlantic margin basins of the British Isles.
In: Petroleum Geology ofNorthwest Europe. Proceedings
of the 4th Conference (Ed. Parker, J.R.), pp. 2 8 5-297.
London.
MozLEY, P.S. ( 1 989) Relation between depositional envi­
ronment and the elemental composition of early dia­
genetic siderite. Geology, 17, 704-706.
NYSTUEN, J.P. & FA LT, L.-M. ( 1 995) Upper Triassic­
Lower Jurassic reservoir rocks in the Tampen Spur area,
Norwegian North Sea. In: Petroleum Exploration in
Norway, Vol. 4 (Ed. Hanslien, S. et al.). Norwegian
Petroleum Society. Elsevier, Amsterdam, 4 1 pp.
NYSTUEN, J.P., KNARUD, R., JORDE, K. & STANLEY, K.
( 1 989) Correlation of Triassic to lower Jurassic se­
quences, Snorre Field and adjacent areas, northern
North Sea. In: Correlation in Hydrocarbon Exploration
(Ed. Collinson, J.D.), pp. 2 7 3-289. Norwegian Petro­
leum Society. Graham & Trotman, London.
PHILLIPS, S.E., MILNES, A.R. & FOSTER, R.C. ( 1 98 7) Calci­
fied filaments: an example of biological influence in the
formation of calcrete in South Australia. A ust. J. Soil
Res. , 25, 405-428.
PURVIS, K. & WRIGHT, V.P. ( 1 99 1 ) Calcretes related to
phreatophytic vegetation from the Middle Triassic
Otter Sandstone of southwest England. Sedimentology,
38, 5 39-5 5 1 .
REEVES, J.C.C. ( 1 976) Origin, classification and geologic
history of caliche on the southern High Plains, Texas,
and eastern New Mexico. J. Geoi., 78, 3 52-362.
ROSENBAUM, J.M. & SHEPPARD, S.M.F. ( 1 986) An isotopic
study of siderites, dolomites and ankerites at high
temperatures. Geochim. Cosmochim. Acta, 50, 1 1 47-
1 1 50.
SAIGAL, G.C. & BJ0RLY KKE, K. ( 1 98 7 ) Carbonate cements
in clastic reservoir rocks from offshore Norway- rela­
tionships between isotopic composition, textural devel­
opment and burial depth. In: Diagenesis ofSedimentary
Sequences (Ed. Marshall, J.D.). Spec. Pub!. geol. Soc.
I..ondon, 36, 3 1 3-324.
SALOMONS, W. & MOOK, W.G. ( 1 986) Isotope geochemis­
try of carbonates in the weathering zone. In: Handbook
of Environmental Isotope Geochemistry, Vol. 2 (Eds
Fritz, P. & Fontes, J.C.), pp. 2 3 9-269. Elsevier, Amster­
dam.
SALOMONS, W., GoUDIE, A.S. & MooK, W.G. ( 1 97 8 ) Isoto­
pic composition of calcrete deposits from Europe,
Africa and India. Earth Surf Process. , 3, 43-57.
SCHLESINGER, W.H. ( 1 98 5 ) The formation of caliche in
soils of the Mojave Desert, California. Geochim. Cos­
mochim. Acta, 49, 57-66.
SCHMIDT, V. & McDONALD, D.A. ( 1 979) The role of
secondary porosity in the course of sandstone diagene­
sis. In: Aspects of Diagenesis (Eds Scholle, P.A. &
Schluger, P.R.). Spec. Pub!. Soc. Econ. Paleont. Miner.,
Tulsa, 29, 1 7 5-207.
SCHULTZ, J.L., BOLES, J.R. & TILTON, G.R. ( 1 989) Tracking
calcium in the San Joaquin basin, California: a stron­
tium isotopic study of carbonate cements at North
Coles Levee. Geochim. Cosmochirn. Acta, 53, 1 99 1 -
1 999.
SPOTL, C. & WRIGHT, V.P. ( 1 992) Groundwater dolocretes
from the Upper Triassic of the Paris Basin, France: a
case study of an arid, continental diagenetic facies.
Sedimentology, 39 , 1 1 1 9- 1 1 3 6.
STEEL, R. ( 1 993) Triassic-Jurassic megasequence stratig­
raphy in the Northern North Sea: rift to post-rift
evolution. In: Petroleum Geology of Northwest Europe.
Proceedings of the 4th Conference (Ed. Parker, J.R.),
pp. 299-3 1 5 . London.
STEEL, R. & RYSETH, A. ( 1 990) The Triassic-early Jurassic
succession in the northern North Sea: megasequence
stratigraphy and intra-Triassic tectonics. In: Tectonic
Events Responsible for Britain 's Oil and Gas Reserves
(Eds Hardman, R.F.P. & Brooks, J.). Spec. Pub!. geol.
Soc. London, 55, 1 39- 1 68 .
TALMA, A.S. & NETTERBERG, F . ( 1 98 3 ) Stable isotope
abundances in calcretes. In: Residual Deposits: Surface
Related Weathering Processes and Materials (Ed. Wilson,
R.C.L.). Spec. Pub!. geol. Soc. London, 1 1 , 22 1 -2 3 3 .
THIRY, M. & MILNES, A.R. ( 1 99 1 ) Pedogenic and ground­
water silcretes at Stuart Creek opal field, South Austra­
lia. J. sediment. Petrol. , 61, 1 1 1 - 1 2 7 .
TILLEY, B.J. & LONGSTAFFE, F.J. ( 1 989) Diagenesis and
isotopic evolution of porewaters in the Alberta Deep
Basin: the Falher Member and Cadomin Formation.
Geochim. Cosmochim. Acta, 53, 2 529-2546.
VOGEL, J.C. & VAN URK, H. ( 1 97 5 ) Isotopic composition
of groundwater in semi-arid regions of Southern Africa.
J. Hydro!. , 25, 23-36.
VOLLSET, J. & DORE, A.G. (Eds) ( 1 984) A revised Triassic
and Jurassic lithostratigraphic nomenclature for the
Norwegian North Sea. Norwegian Petrol. Directorate
Bull. 3, 53 pp.
WALKER, T.R., WAUGH, B. & CRONE, A.J. ( 1 978) Diagene­
sis in first-cycle desert alluvium of Cenozoic age, south­
western United States and northwestern Mexico. Geol.
Soc. Am. Bull. , 89, 1 9-32.
WARD, W.C. & HA LLEY , R.B. ( 1 985) Dolomitization in a
mixing zone of near-seawater composition, Late Pleis­
tocene, Northeastern Yucatan Peninsula. J. sediment.
Petrol. , 55, 407-420.
WATTS, N.L. ( 1 980) Quaternary pedogenic calcretes from
the Kalahari (South Africa): mineralogy, genesis and
diagenesis. Sedimentology, 27, 66 1 -686.
WHITE, D., HEM, J.D. & WARING, G.A. ( 1 96 3 ) Chemical
composition of subsurface waters. In: Data of Geochem­
istry, 6th edn (Ed. Fleischer, M.). US Geol. Surv. Prof.
Paper, 440, Chapter F.
 Sheet-flood sandstones in the Snorre Field 85

WHITICAR, M.J., FABER, E . & ScHOELL, M . ( 1 986) Biogenic
methane formation in marine and freshwater environ­
ments: C02 reduction vs. acetate fermentation­
isotop ic evidence. Geochim. Cosmochim. Acta, 50, 693-
709.
ZIEGLER, P.A. ( 1 988) Evolution ofthe Arctic-North A tlantic
and the Western Tethys. Mem. Am. Ass. Petrol. Geol. ,
Tulsa, 43, 1 9 8 pp .
Spec. Pubis int. Ass. Sediment. ( 1 998) 26, 87- 1 05

Carbonate diagenesis in non-marine foreland sandstones

at the western edge of the Alleghanian overthrust belt,
southern Appalachians

K . L . M I LLIK E N
Department of Geological Sciences, University of Texas, A ustin, TX 78712, USA,
e-mail kittym@mail. utexas. edu

ABSTRACT

The late Palaeozoic synorogenic foreland sandstones of the southern Appalachian basin are relatively
carbonate poor (average < 3 vol%) but locally contain siderite, calcite and ferroan dolomite/ankerite as
cements and grain replacements. Petrographic evidence shows that siderite is an early precipitate,
followed by a generation of Mg- and Fe-rich calcite (average Ca95.6Mg1 . Fe2.4Mn0.2C03) that preceded
8
quartz cementation. Ferroan dolomite/ankerite postdates quartz cementation and is followed in tum
by a generation of relatively Mn-rich calcite (average Ca97 .4M�.6Fe15Mn0.5C03). Early calcite is
highly localized at the outcrop scale, though pervasively distributed at the thin section scale, and
preserves intergranular volumes (IGVs) (30-40%) characteristic of relatively early stages of compac­
tion. Other carbonates are highly localized in their distribution in thin section. Siderite is localized on
expanded detrital micas. Ferroan dolomite and late calcite have a strong spatial affiliation with partially
dissolved silicate grains, and are found in sandstones with markedly reduced IGVs ( < 20%).
Elemental and isotopic values for all the authigenic carbonates suggest that fluids responsible for
carbonate precipitation were most likely 180 depleted, enriched in organic carbon, and contained Mg,
Fe, Mn and, in some cases, Sr mobilized by the alteration of detrital components in the sandstones and
associated mudrocks. 87 Sr/86Sr values fall into a range outside that of marine Sr, supporting a
prominent role of silicate-derived components in carbonate precipitation. Temporal variation in
carbonate mineralogy and compositions probably reflects the changing character of elemental sources
during burial. Early phases (siderite and early calcite) may reflect the reaction of highly unstable Fe- and
Mn-oxyhydroxides and clay-adsorbed Mg. Ferroan dolomite and ankerite may represent Mg, Fe and
Mn mobilized by subsurface alteration of detrital clays and their surface coatings at elevated
temperatures. The late generation of calcite formed after the sources of Mg and Fe were depleted,
though Mn remained relatively high. Manganese in late carbonates was possibly derived from relatively
resistant and late-reacting heavy minerals such as garnet or, alternatively, from fluids derived from
deeper in the basin.

INTRODUCTION

This paper describes the petrography and chemistry
of detrital and authigenic carbonates in a non­
marine foreland setting in the southern Appala­
chian basin. Numerous workers have postulated
significant fluid flow related to the late Palaeozoic
Alleghanian orogeny in the southern Appalachian
mountains. Widespread potassic metasomatism, si­
licification, ore mineralization and magnetization
in lower Palaeozoic rocks of the Appalachian fold
and thrust belt (e.g. Hearn & Sutter, 1985;
Bachtadse et a!., 1987; Elliot & Aronson, 1987;
Schedl et a!., 1992) suggest the passage of orogenic
fluids that acquired a distinctive composition
through reaction with crustal materials at elevated
temperatures. What pathways did these fluids fol­
low as they exited the thrust belt? Did they leave a
record in authigenic phases contained in the fore­
land basin? What was the relative significance of
'rock-dominated' orogen-derived fluids (e.g. Oliver,
1986; Cathles, 1990; Schedl et a!., 1992) versus

Carbonate Cementation in Sandstones Edited by Sadoon Morad 87

© 1998 The International Association of Sedimentologists
ISBN: 978-0-632-04777-2
88 K.L. Milliken
deeply penetrating meteoric fluids (e.g. Bethke &
Marshak, 1990; Deming et al., 1990) in causing
Alleghanian-age rock alteration in the foreland? The
synorogenic siliciclastic sediments preserved at the
westernmost edge of the Appalachian thrust zone
offer an excellent opportunity to examine these
questions unambiguously, without complicated
overprinting by pre-Alleghanian alteration.
The carbonate minerals in these sandstones are
especially interesting in this regard because it is
possible to compare their compositions with those
of similar phases of documented Alleghanian asso­
ciation in the thrust Lower Palaeozoic carbonate
units to the southeast. Marked differences between
carbonate mineral compositions in the relatively
undeformed foreland (reported here) and those in
the fold and thrust belt (reported in the literature),
suggest constraints on models for fluid flow during
Alleghanian orogenesis.
.
1'-
�----
! 400 km
GEOLOGICAL SETTING
The study area (Fig. I) is in a region around the
westernmost edge of preserved Alleghanian-age
thrust faults. Samples include sandstones mapped
as the Lee and Breathitt formations (Middle Penn­
sylvanian) that were deposited in the foreland
region of the Alleghanian orogen (Quinlan & Beau­
mont, 1984; Tankard, 1986). Ultimately, these
synorogenic sediments were themselves involved in
compressional deformation around and above the
Pine Mountain overthrust (PMO) (Mitra, 1988). A
portion of the sample set for this study was collected
in relatively undeformed areas northwest of the
PMO. The fluvjal/deltaic to marginal marine char­
acter of this classic coal-bearing sequence has been
documented in nup1erous studies (e.g. Ferm, 1974;
Cobb et al., 1981). Stratigraphical subdivision
within the Breathitt and Lee formations takes ad-
Atlantic Ocean
Fig. I. Location of the study area
within the Appalachian basin. Key
tectonic elements of the Palaeozoic
Appalachian orogen are indicated.
Modified from Boettcher &
Milliken ( 1 994).
 Non-marine foreland sandstones in the Appalachians
vantage of several thin, but regionally extensive
marine shales (Cobb et a/., 198 1 ). The thickness of
Pennsylvanian units in this area is approximately
I km, being somewhat greater in the Middlesboro
syncline above the PMO and less toward the west­
em limit of outcrop (data summarized from geolog­
ical quadrangle maps of the US Geological Survey).
The stratigraphical section subjacent to the Penn­
sylvanian siliciclastics consists primarily of lime­
stones and dolomites of Mississippian age and
older.
The maximum temperature experienced by
Pennsylvanian rocks in the study area is constrained
by vitrinite reflectance data in the range of 1 1 5-
130·c (O'Hara et a/., 1990; Hower & Rimmer,
199 1 ). Apatite fission track study reveals relatively
young ages of apatite cooling (92- 137 Ma), com­
patible with maximum temperatures that com­
pletely annealed tracks in apatite (Boettcher & Mil­
liken, 1994). Modelling based on these young ages
and on the highly shortened track lengths in the ap­
atites suggests a protracted period of cooling such
that the sandstones did not reach near-surface tem­
peratures until after a final rapid pulse of uplift in the
Miocene (Boettcher & Milliken, 1994). Prolonged
maintenance of elevated temperatures (> 6o·q
through much of the postorogenic stage of burial
limits the usefulness of thermal constraints for
determining the timing of diagenetic processes. On
the other hand, elevated temperatures persisting into
the Mesozoic suggest the possible involvement of
gravity-driven meteoric fluids, which may have
played a role in altering the sandstones. Because of
uncertainties in the range of the geothermal gradient
during and since the Alleghanian orogeny, estimates
of burial depth are even more uncertain than tem­
perature estimates. Based on modelling the isostatic
loading of thrust sheets, Beaumont et a/. ( 1987) es­
timated maximum burial of 3-4 km in the region of
the PMO.
SAMPLING AND METHODS
This study is based on a subset of samples from 3 17
localities in eastern Kentucky, northwestern Ten­
nessee, western West Virginia and southwestern
Virginia. Samples were collected primarily from
channel sandstones in order to control, as closely as
possible, for grain size and depositional environ­
ment. Most samples are medium sandstones. In a
regional petrographic survey, 275 standard blue
89
epoxy-impregnated thin sections were point counted
(200 points per slide). From this survey, 8 1 samples
with > I o/o carbonate by volume were identified.
Only 30 of these contain carbonate in excess of
5 volo/o. This study focuses on the trace elemental
and isotopic results obtained from this set of
carbonate-rich samples. Among the more highly
cemented samples (> 20 volo/o carbonate) are six
concretions for which adjacent, less cemented
host-rock samples were examined for comparative
purposes.
Semiquantitative estimates of the relative abun­
dances of siderite, calcite and ferroan dolomite/
ankerite were made by comparing corrected XRD
(X-ray diffraction) peak intensities using the
method of Lynch ( 1997).
Carbon and oxygen isotopic analyses were per­
formed using the extraction method of McCrea
( 1950). Samples were weighed prior to reaction and
the weight per cent of calcite was determined mano­
metrically. Multiple extractions were attempted for
samples containing significant mixtures of calcite,
dolomite and/or siderite. 'Calcite' gas was extracted
after 2 h of reaction with I 00% phosphoric acid at
2 5·c; 'dolomite' gas was evolved at 2 5·c over 3
days of reaction; 'siderite' gas was evolved at 5o·c
and extracted after visible reaction had ceased
(typically 2-3 days). Most of the gases were ana­
lysed on a Nuclide gas-source mass spectrometer; a
few were analysed on a VG Prism gas-source mass
spectrometer.
87Sr/86Sr was analysed on a Finigan MAT 26 1
mass spectrometer operated in static collection
mode. Procedures described by Awwiller ( 1992)
were employed to minimize contamination by ex­
changeable Sr from silicates.
Analysis of Ca, Mg, Fe, Mn and Sr in calcite,
ankerite and siderite was performed on a JEOL 733
electron microprobe. Acceleratin� voltage was
15 kV; sample current was 12 nA, stabilized on
brass. Spot size was I 0 Jlm. Counting time for all
elements was 20 s, except for Sr, which was anal­
ysed for 60 s. Detection limits are approximately
340 ppm for Mg, 450 ppm for Fe, 3 10 ppm for Mn
and 18 5 ppm for Sr. Totals between 97 and 103%
were accepted. Standards were carbonate minerals
(calcite for Ca; dolomite for Ca, Mg; siderite for Fe,
Mn; and coral for Sr) in the standard collection at
the University of Texas electron microprobe labo­
ratory. Beam placement was guided by back­
scattered electron imaging. Si was routinely counted
by WDS to check for possible contamination from
90 K.L. Milliken

adjacent silicate grains that were not apparent in the
back-scattered image.
GENERAL PETROGRAPHIC
FEATURES
Sandstones in the Lee Formation are quartzarenites
and sublitharenites; Breathitt sandstones are domi­
nantly sublitharenites and range into the quartz-rich
end of the litharenite field (Table 1). Rock fragments
in both units are dominantly metamorphic (MRFs),
somewhat higher-rank MRFs dominating in the
Breathitt. Detrital feldspar assemblages in both units
include K-rich K-feldspars and Na-rich plagioclase
(albite up to about An20). lntergranular volumes are
generally low in both units (Table 1). Visible poros­
ities are correspondingly low, ranging as high as 20%
in both units, but averaging less than 10% and 5% in
Lee and Breathitt, respectively, both above and be­
low the PMO.
Apart from the authigenic carbonate assemblage,
quartz and kaolinite (possibly dickite) are the only
volumetrically significant cements in Lee and
Breathitt sandstones (Table 1). Dissolution and
replacement of detrital feldspars is significant,
though difficult to assess because unequivocal pri­
mary grain assemblages are not preserved. Intra­
granular secondary porosity makes up a significant
portion of the total porosity. Other authigenic
phases in addition to carbonates are localized at
sites of feldspar dissolution, at least locally. Quartz,
kaolinite, albite, sphalerite and barite all replace
detrital feldspars. There is little regional geographi­
cal variation in the distribution of authigenic min-
erals. Other than IGV and total porosity, there is
little contrast in samples above and below the PMO.
A tendency toward 'higher-grade' diagenetic fea­
tures (quartz replacement of feldspars, K-feldspar
loss, albitization, carbonate loss) is observed in a
local area around the northwest end of the PMO
(Milliken, 1992). Because of the tendency of these
more altered samples to lack carbonate, only one
sample (VA21) from this area of more intense
alteration is described in the present study.
PETROGRAPHY AND
GEOCHEMISTRY OF
CARBONATE COMPONENTS
General
On a regional basis the overall carbonate content of
Lee and Breathitt sandstones is low, averaging
approximately 3 volo/o below the PMO and slightly
less above (Table 1). Only around 10% of the 281
samples examined in the general petrographic sur­
vey (see Methods section above) contain carbonate
in excess of 5 volo/o. Carbonate enrichment is a
highly localized phenomenon at the outcrop scale,
and examples of carbonate-rich samples are found
both above and below the PMO.
Quantitative XRD estimates of proportions of
the calcite, ferroan dolomite and siderite in the
carbonate assemblage shows that each of these
minerals locally dominates the carbonate assem­
blage, and also occurs in various combinations with
the other carbonates (Table 2). Minor siderite is
nearly ubiquitous, making up at least a few per cent

Table I. Averaged petrographic properties of Lee and Breathitt sandstones, contrasted in undeformed areas versus
above the Pine Mountain overthrust (PMO)

Quartz Carbonate Kaolinite Total Q Total F Total L

F/ cement cement cement <Ptotal cJlse/ in Q,FL in Q,FL in Q,FL IGV
n P/Ft (F + Q) (vol%) (vol%) (vol%) (vol%) <Ptotal (%) (%) (%) (vol%)

Lee Formation
Above PMO 44 0.40 0.01 7.9 1.2 0.8 5.8 0.22 90.4 1.3 8.3 14.7
Below PMO 31 0.63 0.04 5.9 3.0 1.0 11.8 0.14 84.9 3.3 11.8 20.1

Breathitt Formation
Above PMO 52 0.68 0.14 2.4 2.0 0.5 1.6 0.60 64.8 10.3 25.0 5.5
Below PMO 148 0.59 0.17 2.1 2.9 1.5 5.8 0.38 64.3 12.5 23.2 10.5
Total 275

F, feldspar; F,, total feldspar; IGV, intergranular volume; L, lithics; P, plagioclase; cp, porosity; Q, quartz; Q., total
quartz.
 Non-marine foreland sandstones in the Appalachians 91

Table 2. Normalized percentages of calcite, dolomite and siderite within the carbonate assemblages; determined using
the XRD method of Lynch ( 1 997)

Sample* Calcite (%) Dolomite (%) Siderite (%) Sample* Calcite (%) Dolomite (%) Siderite (%)

Calcite assemblages Calcite-siderite assemblages

48 97 0 3 1 39 85 0 15
68 97 0 3 1 43 42 0 58
76 99 0 I 1 50 70 0 30
87 98 0 2 185 85 0 15
1 34 1 00 0 0 239 80 0 20
225 99 0 I 1 2 5C 65 0 35
54C 95 0 5 1 2 5B 67 0 33
lO l A 94 0 6 54D 80 0 20
1 2A 97 0 3 W4B 63 0 37
2 1 8A 96 0 4
2 1 8C 1 00 0 0 Calcite-dolomite-siderite assemblages
220A 93 0 7 53 34 56 10
2A 96 0 4 1 42 24 63 13
47A 97 0 3 1 49A 37 51 12
54E 98 0 2 VA2 1 66 25 10
64A 94 0 6
W4A 95 0 5 Dolomite assemblages
1 25A 89 9 3 69 0 97 3
1 56 0 93 7
Dolomite-siderite assemblages 1 30F 4 96 0
56 0 70 30
70 0 67 33 Siderite assemblages
1 47 0 60 40 49 0 0 1 00
227 0 79 21 72 0 0 1 00
lOB 0 67 33 209 0 0 1 00
1 2 8A 0 40 60 1 82A 0 0 1 00
1 82A 0 34 66
260B 0 27 73
26C 0 57 43
41A 0 45 55
42A 0 49 51
43A 0 73 27
44A 0 40 60
77A 0 59 41
89A 8 75 17
9C 0 78 22
VA7 0 28 72

*All sample numbers have prefix KY unless otherwise specified.

of the total carbonate content in most samples.
Carbonate assemblages dominated by calcite are
most common. Geographical trends in the make-up
of the carbonate assemblage are lacking and all
combinations of the various carbonate minerals
have examples both above and below the PMO.
Siderite
Siderite has a highly localized distribution at the
thin section scale, rendering temporal assessment of
its formation relative to other authigenic phases
somewhat problematic. The persistence of small
amounts of siderite in all the various combinations
of calcite and fcrroan dolomite (Table 2) argues for
relatively early timing of sideri te precipitation.
Petrographic relationships with early calcite sup­
port this view. Detrital phyllosilicates, mostly mus­
covite, but also biotite and chlorite, are the most
common loci of siderite precipitation, analogous to
the observations of Boles & Johnson ( 1 98 3)
(Fig. 2A). Siderite precipitated between phyllosili­
cate layers that appear to have been dramatically
expanded. This expansion further supports the no­
tion that siderite precipitation is relatively early,
pre-dating the significant compaction that has af­
fected these sandstones. To a lesser degree, siderite
is localized on partially dissolved and replaced K-
92
Fig. 2. Thin-section scale localization of authigenic
siderite. Back-scattered electron images. Scale bars
I 0 J.lm. (A) Zoned siderite (bright mineral) localized
around and within a detrital chlorite (KY2 1 8B). (B)
Siderite (s) localized on partially dissolved and
kaolinized K-feldspar (f).
feldspars, both in sandstones and in associated
shales (Fig. 28).
Most of the siderite is too finely crystalline
( < 20 Jlm) to be analysed by the electron micro­
probe. Even very small siderite crystals display
prominent zoning in back-scattered electron im­
ages, further adding to the difficulty of assessing
siderite composition (Fig. 2A). Typical siderite
crystals have centres that are Fe rich compared with
the more Mg- and Ca-rich outer zones, although the
opposite trend is also observed. Analyses of some of
the larger siderite crystals reveals a range of compo­
sitions consistent with a non-marine origin (Table
K.L. Milliken
Fig. 3. Pre-quartz, early calcite. Back-scattered electron
images. Scale bars l 00 J.lm. (A) Early calcite (c) that
postdates early siderite (bright rhombs); KY64A.
(B) Early calcite (c) locally replaces detrital K-feldspar
(arrows) but is otherwise pervasive in its distribution.
3). Mn contents in particular ar.e higher than those
reported for marine siderite, whereas Ca contents
are generally less (see Mozley, 1989). The observed
range of stable isotopic compositions for siderite in
this study, although not strictly definitive of either
marine or freshwater origin (see Mozley & Wersin,
1992) is compatible with precipitation from 180-
depleted fluids ( -7 to -15o/oo) at low temperature
( 15-25 ·q according to the oxygen isotopic frac­
tionation equation of Carothers et a!. ( 1988).
Calcite
Two generations of calcite are recognized on the
 Non-marine foreland sandstones in the Appalachians 93

Table 3. Electron microprobe analyses of siderite

3
Sample CaC03 (mol%) MgC03 (mol%) FeC03 (mol%) MnC03 (mol%) 0180sid (PDB) 01 Csid (PDB)

KY- l l C 1 . 75 1 9. 3 9 77 . 1 1 1 . 62
KY-54C 3.74 0.23 95.08 0.63
KY-54C 0.36 0.56 97.91 0.85
KY232A 5. 1 0 20. 35 73.01 1 . 54
KY232A 3.96 1 2.45 82. 5 3 1 .06
KY232A 3 . 90 1 1.13 83.35 1 . 62
KY260B 5.88 1 7.29 74. 2 1 2.61
KY260B 4.76 1 7 . 71 74.69 2.84
VA-5 3.24 1 5 . 84 78.32 2.40
VA-5 2.59 1 6.02 78.73 2 . 34
VA-5 2.36 1 5.41 79.85 2.38
W4B 3.75 1 0.48 82.53 3.24
W4B 2.99 8.01 8 5 . 49 3.50
KY 1 85sid - 1 1 . 36 -6 . 6 1
KY239sid -9.48 -3 . 1 3
KY4 1 Asid -6.43 1 .96
KY9Csid -6. 6 1 1 . 74
VA7sid - 1 0.50 -3 . 1 5
KY 1 2 5Bsid -6. 7 1 2.75
W4Bsid -7.30 -2. 3 3

basis of petrographic and geochemical evidence
(Table 4). One type (here designated 'early' calcite)
occurs as generally cemented beds and also as
concretions, discrete bodies of localized cementa­
tion surrounded by host rocks with markedly
smaller amounts of calcite cement (Fig. 3). Only
eight concretions were identified in sandstones dur­
ing the course of field work; concretionary bodies in
shales are somewhat more abundant in this region,
but are not considered here. The sampled concre­
tions were typically large (� I m diameter) and
roughly spherical. Factors controlling the localiza­
tion of concretions, for example shell lags, are not
apparent in the field. Most concretions occur as
isolated bodies, many outcrops containing only one
or a few observable concretions.
Early calcites (both generalized and concretion­
ary) preserve IGVs typically in the range of 3 0-
40%. In terms of cement stratigraphy, early calcite
precedes quartz cement but postdates siderite and
some of the kaolinite (Fig. 3A). Early calcite is
typically poikilotopic. Extensive replacement of
detrital feldspars by early calcite is observed
(Fig. 3B), but the calcite is pervasive in its distribu­
tion and not strongly localized on the feldspars.
In order to examine the role of early calcite
precipitation in the preservation versus alteration
of the detrital grain assemblage, adjacent unce­
mented samples were obtained for six of the concre­
tions. Comparison of detrital feldspar assemblages
and a variety o f other petrographic parameters
reveals little consistent variation between concre­
tions and host rocks beyond the obvious differences
in carbonate content and IGV (Table 5). Quartz
cementation is greater in the host rocks, consistent
with the pre-quartz timing of the early calcite
precipitation. There is no evidence that concretion­
ary calcite has consistently either preserved a less
altered feldspar assemblage or produced a more
altered one through replacement. Only one sample
(KY220A) contains significantly more calcic plagio­
clase in the concretion, as reflected consistently by
the average composition of the Ca-plagioclase, the
ratio of Ca-plagioclase to Ca-plagioclase + albite,
and the composition of the most calcic feldspar
grain. There are also no systematic differences
between concretions and host rocks in terms of total
feldspar content or the ratio of plagioclase to
K-feldspar. The lack of any striking contrast be­
tween concretions and host rocks recalls the similar
observations of Cibin et a!. (1993) for carbonate
concretions in thrust-faulted siliciclastic rocks in
the northern Apennines.
A second generation of calcite ('late' calcite)
postdates quartz cementation (Fig. 4) and is found
within more highly compacted sandstones (IGVs
reduced to around the average for the units). Crys­
tals of this calcite type tend to fill isolated pores.
Early and late calcites fall into distinct fields on
the basis of Mg content and Mn/Fe ratios (Fig. 5).
 '-0
Table 4. Average compositions of early and late calcite cements �

3
Sample Location Petrography n CaC03 (mol%) MgC03 (mol%) FeC03 (mol%) MnC03 (mol%) Mn/Fe o180sid (PDB) o1 Csid (PDB)

'Early' calcite-concretionary and generalized cements

W4A Below PMO Concretiont 12 95.88 1 . 66 2.16 0.30 0. 1 4 - 1 1 .4 -6.75
KY1 1 G Below PMO Concretion 5 94.75 2.69 2.44 0.06 0.02
KY43C Below PMO Concretiont 5 96.28 1 .72 1.71 0. 1 3 0.08
KY-54C Below PMO Concretiont 10 95.26 2.20 2.2 1 0. 1 4 0.07 - 1 0. 9 -4.68
KY-540 Below PMO Concretion -9.4 - 1 .30
KY-54E Below PMO Concretion 5 94.95 2 . 00 2.91 0.09 0.03 -9.4 5.79
KY47A Below PMO Cement 35 96.61 1 . 34 1 . 79 0.20 0. 1 1 - 1 1 .8 -2.6 1
KY2 1 8A Below PMO Concretiont 13 9 7.02 0.7 1 2.00 0.26 0. 1 3 - 1 0.4 - 1 .04
KY64A* Below PMO Concretiont 10 9 3 . 95 2. 1 8 3.48 0.39 0. 1 1 - 1 0.5 -5.66
KY 1 25A Above PMO Concretiont 27 93. 10 4.43 2.36 0.09 0.04 - 1 0.4 1 . 26
KY225 Below PMO Concretion 16 96.33 1.18 2.39 0.08 O.o3 -7.2 3.44
KY220A Below PMO Concretiont 15 96.00 1 .6 1 2.32 0.06 O.Q3 -8.0 3.66
KY-2A* Below PMO Cement 12 94.30 2.00 3.09 0.30 0. 1 0 - 1 1 .7 -6.82
TN 1 1 * Above PMO Cement 11 97.00 0.61 1 . 97 0.26 0. 1 3 - 1 2.6 -5. 5 2
KY68 Below PMO Cement 10 96.44 1.13 2.04 0.29 0. 1 4 - 1 2.7 -2.35
KY87 Below PMO Cement 8 95.27 2.54 1 . 97 0.07 O.o3 -9.0 3.7 1
�
Total: 1 94 Average 1 . 87 2.32 0. 1 8 0.08 - 1 0.38 - 1 .35 !:""<
Average (ppm) 4475 1 2827 982
�
'Late' calcite-/ow IGV, post-quartz
�
KY 1 0 1 A Below PMO Cement/gr. rep!. 14 9 7 . 92 0.43 1 . 20 0.45 0.37 -1 1 .6 -5.04 �
;::,;
KY 1 85 Above PMO Cement/gr. rep!. 8 97.26 0.52 1 . 66 0.56 0.34 - 1 3.6 -6.55
KY- 1 2A Below PMO Cement/gr. rep!. 28 97. 1 9 0.49 1 .54 0.70 0.45 -3.5 -9. 73
KY 1 3 9 Above PMO Cement/gr. rep!. 17 96.08 0.46 1 . 20 0. 3 1 0.26 -2.2 -4.50
KY48 Below PMO Cement/gr. rep!. 26 96.96 0.74 1 . 74 0.40 0.23 - 1 4.3 -4. 1 0
VA-5 Above PMO Cement/gr. rep!. 3 97.09 0.64 1 . 63 0.57 0.35
K Y 1 30F* Above PMO Cement/gr. rep!. 8 9 7 . 95 1.01 0.76 0.25 0.33
W4B Below PMO Cement/gr. rep!. 18 9 7 . 34 0.49 1 .48 0.63 0.42 - 1 1 .8 -3.65
K Y 1 25B Above PMO Cement/gr. rep!. 20 96.95 0.66 1 . 86 0.41 0.22 - 1 3.5 -2 .23
KY76 Below PMO Cement/gr. rep!. 14 97.73 0.30 1 .50 0.49 0.33 -8.8 - 1 .68
KY2 1 8C Below PMO Cement/gr. rep!. 3 98 .00 0.38 1 . 25 0.38 0.30 -1 1 .6 -3.98
KY2 39 Below PMO Cement/gr. rep!. 20 97 .26 0.65 1 .49 0.58 0.39 -1 1 .8 -5 .75
KY 1 42 Below PMO Cement/gr. rep!. 1 96.92 0.80 1 . 88 0.39 0.2 1
KY53 Below PMO Cement/gr. rep!. 3 97.21 0 . 70 1 . 63 0.46 0.28
VA2 1 Above PMO Cement/gr. rep!. 10 97.05 0.60 1 .5 7 0.5 1 0.32
Total: 1 93 Average 0.59 1 .49 0.47 0.32 - 1 0.27 -4.72
Average (ppm) 1 424 82 1 5 2547

gr. rep!., grain replacement; n, number of electron microprobe analyses; PMO, Pine Mountain overthrust.
*Lee Formation.
t Paired with a host rock.
Table 5. Comparison of detrital compositional parameters in concretions and host rocks

Quartz Carbonate
�
;:s
cement cement Per cent Q Per cent F Per cent L �
.::.
Sample Petrography An* An(-Ab)t Ort P/F1 An/F1 Anmax Qp! Q, F/(F + Q) (%) (%) in Q1FL in Q1FL in Q1FL IGV% ....
;:s·
KY54C Concretion 2.8 5.1 94.2 0.77 0.40 1 6.5 0.36 0. 1 8 0.0 33.0 6 1 .4 1 3 .2 n,
25.4 33.0
KY54B Host rock 2.6 5.6 95.2 0.69 0.45 1 3.4 0.25 0. 1 7 2.0 0.5 59.2 1 2. 1 28.7 5.0 '0>
....

KY2 1 8A Concretion 2.6 8.2 93. 1 0.54 0.26 1 3.0 0. 1 5 0. 1 3 0.0 29.8 6 1 .3 9.0 29.7 29.8 �
;:s
KY2 1 8B Host rock 1 .2 3.0 94.5 0.66 0.25 6.2 0.25 0.26 2.5 0.0 54.9 1 9. 1 25.9 7.5 $:l...
"'
KY220A Concretion 4.8 7 .2 94.7 0.71 0.62 1 7 .0 0. 1 3 0.23 0.0 45 .5 62.4 1 8.8 1 8.8 45 .5 .::.
;:s
KY220B Host rock 1 .5 3.6 93.7 0.60 0.29 7.2 0.37 0.28 0.5 1 .0 52.9 1 9. 7 27.4 8.5
�
W4A Concretion 2.7 6.7 93.2 0.68 0.33 1 2.0 0.25 0. 1 6 0.0 27.5 62.0 1 1 .6 26.4 27.5 0
;:s
W4B Host rock 2.7 7.6 94.4 0.7 1 0.29 1 2.4 0.28 0. 1 7 0.0 1 .5 63.4 1 3. 1 23.4 4.0 i;;
KY64A Concretion 0.04 0.04 0.0 43.5 90. 3 3.5 6.2 43.5 ;:s·
KY64B Host rock 0.08 0.04 9.5 2.0 85.5 3.4 1 1 .0 23.6 s.
n,
K Y 1 25A Concretion 4.4 8.9 95.0 0.84 0.45 21.1 0.26 0.24 0.5 36.0 52.0 1 6 .0 32.0 37.0 �
K Y 1 25B Host rock 3.3 8.3 94.8 0.79 0.36 1 5.8 0.27 0.24 0.0 0.6 59.6 1 9. 1 2 1 .3 0.6 �.::.
A n , Ca-plagioclase; Anmax' composition of the most calcic plagioclase grain; F, feldspar; F" total feldspar; IGV, intergranular volume; L, lithics; P, plagioclase; iS'
'"'
;:,-
Q, quartz; Qp, polycrystalline quartz; Q" total quartz.
i:;•
*Average anorthite (An) content of plagioclase. ;:s
"'
tAverage An content of plagioclase, excluding albite (Ab).
tAverage Or content of K-feldspar.

\0
v.
96 KL. Milliken
Fig. 4. Post-quartz late calcite (c) localized on K-feldspar
remnants (k). Arrow indicates euhedral terminations of
quartz overgrowths. Back-scattered electron image. Scale
bar 1 00 Jlm.
Overlap of the compositional fields for the different
generations of calcite occurs because some analyses
of carbonate identified as 'early' fall into the field of
the 'late' calcite, whereas the opposite trend (appar­
ent 'late' calcite with 'early' composition) is not
observed. This suggests that some of the samples
with early calcite may have experienced the addi­
tion of a minor component of late calcite. Strong
spatial localization of the early calcite, however,
makes it less likely that samples dominated by late
calcite are similarly admixed.
Within some individual samples, including both
early and late calcites, Mg and Fe display a positive
covariation that may be related to zoning (Fig. 6). If
analogy can be made to the between-sample covari­
ation, it is possible that this zoning relates to a
temporal trend of relatively high contents of both
Mg and Fe in early precipitates and lower concen­
trations in later ones; however, there is no direct
petrographic evidence to support the existence of
this trend. Similar positive covariations between
Mg and Fe have been reported in other studies of
carbonate cementation in sandstones (e.g. Prosser
et a!., 1993; Milliken et a!., this volume).
Ferroan dolomite and ankerite
Ferroan dolomite (including some ankerite, > 20
mol% FeC03) (Fig. 7) is widely distributed in the
Breathitt Formation. One sample of relatively early
ferroan dolomite (KY 130F; IGV = 40%) was ob­
served. Most ferroan dolomite, however, occurs in
rocks of markedly diminished IGV ( < 20%). In
contrast to the general distribution of the late
calcite, ferroan dolomite resembles siderite in its
highly localized distribution at the thin section
scale, making its relative timing difficult to ascer­
tain.
One control on the localization of ferroan dolo­
mite is apparent fragments of non-ferroan dolo­
mite, many fractured and showing evidence of
partial dissolution and subsequent replacement by
ferroan dolomite (Fig. 7A). These cores within the
ferroan dolomite crystals are interpreted as detrital
carbonate rock fragments (CRFs), which provided
sites favourable for the precipitation of ferroan
dolomite. Similar overgrown detrital dolomites are
preserved by ferroan dolomite overgrowths in the
Tertiary sandstones of the northern Apennines
(Milliken et a!., this volume; Spadafora et a!., this
volume). CRFs are a plausible component for the
detrital assemblage in the foreland sequence of the
southern Appalachians because a significant vol­
ume of the rocks involved in thrusting are lower
Palaeozoic limestones and dolomites. The absence
of CRFs outside the cores of the ferroan dolomite
crystals suggests either that these grains generally
were not preserved in significant number during
transport, or that subsequent diagenesis has largely
removed these grains from the rocks.
Ferroan dolomite precipitation is also promi­
nently localized around partially dissolved detrital
K-feldspars (Fig. 7B). Localization is not strictly
within the volume formerly occupied by the
K-feldspar, but rather is crudely centred on the
feldspar, extending also into the surrounding pore
space. Ferroan dolomite clearly postdates quartz
cementation. In a few samples there is petrographic
evidence suggesting that ferroan dolomite pre-dates
the formation of the late calcite:
Most of the dolomites (including the detrital
cores) have calcium-enriched compositions (Table
6; Fig. 8). Zoning within the ferroan dolomite is
clearly visible in back-scattered electron images,
and the higher Fe contents tend to be characteristic
of the later zones (Fig. 7A). The degree of Ca
enrichment, however, is not strongly controlled by
Fe content and more Fe-rich portions of the crystals
display a range in Ca content that is nearly as broad
as that of less Fe-rich portions. Mn content is
correlated positively with Fe enrichment (Fig. 9).
 Non-marine foreland sandstones in the Appalachians 97

(a)
<.00 ..-
---,
3.50

3.00

• .. .
-2.50
;f. •• •
�
0 ..
5 2.00

c5
u
.!' 1.50

1.00

0.50
•
0.00
0 !;I0
0.00 1.00 2.00 3.00 <.00 5.00 6.00 7.00 R.nn

MgC03 (mole%)
(b)
1.20

1.00 0 0 • early calcite.

a late calcite

0 0 0<a 0
0.80

�
0
v
o 0 B o o 0o�oo Jir5l�o o o
B
oB

5 0.60

c5
'<! / o �0��� Cb 0

:>:
OAO
o o oDo
QJ
oo aoll��
o r:Po
t!
QrP QJ
0 • • 4

• • . • ...
o o .�.····
0 � •
•

o B'; • • •
0
ct••• • oa,t1�1- •
\
0.20 ..•
.... . . ..
Fig. 5. (a) MgC03 versus FeC03
.• Z:··· • . .. . .
content of early and late calcites. 0.00
� -�.:...
-1-----+------+---''--+------l-.=..:.
.
... : iY•.
. •
.:...
._ .:..,. --l-___,
._ ..,:....__..__
(b) FeC03 versus MnC03 content 0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00

of early and late calcites. FeC03 (mole %)

calcite postdate quartz precipitation, suggesting

Oxygen and carbon isotopic trends for calcite
and ferroan dolomite/ankerite
A generally positive covariation between 8180 and
o 13C is apparent across the data set as a whole
(Fig. 10). However, covariations between 8180 and
o13C and other parameters, e.g. carbonate content,
trace element content or trace element ratios, are
absent or very weak. Oxygen isotopic values for
both early calcite and early dolomite, combined
with temperature constraints inferred from IGVs,
are consistent with precipitation from mostly 180-
depleted fluids (Fig. 1 1). Ferroan dolomite and late
that the late carbonates generally precipitated at
temperatures at least greater than 60oC (McBride,
1989). Because temperatures of precipitation are
not well constrained between this probable mini­
mum and the maximum temperature suggested by
vitrinite reflectance data (Hower & Rimmer, 1991),
8180 values for ferroan dolomite/ankerite and, in
particular, late calcite are permissive of a wide
range of o180water compositions (Fig. 11). Two
samples (KY12A and KY 139) have petrographic
characteristics and trace elemental compositions
identical to the late calcites, but have stable isotopic
98 K.L. Milliken

3.50 ,.-
---,
•

•
3.00
0
0 0
•
•
• oo
2.50
0
0 0
0
i' • •
0
Cl) 2.00 Oi:!1J
0
.s .r?' � 0
•
o
8., 1.50 • B •
u.. 0
0

1.00 0 B
0
om OKY225
+KY220A
0.50 OKY12A
0
Fig. 6. Positive covariation between Mg and
0.00 1-------+--+--<---' Fe in two samples of early calcite (KY220A
0.00 0.50 1.00 1.50 2.00 2.50 and KY225) and one sample of late calcite
MgC03 (mole%) (KY 1 2A).

values that fall well outside range of the other
samples. Relatively 'sO-depleted fluids are possible
for most of the late calcite as well as the ferroan
dolomite, though values for D1s0wa as heavy as
+ 7o/oo cannot be ruled out if the leasb 'sO-depleted
late calcite (excluding the anomalous samples) pre-
cipitated at the maximum temperature. The afore­
mentioned covariation between D 'so and D13C
supports the notion that overall, more 180-depleted
r values correspond to later precipitates in which the
calcite contains carbon mobilized from organic
components. If the most 'sO-depleted late calcite

Table 6. Average compositions of authigenic ferroan dolomite (including ankerite)

CaC03 MgC03 FeC03 MnC03 o180 o13 C

Sample n (mol%) (mol%) (mol%) (mol%) (PDB) (PDB)
KY- 1 1 C 10 52.0 26.0 2 1 .0 0.8
KY-4 1 A 20 57.0 2 1 .4 20.3 1 .3 -9.8 -2 . 8
KY 1 0B dolo - 1 1 .0 -2 . 1
KY 1 30F 9 54.7 36.8 7.7 0.4 -6.8 -0.4
KY 1 39 10 56.9 25.5 1 6.4 1 .2
KY 1 42 6 59.0 26.7 1 3. 3 0.6
KY 1 47 - 1 2. 6 -4.3
KY 1 56 11 56.6 28. 1 1 4. 6 0.7 -9. 1 - 1 .2
KY227 12 57.2 20.6 20. 7 1.5 -9.6 -4.0
KY260B 4 56.5 1 9. 4 22.6 1.5 - 1 0. 2 -4.5
KY42A 15 56.7 20.3 2 1 .7 1.3 -9.6 - 1 .3
KY43A 24 55.8 21.1 22.0 1 .0 -1 1 .7 -6.0
KY47A I 56.2 20.7 2 1 .9 1 .2
KY 53 8 57.8 22.2 1 8.8 1.1
KY56 9 57.7 21.1 20. 1 1.1 - 1 0.3 -4.9
KY70 - 1 1 .0 -2.0
KY77A 10 55.3 1 9.8 23.3 1 .6 -7.5 0.4
KY89A - 1 1. 1 -7.6
KY9C 10 58.7 25.2 1 5 .1 0.9 -9.2 -1.3
VA-5 10 57.6 22 . 1 1 8.8 1.5
VA7 dolo - 1 0. 7 -5.0
Total average 1 69 56.6 23.6 1 8 .6 1.1 - 1 0.0 -3. 1
 Non-marine foreland sandstones in the Appalachians
Fig. 7. Thin-section scale localization of authigenic
ferroan dolomite. Back-scattered electron images. Scale
bars I 00 j.lm. (A) Ferroan dolomite (f) localized on
non-ferroan detrital dolomite (d) that has been fractured,
partially dissolved, and partially replaced by the
authigenic overgrowth. Note that the outer zones of the
ferroan dolomite (f1) are brighter, reflecting their greater
Fe content. (B) Ferroan dolomite (ankerite) (f) localized
in the vicinity of a partially dissolved and replaced
K-feldspar (k). The arrow indicates euhedral termination
on a quartz overgrowth.
precipitated at the maximum temperature, the cor­
responding 0 180water would be around + 1. 7o/oo.
Furthermore, the fact that the late calcite postdates
the ferroan dolomite, whereas the most 180-
depleted dolomite necessarily precipitated at higher
temperatures than the most 180-depleted late cal­
cite (Fig. 11), provides rather weak evidence that
the late calcite precipitated after the thermal peak in
the basin.
99
Sr concentration and 87Sr/86Sr ratio in calcite
and ferroan dolomite/ankerite
Sr concentrations and 87Sr/86Sr ratios were deter­
mined for a limited set of carbonate-rich samples
(Table 7). 87Sr/86Sr ratios in early and late calcites
and in dolomite/ankerite are well above the range
for marine Sr during the Phanerozoic (based on a
comparison with the data of Burke et al. , 1982),
suggesting a predominantly silicate derived (i.e.
radiogenic) source of Sr, in both early and late
diagenesis. Sr concentration in ferroan dolomite/
ankerite is uniformly below the detection limit. In
calcite, Sr concentration, though well above the
detection limit in several samples, does not vary
systematically between early and late calcites, with
87Sr/86Sr ratios, or with other trace elements in the
calcites, either within or between samples.
DISCUSSION
Preservation of consistent trace elemental and iso­
topic variations among petrographically distinct
early and late calcites and ferroan dolomites shows
that complete chemical homogenization through
replacement (recrystallization) of the carbonate as­
semblage has not occurred, despite a significant
history of syn-Alleghanian burial and prolonged
postorogenic uplift. Survival of early precompac­
tional carbonates of distinctive chemistry both
above and below the PMO further supports the
notion that the chemical history of carbonate pre­
cipitation in these rocks has not been obscured or
erased through tectonic overprinting. Preservation
of 'primary' compositions in the strictest sense
cannot be verified, however, because variable de­
grees of partial restabilization through dissolution
and precipitation (recrystallization) of an initial
precipitate (early or late) may yield a wide range of
trace elemental and isotopic compositions (Banner
& Hanson, 1990), but leave little discernible petro­
graphic evidence (e.g. Milliken & Land, 1993;
Lynch & Land, 1996).
Possible control of trace element partitioning by
factors other than elemental concentration in the
fluid, for example by sector zoning (Reeder &
Prosky, 1986; Reeder & Grams, 1987; Reeder &
Paquette, 1989), precipitation rate (e.g. Lorens,
1981), temperature (e.g. Mucci, 1987), competing
ion effects (e.g. Pingitore & Eastman, 1986), or
bacterial processes (Coleman, 1993; Folk, 1993;
 100 KL. Milliken

50.00
,_
.... .. o FeC03 <1 5 mole� �
I
..
.. o FeC03 > 1 5 mole%
45.00 ... • detrital dolomite

0
40.00 0 0
0
0
0
i 0
0
.!! 35.00 0
0 0
! 0
0 0
0

8 30.00 oo
"'
:lE 6oo »oo
oo eo
o\3'0 0
0 0 �
25.00 0
e oi'rP

�
0 0
0 l:l
0 (6l
20.00
'b 0

15.00 +---+-----<>---<--t---+--1 Fig. 8. Ca content versus Mg content for detrital

48.00 50.00 52.00 54.00 56.00 58.00 60.00 62.00 and authigenic dolomite, plotted for different Fe
caco, (mole 96) contents.

Vasconcelos et al., 1995), unfortunately, remains that neither seawater nor strongly rock-dominated
unconstrained and unassessed in this example. c 80-enriched) thrust-derived fluids were signifi­
Within these limitations it is only possible to cantly involved in supplying material for carbonate
speculate broadly concerning the significance of precipitation, though some component of 180-
elemental variation in the carbonate. enriched fluids cannot be ruled out, especially in the
The relative temporal variation in carbonate case of the late calcite. Nevertheless, radiogenic Sr
mineralogy, with siderite followed sequentially by compositions implicate reacting detrital silicates as
early Mg- and Fe-enriched calcite, ferroan dolo­ a significant factor, even for the early calcites.
mite, Mn-enriched ankerite, and finally late Mn­ The lack of strong covariation between isotopic
enriched calcite, documents apparent fluctuations values and trace element contents argues that the
in the availability of components for carbonate controls on these parameters were at least some­
precipitation. The 180-depleted nature of carbon­ what independent. The absence of covariation be­
ates across this entire temporal sequence suggests tween Mg content and b180calcite' or between the

1.40 .-------,

':)
•
1.20 .
•
• .
.
.

•
:·1- i·=.. .
1.00 .· -�}.: ,

i •. , . .. .

.!! 0.80 I
.
•

·'' ..
••

! •
.

8 0.60 .•
•
c:
:lE
....
0.40 ,;t.
'
.
0.20 .
.

0.00 t-----+----+--->--1
0.00 o.so 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 Fig. 9. Fe/Mg molar ratio versus Mn
Fe/Mg (molar ratio) content for ferroan dolomite/ankerite.
 Non-marine foreland sandstones in the Appalachians 10 1

6.0
•

4.0
• • •
� early calcite 'I
2.0

0.0
I o late calcite
• ferroan do1omite

•
·� •
J;J 0
-2.0
-<0 • 00.

0
Fig. 10. o 1 80 versus o 1 3C for calcites and --4.0 0 0
0
ferroan dolomite/ankerites. For the purpose of .
. . 0
comparison with calcite, o 1 80 values for • 0 •
-6.0
dolomite/ankerite are adjusted by -3%o, to 0
••
compensate for approximate differences in
-8.0
dolomite-water fractionation relative to calcite
(Land, 1 980). This correction allows for a better
0
separation of the calcite and dolomite/ankerite -10.0

-16.0 - 1 -4.0 -12.0 -10.0 -8.0 ·6.0 ·4.0 -2.0 0.0

data points, better revealing the contrasts in the
conditions of precipitation of these phases. o'6o

timing of calcite precipitation and Sr content or
87Sr/86Sr, tends to rule out simple mixing between
seawater and meteoric water as a factor in the
observed compositional variations. The strong posi­
tive covariation between Mg and Fe tends to rule
out both a simple mixing model between marine
and meteoric fluids and also recrystallization of an
initially marine precipitate, both of which would
produce a negative covariation for Mg and Fe (see
Veizer, 1 983).
Sources for Ca, Mg, Fe and Mn �re most likely
ones affiliated with reacting detrital components on
an intrabasinal, if not an intraformational, scale.
Early in diagenesis, unstable Fe and Mn oxyhydrox­
ides (e.g. Barnaby & Rimstidt, 1 989, and references
therein) are a plausible source of materials for
siderite. Mg adsorbed on to clays is a possible
source for easily mobilized Mg for the early siderite
and calcite. The shift from siderite to calcite precip­
itation may reflect the waning of these easily mobi-

1 4 0 r------.---.

Fig. 1 1 . Range of possible

temperature and 0 1 80wa ter 1 20

conditions for calcite and dolomite

(ankerite) precipitation. Calculated maximum temperature

from the calcite-water 1 00

,
\
�·
fractionation equation in
late dolomite/ankerite � " \

l
Friedman & O'Neil ( 1977). As in
1
late calcite

Fig. 1 0, o 1 80 values plotted for

f 8
0
min. temp of
dolomite/ankerite are adjusted
.;! quartz ppt. ??
-3%o from measured values to f!
compensate for differences in Gl ,
sa
E
- - - - - - - - - - - - - - ..of: - --

fractionation relative to calcite

GI
(Land, 1 980), thus allowing the 1-

\
same fractionation equation to be
40
used for both minerals. (This
correction effectively turns values
for dolomite/ankerite into 'calcite anomalous late calcite
values' for the purpose of 20

examining the range of

temperature and o 1 80water early calcite
early dolomite
conditions in effect during
precipitation, and is approximate -15.0 -10.0 -5.0 0.0 5.0 10.0

only.) o"Ow"" (SMOW)

1 02 K.L. Milliken

Table 7. Sr concentration and Sr isotopic composition of selected calcites and ferroan dolomites

Sample Location Petrography n SrC03 (mol%)

'Early' calcite concretionary and generalized forms

KY l l G Below PMO Concretion 5 0.06
KY43C Below PMO Concretion 5 0. 1 6
KY47A Below PMO Cement 10 0. 1 8
KY68 Below PMO Cement 10 0.09
KY 1 2 5A Above PMO Concretion 10 0.08 0. 7 1 1 6
KY225 Below PMO Concretion 4 0.05 0.7 1 0 1
TN 1 1 1 Above PMO Cement 11 0. 1 5
KY68 Below PMO Cement 10 0.09
KY87 Below PMO Cement 8 0. 1 6
Average 0. 1 1
Average (ppm) 917

'Late' calcite-low IGV, post-quartz

KY- 1 2A Below PMO Cement/gr. rep!. 5 0.04 0.7 1 20
KY48 Below PMO Cement/gr. rep!. 12 0.20 0.7 1 03
Y l 30F Above PMO Cement/gr. rep!. 4 0.06
W4B Below PMO Cement/gr. rep!. 4 0.05 0. 7 1 0 1
KY 1 25B Above PMO Cement/gr. rep!. 10 0.23 0. 7 1 1 4
KY239 Below PMO Cement/gr. rep!. 6 0.07 0.7096
VA2 1 Above PMO Cement/gr. rep!. 10 0.27
Average 0. 1 3
Average (ppm) 1 1 36

Early dolomite
KY1 30F Above PMO Cement/gr. rep!. bd1 0.7089

Late dolomite
KY26C Above PMO Cement/gr. rep!. bdl 0.7 1 46

bdl, below detection limit; gr. rep!. , grain replacement; PMO, Pine Mountain overthrust.

lized sources. Later diagenesis requires renewed
sources of Mg, Fe and Mn for precipitation of
ferroan dolomite and ankerite. One possible source
of these elements is reacting detrital clay in the
associated mudrocks (e.g. Boles, 1 978; Boles &
Franks, 1 979), now highly illitic. Fluid expelled
from deformed and altered dolomitic rocks to the
southeast represents an alternative source for anker­
ite components. However, the average composition
of ferroan dolomite/ankerite in these sandstones is
dissimilar to that observed in the deformed belt in
which late dolomites are more enriched in 1 80 and
13C, less Ca enriched, and relatively depleted in Fe
and Mn (see Barnaby & Read, 1 992; Montanez &
Read, 1 992; Schedl et al., 1 992). If a tectonically
derived fluid constituted a component of the fluid
responsible for ferroan dolomite/ankerite precipita­
tion, an additional, perhaps more local, source for
Fe and Mn is required in addition to the consider­
able admixture of 180-depleted water.
Late calcite precipitation suggests the ultimate
waning of the late diagenetic source of Fe and Mg,
although Mn concentrations remained relatively
high, perhaps reflecting derivation of this element
from a source that, at least initially, was rather
resistant to reaction, for example Mn garnets. Late
calcites of deep burial origin have also been identi­
fied in the underlying Newman Limestone (Missis­
sippian) (Niemann & Read, 1 98.8; Nelson & Read,
1 990). Although enriched in Fe and Mn relative to
early diagenetic calcites in the limestones, late
calcites in the Newman differ in composition from
the late calcites described here, having Mg
< I 000 ppm, Fe < 6000 ppm, and Mn < I 000 ppm
(Niemann & Read, 1 98 8 ; Nelson & Read, 1 990).
This further supports the contention that sources of
these elements were at least partially internal to the
Pennsylvanian siliciclastic section.
Sources of Ca are very uncertain. No detrital
carbonate (apart from the trivial volume of detrital
dolomite described above) survives in the sand­
stones (including the early concretions). Carbonate
 Non-marine foreland sandstones in the Appalachians
content is also limited in associated shales, being
most abundant in the form of skeletal debris in the
thin but regionally extensive marine units (Cobb et
a!., 1981). Alternatively, if Ca was derived in
significant amounts from, say, pressure dissolution
of underlying limestones and dolomites, it is again
not strongly reflected in the oxygen, carbon or
strontium isotopic data. There is no petrographic
evidence for the dissolution of early-formed calcite
cements to provide materials for precipitation later
in diagenesis. The lack of strongly contrasting com­
positions between Ca plagioclases in concretions
and host rocks certainly argues against local silicate
sources of Ca. In general, the timing of Ca­
plagioclase alteration in basinal diagenesis (e.g.
Milliken et a!. 1989) would allow this as a plausible
source for some of the Ca in the late calcite, but the
lack of preserved information on the composition
of the initial feldspar assemblage in these sand­
stones makes any attempt at quantifying this source
highly speculative.
Localization of ferroan dolomite/ankerite and
calcite on partially replaced (i.e. dissolved) detrital
feldspars provides some clues to the nature of the
chemical system responsible for carbonate emplace­
ment in late diagenesis. Partially replaced K-felds­
pars are the most common locus for late carbonate
precipitation. Because K-feldspars do not share in
elements in common with the authigenic carbon­
ates, their role in the localization of carbonate is
necessarily more indirect than simply providing
locally higher concentrations of elements required
for carbonate precipitation. Local pH buffering
through dissolution (in essence, subsurface weath­
ering) is one mechanism through which detrital
feldspars could serve to localize carbonate precipi­
tation. The very strong spatial affiliation of late
carbonates with partially dissolved silicates indi­
cates a chemical system in which carbonate precip­
itation was not possible outside the very local region
of this buffering action, suggesting that the fluids
involved were prominently acidic.
CONCLUSIONS
Pennsylvanian deltaic sandstones of the Appala­
chian basin contain relatively minor amounts of a
multimineralic carbonate assemblage. 1 so-depleted
fluids are implicated in the precipitation of most of
these phases, with the possible exception of the
latest generation of calcite. Elemental variation
103
survives between early and late generations of
calcite, and is manifest as zoning in back-scattered
electron images of siderite and ferroan dolomite/
ankerite. The absence of covariation between trace
elemental and isotopic values suggests that the
controls on these parameters are not strongly
linked. Numerous unconstrained controls on the
partitioning of trace elements into carbonates ren­
der any interpretation of specific fluid compositions
speculative. Together, the likely I so-depleted char­
acter of most of the fluids, the radiogenic nature of
the Sr in the carbonate, and the post-quartz cement
timing of ferroan dolomite/ankerite and late calcite,
argue that materials for carbonate precipitation
were largely derived from the detrital fraction. No
strong role for rock-buffered orogen-derived fluids
is evident, suggesting that any such fluids either
emerged from the thrust belt within the more
deformed parts of the orogen or were masked by
dilution with gravity-driven meteoric fluids in the
foreland.
AC KNO WLEDGEMENTS
This study was supported by the Donors of the
Petroleum Research Fund of the American Chemi­
cal Society (ACF-PRF 22805-AC8). Partial support
of publication costs was provided by the Owen­
Coates Fund of the Geology Foundation, Univer­
sity of Texas at Austin. Field sampling was assisted
at various times by the good company of Steve Seni,
Rachel Eustice, Earle McBride and Katy Milliken. I
am grateful to Leo Lynch for the XRD analyses of
carbonate percentages, to Guoqiu Gao for Sr­
isotopic analyses, and to Rachel Eustice and Lynton
Land for their assistance with the carbon and
oxygen isotopic analyses. An early version of this
manuscript benefited from commt;nts by Earle Mc­
Bride and Lynton Land. The many constructive
suggestions of reviewers Phillipe Muchez and
Calum Macaulay are also gratefully acknowledged.
REFERENCES
AwwiLLER, D.N. ( 1992) Geochemistry, mineralogy, and
burial diagenesis of Wilcox Group shales. PhD disserta­
tion, University of Texas, Austin.
BACHTADSE, V., VAN DER Voo, R., HAYNES, F.M. & KESLER,
S.E. ( 1 987) Late Paleozoic magnetization of mineral­
ized and unmineralized Ordovician carbonates from
east Tennessee: evidence for a post-ore chemical event.
1 04 K.L. Milliken
J. Geophys. Res. , 92 , 1 4 1 6 5- 1 4 1 76.
BANNER, J.L. & HANSON, G.N. ( 1 990) Calculation of
simultaneous isotopic and trace element variations dur­
ing water-rock interaction with applications to carbon­
ate diagenesis. Geochim. Cosmochim. Acta, 54 , 3 1 23-
3 1 37.
BARNABY, R.J. & READ, J.F. ( 1 992) Dolomitization of a
carbonate platform during late burial: Lower to middle
Cambrian Shady Dolomite, Virginia, Appalachians. J.
sediment. Petrol. , 62 , 1 023- 1 043.
BARNABY, R.J. & RIMSTIDT, J.D. ( 1 989) Redox conditions
of calcite cementation from Mn and Fe contents of
authigenic carbonates. Geol. Soc. Am. Bull., 1 0 1 , 795-
804.
BEAUMONT, C., QUINLAN, G.M. & HAMILTON, J. ( 1 987) The
Alleghenian orogeny and its relationship to the evolu­
tion of the eastern interior, North America. In: Sedi­
mentary Basins and Basin-forming Mechanisms (Eds
Beaumont, C. & Tankard, A.J.). Can. Soc. Petrol. Geol.
Mem., 12, 425-445.
BETHKE, C.M. & MARSHAK, ( 1 990) Brine migrations across
North America-the plate tectonics of groundwater.
Ann. Rev. Earth Planet. Sci. , 18, 287-3 1 5 .
BOETTCHER, S.S. & MILLIKEN, K.L. ( 1 994) Mesozoic­
Cenozoic unroofing of the southern Appalachian basin:
apatite fission track evidence from middle Pennsylva­
nian sandstones. J. Geol. , 102, 65 5-663.
BOLES, J.R. ( 1 978) Active ankerite cementation in the
subsurface Eocene of southwest Texas. Contrib. Min­
eral. Petrol. , 68 , 1 3-32.
BoLES, J.R. & FRANKS, S.G. ( 1 979) Clay diagenesis in
Wilcox sandstones of southwest Texas: implications of
smectite diagenesis on sandstone cementation. J. sedi­
ment. Petrol. , 49, 5 5-70.
BOLES, J.R. & JOHNSON, K.S. ( 1 983) Influence of mica sur­
faces on pore water pH. Chern. Geo/. , 43, 303-3 1 7.
BuRKE, W.H., DENISON, R.E., HETHERINGTON, E.A. et a/.
( 1 982) Variation of seawater 8 7 Sr/86Sr throughout
Phanerozoic time. Geology, 10 , 5 1 6-5 1 9 .
CAROTHERS, W.W., ADANI, L.H. & ROSENBAUER, R.J.
( 1 988) Experimental oxygen isotopic fractionation be­
tween siderite-water and phosphoric acid liberated
C02-siderite. Geochim. Cosmochim. Acta, 52 , 2445-
2450.
CATHLES, L.M ( 1 990) Scales and effects of fluid flow in the
upper crust. Science, 248 , 323-329.
ClBIN, U., CAVAZZA, W. , FONTANA, D., MILLIKEN, K.L. &
McBRIDE, E. F. ( 1 993) Comparison of composition and
texture of calcite-cemented concretions and host sand­
stones, northern Apennines, Italy. J. sediment. Petrol. ,
63, 945-954.
COBB, J.C., CHESNUT, D.R., HESTER, N.C. & HOWER, J.C.
( 1 98 1 ) Coal and coal-bearing rocks of eastern Kentucky.
Geol. Soc. Am. Coal Division, Annual Field Trip,
1 69 pp.
CoLEMAN, M. ( 1 993) Microbial processes: controls on the
shape and composition of carbonate concretions. Mar.
Geol. , 1 13, 1 2 7- 1 40.
DEMING, D., NuNN, J.A. & EvANS, D.G. ( 1 990) Thermal
effects of compaction-driven groundwater flow from
overthrust belts. J. Geophys. Res. , 95, 6669-6683.
ELLIOT, W.C. & ARONSON, J.L. ( 1 987) Alleghanian episode
of K-bentonite illitization in the southern Appalachian
Basin. Geology, 1 5 , 7 3 5-739.
FERM, J.C. ( 1 974) Carboniferous environmental models in
eastern United States and their significance. In: Car­
boniferous of the Southeastern United States (Ed. Briggs,
G.). Geol. Soc. Am. Spec. Paper, 148, 79-9 5 .
FoLK, R.L. ( 1 993) SEM imaging o f bacteria and nanno­
bacteria in carbonate sediments and rocks. J. sediment.
Petrol. , 63, 990-999.
FRIEDMAN, I. & O'NEIL, J.R. ( 1 977) Compilation of stable
isotope fractionation factors of geochemical interest. In:
Data of Geochemistry, 6th edn, (Ed. Fleischer, M.). US
Geol. Surv. Prof. Paper, 440K, Chapter KK.
HEARN, P.P. & SUTTER, J.F. ( 1 985) Authigenic potassium
feldspar in Cambrian carbonates: evidence of Al­
leghanian brine migration. Science, 228, 1 529- 1 5 3 1 .
HowER, J.C. & RIMMER, S.M. ( 1 99 1 ) Coal rank trends in
the central Appalachian coalfield: Virginia, West Vir­
ginia, and Kentucky. Org. Geochem. , 17, 1 6 1 - 1 73 .
LAND, L.S. ( 1 980) The isotopic and trace element chemis­
try of dolomite: the state of the art. In: Concepts and
Models of Dolomitization (Eds Zenger, D. H. Dunham,
J.B. & Ethington, R.L.). Soc. Econ. Paleontol. Mineral.,
Tulsa, 28, 87- 1 1 0.
LORENS, R.B. ( 1 98 1 ), Sr, Cd, Mn and Co distribution
coefficients in calcite as a function of calcite precipita­
tion rate. Geochim. Cosmochim. Acta, 45, 5 5 3-56 1 .
LYNCH, F. ( 1 997) Mineralogy of Frio Formation shales
and the stoichiometry of the smectite to illite reaction­
the most important reaction in sedimentary diagenesis.
Clays Clay Miner. , 45, 6 1 8-63 1 .
LYNCH, F.L. & LAND, L.S. ( 1 996) Diagenesis of calcite
cement in Frio Formation sandstones and its relation­
ship to formation water chemistry. J. sediment. Res. ,
A66, 439-446.
McBRIDE, E.F. ( 1 989) Quartz cementation in sandstones:
a review. Earth Sci. Rev., 26, 69- 1 1 2 .
McCREA, J.M. ( 1 950) O n the isotope chemistry o f carbon­
ates and a paleotemperature scale. J. Chern. Phys., 18 ,
849-8 5 7 .
MILLIKEN, K.L. ( 1 992) Regional diagenetic variations in
middle Pennsylvanian foreland basin sandstones of the
southern Appalachians. Geo/. Soc. Am. Ann. Meeting
24, 57-58 (Abstract).
MILLIKEN, K.L. & LAND, L.S. ( 1 993) The origin and fate of
silt-sized carbonate in subsurface Miocene-Oligocene
mudrocks, South Texas Gulf Coast. Sedimentology, 40,
1 07-1 24.
MILLIKEN, K.L., McBRIDE, E.F. & LAND, L.S. ( 1 989)
Numerical assessment of dissolution versus replace­
ment in the subsurface destruction of detrital feldspars,
Oligocene Frio Formation, South Texas. J. sediment.
Petrol. , 59, 740-757.
MITRA, S. ( 1 988) Three-dimensional geometry and kine­
matic evolution of the Pine Mountain thrust system,
southern Appalachians. Geol. Soc. Am. Bull. , 100, 72-
95.
MONTANEZ, I.P. & READ, J.F. ( 1 992) Fluid-rock interac­
tion history during stabilization of early dolomites,
Upper Knox Group (Lower Ordovician), US Appala­
chians. J. sediment. Petrol. , 62, 7 5 3-778.
MozLEY, P.S. ( 1 989) Relation between depositional envi­
ronment and the elemental composition of early diage­
netic siderite. Geology, 17, 704-706.
 Non-marine foreland sandstones in the Appalachians
MOZLEY, P.S. & WERSIN, P. ( 1 992) Isotopic composition of
siderite as an indicator of depositional environment.
Geology, 20, 8 1 7-820.
Mucci, A. ( 1 9 87) Influence of temperature on the compo­
sition of magnesium calcite overgrowths precipitated
from seawater. Geochim. Cosmochim. Acta, 5 1 , 1 9 77-
1 9 84.
NELSON, W.A. & READ, J.F. ( 1 990) Updip to downdip
cementation and dolomitization patterns in a Mississip­
pian aquifer, Appalachians. J. sediment. Petrol. , 60,
379-396.
NIEMANN, J.C. & READ, J.F. ( 1 988) Regional cementation
from unconformity-recharged aquifer and burial fluids,
Mississipian Newman Limestone, Kentucky. J. sedi­
ment. Petrol. , 58 , 688-705 .
O'HARA, K., HowER, J.C. & RIMMER, S.M. ( 1 990) Con­
straints on the emplacement and uplift history of the
Pine Mountain thrust sheet, eastern Kentucky. Evi­
dence from coal rank trends. J. Geol. , 98, 43-5 1 .
OLIVER, J. ( 1 986) Fluids expelled tectonically from oro­
genic belts: their role in hydrocarbon migration and
other geologic phenomena. Geology, 14, 99- 1 02.
PINGITORE, N.E. & EASTMAN, M.P. ( 1 986) The coprecipi­
tation of Sr with calcite at 25 ·c and 1 atm. Geochim.
Cosmochim. Acta, 50 , 2 1 95-2203.
PROSSER, D.J., 0AWS, J.A., FALLICK, A.E. & WILLIAMS,
B.P.J. ( 1 993) Geochemistry and diagenesis of stra­
tabound calcite cement layers within the Rannoch
Formation of the Brent Group, Murchison Field, North
Viking Graben (northern North Sea). Sediment. Geol. ,
105
87, 1 3 9- 1 64.
QUINLAN, G.M. & BEAUMONT, C. ( 1 984) Appalachian
thrusting, lithosphere flexure, and the Paleozoic stratig­
raphy of the eastern interior of North America. Can. J.
Earth Sci., 2 1 , 973-996.
REEDER, R.J. & GRAMS, J.C. ( 1 987) Sector zoning in calcite
cement crystals: implications for trace element distribu­
tion in carbonates. Geochim. Cosmochim. Acta, 5 1 ,
1 8 7- 1 94.
REEDER, R.J. & PAQUETTE, J. ( 1 989) Sector zoning in
natural and synthetic calcites. Sediment. Geol. , 65,
239-247.
REEDER, R.J. & PROSKY, J.L. ( 1 986) Compositional sector
zoning in dolomite. J. sediment. Petrol. , 56 , 237-247.
SCHEDL, A., McCABE, C., MONTANEZ, 1., FULLAGAR, P.O. &
VALLEY, J.W. ( 1 992) Alleghenian regional diagenesis: a
response to the migration of modified metamorphic
fluids derived from beneath the Blue Ridge-Piedmont
thrust sheet. J. Geol. , 100, 339-352.
TANKARD, A.J. ( 1 986) Depositional response to foreland
deformation in the Carboniferous of eastern Kentucky.
Bull. Am. Ass. Petrol. Geol. , 70, 8 5 3-868.
VASCONCELOS, C., MCKENZIE, J.A., BERNASCONI, S., DRUJIC,
D. & TIEN, A.J. ( 1 995) Microbial mediation as a possi­
ble mechanism for natural dolomite formation at low
temperatures. Nature, 377, 220-222.
VEIZER, J. ( 1 983) Trace elements and isotopes in sedimen­
tary carbonates. In: Carbonates. Mineralogy and Chem­
istry (Ed. Reeder, R.J.). Rev. Mineral., Mineral. Soc.
Am., 1 1 , 265-299.
Spec. Pubis int. Ass. Sediment. ( 1 998) 26, 1 07- 1 40

Palaeogeographical, palaeoclimatic and burial history controls

on the diagenetic evolution of reservoir sandstones:
evidence from the Lower Cretaceous Serraria sandstones in the
Sergipe-Alagoas Basin, NE Brazil

2
A . J . V. G A R C I A* 1, S. MORA D*, L.F. DE ROS* and I . S . A L - A A S M t
*Sedimentary Geology Research Group, Institute o fEarth Sciences, Uppsala University, ·

Norbyvagen 18B, 5-752 36, Uppsala, Sweden,

e-mail sadoon. morad@geo. uu.se; and
tDepartment ofEarth Sciences, University of Windsor, Windsor, Ontario N9B 3P4, Canada,
e-mail alaasm@delta. uwindsor. ca

ABSTRACT

The Serraria Formation (Early Cretaceous) was deposited prior to the Proto-Atlantic Rift rupture b y a
braided fluvial system crossing a wide cratonic basin from N to NW, and is an important hydrocarbon
reservoir in several oilfields in the Sergipe-Alagoas Basin, northeastern Brazil. Four diagenetic
domains with different palaeogeographical, palaeoclimatic and lithofacies settings, as well as burial
histories, were distinguished, which roughly correspond to original locations in the fluvial depositional
system . During eodiagenesis ( 1 44- 1 40 Ma) the palaeoclimate was arid to semi-arid and resulted in the
precipitation Of vadose and phreatic calcite (013CPDB -6.7%o; 018QPDB -8.2%o) in the proxima}
= =

deposits and phreatic dolomite (013CPDB -8.5 to -3. 1 %o; 0180pos -8.7 to -6. 7%o) in the distal
= =

deposits. Both calcite and dolomite formed in the middle deposits. During the syn-rift subsidence and
mesodiagenesis (>::: 1 40- 1 1 8 Ma), new generations of calcite (o13CPDs - 1 1 .2 to - 3 .5%o; o180p08
= =

- 1 3.6 to -JJ. 5%o) and dolomite (013CPDB - 1 1 . 2 to -2.6%o; 0180PDB - 1 0.6 to -4. l %o) precipitated
= =

in the proximal (maximum T>::: too·q and distal (maximum T>::: 70- l OO·q domains, respectively.
The best reservoirs of the Serraria Formation are sandstones of the distal area that were affected by
pervasive telogenetic dissolution of carbonate cements and silicates, and the formation of kaolinite.
This occurred mostly at the beginning of the post-rift uplift ( 1 1 4-74 Ma), when warm and humid
conditions prevailed. The post-rift subsidence and mesodiagenesis ( 1 1 5 Ma until now) resulted in
further precipitation of calcite and dolomite with similar isotopic values to the syn-rift cements
(013CPDB -J2.3%o; 0180pos -8.2 to -6.9%o) in the proxima} and distal domains, respectively.
= =

Other mesogenetic processes include dickitization and illitization of kaolinite, illitization and
chloritization of smectite, albitization of feldspars, and precipitation of quartz cement. Recent
exposure and telodiagenesis, �ffecting sandstones along margins of the rift basin in the middle domain,
is resulting in dissolution of silicates and precipitation of quartz, chalcedony and iron oxides.

I NT R O D UCT I O N

The distribution patterns and geochemical compo-
' Present address: Universidade do Vale do Rio dos
Sinos-UNISINOS, Sedimentary Geology Program, Av.
Unisinos, 9 50, CEP 9 3 . 022-000, Sao Leopolda, RS,
Brazil, e-mail garcia@dgeo .unisinos.tche.br.
2 Present address: Universidade Federal do Rio Grande
do Sui, lnstituto de Geociencias, Av. Bento Goncalves,
9 500, CEP 9 1 50 1 -970, Porto Alegre, RS, Brazil, e-mail
ifderos@if.ufrgs.br.
sition of diagenetic minerals, such as carbonates
and clay minerals, in continental sandstones play an
important role in reservoir quality evolution and
give important clues to the palaeoclimatic and
palaeohydrological conditions (e.g. Carlisle, 1 983;
Dickinson, 1 987; Suchecki et a!., 1 988; Mozley &
Hoemle, 1 990; Dutta, 1 992; Spotl & Wright, 1 992;
Wright, 1 992; Marriot & Wright, 1 993; Thyne &
Gwinn, 1 994). The origin, timing and importance

Carbonate Cementation in Sandstones Edited by Sadoon Morad 1 07

© 1998 The International Association of Sedimentologists
ISBN: 978-0-632-04777-2
1 08 A.J. V Garcia et al.

of secondary porosity formed by grain and cement
dissolution have b een strongly debated for more
than two decades (Schmidt & McDonald, 1 979;
Curtis, 1 98 3 ; Franks & Forester, 1 984; Lundegard
& Land, 1 986; Giles & Marshall, 1 986; Surdam et
al., 1 989a; Bj0rlykke et al., 1 989). Unravelling these
aspects is important in understanding and predict­
ing porosity distribution and the overall diagenetic
evolution of siliciclastic sequences. Most of these
earlier studies have poorly constrained the com­
bined effects of basinwide variations in the palaeo­
geographical and palaeoclimatic settings and burial
histories on diagenetic and porosity evolution.
The aims this study are to elucidate the factors
controlling the distribution pattern, mineralogy
and geochemical composition of calcite and dolo­
mite cements, and of grain dissolution and kaolin­
ization in the fluvial sandstones of the Serraria
Formation, northeastern Brazil (Fig. 1 ). This unit
provides an excellent opportunity to examine the
influence of variations in the palaeogeographical
setting, palaeoclimate and burial history on diage­
netic processes.
Variations in palaeogeographical and palaeocli­
matic settings, lithofacies and burial histories con­
trolled the distribution and intensity of diagenetic

o·
20"8��
Lower Cretaceous
Paleolatitude

5 E
X

Sergipe-Aiagoas Basin
s��� )( E
Serraria Fm
Bananeiras Fm

0
Legend
Mainly Fluvial and
Eolian Sandstones
Mainly. Fine Lacustrine
Deposits
Fig. 1. Sedimentary records of
pre-rift sequences in the
northeastern Brazil and adjacent
0 50 1 00 150 km
African regions. Modified from
Szatmari et al. (I987), and Ponte
& Asmus (I 976, 1978).
 Lower Cretaceous Serraria sandstones
Fig. 2. Structural compartments of
the Sergipe-Alagoas basin and
studied areas. Distal domain: CB,
Caloba arena (including Divina
Pastora and Japaratuba lows, and +
Aracaju high); RB, Robalo area;
middle domain: FD, Feliz Deserto
area; JP, Japoata-Penedo area
,,.
(including outcrops); proximal
domain: FU, Furado; SMC, Sao
Miguel dos Campos; and CSMC,
Cidade de Sao Miguel dos
Campos, oilfields of Furado area.
processes in different areas of the basin. Four
diagenetic domains were recognized, which roughly
correspond to original locations in the fluvial de­
positional system (Fig. 2): (i) proximal domain
(Furado, Sao Miguel dos Campos and Cidade de
Sao Miguel dos Campos oilfields area); (ii) middle
domain (Japoata-Penedo area); (iii) shallow distal
domain (Caioba oilfield area); and (iv) deep distal
domain (Robalo area).
GEOLOGICAL S E T T I N G
Palaeogeography and palaeoclimate
During the Late Jurassic/Early Cretaceous, a large
2
area (�500 000 km , 20-25.S) of the Gondwana
palaeocontinent, currently northeastern Brazil and
western Africa, was covered by a vast depression
1 09
Sao Miguel dos
Campos Platform
+
""" �osqueiro Low 40 km
(Afro-Brazilian depression) (Ponte, 1 97 1 ). The sedi­
mentary sequences deposited in this depression are
now preserved in several smaller basins which re­
mained after the fragmentation, uplift and erosion
that later affected this area, including the Araripe,
Almada, Camamu, Rec6ncavo, Tu�ano, Jatoba, and
Sergipe-Alagoas Basins in Brazil, and the Gabon
and Congo-Cabinda Basins in Africa (Fig. 1 ). Dur­
ing the Upper Jurassic most of this depression was
occupied by a shallow lacustrine system, which re­
sulted in the deposition of Alianca and Bananeiras
mudstone formations (Figs 1 and 3A). Climatic
changes caused frequent expansion and contraction
of the lacustrine system. At the beginning of the Early
Cretaceous (Berriasian), an endorheic drainage sys�
tern partially filled the depression. In the northern
part, extensive braided fluvial systems ( ,;;:; 400 km
long) which flowed from the N-NW margins to a
shallow lacustrine complex in the centre of the de-
1 10 A.J V Garcia et a!.

pression (Fig. 3A) deposited the braided fluvial, ae­ The prevailing palaeoclimatic conditions in
olian and ephemeral lacustrine deposits of the Sergi Gondwana, the sedimentary deposits in the basin,
and Serraria Formations (Fig. 3B). and the regional stratigraphical relationships charac-

2oosoo�
�
Afro-Brasilian Depression

Fluvial and Eolian (<} ) Deposit

(Serraria +Sergi Fm.)
Shallow Lacustrine Deposits
with Subaereal Exposure ( �
during Dry Periods (Bananeiras Fm)
Lower Fluvial Sequence
(Candeeiro Fm +Boipeba/Aiian�a Fm)
Paleozoic Sediment Rocks
Igneous-Metamorphic Basement
Lacustrine Depocenters
Conifer Forest

Fig. 3. Palaeogeographical reconstructions of Serraria Formation. Early Cretaceous pre-rift sedimentation. (A) At the
time of maximum extension of the lacustrine system and beginning of the fluvial sedimentation of the Serraria
sandstones; (B) at the time of maximum expansion of the Serraria fluvial system. Modified from Garcia ( 1992).
 Lower Cretaceous Serraria sandstones 111

terize the Afro-Brazilian depression as a peridesertic
region with endorheic and asymmetric drainage
(Garcia & Wilbert, 1 995). At the northern margin of
the depression, pine forests developed under favour­
able palaeoclimatic conditions. In contrast, a more
arid climate to the south led to the development of
widespread aeolian deposits. More extensive fluvial
systems spread from the northern rather than from
the southern region into the depression, owing to
higher rainfall (Garcia, 1 9 9 1 b). Significant flora, rep­
resented by Early Cretaceous palynomorphs, coali­
fied fronds and pinnae, and silicified conifer logs,
and dinosaur fauna inhabited the proximal and
middle fluvial areas (Fig. 3B).
Still during the Early Cretaceous, the establish­
ment of the initial outlines of continental margins
in the African and South American plates (proto-rift
phase) caused subsidence of the central part of the
depression and the development of a new shallow
lacustrine environment. After this, narrow rift ba­
sins developed and Gondwana was fragmented.
Stratigraphy and depositional evolution
A thick sedimentary column (up to 1 0 km in the
depocentres) accumulated in the Sergipe-Alagoas
basin (Fig. 4) from the Carboniferous to the Qua­
ternary, recording the five phases of structural

L ITHOSTRATIGRAPHY
PERIOD SEQUENCE

QUATERNARY

TERTIARY

OCEAN
(PASSIVE
MARGIN)
LATE
CRETACEOUS

GULF
(TRANSITIONAL)

APTIAN

R IFT
BERREMIAN

HAUTERIVIAN

VALANGNIAN

BERRIASIAN CONT.
(PRE-RIFf)
JUR A S S IC

Fig. 4. Stratigraphical column of PALEOZ. INTRAC.

Sergipe-Alagoas Basin. Serraria
Formation (arrow) is part of the PRE-CAMB. BASEMENT
continental pre-rift sequence.
1 12 A.J V Garcia et al.

Fig. 5. Lithofacies intervals of the Serraria Formation, showing the respective depositional palaeoenvironments and
palaeoclimatic aspects. Modified from Garcia (1992).

evolution of the Brazilian marginal basins: intracra­
tonic, pre-rift, rift, transitional and drift (Ponte &
Asmus, 1 976; Ojeda, 1 982). A Palaeozoic sequence
was deposited under intracratonic conditions. Dur­
ing the pre-rift phase (Late Jurassic to Early Creta­
ceous) the Afro-Brazilian depression formed as a
product of crustal warping to the southwest and
northeast of the studied area, and fluvial-lacustrine
sequences, including the Serraria Formation, were
deposited. In the rift phase (Early Cretaceous),
when Gondwana ruptured, lacustrine and deltaic
sediments accumulated in deep asymmetric rift
basins. The transitional phase (Aptian) was charac­
terized by the formation of a narrow proto-oceanic
gulf in which transitional clastics and evaporites
were deposited. During the drift phase (Albian to
Holocene), shallow-marine carbonates were ini­
tially deposited, followed by a deep-water clastic
wedge of shales and turbidites.
The Serraria Formation consists of six lithofacies
intervals (Garcia, 1 992). The sedimentary struc­
tures, facies interpretation and thicknesses of each
interval are shown in Fig. 5 . The lower interval 1
and upper interval 5 are composed of interbedded
 Lower Cretaceous Serraria sandstones
distal-fluvial fine-grained sandstones and lacustrine
mudstones. Two of the middle intervals (2 and 4)
are composed of medium-grained to conglomeratic
sandstones and conglomerates deposited by high­
energy braided systems. Interval 3 is composed of
aeolian sandstones formed by reworking of the
fluvial deposits (Garcia, 1 99 1 a, 1 992). The basal
interval 1 transitionally overlies mudstones of the
Bananeiras Formation. In some areas interval 5 is
covered by the uppermost interval 6, with coarse­
grained sandstones which pass transitionally into
the overlying rift phase mudstones of the Barra de
Itiuba Formation (Fig. 5).
Coarse-grained proximal deposits predominate
in the Furado and Japoatii-Penedo areas (Fig. 2),
whereas fine-grained sandstones and mudstones
predominate in the Caioba area. This distribution
pattern represents a portion of the original deposi­
tional association which was preserved within the
late-developed rift basin. The connotation of 'prox­
imal', 'middle' and 'distal' as used in this paper thus
applies to the relative location within the central
part of the fluvial system. The contemporaneous,
most proximal alluvial fan and most distal lacus­
trine deposits are not preserved within the present
basin configuration.
S A M P L E S A N D M ETHODS
Thin sections prepared from 2 5 5 core samples from
40 wells and 40 fresh outcrop samples, blue resin­
impregnated, were examined with a standard petro­
graphic microscope. Modal quantification of the
detrital and authigenic constituents was performed
on 1 89 selected thin sections by counting 300 points
in each. Staining with alizarin-K and ferricyanide
solution was used to characterize the different
carbonate cements (Dickson, 1 965). Twenty-one
representative thin sections were polished, carbon
coated and examined with a Cameca Camebax
SX50 electron microprobe (EMP) equipped with
three crystal spectrometers and a back-scattered
electron (BSE) detector. The operating conditions
during analysis were an acceleration voltage of
1 5 kV, a measured beam current of 8 nA for the
analyses of the carbonates and clays and 1 2 nA for
the feldspars, and an electron beam diameter that
varied between 1 and 1 0 11m. Standards and count­
ing times were: wollastonite (Ca, Si, 1 0 s), MgO
(Mg, 1 0 s), MnTi03 (Mn, Ti, 1 0 s), hematite (Fe,
I 0 s), orthoclase (K, 5 s), albite (Na, 5 s), barite (S,
1 13
Ba I 0 s), strontianite (Sr, 5 s) and corundum (AI,
20 s). Precision during analyses was better than
0. 1 mol%. Small chips of six samples were gold
coated and examined with a JEOL JSM-T330
scanning electron microscope (SEM) with an accel­
eration voltage of 1 0 kV.
Carbon and oxygen isotope analyses of carbonate
cements were performed on 20 samples. Each sam­
ple was reacted with I 00% phosphoric acid at 25 · c
for calcite and at 50·c for dolomite. The evolved
gas for each carbonate fraction was analysed using a
SIRA- 1 2 mass spectrometer. The phosphoric acid
fractionation factors used were 1 .0 1 025 for calcite
(Friedman & O'Neil, 1 97 7), and 1 .0 1 060 for dolo­
mite (Rosenbaum & Sheppard, 1 986). Oxygen and
carbon isotope data are presented in the normal o
notation relative to PDB (Craig, 1 95 7). Precision
( I cr) was monitored through daily analysis of the
NBS-20 calcite standard and was better than
3
± 0.05o/oo for both o 1 C and o 1 8 0. The isotopic data
obtained are for bulk dolomite and calcite sepa­
rately, and it is uncertain whether or not crystal­
chemical zonations are accompanied by strong
shifts in isotopic values.
X-ray diffraction analysis of the < 2 11m size
fraction of 1 2 samples of diagenetic clays from the
sandstones was performed using a RIGAKU RU-
2.00Z X-ray diffractometer equipped with Cu(Ku)
radiation and a graphite monochromator. The
samples were air-dried, later glycolated and heated
to 49o · c for 2 h.
Vitrinite reflectance data provided by PETRO­
BRAS for representative wells were used to analyse
the burial histories of the different domains within
the basin.
D E TRITAL C O M P O S I TION
AND PROVE N A N C E
The framework compositional plots o f the Serraria
sandstones reveal considerable variations between
different areas in the basin (Fig. 6; Table I ). Sub­
arkoses and arkoses dominate in the proximal
domain (Furado area; av. Q 8 0F20L0), whereas
quartzarenites and quartzose subarkoses dominate
in the distal domain (Caioba and Robalo areas; av.
Q9 8 F2L0 and Q 00F0L0, respectively) and in the
1
Japoatii-Penedo area of the middle domain
(Q97F4L0). The overall average framework compo­
sition of the Serraria Formation is subarkosic
(Q9I F9Lo ).
1 14
Caioba Area � current analyses indicate that the source rocks for
f •·-- _f DISTAL DOMAIN
Q
Robalo Area
Quartzarenite f MIDDLE DOMAIN
Q (subsurface + outcrop)
' of the Afro-Brazilian depression (Garcia, 1 992)
Q
3:1 1:1 1:3
Fig. 6. QFL detrital composition of the sandstones of
the diagenetic domains of the Serraria Formation plotted
on a Folk (1968) classification diagram. The proximal
domain is closer to the original detrital composition.
The framework grains are composed predomi­
nantly of monocrystalline quartz (av. Qzmono 49
bulk volo/o, Qzpoty 1 4). Polycrystalline quartz is more
abundant in the proximal domain (av. 20 volo/o)
and in coarse-grained sandstones in general (av.
1 1 - 1 5%). K-feldspars, which include microcline,
orthoclase and perthite (7%), predominate over
plagioclase (av. 4%). Plutonic rock fragments (gran­
ites and gneisses) are common in the coarse proxi­
mal deposits but absent or rare in the distal fine­
grained sandstones (av. 2% and 0- 1 %, respectively).
Biotite-muscovite schist fragments occur in trace
amounts. In some sandstones, particularly in inter­
vals 4 and 6, mud intraclasts are abundant (av. 5%;
up to 48 volo/o). Palaeosol fragments (calcrete, dolo­
crete and silcrete) are scarce (� 1 o/o). Silcrete intra­
clasts occur in distal sandstones of the Caioba area,
some with irregularly alternating bands of chalce­
dony and iron oxides. Other sedimentary fragments
(e.g. siltstone) are more common in the upper
interval 5 (av. 0. 8%). Detrital accessory minerals
include mica, tourmaline, zircon, amphibole, rutile,
magnetite, ilmenite, epidote, sphene, garnet and
chlorite. Silicified and coalified wood remnants are
common in the coarse conglomeratic sandstones of
intervals 2 and particularly 4.
A.J. V Garcia et a!.
The framework grain composition and palaeo­
the Serraria Formation were mainly granitic­
gneissic rocks, schists and quartzites of the Pre­
Cambrian Complex bordering the northern sector
(Fig. 3). Other less important source rocks include
a volcanic suite.
DIA G E N ETIC M I N E RALS
Calcite
Calcite is the main carbonate cement in the proxi­
mal and middle domains along the northern part of
the basin. Its average content is 3 . 8 volo/o, which
L represents �22 volo/o of all the diagenetic constitu­
ents in sandstones of these domains. Calcite is more
abundant in the lower fine-grained sandstones of
interval 1 (av. 4. 7%) than in the coarser-grained
sandstones of intervals 2, 3, 4 and 6 (av. 0.6- 1 .4%).
In the coarse to medium-grained sandstones calcite
occurs mainly as coarse mosaic to poikilotopic
replacive cement. In fine-grained sandstones, calcite
typically occurs as microcrystalline aggregates re­
placing clay intraclasts and palaeosol fragments.
Poikilotopic calcite cement replaces detrital quartz
and feldspars, as well as infiltrated clays and mud
intraclasts/pseudomatrix. Compositional zoning is
indistinct in BSE images.
Packing and paragenetic relationships indicate
that there are two generations of coarsely crystalline
calcite cement. Calcite C 1 occurs as poikilotopic
patches in sandstones with loose packing and dom­
inantly tangential intergranular contacts, indicating
a precompactional precipitation (Fig. 7A). Iron
content varies from 0 to 0.3 mol% (av. 0. 1 %) and
Mn from 0.5 to 2.2 mol% (av. 0.9 %). The o13CPDB
and o180p08 values of this calcite are -6. 7o/oo and
-8.2o/oo, respectively (Table 2; Figs 8, 9 and 1 0).
Calcite C2 fills intergranular pores previously re­
duced by compaction in sandstones which show
predominantly concavo/convex contacts. This cal­
cite commonly engulfs, and thus postdates, diage­
netic albite (Fig. 7B) and chlorite. Compositional
zoning is indistinct, iron content varies from 0.5 to
1 . 3 mol% (av. 0.6%) and Mn from 0.5 to 5.2 mol%
(av. 2.2%). The o13CPDB values of these calcite­
cements vary between - 1 1 .2 and -5 . 1 o/oo and the
o180p08 values range between - 1 3.6 and - 1 1 . 5o/oo
(Fig. 1 0).
Table 1. Average, maxima, minima (189 samples) and representative modal and petrophysical data of Serraria sandstones

Diagenetic domain Distal Distal Distal Distal Distal Distal Distal Distal Distal Distal
Caioba Caioba Caioba Caioba Caioba Caioba Caioba Caioba Caioba Caioba
Well; location CB-3 CB-3 CB-3 CB-3 CB-6 CB-6 CB-6 NAB-3 SM-1 VV-1
Depth (m) 2044.3 205 3 . 5 208 ! .9 209 ! .0 1986.6 1994.9 2009.5 2460.4 2063.7 257 5 .1

Detrital quartz 63 68 50 47 61 50 51 62 56 61
Quartz monocrystalline 60 65 44 43 55 47 43 53 55 57
Quartz polycrystalline 3 3 6 4 6 3 8 9 I 4
Detrital feldspar - - - - 6
K-feldspar - - - - - - - 3
Plagioclase - - - - - - - 3
Plutonic rock fragments - - - - - - - - - -
- - - - - - - -
t--<
Sedimentary rock fragments 0
Metamorphic rock fragments - - - - - - - - - - �
""
- - - - - - - - - - ....
Mica
Mud intraclasts and matrix 2 I 2 2 10 2 - I
2
Kaolin 3 10 7 !3 4 20 16 - 3 2 !::;
,..,
1/S clays - - - - - - - - ""
0
Chlorite - - - - - - - - - - !::
"'
Quartz overgrowths 17 7 3 14 5 13 8 4 16
Feldspar overgrowths - - - - - - - - - �
Calcite - - - - - - - - �
....
Dolomite - - 31 23 4 - 2 13 35 - E;·
Pyrite 2 I 2 II - 9 I - I - "'
$::>
Iron oxides - - - - - 3 - I :=

Titanium oxides - - - - - - - - - �
Total diagenetic phases 22 18 43 47 22 34 32 24 43 19 0
:=

Intragranular porosity 3 I I - I - 5 - - 2 �
Intergranular porosity 10 12 4 13 16 - 12 5 I 17
Shrinkage porosity - - - - - - - - - -

Total macroporosity 13 13 5 !3 17 - 17 5 I 19
Microporosity 9 5 - 8 3 18
Petrophysical porosity (%) 22.4 13.4 - - 20.4
Horizontal permeability (mD) 1123.4 46.4 - - 636.4
Grain size 0 . 37 0.2 0.29 0.12 0.7 3 0.05 0.17 0.03 0.24 0.22
Packing 14 20 10 26 25 27 13 34 20 32
Quartz 100.0 100.0 ! 00.0 100.0 100.0 100.0 100.0 9 ! .2 100.0 100.0
Feldspar 0.0 0.0 0.0 0.0 0.0 0.0 0.0 8.8 0.0 0.0
Lithic 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

Continued
�
Vl
 0\
Table I. (Continued)

Diagenetic domain Distal Robalo Distal Robalo Middle subs. Middle subs. Middle subs. Middle subs. Middle subs. Middle outer.
Well; location RB- 1 8 RB- 1 8 FD-1 FD-2 SN-1 PN- 1 JP- 1 GSTS-6
Depth (m) 42 1 6. 1 4227.4 2548.6 2276. 1 1 6 1 8.6 489 .5 43.5 Outcrop

Detrital quartz 74 68 58 53 66 68 70 65
Quartz monocrystalline 66 62 6 45 52 33 65 45
Quartz polycrystalline 8 6 52 8 14 35 5 20
Detrital feldspar - - 17 3 2 8
K-feldspar - - I I 2 I
Plagioclase - - 10 2 - 7
Plutonic rock fragments - - 6
Sedimentary rock fragments - - - - - - II
Metamorphic rock fragments
Mica I
Mud intraclasts and matrix 7 12 - 16 I 3 6
Kaolin - I - - - 4 2 4 �
liS clays - - - I 3 - �
Chlorite 3 - - II - - - :-.:::
Quartz overgrowths 2 16 7 2 - 6 2 4 �
>::>
Feldspar overgrowths - - - - - ....
("")
- - - - - -
Calcite 13 I jS•
- - - -
Dolomite 2 I 16 2 �
- - - - -
Pyrite 2 >::>
- - - - - :---
Iron oxides 5
Titanium oxides
Total diagenetic phases 7 18 23 28 4 10 8 13
Intragranular porosity - - - - 3 I
Intergranular porosity II 2 2 - 7 8 21 II
Shrinkage porosity
Tota l macroporosity II 2 2 7 II 22 II
Microporosity
Petrophysical porosity (o/o) 14.8 5.6
Horizontal permeability (mD) 9.5 0.6
Grain size 0.32 0.48 2 0.08 0.35 0.5 0.3 I
Packing 36 52 22 44 31 35 27 36
Quartz 1 00.0 1 00.0 77.3 94.6 97. 1 89. 5 1 00.0 85.5
Feldspar 0.0 0.0 22.7 5.4 2.9 10.5 0.0 0.0
Lithic 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1 4.5

Cominued
Table l. (Con/inued)

Diagenetic domain Middle Middle Prox. to Prox. to Prox. Prox. Prox. Prox. Prox.
outer. outer. middle middle Furado Furado Furado Furado Furado
Well; location GAF- 1 GAF-6 CF D-1 CFD- 1 FU-25 FU-25 FU-25 FU-25 PG-1
Depth (m) Outcrop Outcrop 10 12.4 101 3 . 7 1801.9 1803.6 1824.3 1868.9 703.1

Detrital quartz 70 65 79 80 45 52 50 41 57
Quartz monocrystalline 56 52 43 76 41 42 34 37 26
Quartz polycrystalline 14 13 36 4 4 10 16 4 31
Detrital feldspar - - - 15 27 27 3 14
K-feldspar - - - - 12 22 21 2 9
Plagioclase - - - - 3 5 6 1 5
Plutonic rock fragments - - - - - -
- - - - - - -
1:"-<
Sedimentary rock fragments 5 0
- - - - - - - - �
Metamorphic rock fragments 2 "'
....
Mica - - - - - - 1 - -

Mud intraclasts and matrix 3 19 14 - 4 9 22 6 Q

Kaolin - - - - - - - - �
,.,
12 - - - - 'll
liS clays 3 11 0
Chlorite - - - - - - - - - ::::
"'
Quartz overgrowths 1 13 1 1 - 8 2 - 15
Feldspar overgrowths - - - - l - - �
....
....
Calcite - - - - 33 - - 34 - ;:::,
....
Dolomite - - - - - - - - - E;·
- - - - -
Pyrite 1 1 1 lS
Iron oxides 22 22 - - - - - ;:,:

T itanium oxides - - - - - - - - - 1:;-

Total diagenetic phases 26 35 12 14 34 9 2 34 16 8
;:,:

Intragranular porosity 1 - - - - 1 1 - �
Intergranular porosity 6 - 1 4 - 6 20 - 7
Shrinkage porosity
Total macroporosity 7 - 1 4 - 7 21 - 7
Microporos ity
Petrophysical poros ity (%)
Horizontal permeability (mD)
Grain size 0. 37 0.37 0.48 0.08 0.19 0.13 0.37 0.09 1.43
Packing 27 28 36 38 48 60 40 20 51
Quartz 100.0 100.0 100.0 100.0 69.2 64.2 64.9 93.2 80.3
Feldspar 0.0 0.0 0.0 0.0 2 3 .1 33.3 35.1 6.8 19.7
Lithic 0.0 0.0 0.0 0.0 7.7 2.5 0.0 0.0 0.0

Continued
-
-..1
 00
Table 1. (Continued)

Diagenetic domain Total maxima Total minima Total average Av. doma in Av. domain Av. domain Av. domain Av. domain
Well; location Distal Caioba Distal Robalo Middle outer. Middle subs. Prox. Furado
Depth (m) 4227.4 - 1 975.0 1988.4 4222.1 - 1 1 61 . 3 1985.3

Detrital quartz 92 35 63.4 63.7 77.4 64.6 63.8 60. 5

Quartz monocrystalline 88 6 49.5 56.3 59.5 51.3 50.9 39.9
Quartz polycrystalline 52 0 14.0 7.5 1 7.9 1 3. 3 12.9 20.6
Detrital feldspar 37 - 6.5 1 .0 - 0.6 2.5 14.9
K-feldspar 24 0 7.1 2.7 - 0.2 0.8 8.6
Plagioclase 20 0 3.9 1.5 - 0.4 1 .6 4.3
Plutonic rock fragments 16 0 4.0 1 .0 - - - 2.0
Sedimentary rock fragments 11 0 4.5 1 .0 - 1 .2 - 0.1
Metamorphic rock fragments 2 0 1 .2 1.0 - - - 0. 1
Mica 2 0 1.3 1.3 0.3 - 0.1 0.2
Mud intraclasts and matrix 48 0 8.6 8.1 7.3 5.4 7.1 6.7
Kaolin 24 0 5.6 6.0 0.1 2.4 2.0 - �
liS clays 25 0 7.5 1.0 2.6 2.6 3.3 2.3 �
- - - - �
Chlorite 23 0.4 1.7 0.8
Quartz overgrowths 22 0 5.6 6.8 4.3
-
3.8
-
4.4
-
3.9 �
Feldspar overgrowths 9 0 2.0 1 .0 0.6 �
Calcite 34 0 8.6 29.0 - - 4.0 3.6 iS'
Dolomite 38 0 8.7 10.6 1.3 - 0.2 0.2 �
Pyrite 11 0 2.3 2.6 0.1 - 0.1 0.3 $::,
:--
Iron oxides 22 0 4.5 1 .4 0.3 5.6 - -

Titanium oxides 15 0 2.3 8.0 - - - 0.2

Total diagenetic phases 48 0 51.4 68.3 1 0.6 1 4.4 14. 1 13.7
Intragranular porosity 10 0 1.7 1.5 0.2 2.2 1.8 0.6
Intergranular porosity 31 0 11.1 12.8 4.9 1 4. 7 10.7 5.3
Shrinkage porosity 5 0 2.7 5.0 0.2 - 0.2 0.3
Total macroporosity 35 0 1 2. 3 14.8 5.3 16.9 12.6 7.8
Microporosity 18 0 8.2 9.0 - 0.9 0.8 0.3
Petrophysical porosity (%) 23.9 5.2 15.1 18.1 13.3 - - 3.7
Horizontal permeability (mD) 1123.4 <0. 1 1 70.8 327.0 37.2 - - 7.5
Grain size 2.2' 0.0 0.5 0.3 . 0.4 0.6 0.4 0.6
Packing 71.0 10.0 34.2 27.6 4 1 .0 31.2 34.6 40.7
Quartz 1 00.0 56.8 9 1 .3 98.4 1 00.0 97.6 96.4 80.3
Feldspar 43.2 0.0 8.5 1.5 0.0 0.8 3.6 19.4
Lithic 14.5 0.0 0.2 0. 1 0.0 1 .6 0.0 0.2

Av, average; outer., outcrop; prox., proximal; subs., subsurface.

 Lower Cretaceous Serraria sandstones 119

Fig. 7. Back-scattered electron (BSE) images of: (A) sandstone cemented by eogenetic calcite CI, showing loose
packing due to early cementation and marginal grain replacement; (B) sandstone cemented by postcompactional
mesogenetic calcite C2 after albite replacement and overgrowths on plagioclase grain (pi); (C) rhombs of
dolomite/ankerite D I with decreasing Fe zonation towards the edges of the crystals and incipient dissolution;
vermicular kaolinite partially fills the pores; (D) precompactional D l dolomite/ankerite with complex zonation and
displacive texture in relation to the grains; the white spots are framboidal pyrite; (E) oversized pore rimmed by finely
crystalline dolomite, followed by coarse, blocky, thinly zoned dolomite/ankerite DI, and then filled by vermicular
kaolinite; (F) ferroan zoned rhombs of dolomite/ankerite D l partially replaced by low-Fe, poorly zoned pore-filling
dolomite 02.
1 20 A.J V Garcia et a!.

Table 2. Chemical composition from microprobe analyses and isotopic ratios of representative carbonate cements in
Serraria sandstones
3
o' c 8'80
Well (depth, m) MgC03 SrC03 CaC03 MnC03 FeC03 (o/oo PDB) (o/oo PDB) Carbonate phase (%)

Distal domain (Caioba area)

CB"3 (208 1 . 3 m av.) 35.7 0.0 55.6 1.7 7.0 -2.9 -6.9 3 1 % rhombs and replacive
Minimum 30.4 0.0 53.3 0.9 0.0 poikilotopic dolomite 02
Maximum 43.0 0.1 59. 1 2.4 1 2. 8
CB-3 (208 1 .9 m ) -2.6 -6.3 3 1 % dolomite (02)
CB-3 (2 1 02 m av.) 32.5 0.0 56.3 1.7 9.6 I 7% unzoned dolomite (02)
Minimum 27.3 0.0 5 1 .9 1 .2 0.0 and ankerite rims
Maximum 37.6 0. 1 6 1 .3 2. 1 19.6
CB3 (2 1 03 . 1 5 m av.) 32.2 0.0 59.0 1.5 7.4 -7.7 -4. 3 29% rhombs and poikilotopic
Minimum 22.4 0.0 52.5 0.7 0.0 - 1 0. 5 -5.6 dolomite (02), replacive
Maximum 40. 3 0.2 63. 1 2.3 23.4 -8.3 -5.0 ankerite
CB-6 ( 1 956.9 m av.) 29.9 0.0 56.7 2.5 1 1 .0 Zoned rhombs of 0 I and 02
Minimum 2 1 .9 0.0 5 1 .8 1 .7 0. 1
Maximum 36.8 0.0 6 1 .4 3.3 23.0
CB-6 ( 1 988. 7 m) -6.7 -6.8 3% dolomite
CB-60 ( 1 996.4 m av.) 35.3 0.0 55.6 2.0 7.1 -3. 1 -8.7 II% 0 I zoned rhombs
Minimum 29.5 0.0 53.8 1.5 0.1 -8.5 -6.7
Maximum 4 1 .0 0.0 57.4 2.5 14. 1
CB-60 ( 1 988. 7 m av.) 34.3 0.0 54.5 2.6 8.6 3% zoned dolomite rhombs
Minimum 26.4 0.0 52.2 0.4 0.1 and poikilotopic dolomite
Maximum 42.8 0.0 56.8 3.8 16.7 02 (<0 1 )
CB-60 (200 1 . 7 m av.) 3 1 .0 0.0 54.4 2.5 12.1 I % zoned dolomite rhombs of
Minimum 1 0.9 0.0 49.8 0.6 0. 1 02 and 0 I dolomites
Maximum 42.6 0.0 56.9 5.9 33.5
CB-60 (2005.6 m) - 1 1 .8 -4.1 1 2% zoned dolomite 02
CB- 1 1 0 (2 1 75 . 3 m av.) 34.6 0.0 55.0 3.5 6.9 -5.1 -6.4 I 0% (0I) dolomite rhombs
Minimum 1 6. 2 0.0 50.2 0.9 0.0 and poikilotopic dolomite
Maximum 42.4 0.1 57.3 7.3 28.8 (02)
CB- 1 1 0 (2 1 78 . 5 5 m av.) 34. 1 0.0 54.9 3.3 7.7 -6.5 -6.9 1 4% fine zoned dolomite (0 1 )
Minimum 23.5 0.0 52.2 0.8 0.0 and coarse zoned (03)
Maximum 42.0 0.2 58.6 9.4 2 1 .3
SES-62 (2477.9 m av.) 41.1 0.0 55.6 2.7 0.6 - 1 2.4 -6.9 22% zoned poikilotopic
Minimum 37.3 0.0 54.8 1 .4 0.0 dolomite (03) engulfed by
Maximum 43.6 0. 1 56.9 5.4 3.2 late quartz overgrowth
SES-62 (24 78. 1 m) -1 2.3 -7.0 Zoned dolomite (03)

Distal domain (Robalo area)

RB- 1 8 (422 1 . 35 m) -12.2 -8.2 I% zoned dolomite (03)
RB- 1 8 (4222. 1 m av.) 28.0 0.0 53.2 2. 1 1 6.7 I% zoned rhombs partially
Minimum 1 5. 2 0.0 52.0 0.8 1 0.2 dissolved dolomite 02,
Maximum 35.3 0.1 54.2 4.9 26. 1 ankerite rim
RB- 1 8 (4222.45 m) -2.7 -9.7 I% dolomite 02/03
RB- 1 8 (4226.65 m) -2.4 - 1 0.6 I% dolomite 02/03

Middle domain
JP- 1 (95.6 m av.) 34.9 0.0 62.2 2.6 0.3 Zoned dolomite filling
Minimum 39.9 0.2 66.5 3.6 0.7 shrinkage in a mudstone
Maximum 29.9 0.0 59.0 1.4 0.0 (paleosol)
F0- 1 (2548.6 m av.) 27.8 0.0 53.1 3.8 1 5. 3 1 6% zoned poikilotopic
Minimum 1 2.4 0.0 50.8 0.9 6.1 dolomite ( 0 3 ) with
Maximum 38.2 0. 1 54.7 9.8 26.1 ankerite rims

Continued
 Lower Cretaceous Serraria sandstones 121

Table 2. (Continued)
3
ol c 0180
Well (depth, m) MgC03 SrC03 CaC03 MnC03 FeC03 (o/oo PDB) (o/oo PDB) Carbonate phase (%)

Proximal domain (Furado area)

FU-25 ( 1 8 1 1 . 1 0 m av.) 0.0 0.5 96.5 2.8 0.3 -5.4 -1 2.3 4 % unzoned (C2) calcite
Minimum 0.0 0.5 96.2 2.5 0.2
Maximum 0.0 0.5 96.7 3.0 0.4
FU-25 (I 8 2 1 .6 m av.) 1.3 0.0 97. 1 1 .2 0.4 Microcrystalline calcite replac-
Minimum 0.0 0.0 95.5 0.0 0.0 ing clay in palaeosol; coarse
Maximum 2.7 0. 1 98.7 3 .4 1 .0 calcite filling vugs and cracks
FU-25 (I 8 32.4 m) - 1 1 .2 - 1 1 .5 C2?
FU-25 ( 1 8 3 3 . 7 m) -5. 1 - 1 1 .7 C2?
FU-25 ( 1 862.6 m av.) 0. 1 0.0 98.9 0.9 0. 1 -6.7 -8.2 1 6% zoned calcite (CI)
Minimum 0.0 0.0 97.7 0.0 0.0
Maximum 0.2 0.0 99.9 2.2 0.3
FU-93 ( 1 7 1 8.6 m av.) 0.0 0. 1 97.7 2.4 0.6 6% poikilotopic calcite (C2)
Minimum 0.0 0.0 93.8 1 .0 0. 1
Maximum 0.0 0.3 98.8 5.1 1.1
CSMC-24 (265 1 .9 m av.) 0. 1 0.0 96.3 2.8 0.8
Minimum 0.0 0.0 93.8 1 .4 0.2
Maximum 0.4 0. 1 98.3 4.5 1 .3
CSMC-24 (2652.2 m av.) 0. 1 0.4 98.3 0.8 0.6 20% poikilotopic calcite (C2)
Minimum 0.0 0.3 97.7 0.5 0.3
Maximum 0.2 0.6 98.6 0.9 1.1

Ca /-- ......

[;]
5 /1 • .....,
o Cl
2 / ··'
cz 4 / I
/ /
•

...-
o Dl C2 /
"' 3 I
• DZ 0 • I
u • •
03 c: • I
�2
+

•'
...---
'
I
/
/
...- ' I
' I
0 ,�_..,

0 0.2 0.4 0.6 0.8 1.2 1.4

Mg Fe+Mn
Fig. 8. Composition of eodiagenetic (CI and 0 I) and
mesogenetic (C2, 02 and 0 3) calcite and
dolomite/ankerite cements.
Dolomite and ankerite
Dolomite and ankerite occur in sandstones of the
distal domain (Caioba and Robalo areas; Fig. 2; av.
5.2%; up to 38%), and are associated with calcite
cement in the middle domain (Japoatii-Penedo and
Feliz Deserto areas). As far as calcite, dolomite is
FeC03
Fig. 9. Plot of manganese and iron contents in calcite
cement (early CI and late C2) in the proximal domain.
more abundant in the lower fine-grained sandstones
of interval 1 (av. 6.9%). It is rare in the lower,
coarse-grained sandstones of interval 2, as well as in
the aeolian sandstones of interval 3 in the Caioba
area. In the fine- to coarse-grained sandstones of
intervals 4, 5 and 6, the average dolomite/ankerite
content varies between 1 . 5% and 2.3%.
Three types of dolomite/ankerite were distin­
guished in the Serraria sandstones based on textural
1 22
-2
-4 D D
D varying from � I to 3% (av. 1 . 8%) (Fig. 1 2). The
D
-6
co D D
0
(L
0
-8 D •
00
c:o
-10
• talline dissolution pores contain authigenic kaolin­
-12
-14 .__._
....._.._._....._.�
........_.�....
'-'--'-'-'--'-'-'-....
..._,
- 14 -12 -10 -8
o dolomite
• calcite
Fig.lO. Plot of the o13CPDB and 0180pos values (%o) for
dolomite/ankerite and calcite.
and compositional aspects. Although compositional
overlap occurs among these types, their paragenetic
relationships allow their positive distinction.
Dolomite/ankerite D I occurs predominantly as
coarsely crystalline, blockly to poikilotopic (40-
2000 IJ.m), thinly and sharply zoned rhombs
(Fig. 7C, D) and, subordinately, as aggregates of
small (�5-20 IJ.m) zoned rhombs surrounding over­
sized pores (Fig. 7E). Samples cemented by D 1
have loose grain packing (Fig. 7D) and in places
show displacive fabric. D 1 commonly shows thin
zoning, with decreasing Fe content toward the pore
centres (Fig. 7C). In some cases dolomite shows
irregular zones and overall low Fe content (Fig. 7D)
The iron content of D I is 6-29 mol% (nearly
one-third of analyses displayed ankerite composi­
tion; av. 20%; Fig. 8), and some zones are highly
enriched in manganese (up to 5%; av. 3.3 mol%).
The 813Cpos values range from -8. 5o/oo to -3. 1 o/oo
and the 01 80PDB ValUeS from -8. 7o/oo tO -6. 7o/oo
(Table 2; Fig. I 0).
Dolomite/ankerite D2 occurs as blocky to poiki­
Jotopic, irregularly zoned rhombs (90-200 IJ.m)
which marginally replace and cover, and thus post­
date, D I (Fig. 7F). Samples cemented mainly by D2
show moderate packing, indicating a dominantly
A.J V Garcia et al.
syncompactional precipitation (Fig. !lA). D2
shows irregular and indistinct zonation (Figs 7F &
!lA). It has an iron content ranging from �7 to
20 mol% (av. 1 2%) (Fig. 8) and manganese content
013Cpos values range from - 1 1 . 8o/oo to -2.6o/oo. The
D D
D 0180p08 values range from -7 .Oo/oo to -4. 1 o/oo (Table
2; Fig. 1 0) in Caioba area and from - 1 0.6o/oo to
-9. 7o/oo in Robalo sandstones.
D
Both D I and D2 dolomite/ankerite types show
D evidence of partial dissolution, which preferentially
D affects the Fe-rich zones (Fig. 1 1 B). Some intracrys­
•
• ite booklets. In some samples of the middle domain
partially dissolved poikilotopic crystals of D 1 and
•
...._._._
....._, D2 are covered, and thus postdated, by thin chlorite
-6 -4 -2 rims (Fig. I I C).
Dolomite/ankerite D3 occurs as discrete rhombs
ol3C PDB (Fig. I I C), (� 1 0-200 IJ.m) and as overgrowths
(Fig. I I D) (up to 300 IJ.m thick) which cover and
engulf dissolution remnants of, and thus postdate,
both D 1 and D2 (Fig. l i E). 03 dolomite/ankerite
also engulfs kaolin (Fig. I I E), postdates chlorite
rims (Fig. 1 1 F) and shows no sign of dissolution. In
general 03 is iron rich (�4-26%; av. 1 7%) (Fig. 8)
and is relatively enriched in. Mn (�0.7- 1 0%; av.
3. 5%) in relation to 02. The 013Cpos values range
from - 1 2.4o/oo tO - 1 2.2o/oo and the 0180PDB Values
from -8 .2o/oo to 7 .0o/oo (Table 2; Fig. 1 0). 02 and 03
replace detrital silicates and early quartz over­
growths. However, in some cases thick quartz over­
growths engulf partially dissolved dolomite rhombs
(Fig. 1 3A).
Infiltrated clays
Anisopachous clay coatings occur in the coarse­
grained sandstones (av. 1 .4%), forming up to 25% in
some from middle and proximal domains, and are
less common in the distal domain ( < I %). These
coatings exhibit features typical of mechanically
infiltrated clays (Moraes & De Ros, 1 990), which
are introduced by episodic floods into coarse allu­
vial sediments (Walker et a!., 1 978). Some coatings
are composed of multiple clay layers, suggesting
multiple episodes of flooding and infiltration. Infil­
trated clays were kaolinized and illitized in the
Caioba area of the distal domain, and were mainly
illitized in the Robalo area, and chloritized in the
proximal (Furado area) and middle domains.
 Lower Cretaceous Serraria sandstones 123

Fig. 1 1. (A) Sandstone cemented by syncompactional, low-Fe poorly zoned dolomite 02 covering quartz overgrowths;
(B) extensive dissolution of Fe-zoned dolomite/ankerite D l leaving crystallographically controlled intracrystalline pores
(arrows); quartz grains with overgrowths; (C) partially dissolved dolomite D l covered by thin chlorite rims (arrow) and
by overgrowths and rhombs of ferroan, brighter ankerite 0 3; (D) bright, ferroan overgrowths of dolomite/ankerite 0 3
o n darker, blocky dolomite D l ; (E) dissolved remnants of dolomite 02 engulfed and overgrown b y bright ankerite 0 3,
which also engulfs vermicular kaolinite; (F) blocky ankerite 0 3 grown on top of chlorite rims. (A,C,D,E) BSE images;
(B) optical photomicrograph, half-crossed polarizers; (F) scanning electron microphotograph.
1 24 A.J. V Garcia et a!.
10
8
6 (Fig. 1 3B) replaces vermicular kaolinite aggregates
0
"'
u
c:
::E 4
2 Kaolinite is illitized in Robalo area sandstones
0
0 5 10 15 20 25 30 35
F e C0 3
Fig. 12. Plot of manganese and iron contents in DI, D2
and D3 dolomite/ankerite cement in the distal and
middle domains.
Kaolin
Authigenic kaolin occurs in Serraria sandstones
both in the kaolinite and in the dickite polymorphs,
as indicated by XRD and SEM analyses (Ehrenberg
et a!., 1 993; Morad et a!., 1 994). Kaolin occurs as
booklet and vermicular aggregates of stacked plate­
lets (20-35 �m across) (Fig. 1 3B) filling primary in­
tergranular pores and secondary pores after dis­
solved carbonate cements and feldspar grains
(Fig. 1 3C) and replacing detrital clays and micas.
Kaolinite platelets, which replace infiltrated clays
and pseudomatrix, are commonly subhedral. Kaolin
is more common in fluvial sandstone of the distal
Caioba area (av. 6.0%). It occurs in smaller amounts
in sandstone of the JapoaHi-Penedo area of the
middle domain (�av. 2.4%) and in deeply buried
(�4.2 km) sandstones of the distal Robalo area (av.
0. 1 %) and proximal Furado area (av. < 0. 1 %). Over­
all, kaolin is most abundant in the upper coarse­
grained sandstones ofinterval 4 (av. 4.9 %) and in the
basal fine-grained sandstones of interval 1 (av.
4.6%). In fluvial sandstones (fine to coarse grained)
of the other intervals the average kaolinite content
is � 1 %. Kaolin represents on average < 1 % in the
aeolian sandstones of interval 3.
Kaolinization and dissolution of feldspar grains
were most extensive in sandstones of the distal
domain (Fig. 1 3C). Kaolinization was intense even
in sandstones previously cemented by dolomite.
Kaolinized micas show the typical expanded tex-
ture, with replacement and precipitation along
traces of cleavage planes. Kaolinization of infil­
trated clays and pseudomatrix was extensive in the
distal Caioba and middle Japoatii-Penedo areas.
In the Caioba area blocky euhedral dickite
without disrupting the original stacking pattern (see
McAulay et a!., 1 994; Morad et a!., 1 994). Kaolinite
and dickite are engulfed by dolomite/ankerite D3
(Fig. 1 1 E) and quartz overgrowths (Fig. 1 3D).
(Fig. 1 3E) and chloritized in proximal Furado area
sandstones.
Illite and chlorite
Chlorite occurs mostly in proximal and middle
domains, as rims around grains and intergranular
secondary pores left by the dissolution of dolomite/
ankerite D 1 and D2 (Figs 1 3F and 1 1 C), and as
rosette aggregates on partially dissolved grains and
calcite cement. Chlorite also replaces pseudomatrix,
infiltrated clay and authigenic kaolinite, mostly in
coarse-grained/conglomeratic sandstones from the
Furado area. Chloritized coatings display an aver­
age composition of Fe 3 .7Mg uA13 . 3 Si4. 0 0(0H)8 .
1 1
Chlorite is closely associated with late diagenetic
minerals such as dolomite/ankerite D3 (Fig. 1 1 F),
albite and quartz (Fig. 1 4A).
Illite occurs in all areas, predominating in the deep
distal domain (Robalo area) and is abundant also in
sandstones from the proximal Furado area. Illite
occurs mostly as a transformation of smectitic infil­
trated coatings and pseudomatrix, displaying an
average composition ofK . 5Mg0.2Fe0. Al4. 7Si6.90 2 0
1 1
(OH)4. The amounts of illite interstratified in mixed­
layer liS at similar depths in the distal Caioba and
proximal Furado areas (� 1 960-2 1 1 0 m and 1 760-
2280 m, respectively) are 7 5-95% in Caioba and
70-80% in Furado. Illite also pseudomorphically re­
places authigenic kaolinite (av. 1 .6%) in samples
from the distal Robalo area (Fig. 1 3E), displaying a
composition of K � .9Mg0. 3 Fe0.4Al4. 8 Si6.5020(0H)4.
In the distal Robalo area illite and minor amounts of
chlorite occur mainly in sandstones of the oil zone.
Quartz
Quartz cement occurs as syntaxial overgrowths on
detrital quartz grains (Fig. 1 4B) heals fractures in
detrital quartz, and forms discrete prismatic crys­
tals (Fig. 1 4A). It is more common in the coarse-
 Lower Cretaceous Serraria sandstones 125

Fig. 13. (A) Optical photomicrograph of large quartz overgrowths engulfing dolomite/ankerite rhombs; crossed
polarizers; scale bar 0 . 1 mm; (B) scanning electron micrograph of euhedral dickite which replaced vermicular kaolinite;
thin remnants of kaolinite (arrows); (C) optical photomicrograph of feldspar remnants after intense dissolution (fp) and
vermicular kaolinite in secondary pores; uncrossed polarizers; scale bar 0. 1 mm; (D) scanning electron micrograph of
quartz overgrowths engulfing dickitized kaolinite booklets; (E) scanning electron micrograph of illitized vermicular
kaolinite with hairy extensions; illitized infiltrated coating in the background; (F) BSE image of bright 0 3 ankerite
rhomb covering and engulfing isopachous chlorite rims.
1 26 A.J. V Garcia et a!.
Fig. 14. (A) Scanning electron micrograph of prismatic quartz outgrowths covering and engulfing chlorite rims; (B)
optical photomicrograph of a sandstone extensively cemented by quartz overgrowths which contain bitumen
inclusions; crossed polarizers.
grained sandstones (intervals 2, 4 and 6, Fig. 5 ; av.
4.4-6.2%) and in the lower fine-grained sandstones
(interval I ; av. 4. 3%), particularly in the distal
Caioba area (av. 6. 1 %). In this area syntaxial over­
growths in some loosely packed sandstones cover or
alternate with thin infiltrated clay coatings, and
hence are of eogenetic origin. Large overgrowths en­
gulf authigenic dickite, chlorite, dolomite/ankerite
and bitumen (Figs 1 3A,D and 1 4B), and are thus of
late mesogenetic origin. Quartz cement is poorly
developed in sandstones with grains covered with
abundant authigenic chlorite rims or with thick
infiltrated clay coatings. In the proximal domain
(Furado-Siio Miguel do Campos areas) and the
middle domain (Feliz Deserto area), prismatic
quartz outgrowths cover and engulf thin authigenic
chlorite rims (Fig. 1 4A). Quartz overgrowths are
abundant below 4226 m in the distal Robalo area,
where they contain bitumen inclusions (Fig. 1 4B).
Feldspars
Authigenic albite occurs as discrete euhedral
crystals ( I 0 Jlm), overgrowths on detrital feldspars,
and most commonly as replacement of detrital
K-feldspar and plagioclase in the proximal and
middle domain. K-feldspar grains in these areas
commonly show partial albitization, whereas the
detrital plagioclase is totally replaced by albite.
Albitized feldspar grains display the typical optical,
textural and chemical characteristics outlined by
Morad ( 1 986) and Morad et a!. ( 1 990). Authigenic
K-feldspars occur as overgrowths around detrital
orthoclase and microcline in the deeper portions of
the Caioba area of the distal domain, where feldspar
kaolinization/dissolution was less intense (av.
< I %). In places overgrowths surround secondary
pores formed by the dissolution of host grains.
Sulphides and sulphates
Authigenic sulphides include pyrite and, less com­
monly, sphalerite. Pyrite is more abundant in the
distal deposits of the Caioba area (av. 1 . 6%; up to
I I %), and occurs as framboids and discrete euhedral
crystals. Framboidal pyrite is rare, being associated
with carbonaceous fragments. Euhedral pyrite oc­
curs as scattered crystals, rims, and as nodules en­
gulfing and replacing framework ,grains and cements.
Pyrite commonly replaces the iron-rich zones in do­
lomite cement (Fig. 7E,F). Euhedral pyrite also re­
places mud intraclasts, micas, infiltrated clays and
detrital Ti-Fe oxides, and engulfs kaolinite. Sphaler­
ite occurs in trace amounts in the proximal Furado
area as pore-filling aggregates of euhedral crystals
replacing detrital feldspars and quartz.
Barite occurs in trace amounts as patchy poikilo­
topic, grain-replacive cement associated with par­
tially dissolved carbonate cement. Anhydrite cement
forms < I volo/o in the fine- to medium-grained sand­
stones that contain micritic carbonate intraclasts
(calcretes/dolocretes), occurring in the upper part of
 Lower Cretaceous Serraria sandstones 1 27

upward-fining cycles in the Japoatii-Penedo area
(middle domain). Anhydrite shares pores with D 1
and D2 dolomite/ankerite and partially replaces the
detrital silicates and infiltrated clays.
Iron and titanium oxides
Iron oxides ( < 1%) occur in sandstones of the distal
Caioba area associated with dissolved. ferroan
dolomite/ankerite and oxidized, coarsely crystalline
pyrite. These sandstones display evidence of exten­
sive dissolution and kaolinization of feldspars. Iron
oxides (2-22%) also fill pores as alternating bands
with quartz/chalcedony (4- 1 3%) in some coarse­
grained sandstones and conglomerates that outcrop
in the middle-domain Japoatii-Penedo area. In
these outcropping sandstones detrital feldspars are
totally dissolved or kaolinized. Authigenic Ti-oxides
( < 1 %) occur as leucoxene, anatase and rutile crystal
aggregates after altered detrital titaniferous minerals
(e.g. ilmenite and titano-magnetite). Euhedral ana­
tase occurs in mud intraclasts, infiltrated clays,
micas, and as overgrowths on detrital rutile.
PALAE O SOLS
Palaeosols were developed in the floodplain (inter­
val 4) (Fig. 15) and marginal lacustrine deposits
(intervals 1 and 5) (Figs 3 and 5). However, only a
few palaeosol horizons have survived erosion. Intra­
clastic palaeosol fragments are common in the

� � / / // / /

�
/ // / / I
// + co2, 02

-_-....:. .._..:... ·-:._·-:_:_.:_ � -:..,. -_-_=_=_=_-_- -

'50 100 km

LEGEND

Vegetation a n d Soil � Eolian Dunes Conglomerates and

Coarse Sandstones
Silicified ® Wood � Alluvial Fan Deposits Coarse to Medium
1·. :;�: : •ISandstones
I Meteoric Influx � Braided Fluvial Deposits I ··· ··• : :.•!·
Fine to Medium Sandstones
G round Water Flow -- Shallow Lacustrine 1-.---_-_-_-_.J Mud
Red Shales with
� Environment Cracks
Water Table d s ete(Proximal) � Basement Rocks
� ���==�:� ��?o��e t� (ofs��� )
Increasing Eh

Fig. 15. Palaeogeographical and palaeoclimatic imprints on the eogenetic fluid composition and the distribution of
early carbonate cements in the Serraria deposits. Modified from Garcia ( 1 992).
128 A.J. V Garcia et a!.

fluvial sandstones of intervals 2 and 4. The poor fine-grained sandstones of intervals 1 and 5, and in
preservation of palaeosols indicates a continuous the intermediate coarse-grained fluvial sandstones
lateral migration of the braided fluvial channels. from intervals 2 and 4, where it forms up to 30%.
In thin section these palaeosols show characteris­ Pseudomatrix is partially to extensively replaced by
tic micromorphological features, such as microcrys­ calcite and dolomite.
talline calcite rims and cement with grains 'floating' Sandstones with extensive or partial but evenly
in a displacive texture, clay cutans, glaebular tex­ distributed cementation have undergone less com­
ture and irregular cracks in micritized mud frag­ paction than uncemented sandstones. Dolomite­
ments. In the in situ calcretized palaeosols in the cemented sandstones of the distal domain (Caioba
proximal domain microcrystalline calcite replaces area) have lower packing indices than calcite­
mud matrix, and coarsely crystalline Fe-zoned cal­ cemented sandstones · of the proximal domain with
cite fills vugs. In the calcrete intraclasts calcite similar ranges of cement contents (Fig. 16). This
occurs mainly as microcrystalline aggregates which indicates a more extensive early dolomite/ankerite
replace mud. In in situ palaeosols of the distal cementation in the distal domain. Eogenetic quartz
domain dolomite occurs as scattered rhombs re­ overgrowths, which are more common in the distal
placing mud, and as zoned shrinkage-crack filling. than in the proximal domain, probably played a
Microcrystalline calcite replacing mud in palaeosols similar role in limiting compaction. Packing index
has low Fe and Mn contents (�0.1-0.3 mol% and varies with lithofacies. The fine-grained fluvial de­
�0.1-0. 9 mol%, respectively), whereas coarse vug­ posits, dominant in the distal domain, and aeolian
filling calcite is relatively enriched in Mn (�1.2- sandstones have lower packing indices (av. 15-27)
3.4 mol%). Zoned dolomites in the palaeosols have than the medium- to coarse-grained/conglomeratic
low Fe and moderate Mn contents (up to 0.7 mol% sandstone (av. 31-34 and av. 26 in the proximal
and �1.4-3.6 mol%, respectively). and distal domains, respectively), apparently due to
more abundant early cementation in the finer­
grained sandstones.
C O MPACTI O N Chemical compaction is evident from intergran­
ular pressure dissolution, which resulted in the
Mechanical compaction resulted i n the re­ development of sutured and concave-convex con­
arrangement and fracturing of brittle grains, and tacts between quartz grains. The intensity of pres­
deformation of ductile grains such as mica and mud sure dissolution is related to the amount and timing
intraclasts to form pseudomatrix. Intraclastic of cementation and/or to the maximum burial
pseudomatrix is common in some medium- to attained by the sediments. Pressure dissolution is

80

70 D

•
60 D D
D Do
D o Proximal Domain
� so D D
o Do (Furado area)
o B
0

Cl D
1:
40
r CD
:;;:
(calcite-cem ented)
D e
" D
.. � EP .
0.. 30

20 .
� . .') � ..
•
• • Distal Domain
. � (Caioba area)
.. ••
•
( d o l o m ite-ce m e n t e d )
10 •

0 Fig. 16. Plot of packing proximity

0 5 10 15 20 25 30 35 40 index (Kahn, 1 956) versus amount
of carbonate cement (vol%) in the
Carbonate cement %
proximal (calcite) and distal
(dolomite) domains.
 Lower Cretaceous Serraria sandstones
more intense in the weakly cemented sandstones of
the proximal Furado area and in the deeply buried
sandstones of the distal Robalo area.
P O RO S ITY
Serraria sandstones contain both primary and
secondary porosity. The average intergranular
petrographic porosity ranges from 4. 9 to 1 4. 7%.
Intragranular, mouldic and oversized pores derived
from partial to complete dissolution of detrital
feldspar are common (Fig. 1 3C). Generally, the
average intragranular porosity ranges from 0.2 to
2.2%. The extent of feldspar dissolution has vari­
ably but profoundly modified the framework com­
position and macroporosity of the sandstones. The
extent of dissolution was greater in the distal sand­
stones of the Caioba area (av. remaining feldspar
content < I %; macroporosity av. I I . 7%) than in the
proximal sandstones of the Furado-Siio Miguel dos
Campos area (feldspars � 1 3%; macroporosity av.
7.3%) (Fig. 6; Table 1 ), assuming that original
amounts of feldspar were similar throughout this
continuous alluvial system. This is indicated by the
remaining amounts of feldspar in the deeper wells
of the Caioba area, which are much less affected by
dissolution. Total macroporosity is higher in the
coarse-grained/conglomeratic fluvial sandstones
(av. 9.4%) and in the lower fine-grained sandstones
(av. 1 0. 5%). Macroporosity values average 5% in
the aeolian interval and 1 .2% in the upper fine­
grained sandstones (Fig. 5). In addition to frame­
work grain dissolution, intergranular secondary
pores have also resulted from the partial to perva­
sive dissolution of carbonate cements. Partially
dissolved dolomite cements in the distal sandstones
display intracrystalline pores (Fig. l i B). According
to the paragenetic relationships between carbonate
and other cements in secondary pores (Fig. I I C),
some samples show evidence of more than one
dissolution phase.
In addition to dissolution porosity, shrinkage of
infiltrated clays has also resulted in the creation of
secondary pores. Shrinkage is believed to result
from the diagenetic transformation of originally
smectitic infiltrated clays into mixed-layer illite/
smectite (liS) and chlorite/smectite (CIS) (Moraes &
De Ros, 1 990). Infiltrated clay content, and hence
the amounts of shrinkage porosity, is higher in
coarse-grained/conglomeratic sandstones (intervals
2 and 4) (Fig. 5). In these two intervals, shrinkage
1 29
pores form up to 6% and 1 3%, respectively, of total
porosity.
D ISCUSSION
Diagenetic evolution and burial history
The paragenetic sequence and porosity/permeability
evolution pathways of the Serraria sandstones were
controlled by multiple factors that include variations
in the sedimentary facies, climatic conditions and
burial history.
Characteristic burial history diagrams and para­
genetic sequences of the distal and proximal do­
mains of the basin are shown in Figs I 7, 1 8 and 20.
However, different areas in the same domain were
subjected to different subsidence or uplift intensi­
ties. There is no available burial history curve for
the middle domain, but the diagenetic evolution
there is shown in Fig. 1 9.
The accumulation of the Serraria Formation
started � 1 44 Ma and lasted for � 1 0 Myr (Garcia &
Wilbert, 1 995), when the syn-rift phase of basin sub­
sidence was initiated (at � 1 40- 1 3 5 Ma). The maxi­
mum depths attained in the Caioba area of the distal
domain during this burial phase were between 7 50
and 1 500 m, with a residence time at this depth of
� 1 3 Myr, and a maximum temperature of �70 " C
(Fig. I 7). I n the distal Caioba and Robalo areas sub­
sidence was rapid during this phase. The maximum
depth attained in Robalo area was 2 5 00 m and the
residence period at this depth was 60 Myr, with a
maximum temperature of � I OO " C (Fig. 1 8). After
the syn-rift subsidence phase (� 1 1 8 Ma) the distal
Caioba area was uplifted and exposed during
� I 0 Myr. The most uplifted blocks were exposed
during 40 Myr. The erosional surface developed
during this exposure was named bY. Fugita ( 1 974) the
pre-Muribeca unconformity (Fig. 2 1 ). As will be dis­
cussed below, the distribution of outcropping areas
played an important role in the extent of meteoric­
water invasion and porosity enhancement of the
sandstones. During this exposure time, Serraria
sandstones in most of the Caioba area were between
surface and 500 m in depth (Fig. 1 7). During the
uplift phase the Serraria Formation in the distal
Robalo area was at depths of 1 200-2500 m. A post­
rift subsidence phase started at � 1 1 5 Ma, bringing
the Serraria Formation in the Caioba area to the
present maximum depths of �2500 m (� l OO o C)
(Fig. 1 7). In the Robalo area during the same time
1 30 A.J. V Garcia et a!.

E o I Meso 1 I T e l o d i a g e n e s i s ; M esod iagenesis 2

Clay i n fi l t r a t i o n -1 I
Quartz - - - 1 - -
I I
K-f e l d s p a r -1 I
!2..li D 2 1 D3 D3
-
Dolomite
Compactio n
mrh a n i c a l :
-, - I
rh e c h a n i c a l __ h e m i. c a.._
.- ...,1 _
,.;c;;;
_

-1 -
Pyrite - 1
-- I� -------------
I degradation
Hyd r o c a rb o n s I
-1
Dissolution
Kaolin
I �---.... I
ka o i n i t e
I - iiiiilll••
d i c kite
I
I I
I ro n oxides -1 1-
Ti oxides -1 I
I I
Illite I - -
I
4 01 70 1
APROX. MAX. T (oC) 1 00
0 r-�-r��-r�----------��------------------�� 0
-0.5 -0 . 5
E'
� - 1 .0 -1 .0

b. - 1 . 5 -1 .5
..<::

� -2.0 .,_
________
j -2.0
- 2 . 5 ������ - 2 . 5
1 so 1 20 90 60 30 0
Time ( M .y.)

Fig. 17. Paragenetic sequence and burial history in the Caioba area of the distal domain. Dashed lines in the burial
diagram represent variation in depth during telodiagenesis within the domain (from PETROBRA S, unpublished data).

interval the Serraria sandstones were buried to
�4200 m (� 1 40 ° C) (Fig. 1 8).
In the proximal Furado area, subsidence was rapid
and continuous from the beginning of the syn-rift
burial phase (i.e. 1 40- 1 3 5 Ma) to 1 25 Ma, bringing
the Serraria Formation to approximately the present
maximum depth of �2500-3000 m (� 1 00- 1 20 o C)
(Fig. 20) in the deeper faulted blocks.
At present the Serraria Formation is exposed in
some parts of the middle domain (Japoata-Penedo
area). In outcrops to shallow burial sectors of this
domain the present burial depth is 50-900 m (Fig.
2 1 ). Depths of up to �2000 m are estimated for the
syn-rift subsidence phase (�60-80oC) in this do­
main. After the Lower Cretaceous uplift, depths of
�3 500 m were attained in deeper blocks (� 1 00 " C).
Eodiagenesis: climatic and palaeogeographical
controls (144 to �140-135 Ma)
During eodiagenesis the overall arid to semi-arid
climatic conditions exerted the most important
control on the mineralogy and distribution patterns
of cements. In the proximal and middle domains,
C 1 calcite cement precipitated mainly as groundwa­
ter (or phreatic) calcretes (Fig. 1 5) under the influ­
ence of episodic rainfall, mainly in the bordering
mountains to the N-NW, with dry periods in
between (Fig. 3B) (Garcia, 1 99 1 b; Garcia &
Wilbert, 1 995). Phreatic cementation was accompa­
nied by grain replacement and displacement, and
hence loose grain packing. Evidence of phreatic,
rather than vadose, pedogenic origin includes the
rarity of inorganic/biogenic palaeosol features (e.g.
clay cutans, meniscus or pendant textures, glaebules
and rhyzoliths). Pedogenic vadose calcite-bearing
sediments were rarely preserved, because of erosion
by laterally migrating channels of the braided river
system. In spite of the limitations of isotopic data
from phreatic C I calcite, some interpretations can
be inferred. The o13C value (-6.7o/oo) of C 1 calcite
indicates derivation of dissolved carbon, mainly
from C3 plants (Cerling, 1 984). The o180 value
(-8.2o/oo) indicates precipitation at �30"C, assum-
 Lower Cretaceous Serraria sandstones 131

EoI Mesodiagenesis
.I
Clay i n filtration
I
Qua r t z 1--- - -
D 11 D2 D3 D3
Dolomite �- - -
mechanical chemical mechanical chemical
Co m pa c t i o n

Pyrite I -
I
H y d r o ca r b o n s I
Dissol ution
I
.,
Kao l i n i te I
I
Iron oxides -t
Ti oxides
.,I -
(+chlorite)
Ill ite -
I
Barite I
APR OX. MAX. T ("C) 4b 1 00 1 40 ± 3 0
0 0
� - 1 -1
E e
� -2 -2 �
.s::
0. - 3 -3
Ql
0 ,_
__
--J �
Ql
0
-4 -4
_____

- 5 I I I I I I I I I I I I I I I I I I I I I I I
-5
1 so 1 20 90 60 30 0
Time ( M .y.)

Fig. 18. Paragenetic sequence and burial history in the Robalo area of the distal domain (from PETROBRA S,
unpublished data).

ing that the o 180sMow of meteoric waters at the
palaeolatitude of the Serraria system (�20 " S) was
-5%o (Lloyd, 1 982).
Conversely, precipitation of D1 dolomite/
ankerite as phreatic dolocrete dominated in the drier
distal domain, and to a small extent in the middle
domain, where it is associated with calcite. Based
on palaeopluviosity and palaeotemperature global
maps during the Lower Cretaceous (Parrish & Cur­
tis, 1 982; Parrish et a/., 1 982), climatic and circula­
tion model simulations ofPangaea (Kutzbach & Gal­
limore, 1 989), as well as Recent climatic settings at
similar latitudes (Namibia and Botswana) (Huggett,
1 99 1 ), the magnitude of rainfall was probably
� 0.5 m/yr in the distal domain and � 1 .0 m/yr in the
mountainous area to the north-northwest. Precipi­
tation of carbonates under these climatic conditions
is enhanced by evaporation, evapotranspiration and
C0 2 degassing. Precipitation of eogenetic dolomite
in the distal domain is attributed to an increase of
the Mg/Ca ratio in groundwater owing to the pre­
cipitation of calcite in the proximal and middle
domains (Fig. 1 5), and evaporative concentration
of these waters (Arakel et a!., 1 990; Made et a/. ,
1 994). These favourable conditions probably ac­
count for a more extensive early precipitation, and
hence a lower packing of the dolomite-cemented
sandstones of the distal domain than the calcite­
cemented sandstones of the proximal domain
(Fig. 1 6). The sources of Ca and Mg are unclear, as
there are no detrital extrabasinal carbonate frag­
ments or carbonate rocks in the source area. How­
ever, the alteration of detrital plagioclase, biotite
and, less commonly, sphene, apatite, amphibole
and pyroxene, could contribute Ca, Mg, Fe and Mn
ions to the groundwaters. Increasing ionic concen­
tration, and hence carbonate cementation, in the
phreatic zone was probably induced by evaporative
1 32 A.J. V Garcia et al.

Eo I Meso 1 I T e l o p ost-rift l M esod iagenesis 2 ITelo Recent

Clay i n fi l t r a t i o n -I I I I +chalced ony
Qu a rt z - - I I
C 1 I C2 I I
-
Calcite -� I I
I I 03 03
Dolomite
��
m erh a n i c a l
I
I
I
Imechanical
-
chemical
Compaction
I I
Pyrite -I - I I -
I degradation I
H y d ro c a rbons _ J - -
I I I
Dissolution -1 I - ----J
I I I
Kaolin I I - I - -
I I -
I
Iron oxides -I 1- I
Ti oxides -I I I -
I I I
C h l orite/ i l l ite I I 1-
A n hy d rite ..J.. I I
I I I
Barite I I I
I -
I ,

Fig. 19. Paragenetic sequence of the Serraria sandstones in the middle domain.

concentration along the groundwater flow pathways
throughout the basin (White et al., 1 96 3).
The 813C values of eogenetic dolomite (-8. 5 %o to
-3. 1 %o) indicate variable sources of dissolved car­
bon. The low value, however, suggests carbon deri­
vation from C3 plants, which contribute dissolved
carbon with a 813C signature of �- 1 2%o (Ceding,
1 9 84). As it is generally agreed that there were no
C4 plants (grasses) prior to the Miocene which
could contribute dissolved C with a 813C signature
of -4 to +4%o (but see Wright & Vanstone, 1 9 9 1 ),
the uppermost 813C values suggest a partial input of
carbon from atmospheric C02 (dissolved C with
813C of �-2%o). The isotopic composition of
phreatic dolocretes is commonly related to substan­
tially modified groundwaters (see Wright & Tucker,
1 99 1 ; Wright & Vanstone, 1 9 9 1 ; Made et al. , 1 994).
The 8180 values ofthis dolomite (-8. 7%o to -6. 7%o)
indicate precipitation from the above-mentioned
meteoric water (av. -5%o at the palaeolatitude) at
45-5 0 " C. These temperatures are relatively high for
near-surface precipitation. A possible explanation
for the low 8180 values of this dolomite would
include the depletion in 180 of groundwaters by the
precipitation of C l calcite in the proximal areas,
partial recrystallization during burial, or undetected
mixture with mesogenetic dolomite cements.
An evaporative increase in ionic concentration of
pore waters may have promoted the precipitation of
minor amounts of eogenetic K-feldspar and quartz
overgrowths in sandstones of the distal domain
(Fig. 1 7). Indeed, White et al. ( 1 963) observed that
groundwaters in sediments of Arizona and Califor­
nia, which have a similar climatic and mineralogi­
cal composition to those of the Serraria, contain
elevated concentrations of Si and K ions and hence
are potentially capable of precipitating quartz and
K-feldspar. These ions were derived from the disso­
lution of detrital silicates (Fig. 1 5).
The semi-arid climatic conditions promoted clay
infiltration. The greater abundance of infiltrated
clays in the proximal than in the other domams
(Table 1 ) is attributed to more frequent flooding
during episodic rainfall, and infiltration of sus­
pended mud into the coarser-gr<�;ined, more perme­
able proximal sands. The infiltrated clays were
originally smectitic, derived from the chemical
weathering of source rocks under the semi-arid
climatic conditions (Keller, 1 970; Curtis, 1 990).
Syn-rift subsidence and mesodiagenesis
(�140/135 Ma- 1 1 8 Ma)
Rapid burial during the rift phase brought the
Serraria Formation down to a maximum depth of
� 1 500-2500 m. The diagenetic evolution of the
sandstones during this burial phase has varied
between the different domains, mainly due to vari-
 Lower Cretaceous Serraria sandstones 1 33

E o 1 Meso d i ag e n e s i s
.I
Clay i n filtration
I
...._ -
Qu a r t z
ct l C2 C2
Calcite �
mechanical chemical chemical
Compaction ?
pyrite-sphalerite
S u l fi d e s I -
I
Hyd r o c a r b o n s I - - - - -
I (kaolinite)
Dissolution
-I
C h l o rite I - -
I
Albite I -
I - -
Ti oxides
.,
I l l ite I
I
Barite I
APR OX. MAX. T 4 01 1 00 1 1 0
0

,..... ,.....
E - 1 -1 E
� �
� �
.... -2 - 2 ....
c. c.
Ql Ql
0 0

- 3 - 3

1 50 1 20 90 60 30 0
Time ( M .y. )

Fig. 20. Paragenetic sequence and burial history in the Furado area of the proximal domain (from PETRO BRA S,
unpublished data).

ations in the maximum burial depths reached and
the amounts and composition of eogenetic carbon­
ate cements. The Serraria Formation in the distal
domain was buried at depths of �750- 1 500 m
(T � 40-?o · q in the Caioba area and �2500 m in
the Robalo area ( T � 1 oo · q. The most important
diagenetic processes in these areas were partial
dissolution of eogenetic D 1 dolomite/ankerite ce­
ment followed by precipitation of D2 dolomite with
relatively lower Mn and Fe contents (av. 1 . 8 and
1 2%, respectively) (Fig. 1 2). Using the oxygen iso­
tope values (8 1 80 = -6 . 9 to -4. 1 %o in the Caioba
area and - I 0.6 to -9. 7o/oo in the Robalo area), and
assuming that precipitation occurred at maximum
burial temperatures, the 8 1 80 values of pore waters
from which dolomite has precipitated were �- 1
to + 2o/oo in the Caioba area and �o to + I o/oo in the
Robalo area. As there is no marine water influence
at this stage, these oxygen isotopic values suggest
the evolution of meteoric formation waters caused
by interaction with the silicate minerals during
burial diagenesis. However, the variations in oxy­
gen isotopic values may reflect variations in precip­
itation temperatures within the range of 40-70 " C
mentioned above. If w e assume that precipitation
occurred from mixed, slightly evolved meteoric and
compactional waters with 8 1 80sMoW composition
of -2%o (compared with av. of -5%o for near­
surface meteoric waters), the precipitation temper­
atures would vary between 45 • and 6o·c in Caioba,
and between so· and 90 · c in Robalo.
Similar dolomite did not form in the proximal
1 34 A.J. V Garcia et a/.

+ +
+ + + ; " '- -- ' ....+...._ -
+ _.. ...... -- -7
�

./
-
_ _

PR ES
+ .- -

EN T O U T C R O P
+/ '- / @ Muribeca P-RE. �'-
- - - - - - @
Japoata
,.,.,.. - - - --
- �
-
P e n e do
- -

High J P - � O.,'.)
(SOm) • •

TN · 1 (-i -,'.) VN-1 (4-,'.)

(SOOm) (900m)

a �-
��
M o s q u e i ro Low .. �C B- 1 1 (0.,'.)

��
,
·.·:�B-6(0•!. )

alaia Fault
Sao Francisco Low
10 20

B
A�
S e rr a r i a F o r m a t i o n o ut c ro p s
Geologic section
(at u nconformity t i m e )
8
Basement CB-3 Oilwell
0 "
e

-soo - V e rt i c a l d i s t a n c e o f t o p of S e r r a r i a
F o r m a t i o n f r o m u n co f o r m i t y

U n confo rmity l i m i t
(Q-,;,)

(SOm)
A v e r a g e f e l d sp a r v o l u m e %

P r e s e n t d e p t h in J a p o a ta - P e n e d o A r e a
0
t
5 10
oo m

km
""- Meteoric flow =
--- Major f a u l t
Town
met eo.ric infiltration in Serraria
h a n g i n g wall

Fig. 21. Palaeogeological map of the southern Sergipe-Alagoas Basin at ""74 Ma, showing the areas of exposure of the
Serraria Formation at the maximum development of the post-rift, Pre-Muribeca unconformity (Ojeda, 1 982). The
average remaining feldspar content after meteoric flushing in the studied wells increases with distance from the
unconformity, as illustrated by section A-B.

domain (maximum depth at this burial stage was the 8 1 8 0 values ofpore waters would be 0 to + 1 . 5%o,
2700 m; T ;:::;; 1 00- 1 1 0 · c from 1 25 Ma until now). which again indicates evolution due to interaction
Instead, blocky to poikilotopic C2 calcite ( 8 1 3 C = with silicates during burial diagenesis.
- l l .2%o to -3. 5%o; 8 1 8 0 = - 1 3.6%o to - 1 1 . 5%o), The origin of widely variable, but generally low
which is somewhat more enriched in Mn and Fe than 8 1 3C values of both mesogenetic calcite and dolo­
eogenetic C 1 calcite, precipitated. The relatively mite is poorly constrained, but could be related to
small variations in the 8 1 8 0 values of C2 calcites 1 2 C derivation from several sources, such as soil
compared with the wide range of D2 dolomite/ C02 in modified meteoric waters, the dissolution of
ankerite may be due to precipitation within a nar­ eogenetic carbonates and thermal decarboxylation
rower temperature interval, which in tum could be of organic matter. As there is no correlation be­
related to extended residence time of the Serraria at tween 8 1 3 C and 8 1 80 values (Fig. 1 0), the input of
maximum temperature in the proximal domain. As carbon has not been related to temperature or to
we do not know the 8 1 8 0 composition of formation progressive, systematic variations in fluid composi­
waters during this phase, we are unable to calculate tion.
these precipitation temperatures precisely. How­ In the proximal Furado area other mesogenetic
ever, if we assume that precipitation occurred at minerals that formed included chlorite, illite and
near-maximum burial temperature (i.e. ;:::;; l OO O C), quartz. The abundance of chlorite in the Furado
 Lower Cretaceous Serraria sandstones
area is attributed both to the presence of large
amounts of unstable smectitic infiltrated clays and
pseudomatrix and to elevated temperatures, to­
gether with the availability of Fe and Mg due to the
precipitation of calcite rather than dolomite ce­
ments. Conversely, in the Caioba area Fe and Mg
were preferentially incorporated in dolomite/
ankerite cements, and illite was thus the only
mesogenetic clay formed.
Post-rift uplift and telodiagenesis (�1 14-74 Ma)
Differential crustal thinning during the rift phase
has favoured the uplift of some blocks in the distal
and middle domains, and caused the local exposure
of the Serraria sandstones during the Barremian.
During this time exposed areas were subjected to
humid climatic conditions (Parrish et a/., 1 982),
which caused meteoric water infiltration to promote
profound changes in the detrital and diagenetic
mineralogical composition of the sandstones. The
most important changes include dissolution and
kaolinization of feldspar, mica and pseudomatrix,
as well as the dissolution of dolomite and oxidation
of pyrite. The dissolution of dolomite and feldspars
was more pervasive close to the palaeo-exposure
surface and down to vertical depths of around
500-600 m (Fig. 2 1 ). Feldspars were more exten­
sively dissolved than dolomite. Overall, no feldspar
or carbonate cement remained at < 200 m of lateral
distance from the exposed area of the unit, whereas
no feldspar but 1 4% dolomite remained at �650 m
of lateral distance and �200 m of vertical depth
from the unconformity. Destruction of feldspars
shifted the framework mineralogy of the sandstones
from arkoses and subarkoses to diagenetic quartz
arenites (Fig. 6). The extent of penetration and
dissolution by meteoric fluids in sandstone aquifers
is directly related to the rate and volume of rainfall;
the extent of exposed area accessible for infiltration;
the time of effective infiltration; the hydraulic topo­
graphical gradient; and porosity and permeability.
In the Sergipe-A1agoas Basin the rainfall rate dur­
ing the first Myr of exposure time of the Serraria
sandstones was > 1 .0 m/yr. This high rate (Parrish
et a/., 1 982) provided significant volumes of mete­
oric fluid to infiltration. The exposed zone of the
Serraria sandstones was � 1 -5 km wide and �60 km
long. The time of effective direct exposure was
� 1 0 Myr. The hydraulic head was provided not
only by the differential uplift, but also by the tilting
of the blocks (e.g. Caioba area). The Serraria sand-
1 35
stones had substantial remaining porosity and per­
meability at the time of meteoric infiltration owing
to the limited carbonate cementation during the rift
phase. The telogenetic processes caused a substan­
tial increase in porosity and permeability, and thus
considerably enhanced the reservoir quality in this
particular area, which has up to 26% porosity and
1 1 23.4 mD permeability).
Post-rift subsidence and mesodiagenesis
(start at �ns Ma)
The magnitude of subsidence during the post-rift
stage was larger in the distal domain. In the Caioba
area the Serraria Formation was brought from
subaerial exposure down to �2000 m (maximum
T � 1 00 • C) (Fig. 1 7) in 20 Myr. Because of topo­
graphic variations generated by the uplift of this
area, some parts were buried only later during the
Upper Cretaceous. In the distal Robalo area the
amount of post-rift subsidence was slightly smaller,
bringing the Serraria Formation from a depth of
�2800 m down to 4200 m (maximum T � 1 40 · q
(Fig. 1 8). Conversely, i n the proximal domain no
substantial post-rift subsidence occurred and the
Serraria Formation remained at approximately the
same burial depths as achieved during the syn-rift
phase (i.e. maximum T � 1 oo · q (Fig. 20). Because
of these variations in subsidence history, the post­
rift mesogenetic modifications vary in intensity
between the different domains, and even between
different areas of the same domain. During the
post-rift phase most of the basin was covered by
marine deposits (Fig. 4), which would have pre­
cluded meteoric influence.
In the distal Caioba area the main diagenetic
minerals formed during this phase were D3
dolomite/ankerite cement, dickite, quartz over­
growths, illite and coarse crystalline pyrite. Except
for dickite, these minerals also occur in sandstones
of the Robalo area, which contain more quartz
overgrowths and illite. The predominance of dicki­
tization versus illitization of kaolinite in the Caioba
area compared with the Robalo area is presumably
controlled by the higher maximum burial depths
and temperatures experienced by the latter. Burial
diagenetic transformation of kaolinite into dickite
occurs at temperatures between �so - c and 1 2o · c
(Ehrenberg e t a/., 1 993; McAulay e t a/., 1 994;
Morad et a/. , 1 994), whereas extensive kaolinite
illitization is known to occur at temperatures
greater than � 1 30 " C (Ehrenberg & Nadeau, 1 989;
1 36 A.J V Garcia et a!.
Bj0rlykke & Aagaard, 1 992). The total absence of
detrital K-feldspar in the Robalo area may thus
indicate destruction due to kaolinite illitization,
which can be envisaged as follows:
AI2 Si 205(0H)4 + KA1Si 3 0 8
KA1 3 Si 30 1 0(0Hh + 2Si02 + H2b
=
Besides the conceivable effect of temperature on
illite formation in the Robalo area, it is probable
that the almost complete destruction of detrital
K-feldspar during uplift and telodiagenesis pre­
vented illite formation in the Caioba area (see
Ehrenberg, 1 99 1 ). The extensive quartz cementa­
tion in the Robalo sandstones, which substantially
reduced their porosity and permeability, is perhaps
partially derived from silica supplied from reaction
( 1 ) above. However, part of the silica is believed to
be related to pressure dissolution along intergranu­
lar contacts and stylolites, enhanced by the great
burial depths and temperatures in Robalo, com­
pared with the Caioba region. Pervasive quartz
cementation containing bitumen inclusions below
4226 m in Robalo sandstones (Fig. 1 4B) may indi­
cate the position of the original oil-water contact in
the block.
The 8 1 8 0 values of dolomite/ankerite formed
during the post-rift mesogenetic phase are some­
what higher in the Caioba (:;::;- 7.0%o to -6.9%o) than
in the Robalo area (-8.2%o), but have similar 8 1 3C
values (av. :;::; - 1 2. 3%o). The lower 8 1 8 0 value in the
Robalo area is perhaps due to higher maximum
temperatures than in Caioba. If we assume that
precipitation occurred at maximum burial temper­
atures, the average 8 ' 8 0sMow of pore waters would
be +3%o in Caioba and +8%o in Robalo. These
isotopic values indicate that the formation waters at
this stage were considerably evolved owing to pro­
gressive burial diagenetic interactions with the sili­
cates (Land & Fisher, 1 98 7).
In the Furado area there was no interruption in
burial conditions caused by post-rift uplift, and the
reservoirs remained at similar depths during this
phase. Therefore, it is uncertain whether or not the
mesogenetic constituents were precipitated during
syn-rift or post-rift phases. Some C2 calcite, which
engulfed and thus post-dated albite, chlorite, illite,
quartz and trace amounts of pyrite, barite and
sphalerite, is interpreted to have precipitated dur­
ing this time interval. This C2 calcite is character­
ized by a chemical and isotopic composition similar
to the carbonate cements formed during the syn-rift
subsidence phase.
The total albitization of detrital plagioclase, com-
pared with the partial albitization of K-feldspar, is
probably due to preferential replacement of calcian
plagioclases. Morad et al. ( 1 990) concluded that
plagioclase in Triassic sandstones from the North
Sea off Norway was albitized before K-feldspar.
These authors found that plagioclase albitization
(I)
may contribute small amounts of calcite cement.
Indeed, calcite occurs in dissolution voids of albi­
tized plagioclase of Furado area sandstones.
Recent exposure and telodiagenesis
Sub-Recent (timing is not precisely known) uplift
exposed the Serraria Formation in the Japoatii­
Penedo area of the middle domain (Fig. 2 1 ). The
present-day climatic conditions of coastal NE Brazil
are semi-humid, with heavy rainfall during 3
months followed by dry seasons. The burial history
of this domain is poorly constrained and coring is
limited and fragmentary. The telogenetic modifica­
tions include dissolution of dolomite and calcite
cements, as well as dissolution and kaolinization of
feldspars, infiltrated clays and mud intraclasts. Al­
ternating precipitation of quartz/chalcedony (up to
1 3%) and Fe-oxide (up to 22%) occurs in outcrop
samples. Dissolution of the silicates and carbonates
caused a substantial increase in porosity (av. 1 7%;
up to 30%). Total average porosity in sandstones
cemented by silica and Fe-oxides is 7%. This
silcrete/laterite association suggests a low-relief
landscape and a strongly seasonal hot climate, with
rainfall probably in excess of 1 .0 m/yr (Van de
Graaff, 1 983). The source of Fe and Si for the
precipitation of these cements was probably the
meteoric dissolution of ferroan dolomite/ankerite
cements and detrital feldspars. The incidence of
seasonally humid conditions is indicated by the
presence of kaolinite in these rocks.
Reservoir inferences
The patterns of diagenetic evolution recognized in
this study allow discussion of the conditions for
optimum porosity preservation and/or enhance­
ment in the Serraria reservoirs. The best reservoirs
of the unit occur in the Caioba area of the distal
domain, where porosity was enhanced by dissolu­
tion of detrital feldspars and dolomite cement
during telogenetic influx of meteoric waters. Similar
conditions are expected for other structural blocks
of the basin affected by post-rift uplift and erosion,
or blocks bounded by major fault systems in which
the Serraria Formation was relatively close to the
 Lower Cretaceous Serraria sandstones
exposure surface. These conditions are met in sev­
eral blocks of the middle and distal domains and
along portions of the margins of the rift basin
(Garcia et a/., 1 990).
The more extensive eogenetic carbonate cemen­
tation in the distal domain than in the proximal
areas may have played a role in the preservation of
higher porosity and intergranular volumes. Porosity
enhancement by carbonate cement dissolution due
to telogenetic meteoric influx into the reservoirs of
the distal domain is significant compared with
feldspar dissolution. Other conditions for porosity
preservation which are likely to have played an
important role in the anomalous values of up to
1 8% porosity and 300 mD permeability at 4200 m
in the Robalo area are the presence of clay coatings
and relatively early oil saturation. In these reser­
voirs, quartz cementation was apparently inhibited
by thin coatings of illitized infiltrated clay. Quartz
cementation is extensive beneath the interpreted
palaeo-oil-water contact. The total destruction of
feldspars in this area could be accounted for by the
generation and migration of organic solvents de­
rived from mudstones (Surdam et a/. , 1 989a,b).
Considering that the Serraria Formation was buried
relatively deeply in the distal Robalo area
(�2500 m) at the time of post-rift exposure, teloge­
netic feldspar dissolution and porosity enhance­
ment to an extent similar to that in the distal Caioba
area must be regarded as a remote possibility. As
rift and marine shales are mature in several areas of
the basin (Bruhn et a/. , 1 988), there are two possible
mechanisms for creating the optimum conditions of
mesogenetic porosity enhancement (Garcia et a/.,
1 990): (i) recurrent episodes of organic acid gener­
ation and related dissolution (see Bruhn et a/. ,
1 988), and (ii) migration of the hydrocarbons gen­
erated from deeper source rocks into reservoirs
affected by penecontemporaneous dissolution re­
lated to organic solvents coming from shallower
source rocks (see Moraes, 1 989; De Ros, 1 990).
Conditions of optimum mesogenetic porosity en­
hancement should be further modelled with time­
temperature-integrated evaluation procedures
(Surdam et a/. 1 989a,b) for other deep blocks with
structural settings and burial histories similar to
those of the Robalo area.
CONCLUSIONS
I t has been demonstrated i n this study that a proper
understanding of the diagenetic and porosity evolu-
1 37
tion of sandstone reservoirs can be achieved by
constraining the combined variable effects of the
palaeogeographical settings, palaeoclimatic condi­
tions, depositional environments and burial histo­
ries. Based on these parameters, four major
diagenetic domains were distinguished in the Early
Cretaceous Serraria sandstones. These vary consid­
erably in terms of the distribution patterns and
geochemical composition of carbonate cements, the
extent of grain dissolution and kaolinization, feld­
spar albitization and clay mineral diagenesis.
Eogenetic carbonate precipitation in the vadose
'
and phreatlC zones was largely controlled by the
arid to semi�arid climatic conditions at the time of
deposition C� l 44- 1 40/ 1 3 5 Ma; Berriasian). The
formation of groundwater dolocrete (o 13C � -8. 5%o
to -3. 1 o/oo and 8180 � -8. 7%o to -6. 7%o) in the distal
sandstones was probably related to an increase in
the Mg!Ca ratio of groundwaters due to evapora­
tion and precipitation of pedogenic and ground­
water calcrete (o1 3C � -6. 7%o and 8180 � -8.2%o)
in proximal sandstones. The 8180 compositions of
these eogenetic carbonates are consistent with pre­
cipitation from slightly evolved meteoric ground­
waters (o180sMow -5%o at the palaeolatitude of
=
�2o · s) at temperatures of �30 0 C for calcite and
< 45-500C for dolomite. Ions needed for the for­
mation of these eogenetic carbonates were probably
derived from the groundwater infiltration and detri­
tal mineral alteration in the proximal sediments.
The mesogenetic carbonate cements formed during
the syn-rift subsidence (� 1 40/ 1 3 5- 1 1 8 Ma) inher­
ited the mineralogical composition of eogenetic
cements, with calcite in the proximal (�2700 m of
maximum depth and T � 1 00- 1 1 0 · c from 1 2 5 Ma
until now) and dolomite in the distal sandstones
(buried at depths of � 750-2500 m; T � 40- l OO•C).
The mesogenetic dolomite/ankerite (o 180 -1 0.6%o
=
to -4. 1 o/oo) has precipitated from evolved meteoric
pore water mixed with compacti onal pore fluids
expelled from rift shales in the distal domain
(o180sMow -2%o) at temperatures between 45 and
=
90 • C. Lower o 1 80p08 values (- 1 3 . 6%o to - 1 1 . 5%o)
and tighter packing in sandstones cemented by
mesogenetic calcite (o13Cp08 - 1 1 .2o/oo to -3. 5%o)
=
in the proximal domain indicate that precipitation
occurred at relatively greater burial depths and
temperatures (2700 m and t oo · q than mesoge­
netic dolomite during the syn-rift subsidence.
Subaerial exposure and a more humid climate at
the beginning of post-rift uplift (� 1 1 4-74 Ma) in
part of the distal domain resulted in extensive
feldspar and carbonate dissolution, kaolinite pre-
1 38 A.J. V Garcia et a!.
cipitation and porosity enhancement. Meteoric
water infiltration in this area promoted a strong
modification of the framework composition of the
sandstones, which resulted in the formation of
diagenetic quartz arenites from original arkoses and
subarkoses.
The extent of post-rift subsidence (start at
� 1 1 5 Ma) was variable for different blocks in the
basin. In some areas the Serraria Formation was
brought from subaerial exposure down to �2000 m
(maximum T � 1 00 " C) in 20 Myr. In other distal
areas the amount of post-rift subsidence was slightly
smaller, which brought the Serraria Formation from
a depth of �2800 m down to 4200 m (maximum
T � 1 40 C). Conversely, in the proximal domain
•
no substantial post-rift subsidence occurred and the
unit remained at approximately the same burial
depths and temperatures (maximum T � 1 1 O " C)
achieved during the syn-rift mesogenetic phase. In
the distal domain the main diagenetic minerals
formed during this phase were dolomite, dickite,
quartz overgrowths, illite and coarse crystalline
pyrite. In proximal areas calcite cementation dur­
ing this time interval was accompanied by precipi­
tation of albite, chlorite, illite, quartz and trace
amounts of pyrite, barite and sphalerite. The 8 1 80
values of dolomite in distal areas (-8.2o/oo to
-6.9%o) indicate that the formation waters at this
stage (o 1 80sMow � +3%o to +8o/oo) were consider­
ably evolved due to progressive burial diagenetic
interactions with silicates.
The best reservoir quality potential expected for
the Serraria Formation is in structural blocks in the
distal and middle domains affected by porosity
enhancement through extensive feldspar and car­
bonate cement dissolution in connection with the
post-rift exposure and telogenetic influx of meteoric
waters.
A C K N O W L E D G E M EN T S
We are grateful for the financial support o f the
Brazilian National Council of Research (CNPq;
grants 200465/92.9-GL to L.F.D.R. and 200 1 97 I
95.9-GL to A.J.V.G.), the Swedish Natural Science
Research Council (NFR; to S.M.), and the Natural
Sciences and Engineering Research Council of Can­
ada (NSERC; to l.S.A.), and to PETROBRAS for
access to samples and information, and permission
to publish this work. Comments by reviewers S.P.
Dutton and S. Phillips helped to improve the
manuscript. A.J.V.G. acknowledges P. de Cesero for
initial stimulus to the study of sandstone petrology
and of the Serraria Formation in particular. We also
thank H. Harrysson for aid with the microprobe
analyses, C. Back and B. Gios for photographic
work, and C. Wernstrom for drafting the figures.
REFERENCES
ARAKEL, A.V., JACOBSON, G . & LYONS, w . ( 1 990)
Sediments-water interaction as a control on geochemi­
cal evolution of playa lake systems in the Australian arid
interior. Hydrob iologia, 197, 1 - 1 2.
BJ0RLYKKE, K. & AAGAARD, P. ( 1 992) Clay minerals in
North Sea sandstones. In: Origin, D iagenes is, and Petro­
phys ics of Clay Minerals in Sandstones (Eds
Houseknecht, D.W. & Pittman, E.D.). Spec. Publ. Soc.
econ. Paleont. Miner., Tulsa, 47, 65-80.
BJ0RLYKKE, K., RAMM, M. & SAIGAL, G.C. ( 1 989) Sand­
stone diagenesis and porosity modification during basin
evolution. Geol. Rundsch. , 78, 243-268.
BRUHN, C.H.L., CAINELLI, C. & MATOS, R.M.D. ( 1 98 8)
Habitat do petroleo e fronteiras exploratorias nos rifts
brasileiros. Bol. Geoc ienc. PETROBRAS, 2, 2 1 7-2 5 3 .
CARLISLE, D. ( 1 983) Concentration o f uranium and vana­
dium in calcretes and gypcretes. In: Res idual Depos its:
Surface Related Weathering Processes and Materials
(Ed. Wilson, R.C.L.). Spec. Publ. Geol. Soc. London,
1 1 , 1 8 5- 1 95.
CERLING, T.E. ( 1 984) The stable isotopic composition of
modern soil carbonate and its relationship to climate.
Earth Planet. Sc i. Lett., 71, 229-240.
CRAIG, H. ( 1 9 57) Isotopic standards for carbon and oxygen
correction factors for mass spectrometric analysis of
carbon dioxide. Geoch im. Cosmoch im. Acta, 12, 1 33-
1 49.
CuRTIS, C.D. ( 1 98 3 ) Geochemistry of porosity enhance­
ment and reduction in clastic sediments. In: Petroleum
Geochem istry and Exploration of Europe (Ed. Brooks,
J.). Spec. Publ. Geol. Soc. London, 12, 1 1 3- 1 2 5 .
CuRTIS, C .D . ( 1 990) Aspects o f climatic influence o n the
clay mineralogy and geochemistry of soils, lateosols and
clastic sedimentary rocks. J. Geol. Soc. , London, 147 ,
3 5 1 -3 5 7 .
D E Ros, L.F. ( 1 990) Porosity preservation and generation
in deep clastic reservoirs: a review. Bot. Geoc. PETRO­
BRAS, 4, 3 87-404 (in Portuguese with a summary in
English).
DICKINSON, W.W. ( 1 987) An oxygen isotope model for
interpreting carbonate diagenesis in nonmarine rocks,
(Green River Basin, Wyoming, USA). Chern. Geol.
(!sot. Geosci. Sec.), 65, 1 03- 1 1 6.
DICKSON, J .A.D. ( 1 965) A modified staining technique for
carbonates in thin section. Nature, 205, 5 8 7 .
DuTTA, P.K. ( 1 992) Climat ic Influence o n D iagenes is of
Fluv ial Sandstones, 1 992. D iagenes is Ill. Developments
in Sedimentology, No. 47. Elsevier, Amsterdam, 673
pp.
EHRENBERG, S.N. ( 1 99 1 ) Kaolinized, potassium-leached
zones in the contacts of the Gam Formation, Hatten-
 Lower Cretaceous Serraria sandstones
banken, mid-Norwegian continental shelf. Mar. Petrol.
Geo/. , 8, 250-269.
EHRENBERG, S.N. & NADEAU, P.H. ( 1 989) Formation of
diagenetic illite in sandstones of the Gam Formation,
Haltenbanken area, mid-Norwegian continental shelf.
Clay Miner. , 24, 2 3 3-253.
EHRENBERG, S.N., AAGAARD, P., WILSON, M.J., FRASER,
A.R. & DuTHIE, D.M.L. ( 1 993) Depth-dependent trans­
formation of kaolinite to dickite in sandstones of the
Norwegian Continental shelf. Clay Miner. , 28, 325-
3 52.
FOLK, R.L. ( 1 968) Petrology ofSedimentary Rocks. Hemp­
hill, Austin, TX, I 07 pp.
FRANKS, S.G. & FORESTER, R.W. ( 1 984) Relationships
among secondary porosity, pore-fluid chemistry and car­
bon dioxide, Texas Gulf Coast. In: Clastic Diagenesis
(Eds McDonald, D.A. & Surdam, R.C.). Mem. Am.
Assoc. petrol. Geol., Tulsa, 37, 63-79.
FRIEDMAN, I. & O'NEIL, J.R. ( 1 977) Compilation ofStable
Isotopic Fractionation Factors of Geochemical Interest.
US Geol. Surv. Prof. Paper, 440-KK, 1 2 pp.
FuGITA, A.M. ( 1 974) A geomorfologia da superficie de dis­
cordancia pre-Aptiana na Bacia de Sergipe e sua relacao
com os campos de 6leo. 28th Congresso Brasileiro de
Geologia, Porto Alegre, RS, Brazil, I , 1 2 1 - 1 3 5 .
GARCIA, A.J.V . ( 1 9 9 1 a ) Evoluciio sedimentar d a sequencia
pre-rift das bacias costeiras e interiores do nordeste
brasileiro. Pesquisas, 18, 3- 1 2.
GARCIA, A.J.V. ( 1 99 1b) Paleogeografia do nordeste
brasileiro no Junisico Superior-Cretaceo Inferior. Geo­
ciencias, 10, 37-56.
GARCIA, A.J.V. ( 1 992) Estratigrafia, sedimentat;tioe dia­
genese dos arenitos da Formacao Serraria, Cretdceo
Inferior da Bacia Sergipe-Alagoas, nordeste do Brasil.
Doctorata Thesis, Universidade Federal do Rio Grande
do Sul/UFRGS.
GARCIA, A.J.V. & WILBERT, A. ( 1 995) Paleogeographic
evolution of Mesozoic pre-rift sequences in coastal and
interior basins of northeastern Brazils. In: Pangea:
Global Environments and Resources, 1 995 (Eds Embry,
A.F., Beauchamp, B. & Glass, D.J.). Can. Soc. Petrol.
Geol. Mem., 17, 1 23- 1 30.
GARCIA, A.J.V., DE Ros, L.F., SouzA, R.S. & BRUHN,
C.H.L. ( 1 990) Potencial de reservat6rios profundos
na Formacao Serraria, Bacia de Sergipe-Alagoas. Bot.
Geocienc. PETROBRAS, 4, 467-488 (in Portuguese
with a summary in English).
GILES, M.R. & MARSHALL, J.D. ( 1 986) Constraints on the
development of secondary porosity in the subsurface:
re-evaluation of processes. Mar. Petrol. Geol. , 3, 243-
255.
HuGGETT, H.J. ( 1 9 9 1 ) Climate, Earth Process and Earth
History, 1 991. Springer-Verlag, Berlin, 2 8 1 pp.
KAHN, J.S. ( 1 956) The analysis and distribution of the
properties of packing in sand-size sediments: I. On the
measurement of packing in sandstones. J. Geo/. , 64,
385-395.
KELLER, W.O. ( 1 970) Environmental aspects of clay min­
erals. J. sediment. Petrol. , 40, 788-8 1 3.
KUTZBACH, J.E. & GALLIMORE, R.G. ( 1 989) Pangaean
climates: megamonsoons of the megacontinent. J. Geo­
phys. Res. , 94, 3 34 1 -3 3 5 7 .
LAND, L.S. & FISHER, R.S. ( 1 987) Wilcox sandstone dia-
1 39
genesis, Texas Gulf Coast: a regional isotopic compari­
son with the Frio Formation. In: Diagenesis of
Sedimentary Sequences (Ed. Marshall, J.D.). Spec. Pub!.
Geol. Soc. London, 36, 2 1 9-23 5 .
LLOYD, C.R. ( 1 982) The mid-Cretaceous earth: paleogeo­
graphy, ocean circulation and temperature, and atmo­
spheric circulation. J. Geol. , 90, 393-4 1 3.
LUNDEGARD, P.O. & LAND, L.S. ( 1 986) Carbon dioxide
and organic acids: their role in porosity enhancement
and cementation, Paleogene of the Texas Gulf Coast. In:
Roles of Organic Matter in Sediment Diagenesis (Ed.
Gautier, D.L.). Spec. Pub!. Soc. econ. Paleont. Miner.,
Tulsa, 38, 1 29- 1 46.
MADI':, B., BEN BACCAR, M. & FRITZ, B. ( 1 994) Geochem­
ical modeling of diagenetic reactions: a thermodynamic
and kinetic approach. Miner. Mag., 58A, 549-550.
MARRIOT, S.B. & WRIGHT, V.P. ( 1 993) Palaeosols as
indicators of geomorphic stability in two Old Red
Sandstones alluvial suites, South Wales. J. Geo/. Soc.
London, 150, 1 1 09- 1 1 20.
McAuLAY, G.E., BuRLEY, S.D., FALLICK, A.E. & KuszNIR,
N.J. ( 1 994) Palaeohydrodynamic fluid flow regimes
during diagenesis of the Brent Group in the
Hutton-NW Hutton reservoirs: constraints from oxy­
gen isotope studies of authigenic kaolin and reverse
flexural modelling. Clay Miner. , 29, 609-626.
MORAD, S. ( 1 986) Albitization of K-feldspar grains in
Proterozoic arkoses and greywackes from southern
Sweden. N. Jb. Miner. Mh. , 4, 1 45- 1 56.
MORAD, S., BERGAN, M., KNARUD, R. & NYSTUEN, J.P.
( 1 990) Albitization of detrital plagioclase in Triassic
reservoir sandstones from the Snorre Field, Norwegian
North Sea. J. sediment. Petrol. , 60, 4 1 1 -425.
MORAD, S., BEN ISMAIL, H., DE Ros, L.F., AL-AASM, I.S. &
SERRHINI, N.-E. ( 1 994) Diagenesis and formation water
chemistry of Triassic reservoir sandstones from south­
em Tunisia. Sedimentology, 41, 1 253-1 272.
MORAES, M.A.S. ( 1 989) Diagenetic evolution of
Cretaceous-Tertiary turbidite reservoirs, Campos Ba­
sin, Brazil. Bull. Am. Ass. Petrol. Geo/. , 73, 598-6 1 2.
MORAES, M.A.S. & DE Ros, L.F. ( 1 990) Infiltrated clays in
fluvial Jurassic sandstones of Reconcavo Basin, north­
eastern Brazil. J. sediment. Petrol. , 60, 809-8 1 9.
MOZLEY, P.S. & HOERNLE, K. ( 1 990) Geochemistry of
carbonate cements in the Sag River and Shublik Forma­
tions (Triassic/Jurassic), North Slope, Alaska: implica­
tions for the geochemical evolution of formation waters.
Sedimentology, 37, 8 1 7-836.
·
OJEDA, H.A.O. ( 1 982) Structural framework, stratigraphy,
and evolution of Brazilian Marginal Basins. Bull. Am.
Ass. Petrol. Geol. , 66, 732-749.
PARRISH, J.T. & CURTIS, R.L. ( 1 982) Atmospheric circula­
tion, upwelling and organic-rich rocks in the Mesozoic
and Cenozoic Eras. Palaeogeog. Palaeoclim. Palaeo­
ecol. , 40, 3 1 -66.
PARRISH, J.T., ZIEGLER, A.M. & SCOTESE, C.R. ( 1 982)
Rainfall patterns and the distribution of coals and
evaporites in the Mesozoic and Cenozoic. Palaeogeog.
Palaeoclim. Palaeoecol. , 40, 454-460.
PoNTE, F.C. ( 1 97 1 ) Evolur;tio paleogeogrdfica do Brasil
oriental e da Africa ocidenta/. PETROBRAS, Unpub­
lished Internal Report CPEG 4, 7 1 pp.
PONTE, F. C. & ASMUS, H.E. ( 1 976) The Brazilian marginal
1 40 A.J V Garcia et a!.
basins: current state of knowledge. Anais Acad. Bras.
Cienc., 48, 2 1 5-240.
PoNTE, F.C. & ASMUS, H.E. ( 1 978) Geological framework
of the Brazilian continental margin. Geol. Rundsch., 68,
20 1 -2 3 5 .
ROSENBAUM, J.M. & SHEPPARD, S.M.F. ( 1 986) An isotopic
study of siderites, dolomites and ankerites at high
temperatures. Geoch im. Cosmoch im. Acta, 50, 1 1 47-
1 1 50.
SCHMIDT, V. & McDONALD, P.A. ( 1 979) The role of
secondary porosity in the course of sandstone diagene­
sis. In: Aspects of D iagenes is (Eds Scholle, P.A. &
Schluger, P.R.). Spec. Pub!. Soc. econ. Paleont. Miner.,
Tulsa, 29, 1 75-207.
SPOTL, C. & WRIGHT, V.P. ( 1 992) Groundwater dolocretes
from the Upper Triassic of the Paris Basin: a case study
of an arid, continental diagenetic facies. Sedimentology,
39, 1 1 1 9- 1 1 36.
SUCHECKl, R.K., HUBERT, J.F. & WET, C.C.B.D. ( 1 988)
Isotopic imprint of climate and hydrogeochemistry on
terrestrial strata of the Triassic-Jurassic Hartford and
Fundy Rift Basins. J. sediment. Petrol. , 58, 80 1 -8 1 1 .
SURDAM, R.C., CROSSEY, L.J., HAGEN, E.S. & HEASLER, H.P.
( l 989a) Organic-inorganic interactions and sandstone
diagenesis. Bull. Am. Ass. Petrol. Geo/. , 73, 1-23.
SuRDAM, R.C., DuNN, T.L., HEASLER, H.P. & MAcGowAN,
D.B. ( 1 989b) Porosity evolution in sandstone/shale
systems. In: Short Course on Burial D iagenes is (Ed.
Hutcheon, I.E.), pp. 6 1 - 1 3 3 . Mineralogical Association
of Canada, Montreal.
SZATMARI, P., FRANCOLIN, J.B.L., ZANOTTO, 0. & WOLFF, S.
( 1 987) Evolucao tect6nica da margen equatorial
brasileira. Rev. Bras. Geocienc. , 17, 1 80- 1 88 .
THYNE, G.D. & GwiNN, C.J. ( 1 994) Evidence for a pa­
leoaquifer from early diagenetic siderite of the Cardium
Formation, Alberta, Canada. J. sediment. Res. , A64,
726-732.
VAN DE GRAAFF, W.J.E. ( 1 98 3 ) Silcrete in Western Austra­
lia: geomorphological settings, textures, structures, and
their genetic implications. In: Residual Deposits: Surface
Related Weathering Processes and Materials (Ed. Wil­
son, R.C.L.). Spec. Pub!. Geol. Soc. London, II, 1 59-
1 66.
WALKER, T.R., WAUGH, B. & CRONE, A.J. ( 1 978} Diagene­
sis in first-cycle desert alluvium of Cenozoic age, south­
western United States and northwestern Mexico. Geol.
Soc. Am. Bull. , 89, 1 9-32.
WHITE, D.E., HEM, J.D. & WARING, G.A. ( 1 963) Chemical
composition of subsurface waters. In: Data of Geochem­
istry (Ed. Fleischer, M.). US Geol. Surv. Prof. Paper,
440, 1 8 pp.
WRIGHT, V.P. ( 1 992) Paleosol Recogn it ion: a Gu ide to
Early D iagenes is in Terrestrial Sett ings, 1 992. D iagene­
s is 111. Developments in Sedimentology, No. 47.
Elsevier, Amsterdam, 673 pp.
WRIGHT, V.P. & TucKER, M.E. ( 1 99 1 ) Calcretes, an intro­
duction. In: Calcretes. lAS Reprint Series, Oxford, 2,
1 -22.
WRIGHT, V.P. & VANSTONE, S.D. ( 1 99 1 ) Assessing the
carbon dioxide content of ancient atmospheres using
palaeo-calcretes: theoretical and empirical constraints.
J. Geol. Soc., Lond. , 148, 945-947.
Spec. Pubis int. Ass. Sediment. ( 1 998) 26, 141-162

Carbonate cements in the Tertiary sandstones

of the Swiss Molasse basin: relevance to
palaeohydrodynamic reconstruction

J . MATYA S'
Geologisches Institut, Universittit Bern, Baltzerstrasse 1, CH-3012 Bern, Switzerland,
e-mail matyas@ght04u.unileoben.ac.at

ABSTRACT

Depositional and tectonic variations are not reflected in the diagenetic history of the sandstones of the
Lower Freshwater Molasse and Upper Marine Molasse in the Swiss Molasse basin. Calcite and
pore-lining clays are the main cements in both units, the authigenic mineral assemblage and
paragenesis are similar, and no major differences are detected in the stable isotopic composition of the
authigenic calcites. Evidence from fluid inclusions and stable isotopes suggests that calcites precipitated
early in the diagenetic history, from pore waters composed of variable proportions of the original
marine and fresh formation waters. The mixing of these waters was probably related to compactional
flow during subsidence. The isotopic signature of modern formation waters cannot be recognized
among the diagenetic calcites. These facts emphasize the importance of early fluid flow history on
porosity development in inverted foreland basins.

I N T R ODUCTIO N

This paper discusses carbonate cementation in Ter­

tiary sandstones of the Swiss Molasse basin. The
prime objectives are to reconstruct the postdeposi­
tional evolution of sandstones in the Lower Fresh­
water Molasse and in the Upper Marine Molasse,
and to try to understand the relationship between
porosity development and fluid flow in the basin,
based on the textural and geochemical characteris­
tics of the carbonate cements.
The numerous outcrops, the tunnels and the
extensively cored petroleum, water and geothermal
exploration wells make the Swiss Molasse basin an
ideal candidate for such a study by providing the
opportunity to develop the necessary regional sedi­
mentological framework and allowing detailed sam­
pling in the wells and tunnels to study diagenesis in
the subsurface.
' Present address: HOT Engineering GmbH, Rosegger­
strasse 1 5, A-8700 Leoben, Austria.
Geological setting
Tectonic framework
The Swiss Molasse basin (SMB), located between
the Jura Mountains and the western Alps (Fig. 1 ), is
part of the North Alpine Foreland Basin which
formed as a mechanical response. to the tectonic
load of the northward propagating alpine thrust
wedge (Homewood et a/., 1 986; Schlunegger et a/.,
1 997). Structurally it can be subdivided into the
extensively deformed Subalpine Molasse, including
a stack of imbricate thrust sheets and the classic
triangle zone (Pfiffner, 1 986; Pfiffner et a/., 1 997),
and the relatively undeformed sequences of the
Plateau or Mittelland Molasse. Various thermal
indicators, such as vitrinite reflectance (Schegg,
1 992, 1 993, 1 994), apatite fission track (Matter et
a/., 1 988) and illite/smectite (liS) diagenesis (Mon­
nier, 1 979, 1 982; Schegg, 1 992), indicate that the
Swiss Molasse basin is now in inversion.

Carbonate Cementation in Sandstones Edited by Sadoon Morad 141

© 1998 The International Association of Sedimentologists
ISBN: 978-0-632-04777-2
1 42 J. Mdtyds

Fig. 1. Location map showing the

sampled wells and tunnel sections.

Fig. 2; the key characteristics of the units are as

Five major stratigraphic units
Stratigraphically, the sediments filling the basin can
be subdivided into five major units (Matter et a!.,
1 980). The depositional sequence is shown in
follows:
1 The deep-water turbiditic sediments of the North
Helvetic Flysch, which represents the earliest stage
of basin development, were deposited during Pria-

Litho- Initial
Ma N s stratigraphy formation water

14
Upper Freshwater
Freshwater
Molasse (UFM)

18 Marine with
Upper Marine
Molasse (UMM) local freshwater
lenses

22

Dominantly
Lower Freshwater freshwater
26
Molasse (LFM) with local
brackish influence

30
Marine with
Lower Marine
local freshwater
Molasse (LMM) lenses
34

North Helvetic
Flysch (NHF) Marine

C=:J Conglomerates E3 Siltstones

� Sandstones - Marls Fig. 2. Schematic sedimentological
log.
 Carbonate cements in the Swiss Molasse basin
bonian time (between 40 and 36 Ma). Their maxi­
mum thickness is 4-5 km (Pfiffner, 1 9 86).
2 The overlying Lower Marine Molasse is repre­
sented by storm-dominated beach sediments, off­
shore marls and turbiditic sandstones, with a
cumulative compacted thickness of 500 m or more
(Diem, 1 986; Homewood et al., 1 986). The time­
span of deposition of this unit is relatively short,
being limited to the Rupelian stage (Diem, 1 986).
3 The Lower Marine Molasse and its regressive
counterpart, called the Lower Freshwater Molasse,
form one of the two major coarsening and shallow­
ing upward megasequences filling up the Swiss
Molasse basin. The Lower Freshwater Molasse is
composed of alluvial fan conglomerates and sand­
stones in the proximal areas, and fluvial and lacus­
trine sediments in the distal part (Homewood et al.,
1 986; Platt & Keller, 1 992). The cumulative thick­
ness of the Lower Freshwater Molasse shows signifi­
cant lateral variations. In the proximal areas it may
exceed 4 km (Matter et al., 1 980), whereas in the
northern, distal part, close to the Jura Mountains, it
is only a few hundred metres (Homewood et al.,
1 986). The transgression of the 'Burdigalian' Sea at
the Aquitanian/Burdigalian boundary sets the up­
per limit of the time-span of the deposition of the
Lower Freshwater Molasse, ranging between 20 and
1 9.5 Ma (in the south and north, respectively).
4 The sediments of the marine transgression are
represented by the wave- and tidal-dominated
sandstones and interbay mudstones of the Upper
Marine Molasse. Within the Upper Marine Mo­
lasse, four major facies belts can be recognized: (i)
the proximal fan delta facies, (ii) the coastal facies,
(iii) nearshore facies and (iv) the offshore facies
(Homewood, 1 98 1 ). The cumulative thickness of
the unit may be as much as 1 .3 km (Matter et al.,
1 980).
5 During the middle Miocene the sedimentation in
the basin turned once again to continental, complet­
ing the second major coarsening and shallowing
upward megasequence. The sediments of this pe­
riod are represented by the Upper Freshwater Mo­
lasse. As in the Lower Freshwater Molasse, the
proximal areas are also dominated by coarse­
grained conglomeratic sediments of the northward
prograding alluvial fans which laterally interfinger
with the channel-belt sandstones and floodplain
mudstones of the distal part of the depression. The
maximum compacted cumulative thickness of the
unit is 1 .5 km (in the south), and decreases to a few
hundred metres towards the north.
1 43
Sample selection
This study focuses on the two most important
lithostratigraphical units of the Swiss Molasse ba­
sin, the Lower Freshwater Molasse and the Upper
Marine Molasse. The present-day depth of samples
ranges from 1 5 to 1 300 m; the sample locations are
shown in Fig. 1 .
From the Upper Marine Molasse, samples were
taken from three wells (Tiefenbrunnen- 1 , Altisho­
fen- 1 and Gurten- 1 ) and from a tunnel section
(Sonnenberg Tunnel). The Lower Freshwater Mo­
lasse is represented by samples from three wells
(Bassersdorf- 1 , Murgental and Altishofen- 1 ) and
two tunnel sections (Sonnenberg and Grauholz
Tunnels). Most of the Lower Freshwater Molasse
samples are from two wells (Bassersdorf- 1 and
Altishofen- 1 ) and from the Grauholz Tunnel. The
study includes 260 samples altogether. Quantitative
petrographical and stable isotope geochemical anal­
yses were performed on 96 samples. Detailed sedi­
mentological logs and the sampling programs for
five major locations (Altishofen- 1 , Bassersdorf- 1 ,
Gurten- 1 , Sonnenberg Tunnel, Tiefenbrunnen- 1 )
are given in Fig. 3.
M ETHODS
Samples were impregnated with a high-temperature
blue-dyed epoxy resin before thin-section prepara­
tion. Polished thin sections were examined with a
petrographic microscope and by using a hot­
cathodoluminescence (CL) microscope (Matter &
Ramseyer, 1 985). Most of the samples were stained
using Dickson's ( 1 966) method, and point counted
(300 points per sample).
Samples for clay mineralogical analysis were pre­
pared using standard gravitationa!.technique. Semi­
quantitative estimates of the relative abundance of
clay minerals in the <2 J..Lm fraction were made
using the method given by Moore & Reynolds
( 1 989).
Selected samples were examined on a CamScan
Series 5 scanning electron microscope equipped
with a Tracor Northern 5400 energy-dispersive
spectrometer.
For C and 0 stable isotope analyses, powdered
bulk-rock samples were reacted for 1 2 min (calcites)
and 5 h (dolomite) in 1 00% H 3 P04 at 50±0.2oC.
The isotopic ratios of the released C02 gas were
measured on a VG Prism II ratio mass spectrome-
 Sonnenberg Tiefenbrunnen-1 """
Altishofen-1 Bassersdorf-1 Gurten-1 """
:>, :>, E' :>, :>,
:>,
Lithology/ ..c Lithology/ ..c Lithology/ ..c Lithology/ ..c
Lithology/ ..c
c. c. c. c.
c. Grain size Grain size V> Grain size Grain size Cl
Grain size E Cl E Cl E' E Cl V> E Cl E' E
E' I .!: Cl .!: ., Cl .!: Cl .!:
Cl Cl c c_

II
�

c_ ·;; c_ ·;;
�

·;; ·;; c_ ·;; -""u £

�

£ £ £ E c. E E
c. E c. E E E ...,
c.
., E
..., 0
., Sh FS CSCg
E
...,
Ul
"'
Ul
.,
0
Sh FS CSCg
Ul
..., "'
Ul
E
1--
Sh FS CSCg ...,
Ul
"'
Ul
.,
0
ShFSCSCg
I
Ul
llfl
"'
' ' I I I
Sh FS CSCg
0 Ul I I I ' I I I I I I
I
1700
.... .. ...........
. .

1: {: : .::: : :: .:
_ .

1600
100

11·.:.:.:.:.:.·.1 lluMM IH 100

1500 Q
1: ·: : : : ::: :::::· :.
. 100
200
50 1400

300 1300
200
400
jF IHII 1200
200

100 1100
300

= !�...
1000 300 �· . " ' :.....
900
' 400
�
700 �
150 800 400 "'"
1 �- El·
700 '. .
800 ....
1� 11._ I� 500
600 ...
•.'.•,'.·,·.·
..
900
200 500
LFM ;oo It
400

·���
1100

1200 � '" 700 250

300

200

100
600

1300 f:1:1:!:1:1:1Mn-
l LFM 700

1 I�D 800
� Conglomerate � Very fine- to fine-grained � Limestone
Sh: claystones, siltstones, marls Q: Quaternary

� � sandstone, with shale/marl,

sandstone content i s >50% FS: very fine- and fine-grained sandstone UFM: Upper Freshwater Molasse

� Medium- to coarse -gramed .. Shale/marl, with very fine- to fine

CS: medium- and coarse-grained sandstones UMM: Upper Manne Molasse

� sandstone grained sandstone. Cg: conglomerates u:M: Lower Freshwater Molasse

sandstone content is <50%
K1m.: K1mmendgean

Fig. 3. Lithological profiles of the five sampled sections.

 Carbonate cements in the Swiss Molasse basin
ter. The isotopic reproducibility of standard mate­
rials is better than 0. 1 o/oo for 1 8 0 and 0.05o/oo for
1 3 C. Pyrite o3 4S measurements were performed in
situ on polished samples by the laser microprobe
combustion method described by Kelley & Fallick
( 1 990) and Fallick et a!. ( 1 992); precision is
±0.2o/oo.
Fluid inclusion analysis was performed on
I 00 11m thick rock slices polished on both sides,
using a Linkam semi-automatic freezing-heating
stage. The reproducibility of temperature measure­
ments on standards is ± 0.2 oc.
R E SULTS
Detrital composition
According to McBride's ( 1 963) classification, the
studied sandstones classify mostly as feldspathic
litharenites or lithic subarkoses, with a few samples
falling into the sublitharenite, litharenite and lithic
arkose fields (Fig. 4). There appears to be little
systematic variation between the detrital composi­
tion of different locations.
The dominant detrital constituent in both the
Lower Freshwater Molasse and the Upper Marine
Molasse samples is monocrystalline quartz (ranging
from 1 9. 2 to 50.3%). The dark blue and brown CL
indicates mostly plutonic and/or metamorphic ori­
gin; polycrystalline quartz is much less abundant.
Subarkose
Lithic subarkose
Fig. 4. Ternary plot showing the
detrital composition of the studied
sandstones. Feldspar
1 45
The amount of feldspars varies significantly within
a range of 4.3-2 1 %. Both plagioclase and K-felds­
pars occur.
Dolomite, igneous and metamorphic rock frag­
ments are the dominant lithic grains, but volcanic
grains, micritic cherty limestone, reworked caliche
fragments and shales (including siltstones and flysch
fragments) also occur. The dolomites show bright or
dark red to orange luminescence. Rock fragments
vary in abundance between 8.6 and 4 1 .2%.
Sheet silicates (muscovite, biotite and chlorite)
are absent or occur only in minor amounts in most
of the samples from the Upper Marine Molasse,
whereas in the Lower Freshwater Molasse they can
be locally more abundant. Minor or trace amounts
of opaque and accessory minerals occur in each
sample. In the Lower Freshwater Molasse the acces­
sories are mostly heavy minerals, whereas in the
Upper Marine Molasse they are typically glauconite
grains.
The matrix is mostly clayey, in some cases rich in
feldspars (as shown by the bright spots under the CL
microscope) and variable amounts of carbonate.
Authigenic minerals
Postdepositional processes resulted in a quite sig­
nificant modification of depositional porosity. In
addition to compaction, carbonate and clay min­
eral cementation is the major porosity reducing
factor.
Quartz 0 A.ltishofen
b. Bassersdorf
0 Grauholz
+ Murgental
"' Gurten
+ Sonnenberg
x Tiefenbrunnen
Feldspathic
litharenite
Rock fragments
1 46 J. Mdtyds

Fig. 5. Photographs showing the characteristic authigenic carbonates in the studied sandstones. (A,B) Cross-polarized
and UV-fluorescence photomicrographs showing early dolomite overgrowth on a detrital dolomite grain (note the
inclusion-free thin authigenic rim, arrows), Tiefenbrunnen, 664. 7 1 m, UMM. Scale bar 200 IJ.m. (C)
Cathodoluminescence photomicrograph showing the pervasive mosaic cement, Gurten, 229.8 m, UMM. Scale bar
300 IJ.m. (D) Cathodoluminescence photomicrograph showing the euhedral calcite cement (arrow), Altishofen, 9 5 5 . 7 m,
LFM. Scale bar, 300 IJ.m. (E) Photomicrograph showing pervasive mosaic calcite from Murgental, 1 8. 1 m, LFM. Scale
bar 500 !J.m. (F) Photomicrograph showing early dolomite overgrowth. Note the inclusion-free overgrowth and the
postdating ferroan calcite, stained darker. Tiefenbrunnen, 596.92 m, UMM. Scale bar 300 IJ.m. (See also colour Plate I,
facing p. 1 58.)
 Carbonate cements in the Swiss Molasse basin
Diagenetic calcite
Calcite occurs as intergranular cement and as com­
plete or partial replacement of detrital components,
mostly lithic grains such as volcanic rock fragments.
Intergranular calcite cement occurs as sparry/
microsparry, commonly pervasive, cement (Figs
SC,E; see also Plate 1 , facing p. 1 58) and as small
(1 0-40 J.lm) crystals with euhedral faces (Figs 50,
6B and 7 A-D). These are attached to the surface of
the detrital grains and may occur individually or
form clusters of several crystals. Unlike the sparry
cement, the euhedral cement is rarely pervasive.
Both types of calcite show a bright yellow CL
(Fig. SC,D). The morphology of the euhedral crys­
tals is remarkably similar in samples from Altisho­
fen, Bassersdorf and Gurten (Fig. 7A-D).
Dolomite, ferroan dolomite and ankerite occur
mostly as thin ( < 20 J.lm) overgrowths on detrital
dolomite grains (Figs SA,B,F and 6A) or, rarely, as
Fig. 6. Enlarged views of selected carbonate cements.
(A) Photomicrograph showing early dolomite
overgrowth. Note the inclusion-free overgrowth (white
arrow). Gurten, 1 39.2 m, UMM. Scale bar 1 50 j.lm. (B)
Photomicrograph showing the euhedral calcite cement
(arrows) in Altishofen, 9 5 5 . 7 m, LFM. Scale bar !50 j.lm.
1 47
discrete rhombohedral crystals growing into the
intergranular pore space. In some cases the detrital
grains and the overgrowths are easy to distinguish,
as the overgrowths are inclusion free and often stain
pale light blue (Figs SF and 6A). In general, how­
ever, the identification of diagenetic dolomite is
difficult, as authigenic dolomite occurs as submicro­
scopic overgrowth on the microsparry detrital car­
bonate fragments. Dolomite, ferroan dolomite and
ankerite are dark orange to non-luminescent; the
ferroan dolomite and ankerite stain pale blue. The
overgrowths in the Upper Marine Molasse show
yellowish fluorescence under UV illumination
(Fig. SA,B), indicating the presence of organic mat­
ter (Emery & Robinson, 1 993).
The clay fraction
Authigenic clays (mixed-layer clays, illite, kaolinite)
are common, though volumetrically not abundant
components. Mixed-layer clays occur as highly
crenulated, well-developed pore-lining cements,
both in the Lower Freshwater Molasse (Altishofen,
Bassersdorf ) and in the Upper Marine Molasse
(Gurten, Tiefenbrunnen). Their composition ranges
from nearly pure smectite through smectite/chlorites
and smectite/illites to chlorite-rich chlorite/smectite.
In sandstones of the Lower Freshwater Molasse the
clays are smectites or smectite/chlorites, whereas
in sandstones of the Upper Marine Molasse the
chlorite/smectite dominates (Matyas & Matter, in
preparation).
Filamentous illite occurs in minor amounts in
most samples. Illite crystals are attached to the
edges of the mixed-layer pore-lining clays and to
detrital grains. They grow free into the open pores,
or bridge pores. Although illite is ubiquitous, it is
less abundant than the pore-lining clays. The only
exception is Grauholz, where the pore-lining clays
are absent and the filamentous ill'ite dominates.
Kaolinite occurs only in traces in a few samples
from the Lower Freshwater Molasse in Altishofen.
It occurs as tightly clustered, vermicular aggregates
of pseudohexagonal crystals in pores or between
muscovite plates.
Minor cements
Authigenic K-feldspar is present in the Gurten
samples as thin overgrowths on detrital K-feldspar
grains. Quartz cement was found only in the SEM
in two samples from Altishofen. Sulphates (barite,
148 J Mdtyds

Fig. 7. Scanning electron micrographs showing the characteristic authigenic calcites (C) in the studied sandstones.
(A-D) Euhedral calcite cements, postdating the pore-lining clays: (A) LFM, Bassersdorf, 762.47 m; (B) LFM,
Altishofen, 869.9 m; (C) UMM, Gurten, 227.6 m; (D) UMM, Gurten, 225.3 m. (E) Photomicrograph showing
rhombohedral calcite from Tiefenbrunnen, 66 1 . 52m, UMM. (F) Photomicrograph showing calcite postdating
chlorite/smectite in Tiefenbrunnen, 663.55 m, UMM.
 Carbonate cements in the Swiss Molasse basin 1 49

bariocelestite, anhydrite) occur in trace or in minor

amounts. Barite occurs in all locations as sparry
pore-filling cement, whereas bariocelestite is re­
stricted to a few depth intervals in Bassersdorf and
Altishofen. Anhydrite occurs only in a narrow depth
range in Altishofen, where it is a pervasively dis­
solved cement.
Pyrite is the only authigenic sulphide found in the
studied samples, occurring in several textural types.
In Tiefenbrunnen it is present as framboidal aggre­
gates, whereas in Murgental it occurs as pore-filling
cement (Fig. 8A). In the Altishofen samples, partic­
ularly in those below 1 1 00 m, pyrite occurs as a
nearly complete replacement of mica (Fig. 8B). An
incipient stage of mica replacement occurs in
Tiefenbrunnen and Sonnenberg. Pyrite in Sonnen­
berg occurs as pore-filling cement with euhedral
crystal faces, or as a replacement of lithic grains.

Only slight differences in paragenetic sequence

Overall the paragenetic sequence is not substantially
different between the Lower Freshwater Molasse
(Fig. 9A) and the Upper Marine Molasse (Fig. 9B).
Dolomite, feldspar and pyrite are the only minerals
present in the pervasively calcite-cemented samples
of the Upper Marine Molasse and the Lower Fresh­
water Molasse. This, and the high (over 30%) inter­
granular volume of these samples, suggests that the
first generation of calcite formed early in the diage­
netic history. As opposed to other samples, feldspars
are unaltered in these samples, showing that feldspar
leaching postdates the first generation of calcite.
Barite and bariocelestite occur together in the
Lower Freshwater Molasse, and pre-date the pore­
lining smectites, but overlap or postdate feldspar
leaching, as barite occurs as intragranular cement in
partially dissolved feldspar. The second generation
of calcite is typically euhedral and postdates the
pore-lining clays (Fig. 7F), in both the Upper Ma­
rine Molasse and the Lower Freshwater Molasse.
Stable isotope geochemistry
The measured carbon and oxygen stable isotopic
ratios of calcites are given in Table 1 . In general, the
() 1 3C values of calcites reveal no major variations
between the different locations and lithostratigraph­
ical units, although calcites from the Upper Marine
Molasse are slightly heavier than those from the
Lower Freshwater Molasse. The () 1 80 values of
calcites from the Lower Freshwater Molasse are
Fig. 8. Photomicrographs showing the two major types
of authigenic pyrite (p). (A) Pervasive pyrite, occluding
intergranular pore space, Murgental, 50.86 m, LFM. (B)
Pyrite replacement, Altishofen, 1 279.6 m, LFM.
very similar throughout the studied locations; vari­
ation is slightly greater among the calcites of the
Upper Marine Molasse.
In spite of the overall similarities, the following
variations can be recognized in the () 180 vs. () 1 3C
plots (Fig. I OA,B):
I Among calcites of the Lower Freshwater Molasse
(Fig. l OA), the Murgental samples are the most
depleted in 1 3C.
2 Most of the Altishofen calcites have lower <l C
13
values than those of Bassersdorf, and higher lower
() 13C values than those of Murgental.
18 13
3 Two clusters are recognized on the () 0 vs. () C
plot of the Upper Marine Molasse calcites (Fig.
I OB). Samples falling on to the lower, main trend
are from Gurten, Sonnenberg and Tiefenbrunnen,
and reveal heavier oxygen and lighter carbon isoto­
pic compositions than those of the high trend,
which includes samples from Sonnenberg and
Tiefenbrunnen. The high trend samples from Son­
nenberg have substantially lower mean intergranu-
1 50 J. Mdtyds

(A)
Lower Freshwater Molasse (LFM)

------ Relative time ..

Calcite Ill 1111

Dolomite

Feldspar
Smectite (S'C, S'I)

lllite

Mechanical compaction

(B)
Upper Marine Molasse (UMM)

------ Relative time

Calcite II IIIII

Dolomite/Fe-dolomite

-�niDg�y{Q§)

Pyrite

�i¢11:1! .Jo.
Mechanical compaction

Fig. 9. Probable paragenetic

- volumetrically i mportant
-- volumetrically not i mportant
Q occurrence limited to certain locations
""'"";<M dissolution
uncertai n textural relationship
sequence of major diagenetic
events in (A) the Lower Freshwater
Molasse and (B) the Upper Marine
Molasse.

lar volumes (IGVs) than those of the main trend

samples ( 1 3% and 29%, respectively}.
Sulphur isotopes
Sulphur isotopic ratios were measured on two
samples from Murgental and Altishofen (Lower
Freshwater Molasse). In the pervasive pyrite ce­
ments from Murgental, both the edge and the centre
of a patch were measured, yielding o34 S coT (Can­
yon Diablo Troilite) values of - 1 4.4%o and + 1 . 1 o/oo,
respectively. In Altishofen, two spots of mica re­
placement were measured, yielding o34 S coT values
of -28. 1 o/oo and -22. 1 o/oo.
Fluid inclusion analysis
In spite of the significant effort invested in locating
inclusions large enough for fluid inclusion micro­
thermometry, measurements could only be carried
out on a limited set of samples (Table 2).
One-phase inclusions in the
Lower Freshwater Molasse
Microthermometry was successfully applied to one
pervasively cemented sample from the Murgental
area. The inclusions were all one-phase, primary
aqueous inclusions, yielding a range of final ice
melting temperatures between -2.o·c and -o.s·c.
This range of Tmice values can be converted to
NaCl-equivalent salinities of 3.4% and 0.9%, re­
spectively (Bodnar, 1 992). As these were one-phase
inclusions, the homogenization temperatures could
not be measured.
 Carbonate cements in the Swiss Molasse basin 151

Table 1 Summary of stable isotopic and sedimentological data

Calcite stable isotopes

Lithostratigraphic Facies
Location Sample unit association 013C-(Jlbo PDB) 1)180 (%o PDB)

Altishofen 138.00 UMM n/a -0.88 -12.14

Altishofen 852.10 LFM Mfi -2.47 -10.81
Altishofen 856.70 LFM Sfc -2.25 -9.43
Altishofen 863.85 LFM Sfi -2.29 -9.70
Altishofen 869.90 LFM Sfi -1.35 -8.90
Altishofen 876.05 LFM Sfi -1.83 -9.18
Altishofen 905.10 LFM Sfi -1.04 -8.57
Altishofen 908.05 LFM Mfi -2.14 -9.59
. Altishofen 908.65 LFM Sfi -2.05 -9.64
Altishofen 955.70 LFM Sfi -2.51 -9.23
Altishofen 987.05 LFM Mfi -1.96 -9.65
Altishofen 1028.95 LFM Sfi -1.78 -9.32
Altishofen 1047.80 LFM Sfi -1.96 -9.92
Altishofen 1119.95 LFM Mfi -2.98 -9.87
Altishofen 1176.70 LFM Sfi -2.02 -9.04
Altishofen 1212.90 LFM Sfc -1.40 -7.99
Altishofen 1214.00 LFM Sfc -1.84 -8.28
Altishofen 1240.20 LFM Sfc -2.48 -9.50
Altishofen 1279.60 LFM Sfc -1.38 -7.58
Altishofen 1281.80 LFM Sfc -2.23 -9.00
Bassersdorf 575.51 LFM Sfi -0.36 -8.32
Bassersdorf 576.22 LFM Sfi -0.07 -7.65
Bassersdorf 602.89 LFM Sfi 0.27 -6.71
Bassersdorf 603.35 LFM Sfi -0.08 -7.71
Bassersdorf 604.39 LFM Sfi -0.83 -9.66
Bassersdorf 628.97 LFM Sfi -1.50 -10.84
Bassersdorf 639.63 LFM Sfi -0.87 -9.36
Bassersdorf 662.31 LFM Sfi -1.75 -11.02
Bassersdorf 696.53 LFM Sfi -2.15 -11.00
Bassersdorf 698.78 LFM Mfi -1.75 -11.07
Bassersdorf 746.26 LFM Sfc -1.63 -10.61

Grauholz G14/1 LFM Sfc -1.03 -8.64

Grauholz G17/1 LFM Sfc -1.07 -8.69
Grauholz G18/1 LFM Sfi -1.49 -9.46
Grauholz G2111 LFM Sfc -1.39 -9.07
Grauholz G22/2 LFM Sfc -0.56 -7.49
Grauholz GKN105/1 LFM Sfc -0.69 -8.94
Grauholz GKN124/1 LFM Sfc -0.66 -8.45
Gurten 17.80 UMM Smc -0.35 -6.93
Gurten 41.30 UMM Stmi -0.07 -7.78
Gurten 49.65 UMM Stmi -0.49 -7.56
Gurten 73.40 UMM Smc -0.34 -7.49
Gurten 85.55 UMM Smc -0.10 -7.18
Gurten 95.90 UMM Stmi -0.07 -8.37
Gurten 97.50 UMM Smc -0.40 -8.45
Gurten 130.85 UMM Smc -0.15 -7.96
Gurten 139.20 UMM Smc -0.11 -7.97
Gurten 145.60 UMM Smc -0.32 -7.85
Gurten 162.20 UMM Smc -0.07 -8.20
Gurten 183.60 UMM Smc -0.22 -8.46
Gurten 185.75 UMM Smc -0.21 -8.31
Gurten 197.10 UMM Smc -0.32 -7.78
Gurten 229.80 UMM Smc -1.27 -9.82
Gurten 230.40 UMM Smc -0.78 -8.87
Gurten 237.70 UMM Smc -0.57 -8.36

Continued
1 52 J. Mdtyds

Table I ( Continued)

Calcite stable isotopes

Lithostratigraphic Facies
Location Sample unit association o13C (%o PDB) o18Q (%o PDB)

Gurten 262.30 UMM Sci -2.06 -10.01

Murgental 18.10 LFM Sfc -2.46 -10.21

Murgental 19.70 LFM Sfi -2.38 -8.86
Murgental 22.77 LFM Sfi -2.42 -8.21
Murgental 22.80 LFM Sfi -3.07 -8.75
Murgental 30.48 LFM Sfc -2.66 -13.12
Murgental 39.93 LFM Sfc -1.91 -9.01
Murgental 50.86 LFM Sfc -2.72 -8.54
Murgental 59.68 LFM Sfc -3.53 -7.97

Sonnenberg 620.00 LFM Sfc -0.63 -11.50

Sonnenberg 915.00 LFM Sfc -1.21 -10.86
Sonnenberg 1080.00 UMM Sci -0.41 -8.10
Sonnenberg 1139.00 UMM Smc -0.51 -10.17
Sonnenberg 1150.00 UMM Smc 0.08 -8.36
Sonnenberg 1380.00 UMM Smc -0.29 -11.41
Sonnenberg 1405.00 UMM Smc 0.35 -10.58
Sonnenberg 1495.00 UMM Smc 0.43 -9.63
Sonnenberg 1584.00 UMM Smc 0.38 -10.17
Sonnenberg 1615.00 UMM Smc 0.23 -9.99
Sonnenberg 1666.00 UMM Smc -0.25 -13.23

Tiefenbrunnen 331.95 UMM Smc -0.30 -12.05

Tiefenbrunnen 334.40 UMM Smc 0.02 -11.15
Tiefenbrunnen 343.06 UMM Smc 0.45 -9.83
Tiefenbrunnen 347.35 UMM Smc -0.52 -12.62
Tiefenbrunnen 376.25 UMM Mci -0.87 -12.87
Tiefenbrunnen 380.90 UMM Smc -0.40 -11.62
Tiefenbrunnen 382.37 UMM Smc -0.47 -13.43
Tiefenbrunnen 515.80 UMM Stmi -0.86 -10.32
Tiefenbrunnen 516.30 UMM Stmi -1.28 -11.29
Tiefenbrunnen 519.12 UMM Smc -0.49 -9.23
Tiefenbrunnen 589.49 UMM Smc -1.92 -10.84
Tiefenbrunnen 592.92 UMM Mci -1.48 -10.40
Tiefenbrunnen 595.59 UMM Mci -1.67 -10.94
Tiefenbrunnen 597.74 UMM Smc -1.25 -9.64
Tiefenbrunnen 656.63 UMM Smc -0.49 -9.88
Tiefenbrunnen 661.52 UMM Smc -0.48 -9.40
Tiefenbrunnen 662.54 UMM Smc -0.37 -9.17
Tiefenbrunnen 663.55 UMM Smc -0.48 -10.05

LFM, Lower Freshwater Molasse; UMM, Upper Marine Molasse; Mci, mudstone, coastal, isolated; Mfi, mudstone,
freshwater, isolated; Sci, sandstone, coastal, isolated; Sfc, sandstone, freshwater, connected; Sfi, saqdstone, freshwater,
isolated; Smc, sandstone, marine, connected; Stmi, siltstone, marine, isolated.

temperatures range between -0. 1 ·c and -OA·c

Homogenization temperatures in the
Upper Marine Molasse
Two samples from Sonnenberg and two from
Tiefenbrunnen were analysed. In the Sonnenberg
samples, both one- and two-phase inclusions were
observed. One of the two-phase inclusions yielded a
homogenization temperature of 1 2 1 ·c. Although
no textural evidence was found for stretching, this
possibility cannot be excluded. The final ice melting
and are the highest among the measured values.
These temperatures can be converted into a salinity
range of 0.2-0.9% (NaCl equivalent). The final ice
melting temperatures of the Tiefenbrunnen samples
fall between those from Sonnenberg and Murgental,
ranging from - 1 . 8 to -0. 8 ·c, corresponding to
NaCl-equivalent salinities of 1 .4% and 3 . 1 %, re­
spectively. Some of the two-phase inclusions show
evidence of stretching; the two reliable measure-
 Carbonate cements in the Swiss Molasse basin 153

ments yielded homogenization temperatures of (A)

:::::so·c. Lower Freshwater Molasse (LFM)

DISCUSSION

The three major questions to be discussed in the

()
following section are: = ...
Q .
1 Is the isotopic composition of the calcite cements "" • ...
. .
-1 ...
t . .
. . ..
related to the variations observed in modern forma­
� ...
u ...
. ..,. .
tion waters? ...
-00 -2
2 What was the isotopic composition of the forma­
tion waters at the time of calcite precipitation, and
...
. X '·'\ X
X
-3 • X
what can it tell about the fluid flow pattern in
foreland basins? -4+-----�----�---r--r---�
3 Can the textural and geochemical data be used to -14 -12 -10 -8 -6 -4
constrain the origin of these cements? &"0(%. PDB)

• Altishofen • Bassersdnrf e Grauholz X Murgental +Sonnenberg

Relationship between modem pore waters and
diagenetic calcites
(B)
Modern formation waters Upper Marine Molasse (UMM)

2
It is known from hydrochemical (Schmassmann,
1 990) and stable isotopic (Pearson et al., 1 99 1 )
studies that the present-day formation waters in the
Tertiary aquifers can be subdivided into the follow­
ing three major groups: ....
=
•
1 CaMg-bicarbonate waters, representing shallow Q 0 •
"" • ••• � ......
or moderately deep groundwaters originating from
t
• • _., �·..t
recharge under present climatic conditions. The • • ...
•
C!u -I
813C and 8 180 values range from - 1 4.2 to - 1 0.5%o 00
• ""'
•
PDB and from - 1 0. 6 to -8. 8%o SMOW, respectively. •
2 Na-bicarbonate waters, representing infiltrations -2
• ...
during the last or an earlier glaciation. The carbon
isotopic compositions are between -6.4%o and
-.' +---.r---r---.--.,..--�---.
-3.0%o PDB, slightly heavier than those of the -14 -13 -12 -II -HI -9 -8 -7 -6 -5 -4
CaMg-bicarbonate waters. The 8180 values range
o'"o ('Koo PDB)
from - 1 2 . 7 to -ll .6%o SMOW.
3 Na-chloride waters, representing mixtures of Na­ • Altishofen • Gunen +Sonnenberg e Tic:fenbrunnen

bicarbonate waters and Tertiary marine/brackish

pore waters. The carbon isotopic ratios vary be­
tween -8.7%o and -3. 1 %o PDB; no data for oxygen
isotopic composition are available.
Modern waters more depleted in 13C than
calcite cements
As fractionation between dissolved bicarbonate and
the solid calcite phase is relatively minor
(0. 5 1 ± 0.22%o at 25 ·c; Grossmann, 1 984), and
the carbon isotopic ratios of the modern formation
Fig. 10. Cross-plots showing the stable isotopic
compositions of (A) Lower Freshwater Molasse calcites
and (B) Upper Marine Molasse calcites. Note the
distinctly different cluster of some Tiefenbrunnen and
Sonnenberg samples. This is referred to in the text as
high trend.
waters are known, the carbon isotopic ratios of
calcites from different locations can be directly
compared with the carbon isotopic composition of
the waters (Fig. 1 1 ). It is obvious that all present-
1 54 J. Mdtyds

Table 2 Results of fluid inclusion microthermometry

Location Sample Type Host mineral Tmice ("C) Th (" C) Salinity (% NaCl eq)

Murgental 50.86 !ph, NaCl-wat Calcite >-0. 7; <-0.5 n.o. 0.9-1.2

Murgental 50.86 !ph, NaCl-wat Calcite -0.7 n.o. 1.2
Murgental 50.86 I ph, NaCl-wat Calcite >-2.0; <-1.8 n.o. 3.4-3.0
Murgental 50.86 !ph, NaCl-wat Calcite >-2.0; <-1.8 n.o. 3.4-3.0
Murgental 50.86 !ph, NaCl-wat Calcite -1.7 n.o. 2.9
Murgental 50.86 I ph, NaCl-wat Calcite >-1.9; <-1.7 n.o. 2.7-3.2
Murgental 50.86 I ph, NaCl-wat Calcite >-1.8; <-1.6 n.o. 2.7-3.0
Murgental 50.86 !ph, NaCl-wat Calcite -1.0 n.o. 1.7
Murgental 50.86 !ph, NaCl-wat Calcite -1.6 n.o. 2.7
Murgental 50.86 !ph, NaCl-wat Calcite -1.5 n.o. 2.6
Sonnenberg 1584.00 !ph, NaCl-wat Calcite -0.4 n.o. 0.7
Sonnenberg 1584.00 2ph, NaCl-wat Calcite -0.5 121 0.9
Sonnenberg 1666.00 I ph, NaCl-wat Calcite -0.1 n.o. 0.2
Tiefenbrunnen 595.59 I ph, NaCl-wat Calcite -0.9 n.o. 1.6
Tiefenbrunnen 595.59 2ph*, NaCl-wat Calcite -1.2 Stretched 2.1
Tiefenbrunnen 595.59 2ph*, NaCl-wat Calcite -1.1 Stretched 1.9
Tiefenbrunnen 656.63 2ph, NaCl-wat Calcite -1.5 >75; <80 2.6
Tiefenbrunnen 656.63 2ph*, NaCl-wat Calcite -1.5 Stretched 2.6
Tiefenbrunnen 656.63 2ph*, NaCl-wat Calcite -0.8 Stretched 1.4
Tiefenbrunnen 656.63 2ph, NaCl-wat Calcite -1.2 n.o. 2.1
Tiefenbrunnen 656.63 2ph, NaCl-wat Calcite -1.8 n.o. 3.1

I ph, one-phase, all-liquid inclusion; 2ph, two-phase, liquid/vapour inclusion; 2ph*, two-phase, liquid/vapour inclusion
showing evidence for stretching; n.o., not observed; NaCl-wat, sodium chloride-water binary system; Th, temperature
of homogenization into the liquid field; Tm;w temperature of final melting of ice.

day formation waters are more depleted in 13C than

calcite cements, both from the Lower Freshwater
Molasse and from the Upper Marine Molasse. The
only calcite cements with carbon isotopic composi­
tion comparable with the modem formation waters
are from Murgental.
Reasons for o13C shift
The distinct difference between the carbon isotopic
composition of cements and modem formation
waters suggests that either the formation waters
were completely exchanged after precipitation of

LEGEND:
Na-Cl-type 0 Minimum

Na-HC03-type + Maximum

Ca-Mg-HC03-type • Mean

Tiefenbrunnen-1 Upper Marine 0.,. f---< Standard deviation

Molasse
Sonnenberg

Gurten-1 0�
t
Murgental Lower Freshwater o•
Molasse I
Grauholz
�I
Bassersdorf-1
�
0.. 1
Fig. 11. Diagram illustrating the
Altishofen-1 comparison between carbon
isotopic composition of authigenic
-1 5 calcites and the bicarbonates of
-10 -5 0 5
modem formation waters from the
o13C (%o PDB) Swiss Molasse basin.
 Carbonate cements in the Swiss Molasse basin
the carbonate cements, or the isotopic composition
of the dissolved inorganic carbon in the waters was
subjected to significant change by introducing bicar­
bonate depleted in 1 3C. The source of the light
carbon could be: (i) associated with bacterial sul­
phate reduction (Raiswell, 1 987), (ii) oxidation of
bacterially produced methane (Curtis & Coleman,
1 986), (iii) thermal decarboxylation of organic mat­
ter (Irwin et a!., 1 97 7 ), or any combination of these
processes.
Pore water isotopic ratios from calcites
Textural data used to constrain temperature
of precipitation
Because fluid inclusion microthermometry pro­
vided direct temperature constraints only for a
subset of samples, textural data were also used to
constrain the temperature of precipitation of calcite
cements. Pervasively cemented samples probably
preserve their IGV at the time of calcite cementa­
tion. Comparing these IGVs with a general porosity
vs. depth curve for litharenites (e.g. Marco Polo
Software Inc., 1 99 1 ), the depth of precipitation can
be estimated. Knowing the average geothermal
gradient in the area, the temperature corresponding
to this depth can be calculated. The resulting tem­
perature ranges are shown in Table 3. The range of
geothermal gradients used for the calculations was
25.C/km to 35.C/km, which are typical values for
the Swiss Molasse basin (Rybach, 1 992). From
Table 3 it is obvious that the temperature of
precipitation was similar in the Lower Freshwater
Molasse and in most samples of the Upper Marine
Molasse. As no significant difference was found
Table 3 Summary of temperature constraints obtained
from textural data
Temperature (•C)
Formation/location Minimum Maximum
Lower Freshwater Molasse
Altishofen 42 70
Bassersdorf 42 78
Murgental 37 90
Upper Marine Molasse
Sonnenberg (main trend) 44 66
Sonnenberg (low trend) 100 184
Tiefenbrunnen (main trend) 42 88
Tiefenbrunnen (low trend) 43 128
1 55
between the IGV of samples cemented by mosaic
and euhedral calcites, it is assumed that in spite of
the different relative timing, the temperature of
precipitation falls within the ranges given in Table 3
for both textural types. Unfortunately no perva­
sively cemented samples were found in Gurten and
Grauholz; therefore, no temperature constraint is
available for these locations.
The high trend samples from Sonnenberg and
Tiefenbrunnen do not fit into the 40- 80·c precip­
itation temperature range which is typical for all
other calcites. The significance of these samples will
be discussed later.
Fluid inclusion results
In most cases the measured fluid inclusions were
all-liquid, that is, they did not contain a vapour
phase. Upon cooling, a bubble nucleated in one
inclusion from Tiefenbrunnen, on which the total
homogenization temperature was determined. The
fact that the majority of the inclusions did not
nucleate a bubble suggests an entrapment tempera­
ture of < 50·c (Goldstein & Reynolds, 1 994).
Significant metastability and a slightly higher en­
trapment temperature can be assumed for the
Tiefenbrunnen inclusion which showed bubble nu­
cleation. This is consistent with the measured ap­
proximately 8o·c homogenization temperature.
Thus, fluid inclusion analysis predicts formation
temperatures of < 5o·c or slightly higher, which
are comparable with the range of approximately
40-8o·c obtained from textural data.
Pore water isotopic compositions
Figure 1 2 demonstrates the results of pore water
oxygen isotopic composition determination, calcu­
lated using Friedman & O'Neils ( 1 97 7 ) fraction­
'
ation equation for the Upper Marine Molasse
(Fig. 1 2A) and for the Lower Freshwater Molasse
(Fig. 1 2B). The temperature constraints used in the
calculations were those obtained from textural data
and microthermometry.
In the Upper Marine Molasse, clearly distin­
guished ranges of pore water oxygen isotopic com­
positions are suggested for the main and high trend
samples. Among the main trend samples, textural
data from Tiefenbrunnen suggest a range of -5 to
+ 3%o SMOW for 0180waten which can be further
constrained to a range of -5 to + 1 %o SMOW using the
homogenization temperatures from fluid inclusion
1 56 J. Mdtyds

A 200

180

160

140

� 120
...
Q)

B 100
"
...
Q)
c.
80
E
Q)
E-<
60

40

20

0
10 8 6 4 2 0 -2 -4 -6 -8 -10

ol80water (%o SMOW)

B 200

180

160

140

� 120
Q)
....
B 100
"
....
Q)
c.
80 Fig. 12. Diagrams showing the
E pore water isotopic compositions
"'
E-<
60 calculated from the direct and
indirect temperature constraints
40 and the oxygen isotopic
compositions of calcites using the
fractionation equation of
20
Friedmann & O'Neil (1977). As a
result of the uncertain temperature
0
10 8 6 4 2 0 -2 -4 -6 -8 -10 constraints for the UMM high
trend samples; their composition
ol80water (%o SMOW) ranges are probably unrealistic.

microthermometry. This is in good agreement with
the data obtained for the Sonnenberg main trend
calcites (-3 to + 1 %o sMow) using textural tempera­
ture constraints.
Considering the substantial uncertainties in­
volved in the temperature estimates for the Upper
Marine Molasse high trend samples, the obtained
ranges of - 1 to + 1 O%o SMOW and -8 to +6%o SMOW
for 8180water in Sonnenberg and Tiefenbrunnen are
possibly unrealistic. The single sample with homog-
enization temperature suggests a 8180water value of
+ 5%o SMOW.
In the Lower Freshwater Molasse almost the
same range was obtained for 0180water from the
calcites in Bassersdorf and Altishofen (-6 to + 1 %o
SMOW, and -5.5 to +2%o SMOW, respectively). Tex­
tural data predict a slightly wider range (-9 to +4%o
sMow) for the calcites from Murgental. The pres­
ence of all-liquid fluid inclusions, however, sets the
maximum temperature of formation at 50'C,
 Carbonate cements in the Swiss Molasse basin
which would suggest a range of -9 to - 1 o/oo SMOW for
0 1 80water·
In summary; within the resolution of the applied
temperature constraints, the oxygen isotopic com­
position of pore waters was broadly the same at the
time of calcite cementation in both the Upper
Marine Molasse and the Lower Freshwater Mo­
lasse.
History of formation waters: evidence for
early mixing
Sedimentological variations and diagenetic history
In spite of the very different depositional environ­
ments, little difference was found between the
diagenetic evolution of the Upper Marine Molasse
and that of the Lower Freshwater Molasse. The
following similarities can be pointed out:
I Most diagenetic minerals �re the same in the
sandstones of the Lower Freshwater Molasse and
the Upper Marine Molasse, apd there is no signifi­
cant difference between the paragenesis of the two
units. Minor variations are detected in the diage­
netic history preceding the first calcite generation.
2 With the exception of the high trend samples of
the Upper Marine Molasse, stable isotopic compo­
sitions of carbonate cements are comparable, as are
the estimated oxygen isotopic ratios of the pore
waters.
3 Fluid inclusion microthermometry indicates the
presence of moderately saline waters in the calcite
cements of both units.
These phenomena strongly suggest that the chem­
ical and isotopic composition of formation waters
was broadly similar during and after the formation
of early calcites. These formation waters were prob­
ably mixtures of the original marine and fresh pore
waters.
One possible approach to validate the mixing
hypothesis is to show all samples together, distin­
guished by facies, on a o 1 80 vs. o 1 3C cross-plot. If
mixing of the initial formation waters really oc­
curred, the expected pattern is that o 1 80 vs. o 1 3C
values of clean, well-connected sandstones would
correlate, forming a trend which joins the marine
and freshwater end-members represented by the
calcites of isolated sandstone bodies.
Such a plot is shown in Fig. 1 3 . On the plot seven
major facies associations are distinguished, based
on their lithology, origin and interconnectedness:
Sfc (sandstone, freshwater connected) and Smc
1 57
•
• • ••
• •
0
• • Sm�: (UMM)
• Stmi (UMM)
& St:i (UMM)
· 2
e Mci (UMM)
0 o Sf�: (LFM)
·3 0 <> Sti (LFM)
0 O Mti (LFM)
-14 -13 -12 ·II · I ll -9 -8 -7 6
· -5 ·4
8180 (%o PDB)
Fig. 13. Cross-plot showing the relationship between
15 1 3C and 15 1 80 of calcites from different facies
associations in the Lower Freshwater Molasse and
Upper Marine Molasse. See the text for an explanation
of the facies associations.
(sandstone, marine, connected) represent the well­
connected, clean sands of meander belts and cre­
vasse channels of the Lower Freshwater Molasse,
and of the tidal channels, surf and breaker zones,
sandbanks, sandwaves and ripcurrent channels of
the Upper Marine Molasse, respectively. Sfi (sand­
stone, freshwater, isolated) and Mfi (mudstone,
freshwater, isolated) represent the isolated sand
bodies of levees and crevasse splays, and the mud­
-stones and fine-grained sandstones of the overbank
sediments of the Lower Freshwater Molasse. Stmi
(siltstone, marine, isolated) represents the alternat­
ing siltstone/fine-grained sandstone sequences of
sheltered bays, mixed flats and point bars of the
Upper Marine Molasse. Sci (sandstone, coastal,
isolated) and Mci (mudstone, coa�tal isolated) re­
present the subareally exposed, poorly connected,
fine-grained sandstones and mudstones of washover
fans, mudflats and slackwater bays of the Upper
Marine Molasse, in which mixing of marine or fresh
water could occur during or immediately after
deposition. From Fig. 1 3 it is obvious that:
I Clusters of clean, connected sands of the Upper
Marine Molasse and Lower Freshwater Molasse
within the main trend samples overlap, and the
o 1 80 vs. o 1 3C values correlate; this supports the
mixing model.
2 Among the main trend points Sfc calcites are
generally more depleted in 1 3C and 1 80 than Smc
158 J Mdtyds
calcites, suggesting that in spite of the mixing the
proportion of fresh water was generally higher in the
Lower Freshwater Molasse sandstones than in those
of the Upper Marine Molasse. The minor but
systematic differences between the compositions of
pore-lining clays can also be related to the different
proportion of waters in the mixture.
3 Not all of the isolated sandstones plot at the low
or high ends of the main trend, which indicates that
mixing was possibly not restricted to the well­
connected sandstones.
4 Some of the samples from Mci and Sci facies
associations plot together with the fresh water
end-members, suggesting a pore water exchange
during or immediately after deposition.
5 A group of points representing Sfc and Sfi sam­
ples form an isolated cluster, which is distinguished
from the main trend by its more negative 1 3C ratios.
If Figs 1 0 and 1 3 are compared it is seen that these
points correspond to Murgental and Tiefenbrunnen
samples which are rich in isotopically light, authi­
genic pyrite. The presence of this pyrite indicates
bacterial sulphate reduction, which can account for
the isotopically light, organic carbon which is the
most probable explanation for negative shift of
these points.
6 The high trend samples are almost exclusively
clean, connected sands from the Upper Marine
Molasse.
Timing and mechanism of mixing
Although determination of timing and the mecha­
nism of mixing was beyond the scope of this study,
the following constraints can be applied:
1 Mixing appears to be a basinwide phenomenon,
and not restricted to certain depth intervals, forma­
tions or areas. This suggests that the mechanism of
mixing was related to compactional dewatering
during subsidence. However, it is not clearly under­
stood how the marine formation waters of the Upper
Marine Molasse could reach the underlying Lower
Freshwater Molasse: compactional waters flow
typically upwards or laterally, parallel to the deposi­
tional boundaries, but not downward (Berner,
1 980). A possible solution of this controversy is that
a complex interaction could have developed be­
tween the compactional flow regime and other flow
systems, whose nature is so far unknown.
2 The age of the Upper Marine Molasse sets the
upper limit of the time of the mixing process at
approximately 1 7 Ma.
3 Accepting that the mixing is related to compac­
tional dewatering (which requires active subsid­
ence), the lower limit of the mixing is set at
approximately 1 3 Ma by the age of the top of the
Upper Freshwater Molasse (see Fig. 2), which closes
the second megasequence in the basin.
Origin of calcite cements
There is little doubt that in a basin with such a
complex depositional and tectonic history as the
Swiss Molasse basin, several processes could ac­
count for calcite cementation. Based on samples
which-according to their petrographic and isotopic
data-are dominated by one specific type of calcite,
the following types can be recognized:
1 Euhedral calcites from alteration of volcanic
grains. The overall presence of euhedral calcite
suggests that the source of calcium was probably
internal. This calcite is closely associated with
pore-lining smectites or chlorite/smectites. These
pore-lining, authigenic smectites and smectite-rich
chlorite/smectite mixed-layer clays are typical
by-products of hydration of volcanic detritus
(Robinson & Bevins, 1 994). Although there are
2
several potential sources for the Ca + cations in the
studied sandstones (feldspar albitization, biogenic
carbonate, detrital dolomite), only the alteration of
volcanic rock fragments (Morad & De Ros, 1 994)
can explain both the overall presence of the calcite
and the coupled occurrence with the mixed-layer
clay minerals. The volcanic rock fragments which
are common, though not particularly abundant,
components in the studied sandstones show evi­
dence of extensive alteration (dissolution or re­
placement by calcite), which is recognized as a
common mode of decomposition of volcanogenic
detritus (Maim et a/., 1 984; Sturesson, 1 992).
2 Mosaic calcites related to qrganic C0 2 • Unlike
euhedral cements, which are present only in minor
amounts, mosaic calcites have a major influence on
the hydraulic properties of the studied sandstones.
One group of sandstones influenced by the mosaic
cements (those from Murgental, Altishofen and
Bassersdorf from the Lower Freshwater Molasse,
and a few samples from Gurten from the Upper
Marine Molasse) includes those samples which
have the most negative carbon isotopic ratio. These
occur together with abundant authigenic pyrite,
revealing isotopically light sulphur composition
(o 34 S ranges from - 1 4.4 to + 1 . 1 o/oo coT) typical for
pyrite derived from bacterial sulphate reduction
 Carbonate cements in the Swiss Molasse basin
(Coleman & Raiswell, 1 9 8 1 ), or they are intimately
associated with the organic-rich shales of floodplain
fine clastics and lacustrine deposits. The low o 1 3C
values (-3.5 to -2.0o/oo PDB) of these cements sug­
gest input of organically derived C0 2 . This excess
carbon dioxide possibly mobilized the detrital car­
bonates or bioclasts in the shales or in the sand­
stones, resulting in locally pervasive calcite
cementation.
3 Mosaic calcites from redistribution of biogenic
carbonate material. Early pervasive calcite cement
may occur in Tiefenbrunnen and in Sonnenberg in
beds of marine or brackish origin, which are ex­
tremely rich in mollusc shells, some of them show­
ing evidence of leaching. The fact that unstable
biogenic carbonate material can dissolve and pre­
cipitate, forming carbonate concretions and ce­
mented layers in sandstones, is well documented
(Bj0rkum & Walderhaug, 1 990, 1 993), and this
redistribution process is the likely explanation for
the extensive calcite cementation in these samples.
Calcite cementation from evolved
formation waters?
The formation of mosaic cements in some of the
Tiefenbrunnen and Sonnenberg samples (referred
to as high trend samples) cannot be explained by the
processes discussed above. These samples have
some peculiarities:
1 They represent an isolated cluster of data points
on a o 1 8 0 vs. o 1 3C plot and reveal slightly heavier
carbon and lighter oxygen isotopic compositions
than the other marine calcites.
2 Fluid inclusion microthermometry data from the
Sonnenberg samples indicate very low, practically
freshwater, salinities (Table 2).
3 The oxygen isotopic composition of the pore
water was substantially heavier than of those ob­
tained for the main trend cements.
All these facts indicate that these cements precip­
itated from pore fluids whose chemical and isotopic
properties were clearly different from those of other
calcites. There are two plausible explanations for
these observed phenomena.
1 The high trend cements could have precipitated
from evolved formation waters through in situ re­
crystallization of early calcites at high temperature
during deep burial. This process would explain the
presence of isotopically heavy oxygen, which is not
uncommon in evolved deep burial waters (Long­
staffe, 1 994), and-because of possible interaction
1 59
with the isotopically heavy detrital dolomite-the
positive carbon isotopic ratios.
2 These cements could have precipitated from
evolved formation waters flowing towards the north
in confined aquifers, driven by tectonic loading due
to overthrusting of the Alpine nappes in the south.
The facts that high trend samples are mostly from
clean, well-connected sandstones and that they were
found in the vicinity of highly conductive cataclas­
tic zones (Sonnenberg) or conglomerate beds
(Tiefenbrunnen) would support this hypothesis.
However, the available data are insufficient to allow
a final conclusion on this point.
Relationship between calcite genesis and stable
isotopic composition
Figure 1 4 summarizes the possible interpretations
of the three main clusters recognized on the o 1 80
vs. o 1 3 C plot in the light of calcite genesis. The
cluster of main trend samples includes calcites from
both formations: as it is possible for several types of
calcite to be present in many samples, subdomains
of different genetic types cannot be identified within
this trend. Calcites influenced by isotopically light
carbon form another isolated cluster. In these sam­
ples the diagenetic carbonates probably formed by
dissolution/precipitation reactions related to pres­
ence of organic C0 2 • The third cluster includes the
calcites that precipitated from evolved formation
waters.
C O N C LUSI O N S
Carbonate cements, mostly calcites, are volumetri­
cally the most important authigenic minerals in
sandstones of the Lower Freshwater Molasse and
Upper Marine Molasse. Authigenis; clays are locally
significant; other authigenic minerals occur only in
minor or trace amounts.
Textural evidence and fluid inclusion microther­
mometry suggest that most calcites formed at low
temperatures, probably at around 50 " C. The oxy­
gen isotopic composition of formation waters
ranged from -9 to +2o/oo SMOW in the Lower Fresh­
water Molasse and from -5 to + I o/oo SMOW in the
Upper Marine Molasse. Final ice melting tempera­
tures suggest the presence of moderately saline
waters in both formations.
The similarity of pore water salinities and oxygen
isotopic compositions in the two formations sug-
1 60 J.
co -I
0
0..
0
� -2
�u
vo
-3
-4
-14 -13 -12 -II -10 -9
gests mixing of marine and fresh waters prior to or
coeval with calcite cementation. This mixing seems
to be a basinwide phenomenon, and is not restricted
to the well-connected sandstones. The correlation
between the carbon and oxygen isotopic composi­
tions of calcites is consistent with this explanation.
The mixing is most likely related to compactional
flow during subsidence.
There are at least three genetic types of calcite
present in the studied sandstones, but not all of
them are differentiated by texture or stable isotopic
composition. The presence of deep-burial, late cal­
cites precipitated from evolved pore waters is likely,
but these cements are restricted to specific intervals.
The volumetrically most important cements in
the Swiss Molasse basin formed early, and their
isotopic characteristics reflect the fluid flow history
during subsidence. Furthermore, carbon in the cal­
cite cements is isotopically heavier than that in the
bicarbonate of the modern formation waters, and
the variation in modern formation waters is not
reflected in the diagenetic mineral assemblage.
The above facts suggest that porosity modifica­
tion by cementation is much more significant dur­
ing subsidence than during uplift in the Swiss
Molasse basin, therefore conclusions based on dis­
tribution and chemistry of modern formation wa­
ters concerning diagenetic overprinting in other
inverted foreland basins may have to be handled
with some care.
Mdtyds
-8 -7 -6 Fig. 14. Diagram explaining the
three main clusters recognized on
the 8 1 3C and 8180 cross-plot.
A C K N O W L E D G E M E N TS
This project was funded by Swiss NSF grant No.
20-37663.93. The author is grateful to Professor
Albert Matter, who initiated this project, and to Drs
Karl Ramseyer and Stephen Burns for the fruitful
discussions and comments on the earlier versions of
the manuscript. The thorough reviews of the two
lAS reviewers, Professor James Boles and Dr Olav
Walderhaug, and the lAS editor Dr Sadoon Morad,
are gratefully acknowledged.
REFERENCES
BERNER, R.A. ( 1 9 80) Early Diagenesis. Princeton Univer­
sity Press, Princeton, 241 pp.
Bi0RKUM, P.A. & WALDERHAUG, 0. ( 1 9 90) Geometrical
arrangement of calcite cementation within shallow ma­
rine sandstones. Earth Sci. Rev., 29, 145-161 .
BJ0RKUM, P.A. & WALDERHAUG, 0. ( 1 993) Isotopic com­
position of a calcite-cemented layer in the Lower J uras­
sic Bridport Sands, southern England; implications for
formation of laterally extensive calcite-cemented layers.
J. sediment. Petrol. , 63, 67 8-682.
BODNAR, R.J. ( 1992). Revised equation and table for
freezing point depressions of H20-salt fluid inclusions.
IPACROFI IV. Fourth Biennial Pan-American Congress
on Research on Fluid Inclusions, 4. Lake Arrowhead,
CA, 108-1 1 1 .
COLEMAN, M.L. & RAISWELL, R. ( 1 9 8 1 ) Carbon, oxygen
and sulphur isotopic variations in concretions from the
Upper Lias of N.E. England. Geochim. Cosmochim.
Acta, 45, 329-340.
 Carbonate cements in the Swiss Molasse basin
CURTIS, C.D. & COLEMAN, M.L. ( 1 986) Controls on the
precipitation of early diagenetic calcite, dolomite and
siderite concretions in complex depositional sequences.
In: Roles of Organic Matter in Sediment Diagenesis (Ed.
Gautier, D.L.). Soc. Econ. Paleontol. Mineral. Spec.
Publ. Tulsa, 38 , 23-33.
DICKSON, J.A.D. ( 1 966) Carbonate identification and gen­
esis as revealed by staining. J. sediment. Petrol. , 36,
49 1 -505.
DIEM, B. ( 1 986) Die Untere Meeresmolasse zwischen der
Saane (Westschweiz) und der Ammer (Oberbayern).
Eclog. geol. Helv. , 79, 493-559.
EMERY, D. & ROBINSON, A. ( 1 993) Inorganic Geochemistry:
Application to Petroleum Geology. Blackwell, Oxford,
254 pp.
FALLICK, A.E., McCONVILLE, P., BOYCE, A.J., BURGESS, R.
& KELLEY, S.P. ( 1 992) Laser microprobe stable isotope
measurements on geological materials: some experi­
mental considerations (with special reference to o34S
sulphides). Chern. Geol. , 101, 5 3-6 1 .
FRIEDMAN, I . & O NEIL, J.R. ( 1 977) Compilation ofStable
' Oxford University Press, Oxford, 332 pp.
Isotope Fractionation Factors of Geochemical Interest.
US Geol. Surv. Professional Paper, 440-KK.
GOLDSTEIN, R.H. & REYNOLDS, T.J. ( 1 994) Systematics of
Fluid Inclusions in Diagenetic Minerals. SEPM Short
Course Notes No. 3 1 , Tulsa, 1 99 pp.
GROSSMAN, E.L. ( 1 984) Carbon isotopic fractionation in
living benthic foraminifera-comparison with inorganic
precipitate studies. Geochim. Cosmochim. Acta 48,
1 505- 1 5 1 2 .
HoMEWOOD, P . ( 1 98 1 ) Facies e t environnements d e depot
de La Molasse de Fribourg. Eclog. geol. Helv. , 74, 29-36.
HOMEWOOD, P., ALLEN, P.A. & WILLIAMS, G.D. ( 1 986)
Dynamics of the Molasse basin of western Switzerland.
In: Foreland Basins (Eds Allen, P.A. & Homewood, P.).
Spec. Publ. Int. Ass. Sedimentol., 8, 1 99-2 1 7.
IRWIN, H., CURTIS, C. & COLEMAN, M. ( 1 977) Isotopic
evidence for source diagenetic carbonates formed dur­
ing burial of organic-rich sediments. Nature, 269, 209-
2 1 3.
KELLEY, S.P. & FALLICK, A.E. ( 1 990) High precision
spatially resolved analysis of o34S in sulphides using
laser extraction technique. Geochim. Cosmochim. Acta,
54, 883-888.
LONGSTAFFE, F.J. ( 1 994) Stable isotopic constraints on
sandstone diagenesis in the Western Canada Sedimen­
tary Basin. In: Quantitative Diagenesis: Recent Develop­
ments and Applications to Reservoir Geology (Eds.
Parker, A. & Sellwood, B.W.), pp. 223-274. Kluwer
Academic, Dordrecht.
MALM, O.A., BRUNN-CHRISTENSEN, 0., FuRNES, H. et a!.
( 1 984) The lower Tertiary Balder Formation: an orga­
nogenic and tuffaceous deposit in the North Sea region.
In: Petroleum Geology of the North European Margin
(Ed. Spencer A.M.), pp. 1 49- 1 70. Graham & Trotman,
London.
MARCO PoLO SOFTWARE INC. ( 1 99 1 ) Basin Works™ Man­
ual. 42 pp.
MATTER, A. & RAMSEYER, K. ( 1 985) Cathodoluminescence
microscopy as a tool for provenance studies of sand­
stones. In: Provenance of Arenites (Ed. Zuffa, G.G.),
pp. 1 9 1 -2 1 1 . Riedel, Dordrecht.
MATTER, A., HOMEWOOD, P., CARON, C. et a/. ( 1 980) Flysch
161
and Molasse of Western and Central Switzerland. In:
Geology ofSwitzerland: a Guide Book (Ed. Tnlmpy, R.),
pp. 6 1 -293. Wepf, New York.
MATTER, A., PETERS, T. , BLAESI, H.-R. et a!. ( 1 988) Sondi­
erbohrung Weiach. NAGRA Technischer Bericht No.
NTB 86-0 I . Nationale Genossenschaft fiir die Lagerung
radioaktiver Abfalle, Baden.
MATYAs, J. & MATTER, A. (in preparation) Depositional
and diagenetic control of porosity and permeability in
the Tertiary sandstones of the Swiss Molasse Basin.
McBRIDE, E.F. ( 1 963) A classification of common sand­
stones. J. sediment. Petrol. , 33, 664-669.
MONNIER, F. ( 1 979) Correlations mineralogiques et dia­
genese dans Ia Bassin Molassique Suisse. PhD Thesis,
Universite de Neuchatel, 1 43 pp.
MONNIER, F. ( 1 982) Thermal diagenesis in the Swiss
Molasse basin: implications for oil generation. Can. J.
Earth Sci. , 19, 328-342.
MOORE, D.M. & REYNOLDS, R.C. ( 1 989) X-ray Diffraction
and the Identification and Analysis of Clay Minerals.
MORAD, S. & DE Ros, L.F. ( 1 994) Geochemistry and
diagenesis of stratabound calcite cement layers within
the Rannoch Formation of the Brent Group, Murchison
Field, north Viking Graben (northern North Sea)­
comment. Sediment. Geol. , 93, 1 35- 1 4 1 .
PEARSON, F.J., BALDERER, W., LOOSLI, H.H. et a/. ( 1 99 1 )
Applied Isotope Hydrogeology: a Case Study in Northern
Switzerland. Elsevier, Amsterdam, 439 pp.
PFIFFNER, O.A. ( 1 986) Evolution of the north Alpine
foreland basin in the Central Alps. In: Foreland Basins
(Eds Allen, P.A. & Homewood, P.). Spec. Pub!., Int. Ass.
Sedimentol., 8, 2 1 9-228.
PFIFFNER, O.A., ERARD, P.F. & STAUBLE, M. ( 1 997) Two
cross sections through the Swiss Molasse Basin (Lines
E4-E6, W I , W7-W I O). In: Deep Structure of the Alps,
Results ofNRP20 (Eds Pfiffner, O.A., Lehner, P., Heitz­
mann, P., Mueller, St. & Steck, A.), pp. 64-72.
Birkhauser Verlag, Basel.
PLATT, N.H. & KELLER, B. ( 1 992) Distal alluvial deposits
in a foreland basin setting-Lower Freshwater Molasse
(Lower Miocene), Switzerland: sedimentology, architec­
ture and palaeosols. Sedimentology, 39, 545-565.
RAISWELL, R. ( 1 987) Non-steady state microbiological
diagenesis and the origin of concretions and nodular
limestones. In: Diagenesis of Sedimentary Sequences
(Ed. Marshall, J.D.). Spec. Publ. Geol. Soc. Lond., 36,
•
4 1 -54.
ROBINSON, D. & BEVINS, R.E. ( 1 994) Mafic phyllosilicates
in low-grade metabasites. Characterisation using decon­
volution analysis. Clay Miner. , 29, 223-237.
RYBACH, L. ( 1 992) Geothermal potential of the Swiss
Molasse Basin. Eclog. geol. Helv. , 85, 7 3 3-7 44.
ScHEGG, R. ( 1 992) Coalification, shale diagenesis and
thermal modelling in the Alpine Foreland basin: the
Western Molasse Basin (Switzerland/France). Org.
Geochem., 18, 289-300.
SCHEGG, R. ( 1 993) Thermal Maturity and History of
Sediments in the North Alpine Foreland Basin (Switzer­
land, France). Pub!. Dep. Geol. Paleontol. Univ.
Geneve 1 5, l - 1 94.
SCHEGG, R. ( 1 994) The coalification profile of the well
Weggis (Subalpine Molasse, Central Switzerland): im-
1 62 J. Mdtyds

plications for erosion estimates and the paleogeother­
mal regime in the external part of the Alps. Bull.
Schweiz. Verein. Petrol. Geol. Ing. , 61, 5 7-67.
SCHLUNEGGER, F., MATTER, A., BURBANK, D.W. & KLAPER,
E.M. ( 1 997) Magnetostratigraphic constraints on rela­
tionships between evolution of the central Swiss Mo­
lasse Basin and Alpine orogenic events. Geol. Soc. Am.
Bull. 709, 225-244.
SCHMASSMANN, H. ( 1 990) Hydrochemische Synthese Nord­
schweiz: Tertiar- und Maim-Aquifere. NAGRA Technis­
cher Bericht No. NTB 88-07. Nationale Genossenschaft
fiir die Lagerung radioaktiver Abfalle, Baden.
STURESSON, U. ( 1 992) Volcanic ash: the source material for
Ordovician chamosite ooids in Sweden. J. sediment.
Petrol., 62, 1 084- 1 094.
Spec. Pubis int. Ass. Sediment. (1998) 26, 163-1 77

Carbonate cement in the Triassic Chaunoy Formation of the Paris Basin:

distribution and effect on flow properties

R.H. W O R D E N* and J.M. MAT RAYt

*School of Geosciences, The Queen's University of Belfast, Belfast BT7 INN, UK,
e-mail r. worden@queens-bela f st.ac.uk; and
t Bureau Recherche Geologie et Mineralogie, DRIHGT, Orleans cedex 2, France

ABSTRACT

The distribution of mineral cements in oilfields is critical to the spatial variation of porosity and
permeability. The authors have studied the distribution of dolomite cement within fluvial Triassic
Chaunoy sandstones in the Paris Basin using core description, petrography, core analysis (porosity and
permeability) and wireline data interpreted to give mineralogy, porosity and permeability. Petro­
graphic analysis revealed that dolomite and quartz cements are the main diagenetic minerals. Using
sonic transit time, density and neutron density logs we have been able to resolve the overall proportions
of quartz, dolomite and shale, as well as porosity for each depth interval. Petrographic and core analysis
data showed that permeability could be calculated from wireline-derived porosity and mineralogy data.
There is excellent correlation between core analysis porosity and permeability and their wireline­
derived equivalents. There is also excellent correlation between wireline-derived mineralogy data and
quantitative petrographic mineralogy data. The wireline-derived mineralogy data show that dolomite
is preferentially concentrated at the tops of most sandbodies. Porosity and permeability are
consequently lowest at the tops of individual sandbodies, owing to the localized dolomite cement.
There are a number of potential causes for this distribution pattern, although a combination of early
pedogenetic dolomite cementation and later recrystallization, possibly due to an influx of organically
derived C02, is most likely.

INTRODUCTION

Knowledge of the way in which porosity and per­
meability are distributed throughout an oilfield is
an important building block in a reservoir model.
The key factors controlling porosity and permeabil­
ity in sandstones are depositional characteristics
such as grain size and sorting, and diagenetic
features such as cements and secondary porosity.
Most reservoir simulation models incorporate sub­
units of common primary sedimentary origin. The
distribution of reservoir quality is thus usually
defined in terms of the morphology of the sedimen­
tary architecture. However, reservoir rocks seldom
retain their depositional porosity. Instead, porosity
is usually degraded by a variety of diagenetic pro­
cesses, the effects of which are not necessarily
confined to the boundaries of depositional sedimen­
tary units. Common diagenetic processes either
may transcend sedimentary architecture or may
lead to the subdivision of self-contained sedimen­
tary units in terms of porosity and permeability.
There is no framework for predicting diagenetic
cement distribution in sandstones on the reservoir
scale. It is not yet generally possib)e to predict or
model reservoir porosity and permeability varia­
tions over the distribution of the primary sedimen­
tary units. This is clearly unsatisfactory and may
lead to systematically incorrect reservoir models.
One of the key problems in describing the distri­
bution of cement is the cost (in terms of time and
money) of acquiring the data. Petrographic data are
usually collected at a far lower density than core
analysis data (if at all), are harder to quality-control
and are highly operator-dependent. In this paper we
describe a way to assess carbonate cement distribu­
tion in sandstones using petrophysical logs (here­
after known as wireline logs). We use this method to

Carbonate Cementation in Sandstones Edited by Sadoon Morad 163

© 1998 The International Association of Sedimentologists
ISBN: 978-0-632-04777-2
164 R.H. Worden and J.M. Matray
describe the distribution of dolomite cement in
Triassic fluvial clastic sediments of the Chaunoy
Formation in the Paris Basin, France. We address
the controls on dolomite cement distribution, de­
fine the effects of dolomite cement (and, by infer­
ence, quartz cement) on reservoir flow properties,
and then explore possible mechanisms that con­
trolled the carbonate cement distribution pattern.
GEOLOGICAL SETTING
The Paris Basin is an intracratonic basin with an
areal extent of approximately 6000 km2 and about
3000 m of sedimentary infill deposited on Hercyn­
ian basement (Fig. I; Pommerol, 1974, 1978).
There are two main permeable petroleum-bearing
reservoir units in the central part of the Mesozoic of
the Paris Basin: the Late Triassic (Keuper) fluvial
sandstones and the Middle Jurassic marine carbon­
ates (Pages, 1987). The Paris Basin experienced a
simple subsidence history that included periods of
relatively rapid burial. Rifting started in the Trias­
sic, followed by thermal subsidence in the Jurassic
and Cretaceous (Pommerol, 1978; Brunet & Le
Pichon, 1982; Loup & Wildi, 1994; Megnien,
1980a,b). Maximum burial in the central part of the
basin occurred during the Oligocene-Miocene and
was followed by minor uplift during and following
Alpine and Pyrenean tectonism (Megnien, 1980a,b;
Brunet & Le Pichon, 1982; Pages, 1987). Triassic
sediments in the central part of the basin reached
Fig. 1.Geological map of the Paris Basin with the
approximate extent of the Triassic sandstones. The well
sampled for core (L) is marked.
maximum burial depths of about 3000-4000 m.
Sandwiched between the Triassic sandstones and the
Mid-Jurassic carbonates are organic-rich Liassic
shales. These are mature to the point of oil genera­
tion and expulsion at the base of the Lias, in the
centre of the basin (Herron & Le Tendre, 1990). This
source rock reached maturity at the time of maxi­
mum burial and charged both Triassic and Mid­
Jurassic reservoirs with oil (Poulet & Espitalie,
1987).
The Triassic sandstones are composed of several
reservoir units. The Late Carnian to Norian
Chaunoy Formation has limited areal extent, lies in
the deepest part of the basin, slightly to the west of
the basin centre, and has no outcrop (Fig. I)
(Bourquin & Guillocheau, 1993; Bourquin et a!.,
1993; Fontes & Matray, 1993; Matray et a!., 1993).
The Chaunoy was deposited as a minor trans­
gressive-regressive cycle within an overall trarts­
gressive phase that ended with Rhaetic marine
sediments (Bourquin & Guillocheau, 1993). It is
composed of alluvial fan conglomerates, coarse­
grained channel-fill fluvial sandstones and flood­
basin siltstones. It was deposited in an arid environ­
ment as an alluvial and fluvial fringe to the western
rifted margin of the basin (Bourquin & Guillocheau,
1993; Bourquin et a!., 1993). Locally important
pedogenic arid phreatic dolomite cements are found
within the Chaunoy Formation (Spot! & Wright,
1992). Burial diagenesis resulted in the precipitation
of abundant quartz and dolomite, and less common
calcite and saddle dolomite cements (Demars &
Pagel, 1994; Worden & Matray, 1995).
Previous diagenetic studies of the Chaunoy
Formation showed that quartz cement grew at
temperatures a little lower than those attained at
maximum burial, whereas sparry, rhombic dolo­
mite cement grew at maximum burial (Demars &
Pagel, 1994). The pedogenic dqlocrete has a limited
range of 813C values (-7 to Oo/oo) (Spot! & Wright,
1992), although the later burial dolomite cements
had significantly more isotopically-depleted carbon
(as low as -14o/oo) (Worden & Matray, 1995).
METHODS
Core description and petrography
Slabbed core from the well was examined for
general lithology, facies variations, sedimentary
structures and grain size. The grain size of the core
 Carbonate cement in the Triassic Paris Basin 165

was measured at regular intervals by comparing Table I. Definition of terms and units used in equations
core with standard grain size charts under a binoc­ (1)-(4)
ular microscope. Petrographic analysis was per­ Term Definition
formed on 22 thin sections stained for carbonates
and feldspars and impregnated with blue-dyed ep­ Sonic transit time recorded by log (!!sift)
oxy resin. Grain sizes and sorting class were as­ Sonic transit time of mineral X (!!sift)
Sonic transit time of fluid in pore space (!J-slft)
sessed quantitatively in thin section by measuring Density recorded by log (glcm3)
the sizes of 100 grains per section. Detrital grains, O'minX Density of mineral X (glcm3)
cements and porosity were quantified by point ()'<I> Density of fluid in pore space (glcm3)
counting using 300 grain counts per section. <f>n Neutron density recorded by log (porosity units)
<l>nminX Neutron density of mineral X (porosity units)
<f>n<l> Neutron density of fluid in pore space (porosity
Petrophysical (wireline and core analysis) data units)
minX Proportion of mineral X (as fraction of total
Porosity and permeability core analysis data for the rock volume)
sampled well were made available to the authors by Porosity (as fraction of total rock volume)
Elf (99 data points from the interval under investi­
gation). Core porosity data have an uncertainty of
less than 0.5%, which arises from the variable signal at any given depth to solid grain volume
amount of stress relaxation following withdrawal of (occupied by the three minerals) and the assump­
the core from the subsurface. Analytical errors are tion that the sum of the three mineral fractions plus
insignificant. Sonic transit time, neutron density, porosity equals unity. This also assumes linear
density and other wireline data recorded at 5 em relationships between mineral proportions and
intervals by petrophysical logging methods were their contribution to the petrophysical signal. Thus,
also made available by Elf. These data were used to with four equations and four unknowns (propor­
derive porosity and mineral proportions using tions of three minerals plus porosity), the following
methods outlined by Doveton (1994) and Hearst & algorithms can be solved simultaneously at each
Nelson (1985). depth interval:
The gamma log is commonly used to define the
!t,t =min l .!t.tmin 1 min2.!t.tmin2
+
'shaliness', ( vshale) of sandstones, although this
min3.!t.tmin3 + !t.t¢>
+ (1)
approach is invalid for simple crystal chemical
(J =min l . <Jmin 1 + min2. <Jmin2 + min3. <Jmin3 + (J¢> (2)
reasons. Composite gamma logs record the total
potassium, thorium and uranium contents of the <l>n =min l .<l>nmin 1 + min2.<l>nmin2
+ min3.<l>nminJ + <l>n<t> (3)
rock; spectral gamma logs differentiate the gamma
1 =min1 + min2 + min3 + <1> (4)
radiation from the three elements. However, using
either log for a shale estimate is invalid. Potassium The terms used in the equations above are defined
is commonly held in K-feldspar, illite or micas. in Table 1. The ideal petrophysical responses of
Most other clay minerals do not contain potassium. each mineral were taken from Rider (1986) and
Thus the potassium gamma signal records the rela­ modified slightly according to the distribution of
tive abundance of K-feldspar illite and mica indis­ data on !t.t, o and <l>n cross-plots (fable 2).
criminately, and does not record the shale volume.
The thorium gamma signal, often mistakenly
thought to reflect specific clay minerals, records the Table 2. Petrophysical response characteristics of quartz,
dolomite, shale and the pore fluid as used to calculate
abundance of thorium-bearing trace minerals and
the mineralogy from neutron density, sonic transit time
cannot be used to estimate volumes of clay minerals and density logs
(Hurst & Milodowski, 1996). Consequently, we
have used a multiple log-transformation approach Neutron
density Sonic transit Density
to derive the shale content as well as the dolomite
Rock unit (p.u.) time (�J-slft) (glcm3)
and quartz contents.
The signals from the sonic transit time, neutron Quartz -0.04 51.00 2.66
density and density logs can be integrated and Shale 0.23 82.00 2.75
Dolomite 0.04 43.50 2.88
resolved for three mineral types and total porosity
Pore fluid 1.00 1 89.00 0.95
using three algorithms relating each separate log
166 R.H. Worden and JM. Matray

RESULTS in appearance, very well lithified, and shows abun­

Core description and petrography
Grain size data are shown in Fig. 2. Most of the core
is either fine- (silt/mud, grain size < 62 J..Lm) or
coarse-grained (coarse sand to conglomerate, i.e.
grain size > 1000 J..Lm). Fine-grained core is mottled
dant evidence of pedogenesis with rootlet structures,
rhizocretions and nodules (see, for example, Spotl &
Wright, 1992). Petrographic analysis showed that
fine-grained units are highly dolomitic with a sub­
stantial clay mineral component. The dolomite is
finely crystalline non-ferroan dolomicrite.
The coarse-grained sandbodies are composed of

2455

2460

2465

2470

2475

2480

2485

2490

2495
Fig. 2. Core description and
petrographic data. Grain size is
shown as a continuous log. The
petrographic data are represented
2500 by bars at the appropriate depths,
0 "' "' with mineralogy represented (see
;:;
�
;:; ;:; ;:; ;:;; 0
!=' ;:; ;:; 0
0 0 0 0
0 0 key). Core analysis data are also
Core
Grain size Mineral Core displayed on this diagram. There
p errneab ility
(miaun) proportions porosity are 99 porosity and permeability
(mq
(%) datum points.
 Carbonate cement in the Triassic Paris Basin 167

massive, largely structureless sediments. Petro­
graphic analysis showed that the sandstones are
sublithic to subarkosic (according to Folk, 1974),
with a significant volume of polycrystalline quartz
grains (10-40% of quartzose grains). The feldspar
population is split approximately equally between
plagioclase and K-feldspar. The average sandstone
composition is defined in Table 3.
Sandbodies contain two distinct dolomite mor­
phologies. The top portions of most sandbodies
grade into the overlying silty dolocrete layers; the
proportion of microcrystalline dolomite increases
upwards to the top of sandbodies. A 'floating grain
texture' is present at the tops of sandbodies owing
to mass silicate grain dissolution and replacement
by microcrystalline dolomite (Fig. 3A). Sandbodies
also contain rhombic, pore-filling, ferroan dolomite
crystals that are generally greater than 200 11m in
size (Fig. 3B). The rhombic dolomite is texturally
and mineral chemically distinct from the dolocrete.
The sandstone also contains localized quartz ce­
ment (e.g. minor quartz cement labelled in Fig. 3B).
Textural considerations show that the microcrystal­
line dolomite pre-dated the ferroan rhombic dolo­
mite.
To facilitate the subsequent comparison between
petrographic data and wireline-derived mineralogi­
cal data, we converted the petrographic data into
proportions of quartz, dolomite and shale. In this
manipulation, quartz is the sum of detrital quartz
grains, quartz cement, quartzose lithic fragments
and feldspar; dolomite is the sum of all types of
dolomite and other carbonate minerals; shale is the
sum of clay, micas and micaceous lithic fragments.
There is a broad correlation between grain size and
petrographically defined mineralogy: coarse-grained

Fig. 3. Photomicrographs of (A) microcrystalline non-ferroan dolomite at the very top of a sandbody with partial
replacement of detrital silicate grains, and (B) grain-rimming quartz cement (Q) and pore-filling ferroan dolomite
(DOL) enclosing the quartz cement. Remnant porosity (0) is minor and occupies pore centres. Scale bars 200 Jlm.
168 R.H. Worden and JM. Matray

intervals are mostly quartz rich, the finer intervals
are relatively dolomite rich (Fig. 2). However, the
correlation between grain size, mineralogy and res­
ervoir properties is not perfect. Sandbodies can also
have high dolomite contents (e.g. 2457-8 m, 2470-
2 m, 2482.5-3.5m etc., on Fig. 2). This pattern
shows that dolomite content and grain size together
probably control the reservoir properties of the
sandstone. Dolomite seems to be concentrated in
the top portions of the sandstone units (e.g. 2457 m
and 2472 m), although insufficient samples were
examined petrographically to prove that this pat­
tern was common and predictable.
Core analysis data
Core analysis data are displayed as continuous logs
in Fig. 2. Porosity varies from > 0 to 19%. Perme­
ability varies from < 0.1 mD to > 5000 mD. Poros­
ity and permeability are highest where the rocks are
most coarse grained. However, again the correlation
is not perfect: the tops of the sandbodies tend to
have low porosity and permeability values relative
to the middle and lower portions of sandbodies
(Fig. 2). Consequently, grain size and facies varia­
tions cannot be used in isolation to understand or
predict variations in reservoir quality.
1000
Core analysis data are also plotted on a con­
ventional log-linear diagram (Fig. 4). There is con­
siderable scatter in the data and a wide range of
permeabilities for a given porosity. This probably
means that there is more than one control on
porosity, and thus permeability, evolution.
Wireline log analysis
Wireline log analysis has been used to define poros­
ity and mineralogy (with three components: quartz,
dolomite and shale), and these data have been used
to derive permeability. They will be used subse­
quently to assess dolomite cement distribution
within the reservoir.
Sonic transit time, neutron density and density log
data for the cored interval are presented as functions
of depth in Fig. 5. The same data are cross-plotted in
Fig. 6 with the positions of the three minerals added.
Equations (1)-(4) can be solved for porosity plus
three solid-grain components. The logs have been
converted into fractional porosity and the fractional
quantities of quartz, dolomite and shale. The rock
was thus assumed to consist of three minerals:
'quartz' (all silica minerals and feldspar), 'dolomite'
(all carbonate minerals) and 'shale' (all clay miner­
als). Each group of minerals has approximately uni­
form responses to the three wireline logging tools.
Petrographic analysis shows that the quartz/feldspar
ratio is greater than about three (Table 3), suggesting
that the assumption about the quartz component is

0
100 Table 3. Average petrographic data from the Chaunoy
_§_
Formation sandbodies. Twenty-two samples were
� examined petrographically. The figures illustrate the
:.0
«< importance of dolomite in the Chaunoy Formation
Q) 10
E
Q)
c. Standard
� Grain/cement type Mean deviation
0
()
El >80% quartz
+ 60-80% quartz Polycrystalline quartz 21.1 8.6
a <60% quartz
.1 Monocrystalline quartz 13.0 8.2
- >80% quartz regression
- 60·80% quartz regression K-feldspar 11.9 5.1
• • • · <60% quartz regression Plagioclase 4.6 3.5
.01 Lithic fragments 13.2 8.8
Detrital mica 0.8 1.1
0.0 0.1 0.2 0.3 Detrital clay 3.2 5.4
Core porosity (fractional} Kaolinite 2.5 4.2
Illite 0.4 1.1
Fig. 4.Porosity-permeability data from sandstones. Chlorite 0.5 0.9
Data have been subdivided by petrographically defined Authigenic K-feldspar 1.1 1.4
mineral proportions. High quartz content samples are Authigenic quartz 11.6 8.5
those with greater than 80% quartz, medium quartz Calcite cement 1.4 5.9
content samples have between 60 and 80% quartz, low Dolomite cement 16.4 22.0
quartz content samples have less than 60% quartz.
 Carbonate cement in the Triassic Paris Basin 169

2455

2460

2465

2470

2475

2480

2485

2490

2495

2500
Fig. 5.Wireline sonic transit time, "'
0
m
0
"'
0
0 1\) 1\)
NO "'
density and neutron density
through the cored portion of the Sonic transit time Neutron 0 (p .u.)
Chaunoy Formation. (�secn-1)

reasonable. Feldspar and quartz have similar wire­
line responses (at least for sonic, density and neutron
density logs), so that the arkosic portion of the sand­
stone is probably adequately accounted for. The
lithic portion of the sandstone is probably repre­
sented by 'shale' together with quartz. Dolomite to­
tally dominates the carbonate mineral population
within the rock. Shale represents the sum of all clay
minerals in the rock, although preliminary XRD
data show that these are dominated by kaolinite and
illite. Sonic transit time, neutron density and density
end-member values for shale were taken from cross­
plots; the values lie comfortably within published
bands (Table 2) (Rider, 1986).
170 R.H. Worden and J.M. Matray
2.2r-----------:---,
2.3
,
/: pore
fluid
2.4
12.5
�
� 26
"'
c:
��(; 0
quartz
-:�·:
� 2.7
0 shale
2.8
dolomit..
2.9-t-..-�.
• dolomite
...-�...-.�...-..,...,...j
40 50 60 70 80 90 100 ·0.05 -0.00 0.05 0.10 0.15 0.20 0.25 0.30
Sonic transit time
(A) (�sec/ft) (B)
The wireline-derived porosity data compare fa­
vourably with core analysis porosity data with an
R2 value of 0.74 (Fig. 7). The wireline porosity
values slightly overestimate the porosity (assuming
that the core porosity data are correct). Conse­
quently, wireline-derived porosity data have been
corrected for this slight overestimate by subtracting
0.024 from the fractional wireline porosity values
(Fig. 7). The wireline-derived mineralogical data
also compare favourably with the quantitative
petrographic data, the two having very good corre­
lation coefficients (Fig. 7). Thus despite the paucity
of petrographic data it is possible to derive contin­
uous and credible mineralogical data from wireline
data. Porosity and mineral proportion data were
smoothed by averaging over a 0.6 m interval to
reflect the realistic resolution of the logging tools
(Hearst & Nelson, 1985; Doveton, 1994).
The results of the wireline data transform into
mineral proportions and porosity are given in Figs 8
and 9. There are distinct intervals that are enriched
in dolomite and others enriched in quartz. The
shale fraction tends to be highest in the dolomite
zones. However, the tops of sandbodies have high
dolomite contents in the absence of shale (e.g.
2470-2 m) without any corresponding change in
grain size. This leads to asymmetry in the core
mineralogy. The summary diagram, Fig. 9, shows
that, on average, sandbodies have the most dolo­
mite in the top quarter.
The derivation of permeability from porosity is
not a simple task. Permeability is, of course, af­
fected by porosity, but it is also controlled by the
shape and size of pore throats that connect pores.
The degree of connectivity of the total porosity and
the dimensions of pore throats are critical to perme­
ability. It is not possible to derive permeability from
// pore
fluid
· �·�(
.
Fig. 6. Cross-plots of (A) sonic
transit time against density, and
(B) neutron density against
density. The positions of the three
0 shale minerals used to define the
mineralogy of the formation are
marked on both plots. The
position of the pore fluid is off the
Neutron density scale but the general direction is
(porosity units) marked.
a simple porosity value with any degree of accuracy
using a simple transform. However, recent network
modelling work by Bryant et al. ( 1993) and Cade et
al. (1994) has shown that permeability may be
predicted from porosity if the fundamental control
on porosity evolution is known. The main controls
on porosity loss may be abbreviated to compaction
and cementation. Cementation may be subdivided
further between grain-rimming cements and pore­
filling cements, where the different cement mor­
phologies have different effects upon permeability
for unit porosity loss owing to their different effects
upon the pore network.
Chaunoy sandstones of the same depositional
facies are cemented by both quartz and dolomite
(Fig. 3B). Quartz cement forms approximately equal
thickness overgrowths; whereas dolomite cements
tend to fill pores (Fig. 3B) (Cade et al., 1994). These
two different cement morphologies haye profoundly
different effects upon the pore network.
Core analysis data from the Chaunoy Formation
were subdivided on the basis of the quartz/dolomite
ratios using the wireline-derived mineralogy data.
Regression analysis (Fig. 4) shows that the quartz­
rich (and thus presumably quartz-cemented) sam­
ples have shallower porosity-permeability slopes
( m ) and higher permeability intercepts (c) than
quartz-poor (and thus presumably dolomite­
cemented) samples, in accordance with the network
modelling discussed above. We have thus derived
algorithms for describing the change in both slope
and intercept of the porosity-permeability curves as
a function of total quartz content:
'-
2.777 X 104 X J0<4·55 IO- qtz%
)
C =
x
(5)
-3.0 77 3
I o- qtz%)
m = 30. 7 5 X I o< X
(6)
in which qtz% is the quartz fraction of the rock as
 Carbonate cement in the Triassic Paris Basin 171

0.25 .------------------, c 1.0

Core equivalent 0 = 0.999 wireline derived 0- 0.024 0 correlation coeff. (r2) = 0. 8 40 .. .
x
.
correlation coeft.(r2) = 0.742 n .
0.20
jg 0.8 :=
_...,. N
t
.'
::> ctl
.s r�" ::>
rr
co.15 '0 0.6
·u; Q)
0 >
0 .... ·�
� 0.10 . '0
� 0.4
0 Cii
u ()
. ·. :c
0.05 c. 0.2
�
Ol
e
Q)
0.00 +--+-r----.---r--1 [J_ 0.0
0.00 0.05 0.10 0.15 0.20 0.25 0.0 0.2 0.4 0.6 0.8 1.0
(A) Wireline derived porosity (p.u.) (B) Wireline derived quartz fraction
c

-� 10
' .--,one latio n . ) 8 6-
= 0. _4_
- - (,2_ _-
- - -- -,oeff -----'71

jg
� 0.8
E
0
0
'0
'0 0.6
Q)
-�
Q;
� 0.4
Cii
.\<
.<:
g. 0.2
0,
e
;f 0.0 ¥<�-r----.r---r----.--1
0.0 0.2 0.4 0.6 0.8 1.0
(C) Wireline derived dolomite fraction

Fig. 7. Data quality assurance. (A) comparison of wireline-derived porosity and core analysis-derived porosity. There
is a good correlation between the two data sets. The intercept on the x axis shows that the wireline porosity data are
overestimating porosity by about 0.024. (B) Comparison of petrographically defined quartz and wireline-derived
quartz-the correlation is good and is approximately I: I with a zero intercept. (C) Comparison of petrographically
defined dolomite and wireline-derived dolomite-the correlation is good and has an approximately I: I slope with a
zero intercept.

defined by wireline analysis. It was thus possible to
predict permeability as a function of the wireline­
derived porosity and mineralogy using the follow­
ing algorithm:
Permeability (mD) = c X 1 o<m.<I>)
The results of these calculations are shown in Fig. 8.
Inspection of Figs 2 and 8 shows that the wireline­
derived permeability curve corresponds well with
the core analysis data.
DISCUSSION
Quantitative mineralogical data have been gener­
ated from sonic transit time, density and neutron
density wireline logs. Gamma logs cannot be used
for mineral identification because of the variable
mineralogical location of radiogenic potassium and
the non-concordance between uranium and
thorium and specific minerals ( Doveton, 1994;
(7)
Hurst & Milodowski, 1996). The Triassic sand­
stones and mudstones of the Paris Basin have been
resolved into quartz, shale and dolomite. Dolomite
has a diagenetic (i.e. non-primary) origin, so that
wireline logs can be used to define the spatial
distribution of dolomite cements in these sand­
stones.
The derivation of porosity, mineralogy and per­
meability from wireline data has distinct advan­
tages over core analysis data and petrographic
analysis. Most importantly, wireline mineralogical
172 R.H Worden and J.M Matray

2460

2465

2470

2475

2480

2485

2490

2495

0
N o 1\) .:..
Grain size (microns) Lithology Porosity (p.u.) Permeability (mD)

Fig. 8. Combination diagram of grain size data (derived from core description, Fig. 2) and mineral proportions,
porosity and permeability (derived from wireline log analysis). There is excellent correlation between quartz
proportion and reservoir quality. The correlation of these with grain size is complex. The tops of some sandbodies
have a high dolomite content and correspondingly poor reservoir quality (e.g. 2470-2471 m). Sandbodies are
numbered for reference to Fig. 9. Core analysis porosity and permeability data (dashed and faint) have been added to
the diagram for comparison with the wireline-derived data.

data can be derived for uncored intervals. Petro­ the reservoir and in principle are operator indepen­
graphic data are usually sparse (owing to cost dent. Petrographic data are rarely collected in such
and time constraints) and 'operator' dependent, abundance that cement distribution can be ob­
whereas wireline data are available for the whole of served within reservoir units, whereas such data
 Carbonate cement in the Triassic Paris Basin
Fig. 9. The non-uniform
distribution of dolomite and
porosity in the Chaunoy
Formation sandbodies. The
numbers refer to the sandbodies
numbered in Fig. 8. (A) Dolomite
is preferentially concentrated in
the top quarter of each sandbody.
(B) Conversely, porosity is 0 10
concentrated in the middle two (A) Average dolomite cement content(%)
quarters of each sandbody.
automatically result from wireline mineralogical
analysis. Various features of the dolomite cement
distribution that we have derived for the Chaunoy
Formation will now be discussed.
Amount and distribution of dolomite cement
in the Chaunoy sandstone
Petrographic analysis hinted at the heterogeneous
distribution of dolomite cement in the Chaunoy
Formation sandstones (Fig. 2). Without a major
sampling and petrographic analysis programme it
would be difficult to analyse and describe that
heterogeneity. The interpreted wireline data have
confirmed that dolomite is not homogeneously dis­
tributed throughout the Chaunoy Formation sand­
stone (Figs 8 and 9). Dolomite in the Chaunoy has
either a pedogenic (i.e. very early diagenetic) or
burial diagenetic origin; detrital dolomite may be
discounted as an option. Wireline data cannot be
used to discriminate between different dolomite
grain morphologies (e.g. microcrystalline or coarse
rhombic) or between dolomites of different mineral
chemistry (e.g. non-ferroan or ferroan dolomite).
The wireline data have shown that the tops of most
coarse-grained sandbodies have the most dolomite
(Figs 8 and 9). The dolomite content varies be­
tween, as well as within, sandbodies. Sandbodies 3,
4, 5, 7, 8 and 9 all have significantly more dolomite
in the top quarter than in other quarters. However,
sandbody I has much more dolomite than sand­
body 5. Dolomite can occupy more than 50% of the
solid portion of a sandbody (e.g. sandbody !). The
partially replacive nature of the dolomite within
173
20 30 0 5 10 15
(B) Average porosity(%)
(wireline data) (wireline data)
sandbodies has already been established (Fig. 3A},
so that the high dolomite content probably reflects
partial replacement of detrital silicate mineral
grains as well as precipitation of dolomite into
pre-existing pore spaces.
Effect of dolomite cement upon the reservoir
properties of the Chaunoy Formation sandstone
The porosity in the quartz-rich samples is signifi­
cantly less than the compaction-only porosity typi­
cal for sandstones at these burial depths. We would
expect quartzose sandstones to have approximately
25-30% after compaction (see, for example, North,
1985). The actual porosities even in the quartz-rich
intervals are only as high as 20%, indicating that
some of the quartz in the rock must be quartz ce­
ment. The main control on porosity and thus perme­
ability in the quartz-rich portions of the rock must
be the extent of quartz cementation. The quartz­
rich portions of sandbodies have qetter permeabil­
ities for a given porosity than their quartz-poor
equivalents (Figs 2 and 4). For example, in dolomite­
rich quartz-poor samples with 1 Oo/o porosity, perme­
ability is typically about 1-2 mD. In dolomite-poor
quartz-rich samples with I Oo/o porosity, permeability
is typically about I 0-100 mD. This is reflected by
the slightly steeper permeability-porosity gradient
and higher permeability intercept of the dolomite­
rich quartz-poor samples than the dolomite-poor,
quartz-rich samples in Fig. 4.
This confirms that the main mechanism of poros­
ity loss in the quartz-rich samples (quartz cementa­
tion) is less detrimental to permeability than
174 R.H. Worden and J.M. Matray
porosity loss in the quartz-poor samples (dolomite
cementation), as suggested by Cade et a!. (1994).
Dolomite cement in the Chaunoy Formation sand­
bodies tends to fill pores and block pore throats,
thereby degrading permeability at a greater rate
than quartz cement, which forms equal-thickness
rims to grains. Thus, not only does the dolomite
cement preferentially obscure porosity at the tops of
the sandstone units, it also leads to commensurably
poorer permeabilities than for quartz-cemented
sandstones of similar porosity.
Origin of dolomite cement in the
Chaunoy Formation sandstone
The dolomite-rich fine-grained beds in the Chaunoy
Formation resulted from dolocrete pedogenesis
(Spot! & Wright, 1992) in interchannel facies. The
similarity between the (very fine) crystal size and
texture of the dolomite in the fine beds and the
dolomite at the very tops of the sandbodies
(Fig. 3A) suggests that some of the dolomite in the
sandbodies may be related to the formation of the
dolocrete during pedogenesis. The replacive nature
of the finely crystalline dolomite in the sandbodies,
as indicated by the corrosion of detrital quartz and
feldspar grains (Fig. 3A), supports the development
of this dolomite by 'aggressive' pore waters during
pedogenesis.
However, much of the dolomite within the sand­
bodies does not have the same morphology and
chemistry as the dolocrete material: much occurs as
coarsely crystalline, ferroan dolomite rhombs.
From the textural and mineral chemical evidence,
this must have a different genesis from the dolo­
crete. The timing of the dominant ferroan dolomite
cement growth in sandbodies is difficult to deter­
mine absolutely. Textural evidence proves that
rhombic ferroan dolomite postdates quartz cement
overgrowths (Fig. 3B). Aqueous fluid inclusion tem­
peratures from rhombic dolomite, reported by Spot!
et a!. (1993) and Demars & Pagel (1994) are
somewhat higher than present-day temperature,
suggesting that dolomite grew at maximum burial/
temperature conditions in the Oligocene/Miocene.
This probably coincided with maturation of the
Liassic source rocks and hydrocarbon generation
and migration within the Paris basin (Poulet &
Espitalie, 1987).
Carbon stable isotope data for the burial diage­
netic dolomite cements (Worden & Matray, 1995)
indicate that isotopically depleted carbon has been
added to the dolomite (Spot! & Wright, 1992).
Isotopically depleted carbon is thought typically to
have an organic origin (e.g. Longstaffe, 1989). The
most obvious source of organically derived bicar­
bonate or C02 in the Paris Basin is the Liassic shale
source rock. pH-buffered rocks undergo carbonate
mineral precipitation when the partial pressure of
C02 is increased (Lundegaard & Land, 1989),
suggesting that at least some of the rhombic dolo­
mite cement may be the direct result of bicarbonate
or C02 influx increasing the partial pressure of
C02. Liassic source rocks may have expelled C02
during or before oil generation. The Triassic sand­
stones in the Paris Basin are currently in equilib­
rium with C02, which is partitioned between the
two liquid phases oil and water (Matray et a!.,
1993). The equilibrium partitioning of C02 be­
tween formation water and oil suggests that C02
may have been brought into the reservoir by the oil
in solution. Subsequent partitioning of C02 into the
formation water may then have caused dolomite
supersaturation and precipitation. Alternatively,
C02 may have migrated into the Chaunoy sandbod­
ies as a separate gas phase resulting from the
thermal decarboxylation of organic. matter. What­
ever the mechanism, isotopic data dictate that an
increase in the partial pressure of C02 (from an
organic source) was most likely responsible for the
precipitation of dolomite cement in the sandbodies
during diagenesis at close to maximum burial.
Origin of the dolomite cement distribution pattern
Dolomite cement is generally concentrated at the
tops of sandbodies in the Chaunoy Formation
(Figs 7 and 9). There are several potential generic
controls on dolomite distribution patterns (Fig. 1 0).
1 The cement at the tops of sandbodies may be a
direct result of pedogenesis, whjch occurred at the
same time as the development of the pedogenic
dolocrete in the fine-grained units. This would
occur preferentially at the tops of sandbodies adja­
cent to zones of active dolocrete pedogenesis. This
is probably at least partly responsible for the dolo­
mite cement distribution in the sandbodies.
2 In principle, dolomite distribution in sandbodies
may be due to diffusion from the pedogenic dolo­
cretes that encase the sandbodies. In this case the
dolomite would be redistributed by diffusion from
the dolocrete into the sandbodies. This would influ­
ence the tops and bases of sandbodies equally and
result in a minimum dolomite cement content at
 Carbonate cement in the Triassic Paris Basin 175

Fig. 10. Theoretical dolomite

distributions from four potential

[
�3:::-� �
controlling processes. The model
-=
represents a sandbody sandwiched dolomite -=
between pedogenic dolocrete layers. -=
-=
--= content �-- -=
-=
--
--
-=

J) fl tt
(A) Pedogenic dolomite cement;
there would be most dolomite at the
top of each sandbody. (B) Dolomite Pedogenesis
cement sourced from the dolocrete controlled +-- +- +-
'
during burial, transported by
diffusion; cement should be equally � High-penneability
streak controlled
�
�
--
abundant at the tops and bases of -= -= ===---= -=
===--
-- --
sandbodies, with a minimum at the -= --
-=--
-= --
-=--
centre. (C) Dolomite distribution
controlled by high-permeability

�
streaks allowing input from external
sources; fluvial sandstones usually ::E � -=
fine upwards, leading to high­ � -=:. -= -=
-- -= --
--
-=

-=

� fi �· �r
permeability bases and thus most
dolomite at sandbody bases. (D) ...._.___ C02 gas or
...._.___
oil+ C02
Dolomite distribution controlled by
the relative buoyancy of oil (which Diffusion controlled
may have carried dissolved C02), or ·c . .. t ,, •i • . ( .1 • '
· •

a separate C02 gas phase caused

dolomite cementation and thus led to � � � -� � �
-= -=
controlled

=---= -=
most dolomite at the tops of ===--
--
-=
-- ---=--
-= -=
sandbodies.
-- --

the centre of sandbodies. Note that this is not
observed (Figs 8 and 9) and that rhombic ferroan
dolomite has a carbon isotope signature which is
different from the pedogenic dolomite (Spotl &
Wright, 1992; Worden & Matray, 1995).
3 Cement distribution could be influenced by res­
ervoir quality at the time of cementation. High­
permeability streaks or gradational permeability
may have focused the flow and input of C02 into
specific portions of the rock. Fluvial sandstones
usually fine upwards, resulting in diminishing per­
meability towards sandbody tops. This would lead
to the most extensive dolomite cementation at the
bases of sandbodies. However, note that the
Chaunoy sandstones do not fine upwards (Fig. 2)
and do not have dolomite preferentially at sand­
body bases.
4 Isotope data suggest that C02 has an organic
source and might have come from the oil source
rock (Spotl & Wright, 1992; Worden & Matray,
1995). Oil and C02 may have migrated into the
rock at about the same time (i.e. as C02 dissolved in
oil) (Matray et al., 1993). Alternatively, C02 may
have migrated into the rock separately as a free gas
phase. Because of buoyancy, the top of each sand­
body should be the first part of the sandstone to
encounter either oil (laden with C02) or free C02
gas. In summary, the top of each sandbody may
thus have received C02 preferentially and thus
caused localized dolomite precipitation. However,
it is generally considered that oil emplacement
hinders diagenetic processes, so that the opportu­
nity for this process to operate may be limited to a
window between the onset of oil emplacement and
some elevated level of oil saturation (e.g. Worden et
a/., in press).
The absence of dolomite cement at sandbody
bases and its abundance at sandbody tops, the
reported organic carbon isotope signal in the rhom­
bic ferroan dolomite and the mixture of pedogenic
dolomite textures and burial diagenetic textures in
the sandbodies suggests that options 1 and 4 to­
gether are probably responsible for, the distribution
of dolomite in the Chaunoy sandbodies.
CONCLUSIONS
1 Wireline petrophysical data have been success­
fully manipulated to give mineralogy in terms of the
amounts of quartz, shale and dolomite, as well as
porosity.
2 Core analysis data show that dolomite cement
has a more detrimental effect upon permeability
than quartz cement. Permeability has thus been
calculated from the wireline porosity data using
176 R.H. Worden and J.M Matray
algorithms that account for the variation in miner­
alogy as well as porosity.
3 Petrography and, more importantly, wireline log
data have shown that dolomite cement is not
uniformly distributed throughout the sandstones
within the Chaunoy Formation. Rather, dolomite
cement is localized within the top portions of
individual sandstone units.
4 Reservoir quality in the Chaunoy Formation is a
function not just of depositional facies but also of
localized cement distribution. Building a reservoir
model using primary sandbody architecture alone is
insufficient to correctly describe reservoir quality.
5 Sandbodies contain microcrystalline non-ferroan
and replacive dolomite as well as rhombic ferroan
and pore-filling dolomite. Textural and mineral
chemical data show that the microcrystalline dolo­
mite probably grew during pedogenesis of the over­
lying fine-grained facies. Reported fluid inclusion
and isotope data together with textural evidence
show that the rhombic dolomite probably grew at
close to maximum burial in the mid-Tertiary in the
presence of organically derived C02•
6 Dolomite cement may be localized at the tops of
the sandbodies because of the proximity of overly­
ing fine-grained units when they were undergoing
pedogenesis, and because the tops were the first part
of each sandbody to receive a charge of C02. The
C02 influx may have occurred as a separate buoy­
ant gas phase or as a gas dissolved in oil.
ACKNOWLEDGEMENTS
The authors would like to thank Elf-Aquitaine
(especially Fred Walgenwitz and Gerard Sambet)
for kindly providing the core analysis and wireline
data. Part of the study was the result of a collabo­
rative research programme including BP, B RGM,
Elf-Aquitaine, the University of Paris V I and the
European Community under contract JOUF-
0016c. Jim Hendry, Sadoon Morad, Julian Baker
and Jean-Pierre Girard are thanked for comment­
ing upon various versions of the manuscript and for
identifying key areas for improvement.
REFERENCES
BouRQUIN, S. & GuiLLOCHEAU, F. (1993) Geometrie des
sequences de dep6t du Keuper (Ladinien a Rhetian) du
Bassin de Paris: implications gecdynamiques. C. R.
Acad. Sci. Paris &r. 2, 31 7 , 1341-1348.
BOURQUIN, S., BOEHM, C., CLERMONTE, J., DURAND, M. &
SERRA, 0. (1993) Analyse facio-sequentielle du Trias du
centre-ouest du bassin de Paris a partir des donnees
diagraphiques. Bull. Soc. Geol. France, 164, 177-188.
BRUNET, M.-F. & LE PICHON, X. ( 1982) Subsidence of the
Paris Basin. J. Geophys. Res. , 87, 8547-8560.
BRYANT, S., CADE, C. & MELLOR, D. ( 1993) Permeability
prediction from geological models. Bull. Am. Ass. Petrol.
Geol. , 77, 1338-1350.
CADE, C., EvANS, I.J. & BRYANT, S. (1994) Analysis of
permeability controls: a new approach. Clay Miner. , 29,
49 1-501.
DEMARS, C. & PAGEL, M. (1994) Paleotemperatures et
paleosalinites dans les gres du Keuper du Bassin de
Paris: inclusions fluides dans les mineraux authigenes.
C. R. Acad. Sci. Paris Ser. 2, 319, 427-434.
DovETON, J.H. ( 1994) Geologic Log Analysis Using Com­
puter Methods. AAPG Computer Applications in Geol­
ogy, 2. Am. Ass. Petrol. Geol., Tulsa, 169 pp.
FoLK, R.L. ( 1974) Petrology ofSedimentary Rocks. Hemp­
hill, Austin.
FONTES, J.C. & MATRAY, J.-M. ( 1993) Geochemistry and
origin of formation brines from the Paris Basin. Part 2
Saline solutions associated with oil fields. Chern. Geol. ,
109, 177-200.
HEARST, J.R. & NELSON, P.H. ( 1985) Well Logging for
Physical Properties. McGraw-Hill, New York, 57 1 pp.
HERRON, S .L. & LE TENDRE, L. ( 1990) Wireline source­
rock evaluation in the Paris Basin. In: Deposition of
Organic Facies (Ed. Hue, A.Y.) Am. Ass. Petrol. Geol.,
Studies in Geology, 30, 57-71.
HURST, A. & MILODOWSKI, A. ( 1996) Thorium distribution
in some North Sea sandstones: implications for petro­
physical evaluation. Petrol. Geosci., 2, 59-68.
LONGSTAFFE, F.J. ( 1989) Stable isotopes as tracers in
clastic diagenesis. In: Mineralogical Association of
Canada Short Course in Diagenesis (Ed Hutcheon, I.},
pp. 201-277. Mineralogical Association of Canada,
Montreal.
LouP, B. & WILDI, W. (1994) Subsidence analysis in the
Paris Basin: a key to Northwest European intraconti­
nental basins? Basin Res. , 6, 159-177.
LUNDEGARD, P.O. & LAND, L.S. ( 1989) Carbonate equilib­
ria and pH buffering-response to changes in PC02.
Chern. Geol. , 74, 277-287.
MATRAY, J.-M., FOUILLAC, C. & WORDEN, R.H. (1993)
Thermodynamic control on the ehemical composition
of fluids from the Keuper aquifer of the Paris Basin. In:
Geofluids '93 (Eds Parnell, J., Ruffel, A.H. & Moles,
N.R.}, pp. 12-16. Geological Society of London, Bath.
MEGNIEN, C. ( 1980a) Tectogenese du Bassin de Paris:
etapes de !'evolution du bassin. Bull. Soc. Geol. France,
22, 669-680.
MEGNIEN, C. (1980b) Synthese geologique du bassin de
Paris. Stratigraphie et paleogeographie. Memoire
BRGM, 101, 466 pp.
NORTH, F.K (1985) Petroleum Geology. Allen & Unwin,
Boston, 607 pp.
PAGES, L. (1987) Exploration of the Paris Basin. In:
Petroleum Geology ofNorth West Europe (Eds Brooks, J.
& Glennie, K.), pp. 87-93. Graham & Trotman, Lon­
don.
 Carbonate cement in the Triassic Paris Basin
PoMMEROL, C. ( 1 97 4) Le bassin de Paris. In: Geologie de Ia
France (Ed. Debelmas, J.), pp. 230-258. Doin, Paris.
PoMMEROL, C. ( 1978) Evolution paleogeographique et
structurale du Bassin de Paris, du Precambrian a l'ac­
tuel, en relation avec les regions avoisinantes. Geol.
Mijnbouw, 57, 533-543.
POULET, M. & ESPITALIE, J. ( 1987) Hydrocarbon migration
in the Paris Basin. In: Migration of Hydrocarbons in
Sedimentary Basins (Ed. Doligez, B.), pp. 13 1 - 17 1.
Editions Technip, Paris . .
RIDER, M.H. (1986) The Geological Interpretation of Well
Logs. Blackie, Glasgow, 1 7 1 pp.
1 77
SP6TL, C. & WRIGHT, V.P. ( 1 992) Groundwater dolocretes
from the Late Triassic of the Paris Basin, France: a case
study of an arid, continental diagenetic facies. Sedimen­
tology, 39, 1 1 19-1 136.
SPOTL, C., MATTER, A. & BREVART, 0. ( 1 993) Diagenesis
and pore water evolution in the Keuper reservoir, Paris
Basin (France). J. sediment. Petrol. , 63, 909-928.
WoRDEN, R.H. & MATRAY, J.-M. ( 1995) Cross formational
flow in the Paris Basin. Basin Res. , 7, 53-66.
WORDEN, R.H., SMALLEY, P.C. & OXTOBY, N.H. Can oil
emplacement prevent quartz cementation in sand­
stones? Petrol. Geosci. (in press).
Spec. Pubis int. Ass. Sediment. ( 1 998) 26, I 79- 1 92

Calcite cement in shallow marine sandstones:

growth mechanisms and geometry

0. WALDER HAUG*1 a n d P.A. BJ0R KUMt

*Raga/and Research, PO Box 2503 Ullandhaug, 4004 Stavanger, Norway,

e-mail pab@statoil.no; and
tStatoil a.s., 4035 Stavanger, Norway

ABS T R A C T

Calcite cement in shallow marine sandstones normally cannot be derived from sources outside the
sandstones owing to a lack of viable transport mechanisms for significant amounts of dissolved calcium
carbonate. Within the sandstones the only significant source of calcite cement is usually biogenic
carbonate, which is consequently considered to be the dominant source of calcite cement within
shallow marine sandstones. Influx of carbon dioxide into a sandstone will not lead to precipitation of
additional calcite cement unless a source of calcium other than biogenic carbonate is present, but the
carbon isotopic composition of the calcite cement may be strongly affected.
Geometrically, calcite cementation in shallow marine sandstones typically occurs as continuously
cemented layers, as layers of stratabound concretions, and as scattered concretions. All these forms can
be explained by local diffusional redistribution of biogenic carbonate originally present within the
sandstones. Biogenic carbonate is less stable than calcite cement, and once a calcite cement nucleus has
formed it will lower the concentration of dissolved calcite within its range of influence. Biogenic
carbonate will then dissolve around the growing nucleus, diffuse down the concentration gradient and
precipitate on the surface of the growing nucleus or concretion. This process will continue until the
available biogenic carbonate is consumed or all porosity is filled with calcite cement.
If biogenic carbonate is concentrated in layers, stratabound concretions or continuously cemented
layers form, as calcite cement nuclei are then concentrated within the biogenic carbonate-rich layers.
Nucleation within these layers may take place either because the biogenic carbonate provides
favourable nucleation substrates or because calcite supersaturations are highest within these layers.
Stratabound concretions form when the supply of biogenic carbonate is exhausted prior to merging of
concretions. If more biogenic carbonate is present, concretions merge and form a continuous calcite
cemented layer. Scattered concretions form when biogenic carbonate occurs scattered throughout a
sandstone, as preferred levels of nucleation will then be absent. Concretions occur with a certain
spacing because, once a calcite cement nucleus has formed, the level of dissolved calcite in the pore
water will be reduced around the nucleus, thereby inhibiting the formation of new nuclei within the
range of influence of the first nucleus. Flattening of concretions parallel to bedding, which traditionally
has been ascribed to permeability anisotropy and fluid flow, may rather be a result of more extensive
growth of concretions in the direction of greatest supply of biogenic carbonate.
The presented nucleation and growth model implies that the geometry of calcite-cemented zones is
controlled by the original distribution of biogenic carbonate, and prediction of the geometry of calcite
cementation in subsurface reservoirs therefore largely depends upon an understanding of the
depositional environment.

IN T R O DUCTIO N

Calcite cement in shallow marine sandstones typi- cally occurs as pervasively calcite-cemented volumes
within calcite cement-free sandstone. The calcite­
1Present address: Statoil a.s., 403 5 Stavanger, Norway, cemented sandstone may occur as calcite-cemented
e-mail owald@statoil.no. layers, layers of stratabound concretions, scattered

Carbonate Cementation in Sandstones Edited by Sadoon Morad 1 79

© 1998 The International Association of Sedimentologists
ISBN: 978-0-632-04777-2
1 80 0. Walderhaug and P.A. Bjerkum
concretions, and, more rarely, as calcite-cemented
patches or specks of millimetre to centimetre size
(e.g. Davies, 1967; Fursich, 1 982; Hudson & An­
drews, 1 987; Walderhaug & Bj0rkum, 1992; Mc­
Bride et a!., 1995). This geometrical diversity is one
of the most striking aspects of calcite cementation,
and any model attempting to explain it must conse­
quently be able to furnish an explanation for these
geometrically distinct forms. In a series of previous
papers we presented a nucleation and growth model
designed to explain these modes of calcite cementa­
tion (Walderhaug et a!.; 1 989; Bj0rkum & Walder­
haug, 1 990a,b, 199 3; Walderhaug & Bj0rkum,
1 992), and in the present paper we review and
discuss this model in light of recent developments
and comments from other workers. The establish­
ment of such a model must necessarily also encom­
pass a study of the sources of calcite cement, and
this problem is therefore discussed at some length.
Calcite-cemented zones may have a profound
influence on the performance of hydrocarbon reser­
voirs (Sundal et a!., 1 990; Gibbons et a!., 1 993).
Methods and criteria for predicting the geometry of
calcite-cemented zones in the subsurface are there­
fore of great practical value, and the implications of
the presented nucleation and growth model regard­
ing this problem are therefore briefly reviewed.
Also, published predictions of lateral extents of
calcite-cemented zones in several North Sea reser­
voirs (Walderhaug eta!., 1 989; Bj0rkum & Walder­
haug, 1 990b) are compared with results from
subsequent drilling.
It is emphasized that the model presented here
paper does not purport to be a general explanation of
all types of calcite cementation found in sandstones.
It should not, for instance, be uncritically applied to
settings where calcite cementation is dominated by
evaporation effects, by microbiological activity or
by the presence of abundant volcanic matter. Its
main application is in situations where biogenic
carbonate is the dominant source of calcium and
the pore system is filled by a stagnant or slowly
flowing aqueous fluid, i.e. conditions typical of
shallow marine sandstones during burial diagenesis.
O C CU R R E N C E O F
CALCITE C E M E N T
Calcite-cemented intervals typically account for up
to I 0% of sandstone thickness in shallow marine
reservoir sandstones on the Norwegian shelf, al­
though rare examples of almost complete calcite
cementation also occur. Calcite cement normally
forms pervasively cemented intervals where all
porosity is filled by calcite cement, and usually very
little or no calcite cement is present between the
calcite-cemented intervals (Table 1 ). Study of anal­
ogous outcrops shows that the calcite-cemented
intervals have a variety of forms which, in our
opinion, may be represented by a few geometrical
end-members and combinations of these end­
member morphologies. The most important types
of calcite-cemented volumes in shallow marine
sandstones seem to be: continuously cemented lay­
ers (e.g. Bryant et a!., 1 988; Walderhaug et a!.,
1989) (Plate 1 , facing p. 1 82), layers of stratabound
concretions (e.g. Fursich, 1 982; Wilkinson, 1 992)
(Plate 2), scattered concretions (e.g. Hudson &
Andrews, 1 987; Walderhaug et a!., 1 99 5) (Plate 3),
and patchy or microconcretionary calcite cement
(Walderhaug & Bj0rkum, 1 992) (Plate 4).
Calcite-cemented layers typically have thick­
nesses from around 10 em to a metre or two, and
vary widely in lateral extent. Minimum lateral
extent is a matter of definition, as it is really
determined by where one chooses to start talking of
layers rather than of concretions. Maximum lateral
extents are certainly greater than a few kilometres,
as such extents can be observed in cliff exposures
(Bryant et a!., 1 988; Walderhaug et a!., 1 989), and
lateral extents of tens of kilometres cannot be
excluded. Intermediate lateral extents of tens and
hundreds of metres have also been observed (Mc­
Bride et a!., 1 99 5; Walderhaug et a!., 1 995).
Stratabound concretions are located within the
same stratum, often with a semiregular lateral
spacing that may vary from layer to layer (Plate 2).
Lateral extents and thicknesses are comparable with
lateral extents for continuously cemented layers.
Stratabound concretions may pass laterally into
continuously cemented layers pf varying lateral
extent (Plate 5). The shape of the concretions within
a layer is commonly rather uniform, with some
layers dominated by flattened concretions and
others by more spherical concretions. Where weath­
ering effects permit various stages of concretion
growth to be detected, a progression from spherical
to flattened may be seen (Plate 2).
As the name implies, scattered concretions differ
from stratabound concretions mainly by not being
systematically located along a bedding plane. In
some cases scattered concretions may attain very
large dimensions. The largest calcite-cemented
sandstone concretions known to us are found in the
Lower Cretaceous Dakota Sandstone in Kansas,
Table I. Typical modal compositions of Jurassic shallow marine sandstones from the Norwegian shelf

Plagio- Carbon- Plant Authi- Authi- Grain

Depth Quartz K-feldspar clase Mica Heavy ate frag- Clay Clay Pyrite genic genic Calcite Siderite· Dolomite Quartz size
Well (mRKB) Formation clasts clasts clasts clasts minerals fossils ments clasts matrix cement kaolinite illite cement cement cement cement Porosity (mm)

2/1-4 4044.65 Gyda 57.0 13.3 1.0 Trace 0.7 Trace 0.7 4.3 - 0.7 0.7 1.3 8.0 12.3 0.25
2/1-4 4049.42 Gyda 52.3 17.0 Trace - Trace 0.7 0.3 1.7 Trace - 28.0 Trace Trace - 0.3
7/11-5 4200.01 Ula 53.7 14.7 0.3 Trace Trace - 0.7 2.0 Trace - 0.3 Trace 19.7 8.7 0.3
7/11-5 4230.66 Ula 45.3 8.7 1.3 2.7 0.3 0.3 3.0 2.3 3. 7 - - 32.3 Trace Trace - Trace 0.15
7/12-2 3425.66 Ula 47.0 14.3 1.0 0.3 1.0 0.3 1.0 0.3 0.3 - - 34.0 -
0.3 - - 0.3
7/12-2 3430.22 Ula 46.7 8.0 0.3 Trace Trace 3.7 I. 7 Trace 0.3 - - 39.0 - Trace - 0.3 0.3
7/12-2 3432.69 Ula 57.0 9.7 1.0 0.7 Trace 1.7 - Trace 4.3 1.3 - -
0.3 3.3 20.7 0.3
7/12-2 3433.97 Ula 50.0 12.0 0.7 Trace 0.7 1.0 1.3 0.3 - - 34.0 -
0.3
7/12-3A 3680.70 Ula 52.7 8.3 0.7 Trace Trace Trace - - - 38.3 Trace Trace 0.45
7/12-3A 3706.05 Ula 48.3' L3.7 0.3 Trace Trace Trace 1.3 0.3 Trace 36.0 - - 0.4
7112-3A 3714.38 Ula 48.0 11.3 1.7 1.0 0.7 Trace 1.3 3.0 1.0 - 0.7 3.0 28.3 0.2
7/12-5 3892.17 Ula 39.0 7.7 Trace Trace 0.7 15.0 - - 0.3 - Trace 33.7 -
0.3 3.3 2.0
7/12-6 3446.01 Ula- 54.3 15.0 0.3 Trace 0.7 - 0.7 1.7 0.3 - -
Trace -
0.7 3.0 23.3 0.25
7/12-A15 3542.00 Ula 54.7 12.7 0.3 - 0.3 Trace I. 7 Trace 0.7 -
29.7 - -
Trace 0.35
7112-A 1'5 3640.70 Ula 59.0 14.0 0.3 0.3 0.3 Trace - - 26.0 - - -
0.28
7112-AI5 3644.60 Ula 43.3 15.3 1.0 0.3 0.3 Trace 0.7 - -
39.0 -
Trace -
0.3
24/12-2 4960.70 Heather 48.7 Trace Trace Trace Trace Trace 2.7 - 48.7 - - - 0.35
24/12-2 4963.70 Heather 71.3 - 0.7 0.7 Trace 18.0 Trace 2.7 - -
5.3 1.3 0.4
24/12-2 4966.80 Heather 54.0 Trace Trace 1.3 - - Trace 3.3 - 41.3 - - - 0.4
30/3-2 2904.90 Oseberg 48.0 5.7 - 1.3 Trace - - 0.7 - 2.3 2.7 33.7 1.3 - 4.3 0.5
30/3-2 2909.80 Oseberg 62.3 6.7 0.3 1.0 Trace - - 1.3 2.0 2.3 Trace -
4.0 3.3 16.7 0.6
30/3-2 2928.90 Oseberg 60.0 3.3 1.3 0.3 0.7 0.3 0.3 - 0.7 27.3 Trace - 0.3 5.3 0.7
30/3-2 2936.60 Oseberg 52.7 5.0 1.7 0.3 1.3 0.3 Trace 0.3 Trace 32.0 6.0 Trace 0.3 0.5
30/3-2 2939.60 Oseberg- 50.0 5.0 Trace Trace Trace 0.3 Trace 3.3 0.7 40.0 - 0.7 0.6
30/3-2 2942.50 Oseberg 53.3 6.0 Trace 0.3 Trace - 1.0 1.7 5.3 4.3 - - 3.0 5.3 19.7 0.45
31/4-3 2165.75 Fensfjord 31.7 4.0 0.3 5.3 2.0 0.7 2. 7 0.7 Trace 52.0 0.7 - Trace 0.1
31/4-4 2499.15 Fensfjord 26.3 2.3 Trace 10.7 1.0 0.3 0.3 2.0 0.7 1.7 - - 54.7 -
Trace 0.1
31/4-4 2499.16 Fensfjord 49.0 3.0 0.3 !7.7 0.1 0.3 1.3 1.0 1.0 - - -
26.0 0.08
31/4-5 2153.75 Fensfjord 38.7 5.0 0.3 11.7 1.3 4.7 2.0 1.0 - 35.0 0.3 - - 0.08
31/4-5 2183.95 Fensfjord 37.3 4.3 0.3 13.0 1.0 5.3 1.3 1.7 - 35.7 - - 0.08
31/4-6 2162.25 Fensfjord 34.3 4.0 0.3 5.3 2.0 8.7 1.7 - 43.7 - - 0.1
31/4-7 2097.30 Fensfjord 14.0 0.3 Trace 28.0 0.7 9.0 Trace 1.3 0.7 1.0 Trace 45.0 Trace Trace 0.05
31/4-9 2173.70 Fensfjord 35.7 5.3 Trace 4.3. 1.0 1.0 1.3 3.0· 1.3 0.3 46.7 -
Trace 0.16
31/4-9 2174.65 Fensfjord 48.3 6.3 0.7 1.3 1.3 - - 2.0 1.7 1.0 - - 37.3 0.17
31/4-9 2176.90 Fensfjord 34.7 3.3 0.3 6.3 0.7 7.3 - 0.7 Trace 1.3 43.3 2.0 0.1
31/4- 9 2189.60 Fensfjord 41.7 4.7 0.3 13.7 1.3 - -
5.3 1.0 2.0 - Trace 30.0 0.14
34/8-8 3011.50 Rannoch 41.5 8.5 2.5 9.5 0.5 - - 1.5 - - 3.5 -
1.0 2.5 29.0 0.16
34/8-8 3018.50 Rannoch 31.0 10.0 0.5 9.5 4.0 - - 1.0 - - 1.5 40.5 2.0 0.16
34/8 -8 3028.50 Rannoch 42.5 8.0 1.0 3.0 1.0 Trace - 3.0 - 0.5 40.0 1.0 0.18
34/10-4 1863.60 Rannoch 39.3 1.7 0.3 2. 7 1.0 - - 0.7 - - 47.7 4.3 - - 2.3 0.14
34/10-4 1876.30 Rannoch 37.7 8.0 - 4.0 2.0 - - 5.3 0.3 1.7 - 1.3 - - 39.7 0.11
35/8 -1 3686.02 Rannoch 57.7 3.0 0.3 7.7 4.0 - 3.0 2.7 0.7 - - 0.7 6.0 14.3 0.2
35/8-1 3696.00 Rannoch 54.0 Trace Trace 6.0 2.7 1.3 Trace - -
36.0 - - - 0.09
6407/7- 1 2897.13 Tilje 58.3 5.5 Trace Trace 0.3 0.3 0.3 2.6 - - 9.5 23.2 0.4
640717-1 2897.35 Tilje 56.0 3.7 Trace Trace 0.5 0.7 Trace 2.7 - 32.7 -
3.0 0.7 0.4
6407/7-1 2908.63 Tilje 63.3 4.0 Trace Trace 0.3 -
0.7 Trace 0.3 Trace - 31.3 -
Trace 0.4
6506/12-7 4431.15 Garn 68.0 2.7 - 1.0 Trace - - 0.3 1.0 Trace - 1.7 - 6.3 10.7 8.3 0.35
6506/12-7 4438.45 Garn 50.3 5.3 - 1.0 0.3 - 2.3 2.0 0.7 - 0.3 25.3 - 4.0 8.3 0.4
7120/9-1 1866.80 St0 40.7 0.3 - Trace Trace 4.3 - Trace 0.3 Trace - 40.7 4.7 - 9.0 0.2
7120/9-1 1899.60 St0 64.7 1.0 - 0.3 0.3 - 1.0 0.7 2.0 - 6.3 23.7 0.15

Calcite-cemented samples are overrepresented in the table. Normally calcite-cemented intervals account for less than l 0% of sandstone thicknesses.
182 0. Walderhaug and P.A. Bjerkum
where spheroidal concretions have maximum
diameters of 6 m (Swineford, 1 947). Weathering of
the otherwise relatively unconsolidated sandstone
commonly leaves the concretions of the Dakota
Formation as a residue on the surface, forming
spectacular localities such as 'Rock City' (Bell,
1 90 1 ; Schoewe et al. 1 937; Pettijohn et al., 1 973).
In many cases concretions tend to occur at specific
levels within a sandstone, even if they are not of the
typical stratabound type (Plate 6, between pp. 1 82
and 183), and the distinction between the two
end-members is thus not sharp. The shape of
individual concretions may vary widely. The most
common forms seem to be somewhat flattened and
in some cases roughly spheroidal, with their long
axes parallel to bedding. Moreover, an almost un­
limited variation including perfect spheres, elongate
forms with long axes at right-angles to bedding,
forms with various bumps and protrusions, and
composite intergrown forms also occur (Plates 2, 3,
6, 7 and 8). The presence of tight layers also strongly
influences concretion shape when the outward
growth of concretions is inhibited by such layers or
laminae.
Patchy or microconcretionary calcite cement, i.e.
millimetre- to centimetre-sized specks of calcite
cement with spacings that are also typically in the
millimetre to centimetre range, seems to be less
common than pervasively calcite-cemented concre­
tions and layers, and has only been described from
cores (Walderhaug & Bjmkum, 1 992). The ce­
mented specks may apparently coalesce to form
larger calcite-cemented volumes (Plate 4), but ow­
ing to the lack of outcrop data it is uncertain what
shapes and sizes calcite-cemented volumes formed
by the coalescence of calcite cement patches actu­
ally have.
There seems to be no obvious systematic differ­
ence in the amount or type of calcite cementation in
sandstones that have been buried to depths of
4-5 km and temperatures of more than 1 5o·c
versus otherwise unconsolidated sandstones that
have not been deeper than 1 . 5 km and that have not
been subjected to temperatures of more than
around 1o·c (Plate 9; Table 1 ). This suggests that
most calcite cementation is probably complete at
shallow to moderate burial depths and at relatively
low temperatures, although examples of later calcite
precipitation that postdates initial quartz cementa­
tion have been documented (e.g. Saigal & Bje�r­
lykke, 1987; Walderhaug, 1 990; Taylor & Soule,
1 993). The oxygen isotopic compositions of calcite
cement are also typically compatible with precipita­
tion at temperatures below around 1o·c (e.g. Hud­
son & Andrews, 1 987; Saigal & Bje�rlykke, 1 987;
Wilkinson, 1 992; Bje�rkum & Walderhaug, 1 993)
(Fig. 1). However, because of the difficulty of cor­
rectly estimating the oxygen isotopic composition
of the pore water at the time of calcite cementation,
it is very difficult to give an exact interpretation of
o180 values for calcite in terms of precipitation
temperature.
Calcite cement typically pre-dates quartz cemen­
tation, as indicated by the lack of quartz over­
growths within calcite-cemented zones (Table 1 ). As
abundant fluid inclusion evidence (see review in
Walderhaug, 1 994) strongly indicates that quartz
cementation typically becomes significant at tem­
peratures of 70-8o·c, this is strong confirmation
that most calcite cementation was completed at
temperatures below around 7o·c.
Few fluid inclusion data have been published for
calcite cement in shallow marine sandstones, and
homogenization temperatures in calcite may possi­
bly be reset (Barker & Goldstein, 1 990). However,
the measurements reported :by Saigal & Bje�rlykke
( 1987) are in the range 56-68·c and thus compat­
ible with calcite precipitation prior to deep burial.
Finally, the high intergranular volumes found in
many calcite-cemented sandstone samples (Table I)
also indicate relatively early1calcite precipitation.
��
'J
S O U R C E S OF C A L<HTE C E M E N T
i
The possible sources of calc te cement are either in­
ternal or external relative tol' the sandstone contain­
ing it. Internal sources such as biogenic carbonate
and carbonate rock fragments can be redistributed
as cement by diffusion over ,:hort distances in the
millimetre to metre range. Copyersely, the importa­
tion of calcite cement from sources external to the
sandstone will typically in olve transport distances
of the order of I 00 m or 1 \em, and fluid flow has to
be invoked as a transport agent. However, the fluid
flux required for transporting significant amounts of
dissolved calcite is enormously in excess of what is
typical of compacting sedimentary basins, even if
focusing of compactional flow is postulated. As an
example, conservative estimates show that in order
to transport the amount of calcite cement found in
the Upper Jurassic Fensfjord Formation of the
Brage Field in the Noi:th Sea, compactional water
2
from an area of around I 00 000 km would have to
 Calcite cement in shallow marine sandstones 183

0 I I I I I
I + +
rooo 0 0 +

+
+
0 Cook Fm

-
0 -5 - 0 Fmya Fm

� 0 Paleogene
-
m
c X UlaFm
Q.
0
00
St0 Fm

Oseberg Fm
- 1 0 -
+ Fensfjord Fm

I I
-1 5 I I I I
-50 -4 0 - 30 -2 0 - 1 0 0 10 20

Fig. 1. 8180p08 versus 813Cp08 for 549 calcite cement samples from marine sandstones on the Norwegian continental
shelf.

be focused through the 55 km2 field, i.e. about half
the water available from compaction of all sedi­
ments in the Norwegian sector of the North Sea
(Bj0rkum & Walderhaug, 1990a).
The transport of significant amounts of dissolved
calcite by convection (Wood & Hewett, 1984) can
probably also be excluded, as convection cells are
usually not active except in special settings with
strongly sloping isotherms, such as adjacent to salt
domes (Bj0rlykke eta!., 1988). This point of view is
supported by the fact that we observe no systematic
tendency for calcite cement to be concentrated near
the base of a sandstone (i.e. a convection cell) and
quartz cement near the top, as one would expect if
convection were controlling cementation (Bj0rlykke
& Egeberg, 1993).
Meteoric water flushing could give rise to much
higher rates of fluid flow than those typical of
compaction-driven flow, but meteoric water would
tend to be undersaturated with calcite and net
removal of calcite from the system would often
result. Moreover, oxygen isotopic and fluid inclu­
sion data suggest that much calcite cementation
takes place at depths below the influence of flowing
groundwater (e.g. Saigal & Bj0rlykke, 1987; Giles et
a!., 1992) (Fig. l ).
The above discussion indicates that the transport
of significant amounts of dissolved calcite into a
sandstone from an external source can normally be
disregarded as an important source of calcite ce­
ment, and that this should be sought for within the
sandstones itself. The most obvious internal source
of calcite cement in shallow marine sandstones is
biogenic carbonate. Biogenic carbonate typically
forms a more or less abundant component in
modern shallow marine sands (e.g. Kumar & Sand­
ers, 1976; Einsele et a!., 1977; Frey & Pinet, 1978)
and is also commonly reported from ancient shal­
low marine sandstones, often in 4 partly dissolved
state (e.g. Fursich, 1982; Hudson & Andrews, 1987;
Bryant et a!., 1988) (Table l ). It should also be
borne in mind that if biogenic carbonate is indeed
the source of calcite cement, this implies that most
or all of the original biogenic carbonate present in
the sandstone has dissolved, and it may be ex­
tremely difficult to determine the former presence
of biogenic carbonate since the siliciclastic grains
move into and fill the space formerly occupied by
the biogenic carbonate (Stephens et a!., 1973) (Plate
l 0, between pp. 182 and 183).
The amount of biogenic carbonate necessary to
explain the observed amounts of calcite cement is
1 84 0. Walderhaug and P.A. Bjorkum
actually less than what one might expect. Typically,
only around a third of a calcite-cemented interval
consists of calcite, and the formation of 1 0% calcite­
cemented intervals would thus only require approx­
imately 3% of the deposited grains to consist of
biogenic carbonate. Moreover, approximately I 0%
of the exposed sedimentary rock record consists of
carbonates that mostly originated as some sort of
biogenic carbonate accumulation (Blatt et a!.,
1 980), and biogenic carbonate can therefore hardly
be regarded as an unusual constituent of sediments.
Furthermore, biogenic carbonate typically consists
of aragonite or high-Mg-calcite that is less stable
than calcite cement (Bathurst, 197 5), and is there­
fore expected to dissolve and provide a source for
calcite cement as burial proceeds. Possibly the very
fine crystal size of some forms of biogenic carbonate
may also contribute to its lack of stability, as
crystals smaller than approximately 2 J.lm are signif­
icantly more soluble than larger crystals because of
surface area effects (Bathurst, 1 97 5; Berner, 1 980).
Carbonate rock fragments could also be a source
of calcite cement if they are available in the source
area and if they survive transport. In the Palaeogene
sandstones of the North Sea we have found inter­
vals containing up to 55 volo/o chalk clasts, for
instance, and carbonate rock fragments have also
been reported from sandstones in other basins (e,g.
Richmann et a!., 1 980; Dickinson, 1 988; Taylor,
1 990). However, carbonate rock fragments are
probably more stable than biogenic carbonate be­
cause of their low-Mg-calcite composition and
coarser crystal size.
It could be argued that a combination of inter­
2
nally derived Ca + from, for instance, plagioclase,
and externally derived C02 could be a viable source
2
of calcite cement. However, if Ca + is supplied
from dissolution of plagioclase contained within the
sandstone, then one is faced with two serious mass
balance problems. First, a very plagioclase-rich
sand is required, as dissolution of one volume of
2 2
plagioclase will only produce the Ca + for a frac­
tion of a volume of calcite, whereas many calcite­
cemented sandstones probably contained very little
or no calcic plagioclase originally. Secondly, one is
also faced with the question of the sink for the other
released components, such as alumina and silica. As
dissolution of one volume of oligoclase (An20) and
precipitation of the dissolved material as calcite
plus albite plus kaolinite will produce only 0.07
volumes of calcite but 0.8 volumes of albite and 0.2
volumes of kaolinite, the volume of diagenetic
silicates would be expected to exceed the volume of
diagenetic calcite by a factor of more than I 0 if
calcite was sourced from this reaction. This implies
that calcite-cemented sandstones should consist of
little other than calcite cement and authigenic
silicates if significant amounts of calcite were
sourced from plagioclase. This is simply not com­
patible with the compositions of most calcite­
2
cemented sandstones (Table 1 ), and Ca + from
plagioclase can therefore hardly be regarded as a
maj or source for calcite cement, although minor
amounts of calcite may probably form during albi­
tization of calcic plagioclase (Boles, 1 982; Morad et
a!., 1 990).
It has also been suggested that volcanic matter
may be an important source of calcite cement in
sandstones (Morad & De Ros, 1 994). However,
although considerable calcite cement may form
during diagenesis of volcanic matter (Maim et a!.,
1 979, 1 984), the source of the calcite in these cases
is typically rather obvious as abundant recognizable
volcanic matter, and authigenic clay minerals are
found together with the calcite (Maim et a!., 1 979,
1984). It is consequently not considered probable
that volcanic matter is a significant source of calcite
except in cases where preserved volcanic matter
and/or associated diagenetic silicates are present
within the calcite-cemented interval.
Analysis of the carbon isotopic composition of
calcite cement has shown that carbon from the
decomposition of organic matter is commonly an
important constituent of calcite cement (e.g. Kan­
torowicz et a!., 1 987; Saigal & Bj0rlykke, 1 987;
Giles eta!., 1 992) (Fig. 1 ), an observation which at
first glance might point to an external source for
calcite after all. However, although the abundance
of C02 in many hydrocarbon reservoirs (Smith &
Ehrenberg, 1 989) suggests that C02 might possibly
migrate as a separate gas phase independent of fluid
flow, influx of C02 into a sandstone will not lead to
precipitation of significant amounts of calcite ce­
ment unless a source of Ca + is present within the
sandstone (Saigal & Bj0rlykke, 1 987). Moreover,
unless the system is buffered by minerals other than
calcite, an influx of C02 should actually cause
calcite dissolution (Hutcheon, 1 983). When bio­
genic carbonate is the only significant source of
2
Ca + in the system, this implies that no matter how
much C02 is transported into the sandstone, the
amount of calcite precipitated is essentially limited
by the amount of biogenic carbonate originally
present. The carbon isotopic composition of calcite
 Calcite cement in shallow marine sandstones
cement may, on the other hand, be dramatically
changed from the values typical of biogenic carbon­
ate (Bj0rkum & Walderhaug, 1 990a). Depending
upon the amount of externally sourced carbonate
ions and their isotopic composition, this mixing
process can give rise to a wide variety of o13Cp08
values, commonly ranging from -30%o to + 1 5%o
(Fig. 1 ). It is thus correct to say that carbon from
organic matter is often an important constituent of
calcite cement in shallow marine sandstones, but its
presence is not a necessary condition for calcite
cementation, and volumetrically and geometrically
it is hard to see that influx of C02 leads to a signifi­
cantly different final result compared with a situation
where no organically derived carbon was incorpo­
rated in the calcite cement.
NUCLEATION A N D G R O WTH
OF CALCITE C E M E N T
I f nucleation o f calcite cement is controlled by the
presence of certain types of biogenic carbonate
which act as favourable nucleation substrates, then
calcite cement nuclei will tend to be concentrated
where most biogenic carbonate is present, i.e.
within biogenic carbonate-rich layers. If nucleation
is determined by calcite supersaturations, nuclei
will still tend to be concentrated in biogenic carbon­
ate-rich layers, as these contain the dominant
source of dissolved calcite, and the supersaturations
necessary for calcite cement nucleation will there­
fore normally be first achieved within these layers.
The nucleation ·of calcite cement may be a result of
the difference in solubility between calcite cement
and carbonate fossils, i.e. the concentration of
dissolved calcite 'in equilibrium with' biogenic
carbonate may exceed the supersaturation neces­
sary for nucleating the less soluble calcite cement.
Once a nucleus has formed it will cause a lower­
ing of the concentration of dissolved calcite in the
pore water close to the nucleus, as the calcite
cement has a lower solubility than the biogenic
carbonate. Biogenic carbonate located around the
nucleus will therefore dissolve, diffuse down the
concentration gradient towards the nucleus, and
precipitate on the nucleus as calcite cement. This
process will continue until all biogenic carbonate is
consumed, unless it is interrupted by factors such as
uplift. Some biogenic carbonate may, however, be
preserved within the calcite cement if it is engulfed
by the growing nucleus/concretion. Biogenic car-
1 85
bonate located very close to the nucleus will have
the greatest chance of being preserved as it may
be rapidly covered by calcite cement, which ex­
plains why the content of biogenic carbonate in
some cases can be seen to decrease outwards from
the centre of calcite-cemented concretions (Wilkin­
son, 1 993).
Typical pore water flow rates in compacting
sedimentary basins are probably in the range of
0.0 1 -0.00 1 cm/yr (Bj0rlykke et al., 1 988), which
implies that diffusional mass transport will domi­
nate for distances less than 1 00 m (Berner, 1980),
and the effect of fluid flow on the described growth
mechanism can therefore normally be disregarded.
The radius of the region around the nucleus where
the concentration of dissolved calcite is lowered is
referred to as the range of influence. This will
increase rapidly until the amount of biogenic car­
bonate dissolved within the range of influence per
unit time equals the amount of calcite cement
precipitated per unit time at the surface of the
nucleus/concretion. After this transient period a
semi-steady state will be set up where the range of
influence increases only very slowly as the concre­
tion grows and biogenic carbonate is consumed.
However, expansion of the range of influence will
often stop when the ranges of influence of neigh­
bouring concretions start to overlap (Fig. 2). The
duration of the transient period is very short com­
pared with the duration of the semi-steady state
(Nielsen, 1 96 1), which suggests that nuclei will tend
to have a spacing greater than the range of influence
after the semi-steady state is established. Although
the transient period may have a relatively short
duration, it probably takes millions to tens of
millions of years to grow large concretions with
diameters of around 1 m, depending upon factors
such as the concentration of biogenic carbonate and
' Wilkinson &
calcite supersaturations (Berner, 1 980;
Dampier, 1 990).
The calcite cement within calcite concretions is
polycrystalline, although calcite crystal size may be
up to several centimetres (Hudson & Andrews,
1 987). This implies that new calcite crystals nucle­
ated on the surface of older crystals during concre­
tion growth, even though calcite supersaturations
would be at a minimum at this location. A possible
solution to this problem may be that dislocations on
the surfaces of the growing crystals acted as nucle­
ation points for new crystals.
1 86 0. Walderhaug and P.A. Bjflrkum
range of maximum overlap no overlap of
influence of range of influence range of influence
"--...
I I
- · · · · · ·-- -- � / '
-r·---
( 2n -- v
6 equilibrium
u
u
�
"' with shells
--
' r;
------ 0
"'- equilibrium with
calcite cement
Distance
G E O M ETRY EXPL AI N E D
Spatial confinement
The strong tendency for calcite cement to be very
heterogeneously distributed within shallow marine
sandstones, i.e. volumes of totally calcite-cemented
sandstone surrounded by sandstone lacking calcite
cement, is an obvious consequence of the presented
growth model. When biogenic carbonate is concen­
trated in layers, calcite cement nuclei will be conce­
trated in these layers. Then the biogenic carbonate­
rich layers will be transformed into calcite-cemented
layers with the same extent as the precursor biogenic
carbonate-rich layer by the dissolution and growth
process described above. Alternatively, if the bio­
genic carbonate is homogeneously distributed in
three dimensions as scattered bioclasts, there is no
. reason for nuclei to be concentrated at certain levels
within the sandstone, and a uniform distribution of
nucleation points may arise. The scattered calcite
cement nuclei will then grow into scattered calcite­
cemented concretions as they tap the surrounding
sandstone for biogenic carbonate. This diagenetic
process will thus transform the originally homoge­
neous sand into calcite-cemented concretions and
intervening sandstone lacking both calcite cement
and biogenic carbonate, unless the process is stopped
at some intermediate stage by, for instance, uplift.
Cases where enough biogenic carbonate is present to
totally calcite cement the sandstone form an excep­
tion, as the end result may then be a homogeneous,
totally calcite-cemented sandstone.
Fig. 2. The interaction of the
ranges of influence for several
calcite cement nuclei. Note that
nucleation of calcite cement may
possibly take place at a dissolved
calcite concentration significantly
lower than the solubility of
biogenic carbonate.
Spacing between concretions
The semi-regular spacing observed between indi­
vidual concretions in some layers of stratabound
concretions (Plate 2) (Pirrie, 1 987; McBride et al.,
199 5), and the tendency for some layers to have
systematically larger spacings than others, can both
be accounted for by the range of influence concept.
The reduced supersaturation within the range of
influence of a nucleus will inhibit the formation of
other nuclei, and each nucleus will therefore tend to
be separated from its nearest neighbours by a
distance greater than the range of influence. How­
ever, the distance to the nearest neighbour is not
likely to exceed two ranges of influence, because
part of the bed would then be outside the range of
influence of any nucleus, and new nuclei would be
expected to form at this location owing to high
calcite supersaturation. According to the model, the
spacing between nuclei and concretions would
therefore have a tendency to be between one and
two ranges of influence.
When the type, concentration and specific surface
area of biogenic carbonate is relatively constant
within a bed, the range of influence may show little
variation, and as the distance between concretions
will typically vary between one and two ranges of
influence, as explained above, a semi-regular spac­
ing between concretions can arise. Similarly, varia­
tions in the type, concentration and/or specific
surface area of biogenic carbonate from bed to bed,
plus variable bed thicknesses, could cause the range
of influence to vary systematically between beds,
 Calcite cement in shallow marine sandstones
which in tum could lead to systematic differences in
concretion spacing between beds. Low biogenic
carbonate concentration, low specific surface area,
stable composition of bioclasts and small thickness
of the biogenic carbonate-rich bed would all in­
crease the range of influence and concretion spac­
ing, as a larger area would have to be tapped to
establish a semi-steady state.
When concretion spacing is variable within a
layer of stratabound concretions (Wilkinson, 1 992),
this may largely be due to an originally laterally
uneven distribution of type and/or concentration of
biogenic carbonate. In addition, some concretions
may nucleate during the transient period before the
range of influence has increased to its semi-steady
state value, leading to a closer spacing between
some of the concretions. Nucleation during the
transient period may actually be most probable
when the range of influence and concretion spacing
are large. The time needed to establish the semi­
steady state is then larger than for a smaller range of
influence, thereby increasing the probability of nu­
cleation prior to the establishment of steady-state
conditions.
The spacing between concretions has so far been
treated as a strictly two-dimensional problem. Field
data show that this treatment is usually appropriate,
as little vertical offset is observed between the
concretions (Plate 2). However, in cases where
biogenic carbonate is scattered throughout a bed
that is relatively thick compared with the range of
influence of each calcite cement nucleus, the
problem of concretion spacing becomes three­
dimensional. In such cases the model predicts that
concretions will still tend to have nearest-neighbour
spacings between one and two ranges of influence,
but nearest neighbours may be located in any
direction, not necessarily horizontally. Stratabound
and scattered concretions can be regarded as the
end-members of a series extending from strata­
bound concretions, via layers where concretions
are still located within a certain bed but with
vertical offsets between their centres, to the end­
member situation where concretions occur scat­
tered throughout a sandstone. According to our
model the first situation will arise when biogenic
carbonate is present as relatively thin layers, the
second when biogenic carbonate layers are thicker,
and the third when biogenic carbonate is scattered
throughout the sand.
1 87
Shape of concretions
Calcite-cemented concretions vary in shape from
perfect spheres (Plate 7) to more flattened forms
with their longest dimensions parallel to bedding
(Plate 2), although concretions with their longest
axes at right-angles to bedding also occur (Plate 3)
(McBride et a/., 1 99 5). The common bedding­
parallel flattening of concretions has repeatedly
been suggested to result from permeability aniso­
tropy causing enhanced fluid flow and more rapid
concretion growth parallel to bedding (e.g. Sorby,
1 908; Deegan, 1 97 1 ; Gluyas, 1984; Dix & Mullins,
1 987). However, unless pore throat radius is less
than a few hundred angstroms, diffusion rates for
ions in sediments are almost independent of perme­
ability (Lerman, 1 979), although proportional to
the inverse of the second power of the tortuosity
(Berner, 1980). Anisotropic tortuosity may there­
fore explain at least part of the flattening of concre­
tions in shales due to systematic bedding-parallel
orientation of platy minerals, but this seems far less
likely in shallow marine sandstones, which typically
consist of roughly equidimensional grains and are
commonly homogenized by bioturbation. A more
plausible explanation seems to be heterogeneous
distribution of biogenic carbonate. If a concretion
nucleated within a biogenic carbonate-rich layer,
then the biogenic carbonate might be totally con­
sumed above and below the concretion long before
the supply was exhausted around the concretion in
bedding-parallel directions. Growth of the concre­
tion could therefore continue in horizontal direc­
tions after vertical growth had terminated. In a
setting with a uniform distribution of biogenic
carbonate, on the other hand, growth would termi­
nate at the same time in all directions, and spherical
concretions would form.
Concretions may also have more \rregular forms
(Plates 3 and 8), and may have their longest axes
systematically oriented in the same direction (Pir­
rie, 1 987; McBride et a/., 1 99 5). In some cases
irregular shapes are clearly the result of the coales­
cence of several concretions (Plates 5 and 8) or
termination of growth against tight laminae. If
biogenic carbonate is unevenly distributed, concre­
tions might also develop protrusions or bumps in
the direction of the greatest supply, a situation
similar to the formation of flattened concretions
except for an asymmetric biogenic carbonate distri­
bution around the growing concretion. Concretions
188 0. Walderhaug and P.A. Bjerkum
Fig. 3. Sketch illustrating how the presence of a short
impermeable lamina can cause a concretion to develop
an irregular form. Successive growth stages are
numbered from I to 6.
that engulf tight laminae with restricted lateral
extent may also develop an irregular shape, which at
first glance might be interpreted as a result of
coalescence of concretions (Fig. 3).
Patchy calcite cementation
Stratabound concretions and continuously
cemented layers
After calcite cement nuclei have formed within a
biogenic carbonate-rich layer, they grow into con­
cretions at the expense of the surrounding biogenic
carbonate. The biogenic carbonate-rich layer will
consequently evolve into a layer of stratabound
concretions. Two possibilities then arise: either the
supply of biogenic carbonate is sufficient for concre­
tions to merge and form a continuous calcite­
cemented layer, or the biogenic carbonate is
exhausted before concretions merge, and the bio­
genic carbonate-rich layer ends up as a layer of
stratabound concretions. This mechanism thus ex­
plains why closely spaced continuously cemented
layers and layers of stratabound concretions com­
monly occur within the same sandstone (Plate 2).
This is also supported by detailed isotopic analysis
of continuously calcite-cemented layers where the
8180 values of the calcite cement were found to
decrease outwards from points located near the
centre of the layers, suggesting merging of concre­
tions (Bj0rkum & Walderhaug, 1993). In addition
to the original concentration of bioclasts, the spac­
ing of nuclei may influence whether a biogenic
carbonate-rich layer evolves into a continuously
cemented layer or not. If nuclei are closely spaced
laterally, less calcite cement will be required to form
a continuously cemented layer, as concretion
diameter and therefore layer thickness will be less
than for a large spacing. Similarly, vertical offset
between nucleation points will reduce the chances
of forming a continuously cemented layer, as more
calcite cement is needed for vertically offset concre­
tions to merge laterally.
Stratabound concretions and continuously ce­
mented layers often show impressive lateral homo­
geneity, i.e. relatively constant thickness and lack of
lateral changes from continuous cementation to
concretionary cementation (Plate 1). This is not
always the case, however, and lateral transitions
between continuous cementation and concretionary
cementation are observed (Plate 5). Such transi-­
tions are to be expected when the concentration of
biogenic carbonate varies laterally with stratabound
concretions forming where least biogenic carbonate
was available, and merging of concretions occurring
where more bioclasts were present.
In systems where fluid flow rates are high, diffusion
will dominate mass transport over much smaller
distances than in stagnant pore water, e.g. approxi-­
mately 1 mm for fluid flow rates of 10 m/yr and
1 em for flow rates of 1 m/yr at a temperature of
2s·c and with a whole rock diffusion coefficient
2
(D.) of 3.2 x w-6 cm /s (Berner, 1980). The range
of influence of a calcite cement nucleus in a sand­
stone with rapidly flowing pore water will thus be
much smaller, and nuclei may therefore have a
much closer spacing than in a sandstone with
stagnant pore water. Therefore millimetre­
centimetre-sized specks of calcite cement with spac­
ings in the millimetre-centimetre range (Plate 4)
may form due to calcite cemynt nucleation taking
place in a relatively rapidly flowing pore water
(Walderhaug & Bj0rkum, 1992). The required flow
rates of several metres per year can normally only
be achieved in settings with meteorically driven
groundwater flow (Bj0rlykke et al., 1988), suggest­
ing that patchy calcite cement may form in such
settings. It is, however, emphasized that patchy
cement is probably the least studied and least
understood of the geometrical modes of calcite
cementation discussed here, and the suggested ex­
planation for its genesis is consequently only tenta­
tive. Patchy calcite cementation may also arise by
other mechanisms. Calcite cement formed as a
 Calcite cement in shallow marine sandstones
result of albitization of scattered calcic plagioclase
clasts (Boles, 1 982; Morad et a/., 1990) commonly
occurs as scattered specks, with each speck located
within and/or around an albitized grain.
P R E DICTING TH E G E O M ET R Y
O F CALCIT E - C E M E N T E D
Z O N E S IN TH E SUBSU R F A C E
The nucleation and growth model presented here
has important practical implications regarding pre­
diction of the geometry of calcite-cemented zones in
hydrocarbon reservoirs. The most important gen­
eral conclusion regarding the geometry of calcite­
cemented zones in shallow marine sandstones is
that their geometry is controlled by the original
distribution of biogenic carbonate within the sand­
stones. If biogenic carbonate is concentrated in
layers, nucleation points will also be concentrated
in these layers, and layers of stratabound concre­
tions or continuous calcite-cemented layers with
lateral extents corresponding to the lateral extents
of the biogenic carbonate-rich layers form. If bio­
genic carbonate is scattered throughout a sand­
stone, nucleation points are also scattered and
scattered concretions form. Prediction of the geo­
metry of calcite cementation therefore largely in­
volves answering questions such as: Was the
biogenic carbonate in a given sandstone present as
biogenic carbonate-rich layers, or was it homoge­
neously distributed? What was the lateral extent of
the biogenic carbonate-rich layers?
It is now some time since these principles were
first applied to the problem of predicting the geom­
etry of calcite-cemented zones (Walderhaug et a/.,
1 989; Bj0rkum & Walderhaug, 1 990a,b), and one
may ask whether any successful predictions have
been made. Fortunately, subsequent drilling has
made it possible to test the predictions made con­
cerning the extent of calcite-cemented zones in the
Fensfjord Formation of the Brage Field (Walderhaug
et a/., 1 989) and in the Rannoch Formation (Bj0r­
kum & Walderhaug, 1990b).
In the case of the Jurassic Fensfjord Formation
four wells were studied by Walderhaug et a/. ( 1 989),
and more than 40 calcite-cemented intervals were
encountered. Most of these were interpreted as
belonging to short layers or concretions; only two
calcite-cemented layers were predicted to have
field-wide lateral extents, i.e. 6 km or more (Fig. 7
in Walderhaug et a/., 1 989). The potential great
189
lateral extent of these two layers was largely sug­
gested on the basis of their appearance in cores,
which suggested that they had formed from bio­
genic carbonate-rich layers accumulated during pe­
riods of low siliciclastic input, events that might
have affected large areas. In addition, field studies
had shown that the same type of calcite-cemented
layers had several kilometres lateral extent in out­
crops. Data are now available from 25 wells in the
Brage Field, and calcite-cemented intervals are
present at one of the predicted levels in all of these
and in 20 wells at the second predicted level,
suggesting that the two calcite-cemented layers do
in fact extend across most of the Brage Field.
Bj0rkum & Walderhaug ( 1990b) suggested that
calcite-cemented intervals located within low angle­
or hummocky-laminated shoreface sandstones of
the Jurassic Rannoch Formation are part of concre­
tions. Concretions were predicted as the sediments
probably originally contained scattered biogenic
carbonate rather than distinct biogenic carbonate­
rich layers, that is, calcite nuclei would occur rather
uniformly in three dimensions and scattered
concretions would result. In addition, calcite ce­
mentation in exposures of hummocky-laminated
sandstones was found to be concretionary (Plate 8).
Drilling of sidetracked wells in the Rannoch Forma­
tion of the Gullfaks Field has subsequently con­
firmed that cementation is concretionary. Calcite­
cemented intervals present in the original vertical
well were not encountered in the sidetracked well
only a few metres to the side of the first well. In
addition, cases occur where tight calcite-cemented
intervals are registered on wireline logs but not at
the same depth in cores, indicating that calcite­
cemented zones terminate very close to the well, an
observation most easily explained by the presence
of concretions.
Lastly, studies of outcrops indicate that assuming
a systematic correlation between thickness and
lateral extent does not seem to be a successful
method for predicting the lateral extent of calcite­
cemented zones (Fig. 4).
C O N CLUSIO N S
Calcite cement in shallow marine sandstones is nor­
mally derived from biogenic carbonate contained
within the sandstones, although carbonate ions may
also be supplied by C02 from the decomposition of
organic matter. The formation of continuously
190 0. Walderhaug and P.A. Bjorkum
1 0000 ,-------�
- · ·· 0 Bencliff Grit
e Bridport Sands
1 000 + Valtos Fm, Skye
6. Bearreraig Fm
/). D Valtos Fm, Eigg
+ geothermometer. Geology, 18, 1 003- 1 006.
�
.§_ 1 00 D BATHURST, R.G.C. ( 1 97 5 ) Carbonate Sediments and their
.s::.
c,
c 10
Q)
..J
+
+ proach. Princeton University Press, Princeton, NJ.
, 1 +-----.--.--�
0 2 3 4
T h i c kness (m)
Fig. 4. Lateral extent versus thickness for 4 1 3
calcite-cemented concretions and layers in Jurassic
shallow marine sandstones exposed onshore in England
and Scotland.
calcite-cemented layers, layers of stratabound con­
cretions and scattered concretions can be explained
by local diffusional redistribution of the biogenic
carbonate. Stratabound concretions and continu­
ously cemented layers both form from biogenic
carbonate-rich layers, whereas scattered concretions
form from scattered biogenic carbonate. Continu­
ously cemented layers form from stratabound con­
cretions when enough biogenic carbonate is present
to allow concretions to merge. The growth of flat­
tened concretions is probably a result of a greater
supply of biogenic carbonate in the directions of
flattening, and need not have anything to do with
anisotropic permeability and fluid flow. Prediction
of the geometry of calcite-cemented zones in the sub­
surface should be based on an understanding of the
original distribution of biogenic carbonate within
the sandstone, knowledge of the nucleation and
growth mechanisms for calcite cement, and data
from analogous outcrops.
A C K N O WLE D G E M E N T S
The authors wish to thank Tom Dreyer and Eirik
Graue for providing data concerning calcite cemen­
tation in the Brage and Gullfaks Fields. The manu­
script was also improved by the comments of Niek
Molenaar, Sadoon Morad and Karl Ramseyer.
REFERENCES
BARKER, C.E. & GoLDSTEIN, R.H. ( 1 990) Fluid-inclusion
technique for determining maximum temperature in
calcite and its comparison to the vitrinite reflectance
Diagenesis, 2nd edn. Elsevier, Amsterdam.
BELL, W.T. ( 1 90 1 ) The remarkable concretions of Ottawa
County, Kansas. Am. J. Sci. , 1 1, 3 1 5-3 1 6 .
+
BERNER, R.A. ( 1 980) Early Diagenesis-a Theoretical Ap­
BJ0RKUM, P.A. & WALDERHAUG, 0. ( 1 990a) Geometrical
arrangement of calcite cementation within shallow ma­
rine sandstones. Earth Sci. Rev., 29, 1 45- 1 6 1 .
BJ0RKUM, P.A. & WALDERHAUG, 0. ( 1 990b) Lateral extent
of calcite-cemented zones in shallow marine sandstones.
In: North Sea Oil and Gas Reservoirs-Il (Eds Buller,
A.T., Berg, E., Hjelmeland, 0., et a!.), pp. 3 3 1 -336.
Graham & Trotman, London.
BJ0RKUM, P.A. & WALDERHAUG, 0. ( 1 99 3) Isotopic com­
position of a calcite-cemented layer in the Lower Juras­
sic Bridport Sands, southern England: implications for
formation of laterally extensive calcite-cemented layers.
J. sediment. Petrol. , 63, 678-682.
BJ0RLYKKE, K. & EGEBERG, P.K. ( 1 993) Quartz cementa­
tion in sedimentary basins. Bull. Am. Ass. Petrol. Geol. ,
77, 1 5 38- 1 548.
BJ0RLYKKE, K., Mo, A. & PALM , E. ( 1 988) Modelling of
thermal convection in sedimentary basins and its rele­
vance to diagenetic reactions. Mar. Petrol. Geol., 5,
338-35 1 .
BLATT, H., MIDDLETON, G. & MuRRAY, R. ( 1 980) Origin of
Sedimentary Rocks. Prentice-Hall, Englewood Cliffs,
NJ.
BOLES, J.R. ( 1 982) Active albitization of plagioclase, Gulf
Coast Tertiary. Am. J. Sci., 282, 1 65-1 80.
BRYANT, l.D., KANTOROWICZ, J.D. & LOVE, C. F. ( 1 988) The
origin and recognition of laterally extensive continuous
carbonate-cemented horizons in the Upper Lias Sands
of southern England. Mar. Petrol. Geol., 5, I 08- 1 33.
DAVIES, D.K. ( 1 967) Origin of friable sandstone-calcar­
eous sandstone rhythms in the Upper Lias of England. J.
sediment. Petrol. , 37, 1 1 79- 1 1 88 .
DEEGAN, C .E . ( 1 9 7 1 ) The mode o f origin o f some late
diagenetic sandstone concretions from the Scottish Car­
boniferous. Scot. J. Geol., 7, 3 5 7-365.
'
DICKINSON, W.W. ( 1 988) Isotopic and petrographic evi­
dence for carbonate diagenesis in nonmarine sand­
stones, Green River Basin, Wyoming. J. sediment.
Petrol. , 58, 327-338.
DIX, R.G. & MULLINS, H.T. ( 1 987) Shallow, subsurface
and burial alteration of Middle Devonian calcite con­
cretions. J. sediment. Petrol., 57, 1 40- 1 52.
EINSELE, G., ELOUARD, P., HERM, D., K0GLER, F.C. &
SCHWARZ, H.U. ( 1 977) Source and biofacies of late
Quaternary sediments in relation to sea level in the shelf
off Mauritania, West-Africa. 'Meteor' Forschungsergeb­
nisse, Reihe C, 26, 1 -43.
FREY, R.W. & PINET, P.R. ( 1 978) Calcium-carbonate
content of surficial sands seaward of Altamaha and
Doboy Sounds, Georgia. J. sediment. Petrol. , 48, 1 249-
1 256.
 Calcite cement in shallow marine sandstones
FDRSICH, F.T. ( 1 982) Rhythmic bedding and shell bed
formation in the Upper Jurassic of east Greenland. In:
Cyclic and Event Stratification (Eds Einsele, G. &
Seilacher, A.), pp. 208-222. Springer-Verlag, Berlin.
GIBBONS, K., HELLEM, T., .K.IEMPERUD, A., NIO, S.D. &
VEBENSTAD, K. ( 1 993) Sequence architecture, facies
development and carbonate-cemented horizons in the
Troll Field reservoir, offshore Norway. In: Advances in
Reservoir Geology (Ed. Ashton, M.). Spec. Pub!. Geol.
Soc. Lond., 69, 1-3 1 .
GILES, M.R., STEVENSON, S., MARTIN, S.V. et a!. ( 1 992) The
reservoir properties and diagenesis of the Brent Group:
a regional perspective. In: Geology of the Brent Group
(Eds Morton, A.C., Haszeldine, R.S. Giles, M.R. &
Brown, S.). Spec. Pub!. Geol. Soc. Lond., 61, 289-327.
GLUYAS, J.G. ( 1 984) Early carbonate diagenesis within
Phanerozoic shales and sandstones of the NW European
shelf. Clay Miner. , 19, 309-3 2 1 .
HUDSON, J.D. & ANDREWS, W.J. ( 1 987) The diagenesis of
the Great Estuarine Group, Middle Jurassic, Inner
Hebrides, Scotland. In: Diagenesis of Sedimentary Se­
quences (Ed. Marshall, J.D.). Spec. Pub!. Geol. Soc.
Lond., 36, 259-276.
HuTCHEON, I. ( 1 983) Aspects of the diagenesis of coarse­
grained siliciclastic rocks. Geosci. Can., 10, 4- 1 4.
KANTOROWICZ, J.D., BRYANT, l.D. & DAWANS, J.M. ( 1 987)
Controls on geometry and distribution of carbonate
cements in Jurassic sandstones: Bridport Sands, south­
em England and the Viking Group, Troll Field, Norway.
In: Diagenesis ofSedimentary Sequences (Ed. Marshall,
J.D.). Spec. Pub!. Geol. Soc. Lond., 36, 1 03- 1 1 8.
KuMAR, N. & SANDERS, J.E. ( 1 976) Characteristics of
shoreface storm deposits: modem and ancient exam­
ples. J. sediment. Petrol. , 46, 1 45-1 62.
LERMAN, A. ( 1 979) Geochemical Processes: Water and
Sediment Environments. Wiley, New York.
MALM, O.A., FuRNES, H. & BJ0RLYKKE, K. ( 1 979) Volca­
niclastics of Middle Jurassic age in the Statfjord oil-field
of the North Sea. N Jb. Geol. Palaont. Mh. , 10, 607-
618.
MALM, O.A., BRUUN CHRISTENSEN, 0., FURNES, H . et a/.
( 1 984) The Lower Tertiary Balder Formation: an orga­
nogenic and tuffaceous deposit in the North Sea region.
In: Petroleum Geology of the North European Margin
(Ed. Spencer, A.M.), pp. 1 49- 1 70. Graham & Trotman,
London.
MCBRIDE, E.F., MILLIKEN, K.L., CAVAZZA, W. et a/. ( 1 995)
Heterogeneous distribution of calcite cement at the
outcrop scale in Tertiary sandstones, northern Apenni­
nes, Italy. Bull. Am. Ass. Petrol. Geol. , 79, 1 044- 1 063.
MORAD, S. & D E Ros, L.F. ( 1 994) Geochemistry and
diagenesis of stratabound calcite cement layers within
the Rannoch Formation of the Brent Group, Murchison
Field, north Viking Graben (northern North Seal­
comment. Sediment. Geol., 93, 1 35- 1 4 1 .
MORAD, S., BERGAN, M., KNARUD, R . & NYSTUEN, J.P.
( 1 990) Albitization of detrital plagioclase in Triassic
reservoir sandstones from the Snorre Field, Norwegian
North Sea. J. sediment. Res. , 60, 4 1 1 -425.
NIELSEN, A. E. ( 1 96 1 ) Diffusion controlled growth of a
moving sphere. The kinetics of crystal growth in potas­
sium perchlorate precipitation. J. Phys. Chern. , 65,
46-49.
191
PETTIJOHN, F.S., POTTER, P.E. & SIEVER, R. ( 1 973) Sand
and Sandstone. Springer-Verlag, New York.
PIRRIE, D. ( 1 987) Oriented calcareous concretions from
James Ross Island, Antarctica. Bull. Br. Antarctic Surv. ,
75, 4 1 -50.
RICHMANN, D.L., MILLIKEN, K.K., LOUCKS, R.G. & DODGE,
M.M. ( 1 980) Mineralogy, diagenesis, and porosity in
Vicksburg sandstones, McAllen Ranch Field, Hidalgo
County, Texas. Gulf Coast Ass. Geol. Soc. Trans., 30,
473-48 1 .
SAIGAL, G.C. & BJ0RLYKKE, K. ( 1 987) Carbonate cements
in clastic reservoir rocks from offshore mid Norway­
relationship between isotopic composition, textural de­
velopment and burial depth. In: Diagenesis ofSedimen­
tary Sequences (Ed. Marshall, J.D.). Spec. Pub!. Geol.
Soc. Lond., 36, 3 1 3-324.
SCHOEWE, W.H., HORR, W.H., BURT, C.E. & WOOSTER,
L.D. ( 1 937) Symposium on the geology, flora and fauna
of 'Rock City', a proposed national monument in
Ottawa County, Kansas. Kansas Acad. Sci. Trans. , 40,
1 79-20 1 .
SMITH, J.T. & EHRENBERG, S.N. ( 1 989) Correlation of
carbon dioxide abundance with temperature in clastic
hydrocarbon reservoirs: relationship to inorganic chem­
ical equilibrium. Mar. Petrol. Geol. , 6, 1 29- 1 35.
SORBY, H.C. ( 1 908) On the application of quantitative
methods to the study of the structure and history of
rocks. Q. J. Geol. Soc. Lond. , 64, 1 7 1 -233.
STEPHENS, D.G., EASON, J.E. & P EDLOW , G.W. ( 1 973)
Dissolution of shell material with no disruption of
primary sedimentary structures. J. sediment. Petrol., 43,
6 1 8-620.
SUNDAL, L., KOSSACK, C. A. & KLEPPE, J. ( 1 990) The effect
of low-permeability layers on oil production from ver­
tical and horizontal wells in the Troll Field. In: North
Sea Oil and Gas Reservoirs-// (Eds Buller, A.T., Berg,
E., Hjelmeland, 0. et al.) pp. 2 1 3-22 1 . Graham &
Trotman, London.
SwiNEFORD, A. ( 1 94 7) Cemented sandstones of the Dakota
and Kiowa Formations in Kansas. State Geol. Surv.
Kansas Bull. , 70, 5 3- 1 04.
TAYLOR, T.R. ( 1 990) The influence of calcite dissolution
on reservoir porosity in Miocene sandstones, Picaroon
Field, offshore Texas Gulf Coast. J. sediment. Petrol. ,
60, 322-334.
TAYLOR, T.R. & SouLE, C.H. ( 1 993) Reservoir character­
ization and diagenesis of the Oligocene 64-Zone Sand­
stone, North Belridge Field, Kern County, California.
Bull. Am. Ass. Petrol. Geol. , 77, 1 549- 1 566.
WALDERHAUG, 0. ( 1 990) A fluid inclusion study of quartz­
cemented sandstones from offshore mid-Norway­
possible evidence for continued quartz cementation
during oil emplacement. J. sediment. Petrol., 60, 203-
2 1 0.
WALDERHAUG, 0. ( 1 994) Temperatures of quartz cemen­
tation in Jurassic sandstones from the Norwegian con­
tinental shelf-evidence from fluid inclusions. J.
sediment. Res., A64, 3 1 1 -323.
WALDERHAUG, 0. & BJ0RKUM, P.A. ( 1 992) Effect of mete­
oric water flow on calcite cementation in the Middle
Jurassic Oseberg Formation, well 30/3-2, Veslefrikk
Field, Norwegian North Sea. Mar. Petrol. Geol. , 9,
308-3 1 8.
1 92 0. Walderhaug and P.A. Bjorkum
WALDERHAUG, 0., 8J0RKUM, P.A. & NORDGARD BoLAS,
H.M. ( 1 989) Correlation of calcite-cemented layers in
shallow marine sandstones of the Fensfjord Formation
in the Brage Field. In: Correlation in Hydrocarbon
Exploration (Ed. Collinson, J.D.), pp. 367-3 75. Gra­
ham & Trotman, London.
WALDERHAUG, 0, PRESTHOLM, E. & 0XNEVAD, l.E.l. ( 1 995)
Geometry of calcite cemented zones in shallow marine
sandstones. In: PROFIT Project Summary Reports (Eds
Olsen, J., Olaussen, S., Jensen, T.B., Landa, G.H. &
Hinderaker, L.), pp. 75-90. Norwegian Petroleum Di­
rectorate, Stavanger.
WILKJNSON, M. ( 1 992) Concretionary cements in Jurassic
sandstones, Isle of Eigg, Inner Hebrides. In: Basins on
the Atlantic Seaboard: Petroleum Geology, Sedimento­
logy and Basin Evolution (Ed. Parnell, J.). Spec. Pub!.
Geol. Soc. Lond., 62, 145-1 54.
WILKJNSON, M. ( 1 993) Concretions of the Valtos Sand­
stone Formation of Skye: geochemical indicators of
palaeo-hydrology. J Geol. Soc., Lond., 150, 57-66.
WILKJNSON, M. & DAMPIER, M.D. ( 1 990) The rate of
growth of sandstone-hosted calcite concretions.
Geochim. Cosmochim. Acta, 54, 3 3 9 1 -3399.
Wooo, J.R. & HEWETT, T.A. ( 1 984) Reservoir diagenesis
and convective fluid flow. In: Clastic Diagenesis (Eds
McDonald, D.A. & Surdam, R.C.). Mem. Am. Ass.
Petrol. Geol., 37, 99- 1 1 0.
Spec. Pubis int. Ass. Sediment. ( 1 998) 26, 1 93-2 1 1

Origin of low-permeability calcite-cemented lenses

in shallow marine sandstones and CaC03 cementation mechanisms:
an example from the Lower Jurassic
Luxemburg Sandstone, Luxemburg

N. M O LENAAR
Department of Geology and Geotechnical Engineering (!GG),
Technical University of Denmark (DTU), Building 204, 2800 Lyngby, Denmark,
e-mail iggnm@pop.dtu.dk

ABSTRACT

Calcite-cemented layers and lensoid concretions commonly form low-permeability barriers in shallow
marine reservoir sandstones. In the porous and permeable Lower Jurassic Luxemburg Sandstone such
calcite-cemented lenses form permeability barriers with lateral continuities of a few decimetres to
hundreds of metres. Deposition of these sandstones (�90 m thick) occurred in a wave- and
storm-reworked tidal delta that formed where a seaway through the Ardennes and Rhenish Massifs
entered the shallow Paris basin.
The main cause of tightly cemented layers and lenses is differential early marine calcite/aragonite
cementation. Early cementation took place a few decimetres below the sea floor within the uppermost
sediment layers, where cementing materials were supplied by sea water flowing through the sand.
Occasionally, early lithified layers were exposed at the sea floor after erosion, and typical hardground
features such as borings and encrustations of fauna developed. Precipitation of early marine cement
was controlled by the carbonate grain content and texture of the sand. Cementation began in permeable
structures of the sand which had elevated carbonate grain content. The intensity of early cementation
and the eventual lateral extent of cemented layers were dependent on sedimentation rate as a function
of intermittent storm deposition and reworking. Early diagenesis and the distribution of calcite­
cemented lenses are thus controlled by sedimentary facies.
The local presence of early cement decreased permeability and constrained the flow of pore water
and later diagenetic processes such as dissolution or replacement of carbonate grains and poikilotopic
or blocky calcite cementation, demonstrating their dependence on hydraulic flow. Dissolution of
carbonate grains caused the development of extensive secondary porosity in the host rock. Replace­
ment of metastable frameworks and early diagenetic carbonates by calcite and sparry calcite
cementation took place almost exclusively in the early calcite-cemented lenses. As a result, the initial
differences in detrital mineralogy and early diagenesis of lenses and host rocks were further enhanced
during later diagenesis.

INTRODUCTION

Primary porosity and permeability are mainly a
function of grain size and sorting, and thus reflect
sedimentation processes and hydrodynamic condi­
tions during deposition. This primary signal may,
however, be modified by postdepositional changes.
For instance, bioturbation may affect texture as well
as composition by mixing different grain-size pop­
ulations, by introducing fine-grained matrix and
bioclastic grains, and by changing grain packing and
orientation. Such modifications are partly predict­
able because they are related to specific depositional
environments. Diagenetic alterations can also sig­
nificantly affect porosity and permeability. Cemen­
tation, compaction and dissolution are important
diagenetic processes which modify primary poros­
·
ity and permeability patterns in sandstones and

Carbonate Cementation in Sandstones Edited by Sadoon Morad 193

© 1998 The International Association of Sedimentologists
ISBN: 978-0-632-04777-2
194 N. Molenaar
thus control reservoir properties. Because the main
factors controlling diagenesis are not completely
understood, diagenetic changes in reservoir proper­
ties remain largely unpredictable.
The degree of calcite cementation of most marine
sandstones is heterogeneous. Commonly, such het­
erogeneity is expressed as tightly calcite-cemented
layers or concretions occurring within less ce­
mented sandstone (e.g. Chafetz, 1979; Saigal &
Bj0rlykke, 1987; Walderhaug & Bj0rkum, 1989;
Molenaar & Martinius, 1990). The differential na­
ture of cementation, and the resulting heteroge­
neous distribution of permeability, may reduce the
degree of hydrocarbon recovery. Low-permeability,
laterally extensive calcite-cemented layers inhibit
fluid flow by compartmentalizing potential reser­
voir bodies (e.g. Kantorowicz et a!., 1987; Prosser et
a!., 1993). Also, smaller lenses and concretions act
as baftles and may cause poor recovery. In order to
understand fluid flow through sandstone reservoir
bodies and enhance oil recovery, it is therefore
necessary to understand and predict the distribu­
tion of low-permeability carbonate-cemented lenses
and layers.
The present study concentrates on diagenetic
permeability barriers in siliciclastic carbonate
shallow marine sandstones of the Lower Jurassic
Luxemburg Sandstone (Fig. 1). The formation was
studied and sampled at a number of locations in
Luxemburg, the southeast of Belgium and the north
of France. Most of the sandstones are poorly lithi­
fied. Within these friable host rocks, tightly calcite­
cemented (whitish) sandstones commonly occur in
layers or lenses (Fig. 1). The principal intentions are
to analyse the origin of these diagenetic features and
to understand their distribution.
METHODS
Thin sections from 350 samples were studied by
standard polarized light microscopy. Samples were
impregnated with oil-blue A-stained resin before
thin sectioning. The detrital and diagenetic compo­
nents in 38 samples were quantified by point count­
ing (600-750 points/thin section). Selected samples
were studied by cathodoluminescence microscopy
(polished thin sections). Grain-size analysis of 47
samples, disintegrated by HAc dissolution of calcite
and by ultrasonic treatment, was carried out by a
Malvern laser particle sizer. The <20 Jlm size frac­
tion of 20 samples was used for mineralogical
identification of detrital and authigenic clay miner­
als. Clay minerals were identified by X-ray diffrac­
tometry with CuKu radiation and a nickel filter
using oriented glass-mounted samples (untreated,
glycol saturated and 55o·c heated) and by high­
resolution thermogravimetric analysis of untreated
samples. Carbonate contents were determined by
standard C02 titration. The permeability and po-
Fig. 1. Photograph of a sampled
interval near Luxemburg city along a
1 . 5 km long east-west section (parallel
to direction of progradation) of the
Luxemburg-Trier highway near
Findel. The height of the outcrop is
5 m (see 30 em hammer for scale). In
fresh outcrops the differential
cementation patterns are clearly
visible because of colour differences.
The well lithified lenses form 1 5% of
the sandstones in this location and
appear whitish amongst the darker
host sandstone. The tectonic dip is
towards the SW (to the left of the
photograph). Cosets or bed forms are
composed of low-angle beds forming
laterally extensive lenticular bodies
dipping to the SW. Sedimentation was
dominated by storm deposition and
reworking. The whole delta succession
(approximately 20 m thick) prograded
towards the SW on fossiliferous
bioturbated marls (the Lorraine Marl).
 Low-permeability lenses in shallow marine sandstones 195

rosity of 12 samples were determined by standard
methods, nitrogen permeametry and helium expan­
sion. The stable oxygen and carbon isotopic compo­
sitions of carbonates in 23 bulk samples were
determined. Bulk samples were used because ce­
ment fringes are usually too thin to allow for
separation between fringe cement and late cement.
Carbon dioxide, liberated from 4-12 mg samples
by reaction in vacuum with 100% H3P04 for 44 h
at 2s·c, was analysed on a Micromass 602C mass
spectrometer. After standard method corrections
the results are expressed as %a deviation from the
PDB standard with a reproducibility of 0.1 and
0.05%o for oxygen and carbon, respectively. Com­
positional data were obtained by wavelength disper­
sive electron-microprobe measurements (opera­
tional conditions: spot size 10 Jlm, acceleration
= and interfinger with, a fossiliferous fine-grained
voltage 15 kV, specimen current 50 nA). Detec­
= =
tion limits were around 250 ppm. Sr and Na con­
tents were below the detection limit.
GEOLOGICAL SETTING
Marine sedimentation commenced in the Paris
basin in the middle-late Triassic. During the early
Jurassic the basin was connected to a northern sea
by a seaway (the Eifel depression) between the
topographically high areas of the Ardennes and
Eifel-Hunsnlck (Rhenish) Massifs (Bock, 1989).
The Luxemburg Sandstone (Fig. 2) was deposited in
an embayment (the Gulf of Luxemburg) at the
southern end of this seaway. The formation lies
over the Brabant-Ardennes Massif to the northeast.
The formation has an early Hettangian to early
Sinemurian age (Early Jurassic) in most of central
and eastern Luxemburg, and a Sinemurian age at
the southern Ardennes margin in Belgium and the
adjacent part of France (Fig. 2) (Berners et a!.,
1985; Guerin-Franiatte & Muller, 1987; Muller,
1987).
The Luxemburg Sandstone consists of several
coarsening-upwards quartzose sandstone bodies
that are laterally and vertically stacked to a cumu­
lative thickness of as much as 80-90 m in the
intermediate area near Luxemburg. The sandstone
bodies are bounded towards the south and west by,
facies of the Lorraine marl, consisting of coarse silty
to very fine sandy dark greyish marls with marly
sandstone layers (Fig. 2). Marine fossils in this
fine-grained, intensely bioturbated faces indicate
an open shallow marine environment, as do the
usually transported fossils in the sandstones (Bock
& Muller, 1989). Along the northern coastline,
bordering the Brabant-Ardennes Massif, a more
calcareous facies occurs, with shell beds and small
coral patchreefs (the Arreux limestones; Fig. 2)
(Bock, 1989).
The sandstones are well sorted and have a vari­
able carbonate grain content (0-45% of the frame­
work grains). The grain size decreases from medium

Ammonite Zone S. Belgium Luxemburg

Raricostatum

Oxynotum
z
<(
0:: Obtusum
::J
:::<
w
z
(f) U5
<( BuckIandi
::::;

Rotiforme

z Angulata
<(
(5
z Liasicus

�w Planorbis
:r:

RHAETIAN
Mortinsart Sandstone::::::-··

Fig. 2. Stratigraphic scheme for the Rhaetian, Hettangian and Sinemurian in a east-west transect from Luxemburg to
the north of France. After M uller ( 1 987) and Bock ( 1 989).
196 N. Molenaar
in the proximal area in the north of Luxemburg to
fine-grained sandstones in the intermediate area,
and finally to siltstones in the distal western area in
France. Occasionally, conglomeratic intervals and
pebbly sandstones with quartz granules-pebbles
occur in the proximal and intermediate areas (e.g.
Muller & Rasche, 1971; Berners, 1985). Bioturba­
tion is found mainly in poorly lithified sandstone,
but well lithified sandstone usually contains a few
isolated burrows. Bedding surfaces with clay lami­
nae show more abundant bioturbation, reflecting
the Cruziana ichnofacies in the proximal areas, and
the Skolithos ichnofacies in the distal (Guerin­
Franiatte et at., 1991).
The sandstones were deposited in a tidal delta
complex which prograded from the mouth of the
seaway toward the southwest (Berners, 1983, 1985;
Mertens et al., 1983). Tidal flow was constrained by
the relatively narrow seaway in the proximal north­
eastern area. Here, deposition occurred mainly by
tidal currents flowing towards the SW and minor
countercurrents towards the NE (Berners, 1983,
1985; Mertens et al., 1983). Tidal channels are as
much as several metres deep. Sandstones deposited
by tidal currents show cosets of large-scale, high­
angle to low-angle dipping beds (which resulted
from migrating transverse dunes) and medium­
scale high-angle, and low-angle cross-bedded cosets.
The latter cross-beds have tangential bottomsets
with clay drapes and double clay layers, and display
internal reactivation surfaces typical of tidal envi­
ronments. Foresets of accretionary and lateral ac­
cretionary origin often vary cyclically in thickness,
reflecting neap-spring variations in tidal force
(see Visser, 1980). Topsets commonly have been
eroded. These features are characteristic of a tidally
dominated depositional system with deposition by
migrating dunes in laterally shifting channels. In
this proximal area, where sedimentation rates were
high and the degree of reworking was large due to
erosion and channel cutting, tightly cemented sand­
stones are absent or scarce.
More basinward, the influence of the tidal cur­
rents decreased and the transport and deposition of
fine sand by waves and storms became increasingly
important (Berners, 1985). The sandstone bodies
consist of large-scale (hundreds of metres) low-angle
foresets that generally dip towards the SW. Storm
deposits built SW-prograding bars consisting of
large-scale low-angle bedding with laterally exten­
sive toesets (Fig. 3A). These storm-built shelf sand
ridges show similarities with those described by Carr
& Scott (1990). The low-angle beds consist mainly
of laminated or cross-bedded and hummocky cross­
bedded sandstone with a ripple-laminated upper
part that has wavy clay bedding. The ripple and
wavy bedded upper part can be caused by deposi­
tion under waning flow conditions, but can also be
the result of wave reworking. In the intermediate
area, where the delta complex is storm and wave
dominated, lenses are abundant (Berners, 1985).
Extensive coarse-grained amalgamated erosive
beds (up to 1.5 m thick) with hummocky cross­
stratification and shallow channels occur in the
upper part of the sandstone bodies. Beds and
channel fills are commonly coarse grained and have
lag deposits of quartz granules and pebbles, coarse­
grained carbonate shell material and intraforma­
tional clasts ranging from granule to cobble size (e.g.
Muller & Rasche, 1971). The intraformational
clasts are derived from reworked early lithified
lenses and hardgrounds (Fig. 3B). The coarse­
grained deposits are interpreted as amalgamated
storm beds and storm-surge channels, and occur at
the top of depositional sequences. The conglomer­
atic deposits are interpreted either as lag deposits
from winnowing during storms, or as accumula­
tions of offshore transport of bioclastic material
from shell beds along the coast and on abandoned
parts of the tidal delta (e.g. Aigner, 1982).
Further to the west, along the coast of the Ar­
dennes Massif, longshore sand bars occur with low­
to high-angle cross-bedding, indicating longshore
transport directed towards the west (Fig. 3C). Reac­
tivation surfaces and clay drapes are consistent with
tidal influence. The bars are several metres thick
and tens of metres long. Tightly calcite-cemented
lenses also occur in these deposits.
Extensive, thin storm-deposited silty-sandy
sheets interbedded with fine-grained marls occur in
areas distal to the tidal delta (BocJ<, 1989). Alternat­
ing storm beds and silty fossiliferous marls, both
intensely bioturbated, form rhythmically bedded
successions (Fig. 3D). The grain size of the coarser­
grained beds decreases in a distal direction (Bock,
1989).
DESCRIPTION OF
CALCITE-CEMENTED LENSES
AND LAYERS
The degrees of cementation, porosity and perme­
ability of lenses and host rocks are distinctly dif-
 Low-permeability lenses in shallow marine sandstones 197

Fig. 3. Examples of several facies in the Luxemburg Sandstone. (A) Facies typical for the intermediate area with
storm- and wave-dominated deposition. The low-angle bedding is clearly visible in the upper part of the outcrop,
which is 5 m high. Locality: Findel (Luxemburg). (B) Bored intraformational clasts (granule-cobble size). The clast in
the centre of the photograph is 10 em long. Locality: Brouch (Luxemburg). (C) Example of longshore bars. Outcrop is
approximately 4 m high. Locality: Pin (Belgium). (D) Outcrop in the distal area with an 8 m thick alternation of soft,
fine silty marls and slightly more lithified siltstones, the latter representing periods with more frequen,t storm
deposition. Both lithologies are intensely bioturbated (Skolithos ichnofacies). Locality: Romery (France).

ferent (Table 1; Fig. 4). Host rocks are highly The calcite-cemented lenses are stratabound and
permeable (averaging 1.2 D; n 5) and porous
= parallel to the large-scale, low-angle bedding or to
(averaging 24.5%). The lenses are diagenetic flow coset boundaries of storm-built ridges and tidal­
barriers with a permeability around 0. 1 mD (n 7) = delta lobes. In addition, smaller lenses occur as
and a porosity of 7.9%. In fresh outcrops lenses are tightly calcite-cemented foresets of metre-scale
whitish due to a higher average carbonate content dunes, lateral-accretion sets and channel lags. Such
(44.6% CaC03; n 16) whereas the carbonate-poor
= small lenses are concordant to, or partly coincide
host rocks (on average 10.1% CaC03; n 13) are = with, primary sedimentary structures. Most of the
yellowish. lenses are discontinuous, with a lateral continuity of
198 N. Molenaar

Table I. Composition of lenses and host rocks as 32 "I "'I "I "I "I=
determined by point counting thin sections (outcrop
Findel, Fig. 1). Mean values and standard deviations are host rocks
shown. The compaction is estimated by assuming a 28

primary porosity of 40%

6L:s.
24 6
Property Lenses Host rocks �
0

� 6
6
Siliciclastic grains 41.9 (5.3) 59.6 (5.0) 'iii
0 20
Carbonate grains 17.1 (4.9) 4.8(2.8) ...
0 6
Fringe cement 8.5 (7.2) c.
Quartz cement 0.9(1.4) 6.6 (1.7) 16
E
Late calcite cement 26.8 (7.5) 7.4(5.1) ::I
Total carbonate cement 35.3(3.7) 7.4(5.1) a; lenses
Primary pores 10.5(4.2) ..c::: 12 ...
Mouldic pores 3.7(2.9) 0.6(1.3)
Oversized pores 7.1(3.9) ...
Total pores 3.7(2.9) 18.2(5.1) 8

Total clay (matrix+ authigenic) 1.1(1.3) 3.4(3.2) ......

Total cement+ pores 37.3(3.7) 27.9(4.2) .,[ ,I ..I ..I ,I
4 p.
Compaction (pore loss) 2.7(3.7) 12.1(4.2)
1.00 10.00 100.00 1000.00
0.01 0.10
Original carbonate grains 20.8 (5.1) 12.5(2.6)
% Carbonate grains of total 33.2(7.5) 17.3(3.7) horizontal permeability mD
grains
Number of samples 22 14 Fig. 4. Helium porosity and horizontal permeability of
low-permeability lenses and host rocks.

60 .-------�--,

a few decimetres to several tens of metres (Fig. 1) N=344

that is parallel as well as perpendicular to the 50-
+
+

general transport directions. Occasionally, ce­

mented layers are continuous on outcrop scale with + +
+
a lateral extension of several hundreds of metres to 40 - +
+
+ +

E
kilometres, and extend beyond the scale of single
+ + +
+
� + + .....

outcrops, which are as much as 1.5 km long. They rJ) +

++
+ + +
rJ) 30- ... + +

Q)
are either hardgrounds or calcite-cemented amal­ s::::
-"
gamated storm beds. The coarse-grained fossilifer­ .�
ous lags of such storm deposits are well cemented. ; 20 f-
Hardgrounds and storm deposits commonly occur
at the top of a depositional sequence and separate
stacked depositional units. 10 f-
The thickness of lenses is between 2 and 60 em
(Fig. 5) and they either wedge out concurrent with
0
the low-angle bedding or are disconnected within a 0.01 0.10 1.00 1000 100.00
more continuous layer or set of layers. Most lenses
length (m)
have distinct boundaries formed by bedding planes,
often marked by clay laminae or erosional surfaces Fig. 5. Length and thickness of low-permeability lenses
(in the case of hardgrounds). Gradational bound­ and nodules at Findel (Fig. 1) in a transect NE-SW
parallel to the general transport direction. The
aries also occur, the degree of cementation changing
maximum length measured in this outcrop (30 m) is
over a few millimetres. Stratabound lenses occur in limited by the tectonic dip and the internal foresets,
low-angle or swaley laminated sandstones which both dipping towards the SW, and the height of the
contain only few isolated escape burrows. By con­ outcrop(2-5 m). Other outcrops (Aspelt) show
trast, most host rocks typically are wave and ripple cemented layers with a lateral extension of more than
300 m. The logarithmic relationship between the two
cross-laminated sandstones with wavy clay lamina­ variables in host rocks (expressed by the function
tion and flaser bedding due to wave and storm Y 5.66 log(X) + 17.95, with a correlation coefficient
�

action. R 0.60; n 344), is statistically significant.

= =
 Low-permeability lenses in shallow marine sandstones 199

DETRITAL COMPOSITION packed texture of subhedral-euhedral low-Mg

OF SANDSTONES microsparite. Fragmented bioclasts are well
rounded and some show microborings and
Sandstones are moderately sorted (a0 0.73) fine­ = micritized outer walls, suggesting reworking and
grained quartzarenites to litharenites with a mean exposure at the sea floor before final deposition.
composition Q80F0Lcarbonate20 (Folk, 1968). The Entire and disarticulate bivalve shells occur in
main terrigenous component of the sandstones is channel- and storm-lag deposits. Many carbonate
monocrystalline quartz. K-feldspars and plagioclase grains have been dissolved, leaving mouldic or
are rare ( <1% ) and display dissolution features. oversized secondary pores. The amount of carbon­
Minor amounts of polycrystalline quartz, chert, ate grains is highly variable and is typically related
quartzose metamorphic and sedimentary rock frag­ to the amount of carbonate cement in the sandstone
ments, muscovite and glaucony grains are also (Fig. 6; Table 1). Most of the sandstones are matrix
present. Sandstones have a variable content of free, but in hardgrounds some infiltrated carbonate
intrabasinal carbonate grains (0-45%). These are matrix occurs with characteristic geopetal accumu­
bioclasts (mainly fragments of molluscs and echino­ lation.
derms), ooids with quartz, or occasionally carbon­ Detrital clay-sized material occurs in laminae and
ate nuclei and carbonate peloids. Mollusc fragments wavy laminae. The detrital clay minerals present
that originally consisted of low-Mg calcite have are illite, kaolinite, a mixed-layer smectite-illite
retained their primary textures. Other bioclastic (11-12 A) and, rarely, smectite.
grains have been replaced by blocky calcite or by
microsparite. In peloids this resulted in a loosely
DIAGENETIC MINERALS

45
The temporal relationships between the various
40
..... ... ...
diagenetic processes, inferred from their spatial
...
... ... arrangement and the paragenetic sequence, are
35 \ shown in Fig. 7. A comparison of depositional and
-cQ) ... ... ...
.i.Jt.;.
...
...
diagenetic characteristics of lenses and host rock is
30 ... ... ... shown in Fig. 8.
E to.
Q)
0 25
Q)
-
Cll Calcite cement
c 20
0
-E to. Two calcite cement generations are present in the
Cll to.
lenses. The first is a low-Mg calcite fringe cement
0 15 t:. t:.
';!!. t:. which is found in all lenses (averaging 8.5% ; n 22),
=

10
t:. t:.
t:.
5 variable (ranging from 1.7 to 36.0%). The cement
0 t:. t:.2S �
0 5 10
% carbonate grains
Fig. 6. Percentages of carbonate cement (fringe and late
calcite cement) and carbonate grains in lenses and host
rocks. Host rocks contain only late calcite cement.
Percentages were obtained by point counting thin
sections. In host rocks, a statistically significant linear
relationship exists between the two variables (expressed
by the linear function Y 1.80 X- 0.87; R 0.88;
n 18). This suggests that late calcite cementation
=
occurred relatively late after most metastable carbonate
grains had been dissolved. Cementation only occurred
when carbonate grains were available as nuclei for
cement precipitation.
, t:.
... lens j hardgrounds and intraformational clasts, but absent
host rock in host rocks. The amount of fringe cement is
fringes are isopachous, indicating phreatic precipi­
15 20 25 30 35 tation. Cement fringes are composed of fibrous
(Fig. 9A) or bladed crystals (Fig. 9B) with the long
axes of the crystals perpendicular to the grain
surfaces. The rim is syntaxial around echinoderm
fragments. Occasionally, thin scalenohedral (or dog­
tooth) cement fringes are found. Where abundantly
present, fringe cement consisting of closely packed
radial fibrous crystals (Fig. 9A) occurs both around
= = carbonate and quartz nuclei. Otherwise, bladed
cement crystals only fringe carbonate grains.
A second generation of low-Mg calcite cement
was precipitated in primary and secondary pores in
both lenses and host rocks. This cement is moder-
200 N. Molenaar

I
TIME
SCALE JURASSIC -CRETACEOUS TERTIARY -RECENT

fringe cementation

dolomite precipitation
en -
w mechanical compaction
en - -
en
w quartz cementation
(.) -
0
0:: kaolinite-dickite authigenesis and dissolution of feldspar
0.. -
(.) uplift and influx of fresh water
i=
w
z dissolution of carbonate grains and dolomite/secondary porosity
w
(!)
<t: replacement of carbonate grains by low-Mg calcite
c ---
Fig. 7. Paragenetic sequence in the
sparry-poikilotopic calcite cementation Luxemburg Sandstone as
------ determined from textural
relationships. Present weathering
RELATIVE TIMING processes are omitted in this
diagram.

ately luminescent without visible zonation. It is
composed of equant-shaped crystals in lenses. The
average content of sparry calcite cement in lenses is
26.8% (n 22). In some cases the crystal size in­
=
creases toward the pore centre. In host rocks, the
calcite cement is sparse (9.3%; n 14) and consists of
= Pores with rhombohedral outlines and lined by iron
poikilotopic crystals. The poikilotopic cement was
mainly precipitated syntaxially around carbonate
grains, such as echinoderm and bivalve shell frag­
ments which survived dissolution. Occasionally,
dolomite crystals or crystal clusters were replaced by
calcite and became part of larger poikilotopic cement
crystals (Fig. 9C). The amounts of calcite cement in
the host rocks show a positive correlation with the
amount of carbonate grains (Fig. 6).
All carbonate components are low-Mg calcite
with an average MgC03 mol% of 2.6 ( ± 1.5 a;
n 22) as determined by random powder XRD and
=
3
o1 C = -1.07%o) with respect to lenses (average
13
o C = -0.39%o).
Dolomite
hydroxide or oxide occur dispersed throughout the
host rocks (Fig. 90), suggesting the former presence
of small amounts of dispersed, iron-rich dolomite
crystals (�:d %). The euhedral form of the former
dolomite crystals suggests that ample pore space
was available for growth, implying that dolomitiza­
tion occurred relatively early during diagenesis.
During later diagenesis dolomite was usually dis­
solved, leaving mouldic pores in the host rocks
(Fig. 90), and occasionally reJ?laced by low-Mg
calcite (Fig. 9C).

electron microprobe measurements. Fringe cement,

late calcite cement and carbonate grains have com­
parable ranges of chemical compositions (Fig. 10)
and similar cathodoluminescence characteristics in
lenses as well as in host rocks. The oxygen isotopic
compositions of calcite in lenses and host rocks,
measured in 23 bulk samples, are also similar and
show a small range (-8.4 to -9.3%o) (Fig. 11) around
a mean o180 of -8.7%o PDB (n 23). The calcite in
= present in abundance the growth of quartz cement
2
host rocks is slightly enriched in 1 C (average
Silicates
Small amounts of quartz cement are present as over­
growths on detrital quartz grains and are most abun­
dant within the host rocks (averaging 7%; Fig. 9E),
but are rare in lenses (averaging 1%). Quartz over­
growths do not have fluid inclusions, suggesting a
single phase of precipitation. Where dolomite was
was limited, suggesting that quartz cementation
 Low-permeability lenses in shallow marine sandstones 201

1990). Deformation of ductile grains, such as mus­

FEATURES
covite and carbonate peloids around rigid quartz
LENSES HOST ROCKS
grains, and rarely also around thin quartz over­
growths, is a feature indicative of mechanical com­
cross beddin 9 paction. Additional compaction features are frac­
swaley cross amination
tured fragile shells, large biogenic carbonate grains
wavy or flaser bedding that have been penetrated by quartz grains, and
ripple cross lamination
chemical compaction, indicated by sutured contacts
degree of bioturbation ---------- between quartz grains in clayey laminae. The pres­
ence of grains deformed over quartz grains seems to
degree of lithification ---------- have locally inhibited quartz cement precipitation.
Mechanical compaction began just before, and con­
early fringe cement tinued after, quartz cement precipitation. In sand­
stones with fringe cement or abundant quartz ce­
late calcite cement ---------- ment, the framework of the sand was stabilized and
mechanical compaction was prevented almost com­
quartz overgrowths ---------- pletely. The mean loss of porosity through compac­
tion in the host rock is approximately 12% (Table 1;
secondary pores after estimated by assuming an initial porosity of 40% and
carbonate grains
taking all cements and primary porosity into ac­
oversized pores
count).

secondary pores after

dolomite crystals
Secondary porosity

Much of the porosity in the host rocks is secondary,

compactional features
in the form of mouldic and oversized pores
(Figs 9B,F). Oversized pores are approximately the
porosity/permeability ----- - - - --

same size as adjacent grains and surrounding pri­

Fig. 8. Summary of characteristic textural features and
differences in diagenesis or intensity of diagenetic
processes in lenses and their host rocks. The relative
importance or intensity of processes is indicated by the
thickness of the bar.
occurred later than dolomitization. Quartz over­
growths are covered by authigenic kaolinite/dickite
and/or by sparry calcite cement, which demonstrates
that the latter two originated later.
Up to 2% authigenic kaolinite and dickite occur
in host rocks as loosely packed booklets of pseudo­
hexagonal crystals (5-15J.1m). They occur on authi­
genic quartz overgrowths and on partly dissolved
feldspar grains.
Mechanical compaction
In the area studied the Luxemburg Sandstone
reached a maximum burial depth of 500 m (±50 m)
(Bemers, 1985; Muller, personal communication,
mary pores. The form of the mouldic pores suggests
that they resulted from the dissolution of detrital
carbonate grains such as bivalve shells and ooids
(Fig. 9B). Dissolution of dolomite crystals in the
host rocks resulted in rhombohedral pores
(Fig. 9D). The abundance of mouldic and oversized
pores indicates that most of the carbonate grains
must have been dissolved in the permeable host
rock. Partial dissolution of feldspar grains along
cleavage and twin boundaries also resulted in addi­
tional secondary porosity development. Features
indicative of framework collapse are absent, sug­
gesting that the development of secondary porosity
by dissolution of carbonate grains occurred rela­
tively late after compaction.
Lenses contain few secondary pores, indicating
that fringe cement largely inhibited dissolution.
Mouldic pores in lenses mainly resulted from the
dissolution of ooids, and commonly are partly filled
by sparry calcite cement. Assuming that most of the
secondary porosity was caused by dissolution of
carbonate grains, the lenses originally contained
twice as much detrital carbonate (Q66_8F0Lcar-
202 N. Molenaar
 Low-permeability lenses in shallow marine sandstones 203

1.0 • •
N

0 09 • rl'
;--
-0.8
• • •
•
0.8 • •
Ill •
(.) 0.7 ..
-1.2
Qj • • •
u...
0.6 • •
• • • • ••
.. -1.6
.... "' • •
0.5 • • ••
+ -2.0
. � •
0.4 • •
•
0.3 • • late calcite cement -2.4 l &lens I
• carbonate grain L'>host rock I
+ early fringe cement
0.2 •
• -9.2 -9.0 -8.8 -8.6 -8.4 -8 .2 -8.0
0.1 L__L L__L L__L L__L L__L L__L�
__ __ __ __ __

0.05 0.15 0.25 0.35 0.45 0.55 0.65

-2
Mg/Ca • 10
Fig. 10. Plot of the Mg/Ca and Fe/Ca ratios of sparry
calcite cement, (replaced) carbonate grains and fringe
cement of lenses and host rocks. All carbonate is low-Mg
calcite. Despite the rather large range in chemical
compositions, there is no significant difference in
chemical composition between the various carbonate
components.
b 1b %oPDB
Fig. 11. Plot of oxygen and carbon stable isotopic
compositions of bulk carbonate samples. Note the small
range in isotopic compositions of calcite in lenses
despite variable ratios between early and late calcite
cement. This points to similar physical and chemical
conditions during replacement of early diagenetic
CaC03 cement by low-Mg calcite and late calcite cement
precipitation.

bonate33_2%) as the host rock (Q82.4F0Lcarbon­
ate17_6%).
DISCUSSION
Timing of early marine CaC03 cementation
The development of tightly cemented lenses is due
to the heterogeneous precipitation of marine fringe
cement. That fringe cementation was among the
first diagenetic processes is evident from the para-
genetic relationships, and from indirect evidence
such as the lack of compaction in fringe-cemented
sandstone. Unequivocal evidence for its early na­
ture and the marine conditions is provided by the
occurrence of hardgrounds with distinctive fea­
tures, such as borings and encrustation by oysters
(Hanzo et a!., 1987), and intraformational clasts
(Fig. 38). Such features indicate that the upper
surface of a lithified layer was exposed on the sea
floor (Goldring & Kazmierczak, 1974) and com­
monly eroded. Cementation must have occurred at
or just below the sea floor under marine phreatic

Fig. 9. (Opposite) Photomicrographs of thin sections made under plane polarized light. Scale bars are 0.1 mm. (A)
Calcite fringe cement composed of bladed crystals occurring exclusively around carbonate grains (peloids). The
remaining pores have been filled with late diagenetic sparry calcite cement. (B) Fibrous low-Mg calcite fringe cement
completely filling primary pore spaces. The fringe cement has nucleated on carbonate grains as well as on quartz
grains. Most of the grains in this sample are ooids of which the coatings, and sometimes even the carbonate nuclei,
have been dissolved, resulting in mouldic secondary porosity. The fibrous crystals suggest an aragonite precursor. (C)
An example of dolomite crystals which have been replaced by calcite in a host rock. The calcite is part of large
poikilotopic crystals. (D) Mouldic pores after dolomite in a host rock with a relatively large quantity of late
poikilotopic calcite cement. (E) Quartz cement occurring as overgrowths around detrital quartz grains in a host rock.
(F) An oversized pore (OP) forming secondary porosity which resulted from dissolution of the carbonate grains in a
host rock with quartz cement. This sandstone has become a diagenetic quartzarenite due to dissolution of the
carbonate grains.
204 N. Molenaar
conditions. The latter is evidenced by the isopac­
hous nature of all fringe cement. Geopetal accumu­
lation of carbonate matrix in hardgrounds indicates
a stable framework when the lithified becj was
exposed at the sea floor. In addition, the fibrous and
bladed crystal forms of fringe cement are typical
for marine cements. Although all carbonate is now
low-Mg calcite, the fibrous crystal form suggests an
original aragonite mineralogy, whereas the bladed
cement fringes could point to a high-Mg calcite
precursor.
Most hardgrounds just have borings and encrus­
tations, suggesting a relatively short exposure on the
sea floor. Some of the hardgrounds, notably those at
the top of the Luxemburg Sandstone, have devel­
oped in intensely bioturbated layers enriched in
marine bivalve fossils (Hanzo et al., 1987). This
indicates a long period of abandonment before early
cementation caused lithification, during which the
composition and texture were changed through
biological activity. In this case, cementation must
have proceeded slowly, changing the substrate from
a softground into a firm and eventually into a
hardground (e.g. Fiirsich, 1979).
Commonly, intraformational clasts were re­
worked from hardgrounds and -lenses. These clasts
are usually flattened and unbored, and were eroded
from fringe-cemented layers. LOcally the intrafor­
mational clasts are well rounded and bored (Fig.
3B), suggesting relatively extended submarine
exposure and reworking of the hardground (e.g.
Kennedy & Garrison, 1975). Most fringe-cemented
lenses and intraclasts show no signs of exposure at
the sea floor, lacking not only borings and other
faunal elements typical of hard surfaces, but also
burrows. This suggests that most, if not all, of the
early cementation occurred at a depth of a few
decimetres to metres below the sea floor, out of
range of the burrowing fauna. It is relatively rare for
fringe-cemented lenses or beds to have been eroded,
exposed at the sea floor and subjected to boring and
encrustation. Soft substrate conditions gen,erally
prevailed, as indicated by< the macrofauna assem­
blage that mainly consisted of infaunal or semi­
infaunal bivalves, comprising semi-infaunal species
such as Pinna and Cardinia (Bock & Muller, 1989).
Although this fauna is largely reworked, it is occa­
sionally found in situ. That the Lorraine Mar was
also soft at the sea floor is indicated by its major
fauna element, Gryphea, whose shell form is typical
for a reclining mode of life on a soft marine bottom
(Seilacher, 1984).
Sources for early marine CaC03 cement
Under shallow marine conditions, CaC03 for ce­
ment can be sourced internally by dissolution of
metastable carbonate induced by bacterial decom­
position of organic matter, and externally by sea
water. Mixing of different types of interstitial water,
in particular marine pore water and meteoric
groundwater, can also cause supersaturation with
respect to calcium carbonate, and could evoke
carbonate precipitation in the mixing zone (Run­
nells, 1969; Wigley & Plummer, 1976). This has
been suggested as a mechanism for calcite-dolomite
cementation in sandstones (e.g. Morad et al., 1992)
and it may produce calcite-cemented layers along
the coast that resemble beach rocks (Moore, 1977;
Manor, 1978). In the Luxemburg Sandstone this
process can be ruled out simply because early
cemented lenses are not concentrated in proximal
or uplifted areas.
Internal cement source
A potential internal source for calcite cementation
is the dissolution of bioclastic aragonite and
high-Mg calcite by C02 production during bacterial
oxidation of organic matter in the uppermost oxi­
dizing layers of the sediment. C02 is produced at
high rates during bacterial oxidation of organic
matter. The consequent rise of pC02 and the
decrease of pH in these layers is greater than the rise
in carbonate alkalinity, and may therefore cause
carbonate dissolution (e.g. Cranston & Buckley,
1990). When the pH decreases the interstitial water
becomes undersaturated with respect to relatively
unstable high-Mg calcite and/or aragonite. Subse­
quently, there is dissolution of bioclastic compo­
nents of that mineralogical composition. The
solubility of carbonate grains, powever, also de­
pends on their texture (Walter, 1985). Calcite with
more than 12 mol% MgC03 is more soluble than
aragonite (Walter & Morse, 1984; Bischoff et a!.,
1987). Upon continued dissolution of high-Mg
calcite and aragonite, the interstitial water may
become more saturated with low-Mg calcite and,
finally, this may cause precipitation of calcite ce­
ment. The potential thickness of the cementing
layer is probably limited to the depth to which
oxidation extends, depending on the type and con­
tent of organic matter, the texture and the sedimen­
tation rate.
In modem calcareous sediments in shelf settings,
 Low-permeability lenses in shallow marine sandstones
dissolution in the oxygenated layer has been ob­
served and the degree of dissolution correlates with
the amount of organic matter included in the
biogenic particles (Freiwald, 1995). Precipitation of
low-Mg calcite in deeper shelf limestones is associ­
ated with dissolution of aragonite bioclasts (Melim
et al., 1995), suggesting that early calcite cementa­
tion occurred through an internal dissolution and
reprecipitation mechanism. Apart from microbor­
ing, the conversion of skeletal grains into equant
micritic high-Mg calcite and micritic envelopes is
the result of oxidation of organic matter. Oxidation
can induce crystallization of primary aragonite and
high-Mg calcite in sea water saturated with these
carbonate minerals (Reid et al., 1992). In the
Luxemburg Sandstone, however, micritic envelopes
around shell fragments are covered by cement
fringes, thereby eliminating micritization as a
source of the early cement.
Calcite cementation by this mechanism would
comprise local dissolution of bioclastic material
followed by short, diffusion-controlled transport
and reprecipitation as cement. The main controls
are the redox potential as a function of bacterial
organic matter degradation, and the framework
mineralogy. The control exerted by framework min­
eralogy could explain fringe cementation in the
Luxemburg Sandstone as being related to carbonate
grain accumulations. However, it fails to explain
the observed occurrence of fringe cementation in
permeable sand and the constraints exerted by
impermeable structures. In addition, no traces of
carbon derived from the decomposition of organic
matter have been left in the isotopic compositions
of calcite. Moreover, early dissolution of carbonate
grains has not been observed. On the contrary,
secondary porosity has resulted from late diagenetic
dissolution of carbonate grains with apparently
unstable mineralogy, leaving oversized or mouldic
pores. In lenses, dissolution has often delicately
removed the outermost thin ooid coatings, leaving
the carbonate nuclei as well as the fringe cement
intact (Fig. 9A). This suggests that interstitial water
remained supersaturated during pre-burial condi­
tions with respect to aragonite as well as calcite,
excluding early dissolution and redistribution as a
local internal source for fringe cementation.
External cement source
A potential external source of calcium carbonate for
early cementation is supersaturated sea water. In
205
low-latitude shallow seas the sea water is supersat­
urated with respect to aragonite and calcite, and
precipitation of these minerals as cement can pro­
ceed in siliciclastic sands through external supply.
For kinetic reasons, aragonite and high-Mg calcite
cements usually precipitate, although low-Mg cai­
cite cement has also been noted in shelf sediments
(e.g. Melim et al., 1995). Lithification by sea water
supply can occur within less than I 000 years in the
case of beach rocks and shallow subtidally ce­
mented layers (Sibley & Murray, 1972; Friedman,
1975; Hattin & Dodd, 1978; Dravis, 1979). In the
case of the Luxemburg Sandstone the abundance of
ooids of intrabasinal origin indicates that sea water
indeed was supersaturated with respect to calcium
carbonate. The thickness of subtidally cemented
layers in modem sediments ranges from a few
millimetres or centimetres (Harris, 1978; Dravis,
1979) up to 60 em (Sibley & Murrey, 1972), and is
similar to the measured range in the Luxemburg
Sandstone.
The restriction of calcite fringes to sands of
initially high permeability indicates that permeabil­
ity and fluid flow were factors controlling cementa­
tion. In the Luxemburg Sandstone, lenses are
commonly bound by impermeable clayey, £laser­
bedded layers or clay drapes. The flow of sea water
within the sand was constrained by such imperme­
able barriers, which prevented further cementation.
However, lenses also occur within a textural homo­
geneous sandstone layer and with gradational
boundaries. The large variabilities in the degree of
early cementation, and their distribution, suggest
control of a number of other parameters, such as
hydrodynamic energy and framework mineralogy.
In subtidal environments the amount of early ma­
rine cement is directly related to the hydrodynamic
conditions, and higher-energy environments gener­
ally have more early diagenetic cem!!nt (Marshall &
Ashton, 1980).
In recent subtidal sands cemented layers com­
monly occur below a cover of loose sand. Wave and
current reworking of the uppermost sand layer
inhibits lithification, because the sand must be
stable to permit cementation. Such a stable situa­
tion exists when deposition occurs only by periodic
processes such as storms (Ginsburg, 1953), or when
the surface layer is biogenically stabilized by sea
grass meadows or algal mats (Davies & Kinsey,
1973; Harris, 1978; Dravis, 1979). Otherwise, in
places where sand is continually moved by waves or
tidal currents, cementation takes place at a few
206 N. Molenaar
centimetres to decimetres below the sea floor, where
sand is not being moved (Taylor & Illing, 1969;
Alexandersson, 1972; Sibley & Murray, 1972;
Dravis, 1979). Thus cementation does not necessar­
ily take place at the sea floor itself. In such cases,
straightforward evidence for its early origin in the
form of borings and faunal encrustations will be
lacking, and it may be difficult to prove the early
origin of the cement. The absence of features
indicative of compaction merely is a coarse indica­
tor and can only be used as evidence for cementa­
tion during pre-burial conditions.
It may be argued that the observed variability in
crystal form and mineralogy of the fringe cement is
an intrinsic argument for hydrodynamic flow and the
involvement of sea water. This variability may point
to different growth rates and degrees of saturation of
involved fluids (e.g. Rushdi et a!., 1992), reflecting
locally changing environmental conditions in terms
of energy and changes in sediment texture. This is in
contradiction to a slow diffusion-controlled cemen­
tation mechanism, which is independent of such en­
vironmental controls.
Differential early CaC03 cementation
With the notable exception of laterally extensive
hardgrounds and early cemented storm deposits, all
early cementation resulted in low-permeability
lenses (Fig. 1). These are stratabound or concordant
to sedimentary structures, but often not coincident
with depositional structures such as layers or fore­
sets (Figs I and 3C). In general, early lithification of
marine sediments is heterogeneous. Examples are
nodular lithification of hardgrounds in carbonate
rocks (Bathurst, 1971; Kennedy & Garrison, 1975;
Bromley & Gale, 1982; Garrison et a!., 1987) and in
mixed siliciclastic-carbonate sandstone (Molenaar
& Martinius, 1990; Martinius & Molenaar, 1991).
Continuously cemented and lithified layers develop
by lateral coalescence of nodules or lenses (e.g.
Bathurst, 1971; Kennedy & Garrison, 1975). Often
hardgrounds pass laterally as well as downward into
nodular cemented sediment (Bromley & Gale,
1982). Similarly, subtidal cemented layers pass from
small-scale cemented peloids and aggregate grains
into nodular and, finally, lensoid and layer-like ce­
mentation (Taft et a!., 1968; Harris, 1978). The com­
mon heterogeneity of early marine cementation re­
flects the process involved and the constraining
factors on that process, such as sedimentary texture,
framework mineralogy and time.
In case of external fluid supply, it is to be
expected that the texture and internal structures of
the sandbodies would influence cement distribu­
tion. Vast quantities of sea water are needed for
CaC03 cementation, when dissolved carbonate
must be derived from outside sources (e.g. Blatt,
1979; Scholle & Halley, 1985). Therefore, early
cementation, when dependent on hydrodynamic
flow and external supply, is strongly constrained by
the permeability of the sediment and the presence
and force of a supply mechanism (e.g. Harris et a!.,
1985). Possible mechanisms to move sea water
through the sediment are waves and tidal currents
operating in shallow marine environments and at
platform edges. In this model, cementation can only
be expected in highly permeable sands. This is
confirmed by the occurrence of lenses within the
most permeable parts of the sandstone, and the fact
that they are often bound by impermeable clayey
structures. The texture of the sand determines the
permeability (Beard & Weyl, 1973), whereas gen­
eral fluid flow patterns are dependent on the inter­
nal sedimentary structures and the architecture of
the sandbodies. Even within seemingly homoge­
neous sandbodies, small differences in grain-size
distribution and packing density normally cause
distinct differences in primary permeability (Pryor,
1972, 1973), and could produce heterogeneous
cementation.
The form of carbonate nodules in sandstone often
gives evidence for hydrodynamic flow controlling
cementation. Carbonate nodules are commonly
flattened and elongated in a direction parallel to the
preferred permeability and the palaeogroundwater
flow direction, owing to preferred longitudinal ac­
cretion (e.g. Johnson, 1989; McBride et a!., 1994).
Preferred grain orientations and bedding cause
directional differences in permeability. This con­
trols the growth, and thus the fprm, of nodules in a
certain direction (Colton, 1967; Pirrie, 1987; Moz­
ley & Davis, 1996).
Lenses occur in sandstone that contains abundant
framework carbonate grains, implying that frame­
work mineralogy is an important parameter influ­
encing fringe cementation. In the Luxemburg
Sandstone, early carbonate cementation was re­
stricted to sedimentary structures or portions of
structures with higher carbonate grain contents.
The original carbonate grain content of lenses was
approximately twice that of host rocks (Table 1).
The association between early fringe cement and
carbonate grain accumulation can be explained in
 Low-permeability lenses in shallow marine sandstones
two ways: carbonate grains may be necessary as
nuclei to allow precipitation, or they may form the
source for the cement. Because no evidence for
early dissolution has been found, the necessity for
available nucleation sites is the most likely explana­
tion. Homogeneous nucleation of cement (i.e.
cementation without nuclei) does not occur (Alex­
andersson, 1972), with the exception of peloidal
cementation (Macintyre, 1985); Early cement in
sandstones can develop independent of the primary
mineralogical composition of the sand only under
exceptional environmental or chemical conditions,
when sea water or interstitial water becomes highly
saturated with respect to carbonate. Examples are
lithification of beach rocks, which develops in
siliciclastic sands regardless of the detrital mineral­
ogy (e.g. Strasser et a!., 1989), and cement precipi­
tation caused by bacterial decay or methane
oxidation (e.g. Roberts & Whelan, 1975; Nelson &
Lawrence, 1984). Normally, under subtidal condi­
tions with limited available time, calcite and arago­
nite cement in sand precipitates preferentially on
carbonate nuclei (e.g. Molenaar et a!., 1988; Mo­
lenaar & Martinius, 1990). Given the proper condi­
tions, carbonate grain accumulations in specific
parts of sedimentary structures can evoke differen­
tial cementation (Chafetz, 1979; Kantorowicz et a!.,
1987; McBride, 1988). Thus, in part, the detrital
composition of the sand directly determines the
potential for precipitation of (early) carbonate
cement. Even small changes in hydrodynamic con­
ditions during deposition may cause distinct varia­
tions in the relative proportions of carbonate and
siliciclastic grains, due to selective sorting (see
Calvert, 1982). A common example is formed by
the common accumulation of coarse bioclastic car­
bonate material during storms (e.g. Aigner, 1982).
Based on this, it can be postulated that carbonate
diagenesis may generally be related to facies and be
concordant with sedimentary structures.
As the permeability and framework mineralogy
are spatially variable, cementation in general is
expected to show spatial variability and an initial
tendency to produce heterogeneous lithification. In
particular, when dependent on fluid flow, the effect
of textural and compositional heterogeneities is
expected to prevail during diagenesis. When diage­
netic processes are dependent solely on diffusion,
these effects are less distinct as diffusion is less
dependent on permeability.
Nodular cementation indicates an initial stage of
cementation. In addition to its nodular distribu-
207
tion, cementation usually is incomplete and poros­
ity is partially preserved. This is initially due to the
textural and mineralogical heterogeneity of the
sand. During the cementation process the perme­
ability decreases owing to blocking of pore throats,
and finally fluid flow and cementation cease. More­
over, limited time for cementation and continuing
sediment aggradation ceases the flow of sea water at
depth.
Late calcite cementation mechanism
The source of calcite cement under burial conditions
often remains problematic. Apart from the dissolu­
tion of detrital carbonate, additional potential Ca
sources are feldspar alteration, comprising the dis­
solution and albitization of plagioclase (Morad et a!.,
1990), and illitization of smectite (Boles & Franks,
1979; Wintsch & Kvale, 1994). Besides the dissolu­
tion of detrital Ca-rich feldspar due to influxes of
meteoric water during uplift, Ca liberated during
albitization of plagioclase and diffusional transport
may cause limited calcite cementation in feldspathic
sandstones under burial conditions (Morad et a!.,
1990). The scarcity of feldspar shows that this pro­
cess did not contribute significantly to calcite cemen­
tation in the Luxemburg Sandstone.
The progressive conversion of smectite into illite
in shales, in combination with shale compaction,
may form an important potential source of Ca for
calcite cementation in sandstones (Boles & Franks,
1979; Wintsch & Kvale, 1994). Illitization com­
mences at an approximate burial depth of ::::: 2-4 km.
Therefore, this process cannot have influenced low­
temperature diagenesis in the Luxemburg Sandstone
and the Lorraine Marl, as the maximum burial depth
reached was approximately 500 m. Even in the
underlying Lower Jurassic and Triassic shales, the
minimum burial depth required, for substantial
illitization was never reached. In addition, the
variation in clay mineral composition, comprising
smectite, mixed-layer smectite-illite ( 11- 12 A),
smectite-chlorite and illite, suggests detrital inher­
itance of the clay mineralogy instead of diagenetic
modification (Muller et a!., 1973). Moreover, lenses
are not concentrated along faults or in areas where
the Luxemburg Sandstone and the Lorraine Mar are
interbedded. This excludes large-scale compaction­
driven fluid flow as a cause of late calcite cementa­
tion, implying that most of the calcium carbonate
for late calcite cementation was derived from the
dissolution of detrital carbonate in host rocks.
208 N. Molenaar
Later diagenesis in the Luxemburg Sandstone
enhanced early permeability patterns. Fringe­
cemented sandstone lenses formed local zones of
decreased permeability and thus created barriers to
fluid movements in the sandstone body during
burial diagenesis and uplift. The movement of late
interstitial fluids was limited to the most permeable
sandstones, i.e. the host rock. Subsequent diage­
netic processes, such as dolomitization, quartz ce­
mentation and carbonate grain dissolution, were
restricted to the host rocks, suggesting that these
processes were dependent on hydraulic flow of
diagenetic fluids.
The amounts of calcite cement in the host rocks
show a positive correlation with the amount of
carbonate grains (Fig. 6). This indicates calcite ce­
ment precipitation after dissolution of metastable
carbonate grains. Stable carbonate grains (which
survived dissolution) served as nuclei for late calcite
cement in host rocks. Within lenses, late calcite
cement merely filled all available pores, which can
be ascribed to either an abundance of nuclei or
locally low permeability. The lack of correlation
between calcite cement and secondary porosity
suggests relatively long-range transport of dissolved
CaC03. Late calcite cement precipitated preferen­
tially within lenses that contained more calcite
nuclei than surrounding host rocks.
All framework carbonate and early diagenetic
carbonate cement was dissolved or replaced by
low-Mg calcite during late diagenesis. The isotopic
compositions of calcite therefore merely reflect
temperature and chemical conditions during late
diagenesis. The small range in oxygen isotopic
compositions of calcite in lenses and host rocks
suggests a limited range of environmental condi­
tions during late diagenesis. The relatively large
range in chemical compositions suggests that the
replacement of unstable detrital carbonate and early
fringe cement locally changed the chemistry of the
interstitial . water with respect to trace element
concentrations.
The carbon isotopic composition of calcite re­
flects interstitial water (either sea water or meteoric
water), with carbon in near equilibrium with atmo­
spheric C02 with a variable but small contribution
from organic matter-derived C02. The oxygen iso­
8
topic values are strongly negative. 1 0 depletion of
marine pore water may be caused by several mech­
anisms (e.g. Mozley & Bums, 1993), such as early
diagenetic silicate alteration, including volcanic
grain alteration and smectite authigenesis (e.g.
Lawrence, 1 989), and massive sulphate reduction
in organic-rich marine sediment (Sass et a!. , 199 1 ).
The absence of volcanic material and associated
clay mineral authigenesis, as well as organic-rich
sediments in the stratigraphic succession, rules out
these mechanisms as a cause of the negative oxygen
isotopic compositions. When precipitated from
buried Jurassic sea water, which had an isotope
ratio of about - 1.2o/oo (Shackleton & Kennett,
8
1 975), ce ment would have a o 1 0 of about -4.5o/oo
at the maximum temperature reached. The temper­
ature did not exceed approximately 3o · c even
during maximum burial depth of around 500 m
(Haenel & Staroste, 1988). The negative oxygen
isotope ratios thus cannot be explained by precipi­
tation at high temperatures from buried sea water,
or by extensive mineral authigenesis. The only
8
possibility for o 1 0 depletion is therefore the influx
of meteoric water, which at present has a mean
8
o 1 0 value of -8o/oo in Luxemburg (Yurtsever,
1 97 5). Precipitation at surface temperatures from
meteoric interstitial water explains the measured
oxygen isotope values in lenses and host rocks. The
extensive dissolution prior to and/or contempora­
neous with late calcite cementation was caused by
meteoric water influx. Probably the dissolved detri­
tal carbonate was the source of late calcite cemen­
tation.
CONCLUSIONS
Tightly calcite-cemented lenses in shallow marine
sandstones of the Jurassic Luxemburg Sandstone
are primarily the result of differential cementation
by marine pore waters. Cementation proceeded
through the external supply of cementing materials
from supersaturated sea water. As this cementation
mechanism is dependent on the;: flow of sea water
through a sandbody, the process is controlled by
parameters such as texture and sedimentary struc­
tures. It is also influenced by framework mineral­
ogy: the amount of carbonate grains determined the
susceptibility for early cementation. The lateral
extension of low-permeability lenses is dependent
on the mentioned parameters, as a function of the
depositional facies, and on the sedimentation rate.
It may be expected that in general such diagenetic
processes are concordant with primary sedimentary
structures and facies. The depositional variability in
texture and framework mineralogy were enhanced
by early cementation. Later diagenesis due to the
 Low-permeability lenses in shallow marine sandstones
influx of meteoric water was constrained by the
primary and early diagenetic permeability patterns,
and resulted in further enhancement of the already
existing facies-controlled heterogeneities.
ACKNOWLEDGEMENTS
Fieldwork was financed by the Koninklijke Shell
Exploration and Production Laboratory at Rijs­
wijk. The helpful discussions and guidance in the
field by Professor Dr A. Muller are highly appreci­
ated. Furthermore, I am grateful for the comments
of M.P. Clemente, T.M. McGee, D. Pirry, P.S.
Mozley and S. Morad, and for the porosity and
permeability measurements by G.P. van de Bilt
(Panterra Geoconsults). The Feidt enterprise is
thanked for allowing access to various quarries.
REFERENCES
AIGNER, T. (1982) Calcareous tempestites: storm­
dominated stratification in Upper Muschelkalk Lime­
stones (Middle Trias, SW Germany). In: Cyclic and
Event Stratigraphy (Eds Einsele, G. & Seilacher, A),
pp. 180-198. Springer Verlag, Berlin.
ALEXANDERSSON, E.T. (1972) Mediterranean beachrock
cementation: marine precipitation of Mg-calcite. In:
The Mediterranean Sea: a Natural Sedimentation Lab­
oratory (Ed. Stanley, D.J.), pp. 203-223. Dowden,
Hutchinson & Ross, Stroudsburg, PA
BATHURST, R. (1971) Carbonate Sediments and their Di­
agenesis. Developments in Sedimentology, 12. Elsevier,
Amsterdam.
BEARD, D.C. & WEYL, P.K. (1973) Influence of texture on
porosity and permeability of unconsolidated sand. Bull.
Am. Ass. petrol. Geol. , 57, 349-369.
BERNERS, H.P. (1983) A Lower Liassic offshore bar envi­
ronment, contribution to the sedimentology of the
Luxemburg Sandstone. Ann. Soc. geol. Be/g. , 106, 87-
102.
BERNERS, H.P. ( 1985) Der Einjluss der Siercker Schwelle
auf die Faziesverteilung meso-kii nozoischer Sedimente
im NE des Pariser Beckens-Ein Sedimentationsmodell
zum Luxemburger Sandstein (Lias), spezielle Aspekte
zur strukturellen Anderung der Beckenkonfiguration und
zum naturraumlichen Potential. PhD dissertation,
RWTH Aachen, 321 pp.
BERNERS, H.P., GuERIN-FRANIATTE, S. & MuLLER, A.
( 1985) L'ensablement de l'Hettangien au Luxembourg:
Ia coupe du Findel Aeroport de Luxembourg. Cahiers de
l 'lnstitut Catholique de Lyon, 14, 213-229.
BISCHOFF, W.O., MACKENZIE, F.T. & BISHOP, F .C. (1987)
Stabilities of synthetic magnesian calcites in aqueous
solution: comparison with biogenic material. Geochim.
Cosmochim. Acta, 51, 1413-1423.
BLATT, H. (1979) Diagenetic processes in sandstones. In:
209
Aspects of Diagenesis (Eds Scholle, P.A. & Schluger,
P.R.). Spec. Pub!. Soc. econ. Paleont. Miner., Tulsa, 26,
141-157.
BocK, H. ( 1989) Ein Modell zur Beckenausdehnung und
Fazieszonierung am Westrand der Eifeler Nord-Sud
Zone wahrend der Trias und zur Transgression des
Unteren Lias am Ardennensiidrand. PhD dissertation,
RWTH Aachen, 417 pp.
BocK, H. & MULLER, A { 1989) Environments Sedimen­
laires et £cologiques dans le Quart NE du Bassin Pa­
risien au Trias Terminal et au Lias lnferieur. Atti 3 •
Simp. Ecol. Paleoecol. comunitit bentoniche, Catania,
1985, 157-178.
BOLES, J.R. & FRANKS, S.G. (1979) Clay diagenesis in
Wilcox sandstones of southwest Texas: implication of
smectite diagenesis on sandstone cementation. J. sedi­
ment. Petrol. , 49, 5 5-70.
BROMLEY, R.G. & GALE, A S. (1982) The lithostratigraphy
of the English Chalk Rock. Cretac. Res. , 3, 273-306.
CALVERT, R.W.G. (1982) Some problems with the analysis
and interpretation of mixed carbonate and quartz beach
sands, illustrated by examples from north-west Ireland.
Sediment. Geol. , 33, 35-56.
CARR, D.L. & SCOTT, A.J. (1990) Late Pennsylvanian
storm-dominated shelf sand ridges, Sacramento Moun­
tains, New Mexico. J. sediment. Petrol. , 60, 592-607.
CHAFETZ, H.S. (1979) Petrology of carbonate nodules from
a Cambrian tidal inlet accumulation, Central Texas. J.
sediment. Petrol. , 49, 215-222.
CoLTON, G.W. (1967) Orientation of carbonate concre­
tions in the Upper Devonian of New York. US geol.
Surv. Prof Pap. , 575-B, B57-59.
CRANSTON, R.E. & BucKLEY, D.E. (1990) Redox reactions
and carbonate preservation in deep-sea sediments. Mar.
Geol., 94, 1-8.
DAVIES, P.J. & KINSEY, D.W.(1973) Organic and inorganic
factors in Recent beachrock formation, Heron Island,
Great Barrier Reef. J. sediment. Petrol. , 43, 59-81.
DRAVIS, J. (1979) Rapid and widespread generation of
recent oolitic hardgrounds on a high energy Bahamian
platform, Eleuthera Bank, Bahamas. J. sediment.
Petrol. , 49, 1-12.
FoLK, R.L. (1968) Petrology a/Sedimentary Rocks. Hemp­
hill, Austin, TX, 170 pp.
FREIWALD, A. (1995) Bacteria-induced carbonate degrada­
tion: a taphonomic case study of Cibicides lobatulus
from a high-boreal carbonate setting. Palaios, 10, 337-
346.
FRIEDMAN, G.M. (1975) The making and unmaking of
limestones or the downs and ups of porosity. J. sedi­
ment. Petrol. , 45, 379-398 .
FORSICH, F .T. (1979) Genesis, environments, and ecology
of Jurassic hardgrounds. N. lb. Geol. Palaontol. , Abh.,
158, 1-63.
GARRISON, R.E., KENNEDY, W.J. & PALMER, T.J. (1987)
Early lithification and hardgrounds in Upper Albian
and Cenomanian Calcarenites, Southwest England. Cre­
tac. Res., 8, 103-140.
GINSBURG, R.N. (1953) Beachrock in South Florida. J.
sediment. Petrol., 23, 89-92.
GOLDRING, R. & KAZMIERCZAK, J. (1974) Ecological suc­
cession in intraformational hardground formation.
Palaeontology, 17, 949-962.
2 10 N. Molenaar
GUERIN-FRANIATTE, S. & MuLLER, A. ( 1 987) L'Hettangien
dans le NE du Bassin de Paris: biostratigraphie et
evolution sedimentaire. Ann. Soc. geo/. Be/g. , 109, 4 1 5-
429.
GUERIN-FRANIATTE, S., HARY, A. & MULLER, A. ( 1 99 1 ) La
formation des Gres du Luxembourg, au Lias inferieur:
reconstruction dynamique du paleoenvironnement.
Bull. Soc. geol. France, 162, 763-773.
HAENEL, R. & STAROSTE, E. (Eds.) ( 1 988) At/as of Geother­
mal Resources in the European Community. Commis­
sion of the European Communities, Brussels.
HANOR, J.S. ( 1 978) Precipitation of beachrock cements
mixing of marine and meteoric water vs. C02-
degassing. J. sediment. Petrol. , 48, 489-50 1 .
HANZO, M., HILLY, J., GUERIN-FRANIATTE, S. & MANGOLD,
C. ( 1 987) Duree des discontinuites et des sequences:
example du passage Hettangien-Sinemurien dans Ia
region d'Hettange. Bull. Soc. geo/. France, III, 1 1 07-
1 1 1 2.
HARRIS, P.M. ( 1 978) Holocene marine-cemented sands,
Joulters ooid shoal, Bahamas. Gulf Coast Ass. Geo/.
Soc. , Trans., 28, 1 75 - 1 8 3 .
HARRIS, P.M., KENDALL, C.G.S.C. & LERCHE, I. ( 1 985)
Carbonate cementation-a brief review. In: Carbonate
Cements (Eds Schneidermann, N. & Harris, P.M.). Spec.
Pub!. Soc. econ. Paleont. Miner., Tulsa, 36, 79-95.
HATTIN, D.E. & DODD, J.R. ( 1 978) Molocene cementation
of carbonate sediments in the Florida Keys. J. sediment.
Petrol. , 48, 307-3 1 1 .
JOHNSON, M.R. ( 1 989) Paleogeographic significance of
oriented calcareous concretions in the Triassic Katberg
Formation, South Africa. J. sediment. Petrol. , 59, I 008-
1 0 1 0.
KANTOROWICZ, J.D., BRYANT, J.D. & DAWANS, J.M. ( 1 987)
Controls on the geometry and distribution of carbonate
cements in Jurassic sandstones: Bridport Sands, south­
ern England and Viking Group, Troll Field, Norway. In:
Diagenesis of Sedimentary Sequences (Ed. Marshall,
J.D.). Spec. Pub!. Geol. Soc. London, 36, I 03-1 1 8.
KENNEDY, W.J. & GARRISON, R.E. ( 1 975) Morphology and
genesis of nodular chalks and hardgrounds in the Upper
Cretaceous of southern England. Sedimentology, 22,
3 1 1 -386.
LAWRENCE, J.R. ( 1 989) The stable isotope geochemistry of
deep-sea pore water. In: Handbook of Environmental
Isotope Geochemistry, Vol. 3: The Marine Environment
(Eds Fritz, P. & Fontes, J.Ch.), pp. 3 1 7-356.Elsevier,
Amsterdam.
MACINTYRE, !.G. ( 1 985) Submarine cements-the peloidal
question. In: Carbonate Cements (Ed. Schneidermann,
N. & Harris, P.M.). Spec. Pub!. Soc. econ. Paleont.
Miner., Tulsa, 36, 1 09- 1 1 6.
MARSHALL, J.D. & ASHTON, M. ( 1 980) Isotopic and trace
element evidence for submarine lithification of hard­
grounds in the Jurassic of eastern England. Sedimentol­
ogy, 27, 27 1 -289.
MARTINIUS, A.W. & MOLENAAR, N. ( 1 99 1 ) A coral-mollusc
(Goniaraea-Crassate//a) dominated hardground com­
munity in a siliciclastic-carbonate sandstone (the
Lower Eocene Roda Formation, southern Pyrenees,
Spain). Pa/aios, 6, 1 42- 1 5 5 .
McBRIDE, E.F. ( 1 988) Contrasting diagenetic histories of
concretions and host rock, Lion Mountain Sandstone
(Cambrian), Texas. Bull. Am. Ass. petrol. Geo/. , 100,
1 803- 1 8 1 0.
MCBRIDE, E. F., PICARD, M.D. & FOLK, R.L. ( 1 994) Ori­
ented concretions, Ionian coast, Italy: evidence of
groundwater flow direction. J. sediment. Res. , A64,
5 35-540.
MEUM, L.A., SWART, P.K & MALIVA, R.G. ( 1 995)
Meteoric-like fabrics forming in marine waters: impli­
cations for the use of petrography to identify diagenetic
environments. Geology, 23, 755-758.
MERTENS, G., SPIES, E.D. & TEYSSEN, T. ( 1 983) The
Luxemburg Sandstone Formation (Lias), a tide­
controlled deltaic deposit. Ann. Soc. geol. Be/g. , 106,
1 03- 1 09.
MOLENAAR, N. & MARTINIUS, A.W. ( 1 990) Origin of
nodules in mixed siliciclastic-carbonate sandstones, the
Lower Eocene Roda Sandstone Member, Southern
Pyrenees, Spain. Sediment. Geo/. , 66, 277-293.
MOLENAAR, N., VAN DE BILT, G.P., VAN DEN HOEK 0STENDE,
E.R. & N i o, S.D. ( 1 988) Early diagenetic alteration of
shallow-marine mixed sandstones: an example from the
Lower Eocene Roda Sandstone Member, Tremp-Graus
Basin, Spain. Sediment. Geo/. , 55, 295-3 1 8.
MooRE, C.M., Jr. ( 1 977) Beach rock origin: some
geochemical, mineralogical, and petrographic consider­
ations. Geosci. Man, 18, 1 5 5-1 63.
MORAD, S., BERGAN, M., KNARUD, R. & NYSTUEN, J.P.
( 1 990) Albitization of detrital plagioclase in Triassic
reservoir sandstones from the Snorre Field, Norwegian
North Sea. J. sediment. Petrol. , 60, 4 1 1 -425.
MORAD, S., MARFIL, R., AL-AASM, I.S. & GOMEZ-GRAS, D.
( 1 992) The role of mixing-zone dolomitization in sand­
stone cementation: evidence from the Triassic Bunt­
sandstein, the Iberian Range, Spain. Sediment. Geo/. ,
80, 5 3-65.
MOZLEY, P.S. & BURNS, S.J. ( 1 993) Oxygen and carbon
isotopic composition of marine carbonate concretions:
an overview. J. sediment. Petrol. , 63, 73-83.
MOZLEY, P.S. & DAVIS, J.M. ( 1 996) Relationship between
oriented calcite concretions and permeability correla­
tion structure in an alluvial aquifer, Sierra Ladrones
Formation, New Mexico. J. sediment. Res., 66, 1 1 - 1 6.
MULLER, A. ( 1 987) Structures geologiques et repartition
des facies dans les couches meso- et cenozoiques des
confins nord-est du Bassin Parisien. In: Aspects et
Evolution Geologiques du Bassin Parisien (Eds Cavalier,
C. & Lorenz, J.). Inf. geol. Bass. Paris Bull. , mem. h.-s.
6, 27 1 pp.
MULLER, A. & RASCHE, P. ( 1 9 7 1 ) Der Luxemburger Sand­
stein (Hettangien) im Gebiet Syren, Munsbach, Sand­
weiler, Itzig, Hassel (Luxemburg). Bull. Pub/. Serv. geol.
Luxembourg, 4, 28 pp.
MULLER, A., PARTING, H. & THOREZ, J. ( 1 973) Caractere
sedimentologiques et mineralogiques des couches de
passage du Trias au Lias sur Ia bordure nord-est du
bassin de Paris. Ann. Soc. Geo/. Be/g. , 96, 67 1 -707.
NELSON, C.S. & LAWRENCE, M.F. ( 1 984) Methane-derived
high-Mg calcite submarine cement in Holocene nodules
from the Fraser Delta, British Columbia, Canada. Sed­
imentology, 31, 645-654.
PIRRIE, D. ( 1 987) Orientated calcareous concretions from
James Ross Island, Antarctica. Br. Antarct. Surv. Bull. ,
75, 4 1 -50.
 Low-permeability lenses in shallow marine sandstones
PROSSER, D.J., DAWS, J.A., FALLICK, A.E. & WILLIAMS,
B.P.J. ( 1 993) Geochemistry and diagenesis of stra­
tabound calcite cement layers within the Rannock
Formation of the Brent Group, Murchison Field, North
Viking Graben (northern North Sea). Sediment. Geol. ,
87, 1 39-1 64.
PRYOR, W.A. ( 1 972) Reservoir inhomogeneities of some
recent sand bodies. Soc. petrol. Eng. J., 12, 229-245.
PRYOR, W.A. ( 1 973) Permeability-porosity patterns and
variations in some Holocene sand bodies. Bull. Am. Ass.
petrol. Geol. , 57, 1 62- 1 89.
REID, R.P., MACINTYRE, I. G. & POST, J.E. ( 1 992) Mic­
ritized skeletal grains in northern Belize lagoon: a major
source of Mg-calcite mud. J. sediment. Petrol. , 62,
1 4 5 - 1 56.
ROBERTS, H.H. & WHELAN, T. ( 1 975) Methane-derived
carbonate cements in barrier and beach sands of a
subtropical delta complex. Geochim. Cosmochim. Acta,
39, 1 085- 1 089.
RuNNELLS, D.D. ( 1 969) Diagenesis, chemical sediments,
and the mixing of natural waters. J. sediment. Petrol. ,
39, 1 1 8 8- 1 20 1 .
RusHDI, A.l., PYTKOWICZ, R.M., SuEss, E. & CHEN, C.T.
( 1 992) The effects of magnesium-to-calcium ratios in
artificial seawater, at different ionic products, upon
the induction time, and the mineralogy of calcium
carbonate: a laboratory study. Geol. Rundschau, 81,
57 1-578.
SAIGAL, G.C. & BJORLYKKE, K . ( 1 987) Carbonate cements
in clastic reservoir rocks from offshore Norway- rela­
tionships between isotopic composition, textural devel­
opment and burial depth. In: Diagenesis ofSedimentary
Sequences (Ed. Marshall, J.D.). Spec. Pub!. Geol. Soc.
London, 36, 3 1 3-324.
SASS, E., BEIN, A. & ALMOGI-LABIN, A. ( 1 99 1 ) Oxygen­
isotope composition of diagenetic calcite in organic-rich
rocks: evidence for 1 80 depletion in marine anaerobic
pore water. Geology, 19, 839-842.
SCHOLLE, P.A. & HALLEY, R.B. ( 1 985) Burial diagenesis:
out of sight, out of mind! In: Carbonate Cements (Eds
Schneidermann, N. & Harris, P.M.). Spec. Pub!. Soc.
econ. Paleont. Miner., Tulsa, 36, 309-334.
SEILACHER, A. ( 1 984) Constructional morphology of bi­
valves: evolutionary pathways in primary versus sec­
ondary soft-bottom dwellers. Palaeontology, 27, 207-
237.
211
SHACKLETON, N.J. & KENNETT, J.P. ( 1 975) Palaeotemper­
ature history of the Cenozoic and the initiation of the
Antarctic glaciation: oxygen and carbon isotope analysis
in DSDP sites 277, 279, and 2 8 1 . In: Initial Reports of
the Deep Sea Drilling Project, 29, 743-7 5 5 . Govern­
ment Printing Office, Washington, DC.
SIBLEY, D.F. & MURRAY, R.C. ( 1 972) Marine diagenesis of
carbonate sediment, Bonaire, Netherlands Antilles. J.
sediment. Petrol. , 42, 1 68 - 1 78.
STRASSER, A., DAVAUD, E. & J wou1 , Y. ( 1 989) Carbonate
cements in Holocene beachrock: example from Bahiret
el Biba, southeastern Tunisia. Sediment'. Geol. , 62,
89-1 00.
TAFT, W.H., ARRINGTON, F., HAIMOVITZ, A., MACDONALD,
C. & WOOLHEATER, C. ( 1 968) Lithification of modern
marine carbonate sediments at Yellow Ban, Bahamas.
Bull. Mar. Sci., 18, 762-82 8 .
TAYLOR, J.C.M. & ILLING, L.V. ( 1 969) Holocene calcium
carbonate cementation, Qatar, Persian Gulf. Sedimen­
tology, 12, 69- 1 07.
VISSER, M.J. ( 1 980) Neap-spring cycles reflected in Ho­
locene subtidal large-scale bedform deposits: a prelimi­
nary note. Geology, 8, 543-546.
WALDERHAUG, 0., BJORKUM, P.A. & NORDGARD BOLAS,
H.M. ( 1 989) Correlation of calcite-cemented layers in
shallow-marine sandstones of the Fensfjord Formation
in the Brage Field. In: Correlation in Hydrocarbon
Exploration (Ed. Collinson, J.D.), pp. 367-3 7 5 . Gra­
ham & Trotman, London.
WALTER, L.M. ( 1 985) Relative reactivity of skeletal car­
bonates during dissolution: implications for diagenesis.
In: Carbonate Cements (Eds Schneidermann, N. &
Harris, P.M.). Spec. Pub!. Soc. econ. Paleont. Miner.,
Tulsa, 36, 3- 1 6.
WALTER, L.M. & MORSE, J.W. ( 1 984) Magnesian calcite
stabilities: a reevaluation. Geochim. Cosmochim. Acta,
48, 1059- 1 069.
WIGLEY, T.M.L. & PLUMMER, L.N. ( 1 976) Mixing of
carbonate waters. Geochim. Cosmochim. Acta, 40, 989-
995.
WINTSCH, R.P. & KVALE, C.M. ( 1 994) Differential mobility
of elements in burial diagenesis of siliciclastic rocks. J.
sediment. Res., A64, 349-36 1 .
YURTSEVER, Y. ( 1 975) Worldwide Survey ofStable Isotopes
in Precipitation. Rep. Sect. Isotope Hydro!., IAEA,
40 pp.
Spec. Pubis int. Ass. Sediment. ( 1 998) 26 , 2 1 3-239

Geochemical history of calcite precipitation in Tertiary sandstones,

northern Apennines, Italy

K . L. M I L L I K EN*, E . F. M cB R I D E*, W. CAVAZZAt, U . C I B I Nt,

D. F O N TANA§, M . D. P I CAR D I! a n d G . G . Z U F FAt
*Department of Geological Sciences, University of Texas at Austin, Austin,
TX 787I2, USA, e-mail kittym@mail.utexas,edu; efmcbride@mail.utexas.edu;
tDepartimento di Scienze Mineralogiche, Universita di Bologna,
Piazza Porta S. Donato I, 40I 27 Bologna, Italy,
e-mail cavazza@geomin. unibo. it; zuffa@geomin. unibo. it;
tDepartimento di Scienze Geologiche, Universita di Bologna,
via Zamboni 63-67, 40126 Bologna, Italy;
§Departimento di Scienze della Terra, Universita di Modena,
via Santa Eufemia, I9, 4I 100 Modena, Italy, e-mail dfont@unimo.it; and
IIDepartment of Geology and Geophysics, University of Utah,
Salt L ake City, UT 84 I I2, USA

ABSTRACT

In order to better understand the orig in and controls of calcite cementation in marine sandstone we
studied ten Tertiary lithostratigraphical units exposed in the northern Apennines, I taly, which display
a variety of patterns of calcite cement. F ive of the units studied were deposited in piggy-back (satellite)
basins, and five were deposited in foreland basins. Burial depths of cemented units in piggy-back basins
range from 0 to 1 300 m, whereas burial depths of foreland basin rocks range from 500 to > 7000 m.
Petrographic data and stable isotopes indicate that detrital carbonate particles (rock fragments and
marine skeletal debris) in both sandstones and intercalated mudrocks were the main sources of calcium
and carbon in cement. Cementation took place at or near maximum burial depth in most of the
shallowly buried units, and at somewhat less than the maximum burial depth for the deeply buried
units. Oxygen isotopes indicate that (i) the I ntra- Apenninic Pliocene, Antognola and Ranzano
stratig raphical units were cemented by fluids with negative o 1 80 values (i.e. deeply circulated meteoric
water from nearby mountains); (ii) the Bismantova, Borello and Loiano formations were cemented by
water with a meteoric component; (iii) all the foreland basin units contain calcites with o 1 80 that is
permissive of flu ids that ranged from slightly negative to markedly positive (-2 to + 7%o ). o 1 80-
enriched values of o 1 80watec are compatible with plausible depths and temperatures of cementation of
the three deepest formations, where water evolved from silicate reactions dominated, but not for the
less deeply buried ones. Possible replacement of earlier-formed calcite by higher-temperature material
cannot be ruled out for the deepest-buried formations.
Multiple samples taken from concretions vary less than 2%o (and commonly less than l %o ) in o 1 3C
and o 1 80 from core to margin, without any consistent trend. In contrast, variations in isotopic values
between concretions in the same formation are g reater: oxygen isotopes commonly differ by 4%o
(locally g reater) within a single formation. A combinati on of variations in temperature and water
composition seems to be the cause. In the Loiano Formation there is a sig nificant di· fference between
o 1 80 values of bedding- parallel concretions and fault- parallel concretions, which reflects different times
of cementation.
Variations in Mg , F e, and Mn concentrations reflect zoning in some of the concretion calcites
examined. Covariati ons between these minor and trace elements differ g reatly both between
formations and between samples within a single formation. F rom limited temporal information it
appears that the availability of Mg and Fe was g reatest relatively early in diagnesis, whereas Mn was
mobilized in at least two distinct pulses, one early and one late.
Burial depth and its attendant temperature played a more i mportant role in diagenesis than whether
a formation was deposited in a piggy-back or a foreland basin.

Carbonate Cementation in Sandstones Edited by Sadoon Morad 213

© 1998 The International Association of Sedimentologists
ISBN: 978-0-632-04777-2
214 KL. Milliken et a!.
INTRODUCTION
Ten Tertiary stratigraphical units (Table 1) exposed
in the northern Apennines of Italy (Fig. 1) display a
variety of patterns and distributions of calcite ce­
ment. Five of the units studied were deposited in
small basins developed on top of moving thrust
sheets (piggy-back basins), and five were deposited
in foreland basins. Units in the piggy-back basins
were buried less than 1300 m, whereas units in
foreland basins were buried from 500 to 7300 m.
The three deepest units were probably buried
deeper than 5000 m. The units studied are deposits
of deep-water submarine fans and adjacent basin
plains, except for the Intra-Apenninic Pliocene (IA
Pliocene) and the lower part of the Bismantova
Formation, which are mostly marine-shelf deposits.
Sedimentological interpretations of the formations
are given by Ghibaudo & Mutti (1973), Ghibaudo
(1980), Ricci Lucchi (1981, 1986), Ricci Lucchi et
a!. (1981), Amorosi (1990), Andreozzi (1991), and
De Nardo et a!. (1992). The structural setting of the
study area is summarized by McBride et a!. (199 5).
To document the origin of and controls on calcite
cement in these marine sandstones we have taken a
twofold approach. In the first part of our study
(McBride et a!., 1995), we examined the spatial
distribution of concretionary bodies and completely
cemented beds in seven of these units. We use the
•PIACENZA
Fig. 1. Locality map. See Table I for locality numbers.
term concretion for cemented parts of beds that
contrast markedly with the surrounding unce­
mented or poorly cemented host rock. No particular
shape is implied.
In the units we examined first (IA Pliocene,
Bismantova, Antognola, Ranzano, Loiano, Borello
and upper Marnoso-arenacea), calcite cement oc­
curs as concretions of diverse size and shape em­
bedded in weakly cemented or uncemented host
sands. Concretions comprise from 10 to 30% of an
outcrop, and are characteristic of both the shallow­
marine deposits and submarine-fan deposits that
have thick sequences of sandstone with almost no
mudrock interbeds. Concretions are chiefly equant
or subequant, but some are tabular. Some are
evenly spaced along beds or along faults, whereas
others are random or selective to claystone clasts
(Fig. 2). Beds completely cemented by calcite char­
acterize the lower Marnoso-arenacea, the Monte
Cervarola (M. Cervarola), Monte Modino (M.
Modino ), and Macigno formations, and occur lo­
cally in the Bismantova and Ranzano formations.
Except for the latter two, completely cemented beds
occur in formations that have been more deeply
buried and attained higher temperatures (Table 2).
These deeply buried sandstones are non-channelized
turbidites interbedded with calcitic mudrocks. The
[2] Late Eocene to Pliocene Piggy-back Sequences
12221 Oligocene-Miocene Foreland Sequences
c:::::J Jurassic to Middle Eocene Ligurian Units
• Sampled localities
.50 km
'------'
Table 1. Summary ofs tratigraphy , facies and es timated burial depths of the formations s tudied

Thick nesst Es timated burial

Formation and s ampling areas * Aget (m) Faciest Sand/s halet depth (m)t

Piggy-back basins
Intrapenninic Pl iocene Pliocene 200- 600 Fan-delta congl omerates and s ands tones 20 : I 100 -300
(9) Pianoro Vecchio
( 10 ) Livergnano
( 1 2) Val Zena
Bis mantova Middle Miocene 200- 800 Bioturbated s helf marls and s ands tones , I : 5 to 3 : 1 200- 700
(4 ) Vetto fi ne- to medium-grained
(6) Pavullo unchannelized turbidites
(8) Val Savena
( 1 3) Monterenzio
Antognola-Nivione§ Early Miocene 250- 650 Coars e-grained channelized turbidites 0 to 20 : I 1 250- 1 5 50
0
( ! ) Vall e di Nivione within hemipelagic marls
fS
<�;'
(5) Carpineti ":1
Ranzano Late Eocene- early 500- 1000 Medium- to coars e-grained channelized I : I to 20 : 1 600 -700 �
(2) Val Pess ola Oligocene and unchannelized turbidites �·
(3) Val d' Enza �­
Loiano Late Eocene 300- 1000 Coars e-grained channelized turbidites 20 : I 1 250 - 1 600
(7) Vado
g.
;:s
(8) Val Savena ;:;·
( I I ) Loiano
�
;:c.
Foreland basins ;s·
Borello Pliocene 200 Fine- to medium-grained lobe turbidites I :4 700 - 1000 �
( 1 6 ) Predappio iS
Upper Marnos o-arenacea Middle- Late Miocene 1 100 Fine- to medium-grained lobe to 3 : I to 20 : I 2500 ;:s
( 1 5) Fontanelice channelized turbidites �
Lower Marnos o-arenacea Early- Middle Miocene 2900 Fine- to medium-grained bas inal I : 2 to I : 5 ?4000 (3'
;:s
( 14 ) Moraduccio turbidites and lobes �
M. Cervarola Early- Middle Miocene ? 1000 Fine- to medium-grained bas inal 1:1 ? 3500- 5500
( I 7) Pracchia turbidites and lobes
M. Modino Early Miocene 700 Medium- to coars e-grained bas inal 2: I ? 3000- 5700
( 1 8) Pievepelago turbidites and lobes
Macigno Late Oligocene-Early 2300 Medium- to coars e-grained bas inal 7: I ? 3000- 7300
( 1 9) Gordana Valley Miocene turbidites and lobes

Piggy -back bas in references : Ghibaudo & Mutti ( 1 9 73), Ricci Lucchi et a!. ( 1 98 1 ), Ricci Lucchi ( 1 986), Bettelli et a!. ( 1 987), Cavanna et at. ( 1 989), Amoros i
( 1 990 ), Rio D. In: Carta Geologica dell'Appennino Emiliano-Romagnolo 1 :50 ,0 00 F. 2 1 7 ( 1 990 ), De Nardo et at. ( 1 992).
Foreland bas in references : Ricci Lucchi ( 1 9 8 1 , 1 986), Carta Geologica dell'Appennino Emiliano-Romagnolo I :25,000 F. I00 III NO-NE ( 1 982).
*Numbers refer to Fig. I .
t Data refer t o formations i n the s tudy area.
t Es timated depth to s amples us ed for is otopic s tudy . N
-
§ Nivione is a s mall s ands tone turbidite body enclos ed within the marls of the Antognola Formation.
Vl
216 K.L. Milliken et a/.
A D
c He mispherical objects a re
EXPLANATION
D Sandstone
D Shale
� Calcite cement
• • Clay clast
t Graded bed
- Major bedding plane
- Minor bedding plane
o Concretion <?em in diameter
1m
M. Cervarola, M. Modino and Macigno formations
contain only small amounts of calcite cement, be­
cause intergranular volume (IGV) was reduced to
< 10% before cementation took place.
Cementation occurred chiefly by diffusive supply
of Ca2+ and HC03- derived from detrital carbon­
ate grains uniformly distributed in sandstone beds
and, in formations with mudrock interbeds, from
detrital and biogenic carbonate in mudrocks. Local
factors, many of which remain unidentified, influ­
enced the cementation process and resulted in
substantial heterogeneity in the distribution and
form of calcite cement (Fig. 2).
This paper focuses on the petrographic, isotopic
and trace element compositional characteristics of
associated detrital and authigenic carbonate. The
Fig. 2. Ske tch showing the dive rse
patterns of ca lcite ce me nt in
sa ndstones of this stmly (from
McBride et a!., 1995). (A)
Sa ndstone be ds inte rbe dde d with
mu drock are comple te ly ce mented
by calcite ( I ), ce me nted in two
rows of sphe roidal concretions
(Borello Forma tion only, 2),
cemente d only in the lowe r
one-third (3), or are ce mented in
concretions located chiefly at the
base of the be ds (4). (B) Stacke d
sa ndstone be ds a re ce mente d in
ta bu la r concretions pa ralle l with
be dding (2) or pa ra lle l with faults
( I ). (C) Stacke d sa ndstone be ds a re
ceme nte d in sphe rical concretions
mostly smaller than 7 em (I ), as
subequant, regu la r or irregu la r
concretions, ma ny with nu cle i of
clay rip-u p cla sts (2), or as ta bu la r
concretions possessing a prefe rre d
orientation of long axes (3). (D)
Sta cke d shelf sa ndstone s with
low-angle , wa ve -forme d cross-beds.
mollu scs. Ca lcite cement weakly
encloses pla ce rs of molluscs (I ),
conforms with major
cross-la minations (2) or conforms
with and encloses cross-la minae
pe rfectly (3). (E) Sta cked
sa ndstones with ce me nt in prolate
sphe roids that are uniformly
space d and e ithe r sele ctive to the
middle of be ds ( I ) or not (3), and
a s elongate concretions that a re
selective to the tops of graded be ds
(2).
authigenic carbonate occurs as concretions, com­
pletely cemented beds, grain replacements and
veins. Associated detrital carbonate was examined
to aid the interpretation of whole-rock data.
SAMPLING AND METHODS
We sampled calcite-cemented concretions and beds
and, in places, took samples from host sandstones
immediately adjacent to concretions. Commonly,
more than one sample per concretion and host bed
was taken to evaluate variability at a sample site.
Samples were taken using either a hammer or a core
drill. Host sandstone samples were taken as close as
practical to the concretions and within the same
Table 2. Data from fluid i nclusions, vitrinite refle cta nce , a patite fission tracks, a nd e sti ma ted burial depths a nd te mpe ratures

Vitrinite reflecta nce (R0%)

Flui d inclusions* (number of sa mple s a nalysed)t
AFT data T ("C) esti ma te d
Lithostra tigra phica l units Tmax ("C) Sa li nity (ppt) Study I Study 2 Study 3 (estimate d T("C)) from burial de ptht
Q
Piggy-back basins r;-
lA Plioce ne 0.67(2) 7-11 �-
Bi smantova Un <50-75 9-19 '"<:l
Antognola 0.46(2) 0.39(6) 0.49 (l ) 30-36 �
Ra nza no <60-100 0 0.56(2) 0.54(8) 0.42(2) PA 50-125 17-19 'i:l"
Loia no <60 17 0.42(2) 0.59(4) Un <50-75 30-37 §"
�-
;::::
Foreland basins
Borello 19-25 s·
Marnoso-Are n. (upper) 0.49(3) PA
Marnoso-Aren. (lowe r) 0.52(3)
50-125 55
�
130 10 0.68(13) 0.38(15) 85 �
M. Ce rvarola 225 0-17 1.15(5) 1.00(3) Tot A >100-125 75-115 E;·
M. Modino 140 >200 0.88(2) Tot A >100-125 65-119 �
<60 35-50 iS
;:s
Macigno 265 17 1.86(6) 1.61(2) Tot A >100-125 65-149
�
AFT, apatite fission track; PA, pa rtially a nnea le d; Tot A, totally unanneale d; Un, unannealed. 0
;:s
Flui d inclusion da ta from our sa mple s by Flui ds Inc., not correcte d for pressure . Data are for secondary i nclusions; se cond value s for M. Modino are from �
primary i nclusions. Secondary M. Modino i nclusions contain hydrocarbons. Macigno fluid inclusion da ta are from qua rtz crysta ls. R0 values are averages of
the number of sa mples in pa rentheses.
*Data from ca lci te vei ns that postdate ca lcite ce me nt.
tStudy I from Ruetter et a!. (1983); Study 2 from Fa illa (1987) and Fai lla & Mezzetti(1987); Study 3 from thi s study; da ta by DGSI.
t Assuming thermal gra die nt= 20"C/km; sea -floor T= S"C.

-.J
218 K.L. Milliken et a!.

bed. Particularly friable host rock samples were
taken by excavating a rectangular trench in a bed
and embedding the samples in plaster of Paris
before they were removed from the outcrop.
Oxygen and carbon isotopic analyses were made
on 166 concretion samples, 34 host rock samples, 54
calcite-cemented beds, 33 veins and 38 mudrock
samples. Samples were reacted with I 00% phospho­
ric acid at 2s·c and the extracted gases were ana­
lysed on a VG Prism gas-source mass spectrometer.
Isotopic values (machine reproducibility ± 0.02o/oo
are reported relative to the PDB standard.
The trace elemental composition for carbonate
was determined by WDS on a JEOL 733 electron
microprobe. Accelerating voltage was 15 kV; sample
current was stabilized at 12 nA on brass; spot size
was 10 Jlm; count time for all elements was 20 s,
except for Sr, analysed for 60 s; standards were cal­
cite (Ca), dolomite (Ca, Mg), coral (Sr) and siderite
(Fe, Mn); beam placement was in every case guided
by the back-scattered electron image; Si was rou­
tinely counted by WDS in order to monitor possible
contamination from adjacent silicate grains not re­
vealed by the back-scattered image; totals between
97 and 103 wto/o were accepted. Detection limits are
approximately 340 ppm for Mg, 450 ppm for Fe,
310 ppm for Mn and 185 ppm for Sr.
The carbonate content of mudrocks was deter­
mined by weight loss upon reaction with 1Oo/o HCl
(Table 3). The mineralogy of the mudrocks was
determined by XRD (X-ray diffraction) of ran­
domly oriented whole-rock powders; mineral per­
centages were calculated using the method of Lynch
(in review). Further evaluation of the clay minerals
was accomplished from oriented slides made from
the < 2 Jlm fraction (glycolated and unglycolated;
method described by Lynch, 1997).
Apatite fission track ages and thermal histories
(Boettcher & McBride, 1993) were obtained using
the external detector method (Naeser, 1979).
PETROGRAPHY
General features
Sandstones in all units are mineralogically imma­
ture, and there is considerable diversity in their
compositions (Fig. 3). They range from arkose to
litharenite, with considerable variations in the type
and relative abundance of rock fragments.
Compactional modifications of depositional
grain fabrics range from moderate for the shallowly
buried Pliocene sandstones to severe for the fore­
land basin formations (Cibin et a!., 1993; Lunar­
dini, 1993). Average IGVs in percentages for the
formations are: Pliocene 22.8, Bismantova 19. 7,
Loiano 15.3, Antognola 11.6, Ranzano 10.9, upper
Marnoso-arenacea 13. 5, lower Marnoso-arenacea
15.0, M. Modino 6.4, Macigno 3. 5 and M. Cer­
varola 8.9. Values below 26% indicate that compac­
tion involves more than just grain rotation and
rearrangement (Graton & Fraser, 1935; Fucht­
bauer, 1967). Thin sections show that compaction
also took place by a combination of ductile grain
deformation, pressure dissolution and grain fractur-
Q

Table 3. Acid-soluble carbonate in mudrocks

Formation Carbonate No. of
M�rnoso-arenacea
(wt%) samples

IA Pli ocene ND
Borello 38.5 2
Bismantova 45.8 4
Antognola 45.2 3
Ni vione 50.7 2
Ranzano 33.3 6
Loiano 1 4. 5 I
Marnoso-arenacea (upper) ND
Marnoso-arenacea (lower) 43.1 46 F L
M. Cervarola 29.0 3
M. Modino 1 9. 1 2 Fig.3. QFL triangle for 10 stratigraphical units. Data
Macigno 1 4.8 5 from Valloni et al. ( 1 99 1 ), Bruni et al. ( 1 994) and this
study. Dashed lines indicate foreland basin units. Q ,
Carbonate was determined by weight loss after reaction quartz ; F, feldspar, L , lithics, i ncluding ex trabasi nal
wi th HC
I (ND, no data). carbonate rock fragments.
 Calcite precipitation in Tertiary sandstones 219

ing. Pressure dissolution of carbonate grains is

widespread. Sutured quartz and silicate grains are
widespread in the foreland basin samples. In the
foreland basins, thrust sheets added to the strati­
graphical overburden.

Detrital and authigenic carbonate in sandstones

Detrital carbonate grains, including limestone, bio­

clasts and rare dolomite, are important components
of the sandstones, except for the Loiano, Macigno
and M. Modino formations. These strata contain less
than 5% detrital carbonate, whereas the other forma­
tions contain from 8 to 30% detrital carbonate.
Calcite is the dominant cement in all of the
sandstones except the chlorite cement-rich Ran­
zano. The absence of volumetrically significant
cements other than carbonate precludes a definitive
placement of carbonate cementation within a pro­
gression of diagenetic events. No petrographic evi­
dence marks any particular group of concretions as
temporally distinct from another. IGV provides a
crude estimate of burial at the time of cementation,
and suggests that most of the concretions formed
after considerable compaction.
Detrital carbonate grains (including bioclasts)
serve as nuclei for calcite cement in most of the
sandstones. The size of the cement crystals is con­
trolled in part by the fabric of calcite in the detrital
grains. Thus, micritic limestone grains have micro­
crystalline calcite overgrowths in the first layer of
cement, and crystals become progressively larger
Fig. 4. Localiz ation of authigenic calcite (c) on
away from the grains. Single detrital crystals of spar
dissolving silicate grains. Back-scattere d electron images.
have sparry overgrowths the size of the adjacent (A) Calcite replace s parts of two K-feldspar grains (K)
pores. Limestone clasts composed of unequal crystal from the M. Ce rvarola Formation. (B) Partial
sizes have polycrystalline overgrowths with corre­ calcitization of an epidote grain (e ) in the Bore llo
Formation.
spondingly variable crystal sizes. As a result, the size
of calcite cement crystals in rocks with abundant
detrital carbonate grains is quite uneven, ranging
from 0.03 to 0.2 mm. At the thin-section scale dis­ between calcite and Ca-plagioclase, and also heavy
tribution of calcite cement in concretions tends to be minerals (Fig. 48).
pervasive, showing no clear preference for localiza­ Volumetrically minor authigenic ferroan dolo­
tion on microcrystalline rather than more coarsely mite is present locally as overgrowths on distinctive
crystalline carbonate grains. Poikilotopic cement cores of partially dissolved detrital dolomite
texture is essentially absent. (Fig. 5). Prominent zoning of Fe and Mg in these
Detrital silicate grains also play a role in nucle­ overgrowths is readily observed on back-scattered
ation of authigenic calcite. In samples with non­ electron images. Dolomite precipitation clearly pre­
pervasive cement, much calcite is localized within cedes the formation of calcite cement.
and around partially dissolved detrital silicate Petrographic evidence for pressure dissolution of
grains. K-feldspar is the most common detrital detrital carbonate particles is widespread (Fig. 6),
mineral that is spatially related to calcite in this and is even manifested in the youngest and least­
manner (Fig. 4A); a similar relationship is observed buried units (Pliocene). An overall similarity in the
220 KL. Milliken et a!.

Fig. 5. Au thigenic fe rroan dolomite overgrowth (o) on a

core of fractu red de trital dolomite (d) from the
Bismantova Formation. Overgrowth de velopment
prece de d pre cipitation of calcite ce me nt. Similar
dolomite overgrowths are obse rve d in the Borello
Formation. Back-scattered e le ctron image .

abundance of detrital carbonate i n concretions and

host rocks suggests, however, that any loss of detri­
tal carbonate from the sandstones in the period
postdating concretion formation is not readily de­
tected (Cibin et al., 1993). There is no correlation
between the amount of calcite cement (i.e. the
volume of concretions) and the amount of detrital
carbonate in the host sands of the various forma­
tions.
Fig. 6. Pre ssu re dissolu tion of Foraminife ra. Dark grey
grains are quartz . Calcite ce me nts the grains and also
Calcite in mudrocks
fills the intraskeletal pores. Back-scatte red e lectron
image s. (A) Bismantova Formation. Large grain in lowe r
Mudrocks from both types of basin are grey and right is a mica. (B) Borello Formation.
calcitic, dominated by clay minerals and calcite,
and 5-15% quartz, feldspar (plagioclase >> K­
feldspar) and mica. They are largely hemipelagic more carbonate in their mudrocks than other for­
and pelagic beds, but include some Bouma E inter­ mations. These observations, plus thin-section pe­
vals. Clay minerals are dominated by illite and trography, indicate that carbonate in mudrocks is
chlorite, but mixed-layer illite/smectite, serpentine, both terrigenous and pelagic-biogenic in origin.
corrensite and possibly kaolinite are present in
some samples. Pyrite is ubiquitous.
The carbonate content of mudrock interbeds GEOCHEMISTRY OF
averaged by formation ranges from 1 7 to 59% (the CALCITE CEMENT
latter are more properly clayey limestones) (Table
3), about 3-10 times the amount of carbonate in
Carbon and oxygen isotopes: general trends
sandstones. There is no correlation between the
amount of carbonate in sandstones and associated For whole-rock isotopic data there is major overlap
mudrocks, but formations whose sandstones have in values for most of the shallowly buried forma­
abundant detrital carbonate grains generally have tions (Figs 7 and 8) across a wide range in both 81 3C
 Calcite precipitation in Tertiary sandstones 221

1.0,.-------, 1.0 ,.-------, 2.0,------,

o oe
�o
BORELLO I. A. PLIOCENE
tc 0 .l .....
o .o .... ...... ......... ... . ........ 0.0 -- --- O· · · · · · - 0.�. . . .
-·-····································-···-··············c ···
0.0 . ·············· •··
.o••
,-
. � &8 .
0 . ..o '
• .. -2.0 ·.
· :. \
� 1
- .0 %•.. .. -1.0
.
...._-: .., - ".0
.. . ,_
J? .
K> ·2.0 -2.0 .
.-.o
.

• Concretion
..
MARNOSO-ARENACEA • Concretion
3
- .0 o -3.0
-8.0 o
Upper part
Detrital
• Concretion Detrital
• • Shale
o Detrital
Shale :-
"
- .0+---.--,----.--,.---1 -".0-f-----.-----1 ·10.0-f---'0C,0---.----.---1
-s.o -s.o -4.o -3 .o -2.o -1.o o. o ...a - .0
5 -4
- .0 -3.0 ·2.0 -1.0 0.0 -fi.O -5.0 -".0 3
- .0 -2.0 -1.0 0.0

2.0 ,.-----------------. 1.0 -=---- :-:--- :- -...-- . --

t.4AANOSOAAENACEA : lower part
- ,-= -, 1.0 -------- -
.. • ,--A-N-::- -
T OG-NOLA -. .,.,.
• ,-,
···························································· ··· .......... o..... .
0.0 0.0
81SMANTOVA
1.0
•
• .6
.\ . -1.0
• •
-1.0 0
. ..
.6
.. .... .6 0 ·r\ •
0
.
o .of-----,;- .- . 0 .,
Concretion
-2.0
• • . .6
,---
•
• -, _a__:::_
_ _ _...__::___ !l -2.0 0 'Y
• * :·. ·· • .6 .6

-3.0
-3.0
Authigenic

1 1
- .0 + .
- ".0
•
o
Concretion
-".0
o
.6
Detrital

Shale
.
Detrital
.s? -5.0
- 2- .0 5
- .0
•
- e.o
+
8.0
.. ..
• Bod
.6
"'
6
- .0
Shale
-7.0
.6
-3.0
Shale .
+
7
- .0
Vein
..
+
-(1.0
Vein

-4 - .o
-�to.o-9 .o -s.o -7.0 s
- .o -s.o __._o -3.0 - .o -1.o
2 o.o - --
a - &-f.o -• - ---•3.0
- -- _..• o - --2.
•- o
- --•-1.-o----l .g_�1-o
t -.o- '
5.o o_o - 9- .o -s-' . -o--'7 .-o-'s
- - .o-'-5.-o--•' .-o--'
3.-o--'2.-o- '-1-
. o----i
o .o

2.0 ,.------�--, 10.0,------,---, 5.0,-------,--,

5.0
1.0
RANZANO

•
+ . .. . LOIANO

0.0 ... ........ . . . ... ...................«...�---·

; 0.0 i'"'"
LOIANO
:--·-- . . -· ·-·-

� .... .. ......±.. ......e.. ........... � ........................�.... . .. .

0.0
+ +·.·
·-··
-5 0 ° �.,.
f ,Jr._-,
+-
.• +. · . ..
:
j
-5.0
.. . .
.
-1.0 ... . -10.0 . .
�
....
+• •:, + -10.0
..
+
..
. • -15.0 •
�-2.0 .
Concretion Concretion
+ o
-20.0 -15.0
Detrital " Authigenic
C.() 3
- .0
"'
•Bed o,
-25.0
Detrital
• •
.6
-20.0
B&d parallel
4
- .0
.6
Shale
-30.0
Shale

+ •
+
Fauh parallel
Vein

-2 . o
Vein

·5.0.+ --,s - --. 'T.o ---,-2.0-�-'

_ • .o -.-'-3
O - -
s.o - .o 1.o-.---
o.-j o -3S. -fi.O -s.o -4.0 3
- .0 - .0 ·1.0
2 o. o 1.0 2.0 3.0 s +_ 6 .-0�' -s.o
--�r .o� 3-, -
.o-' 2.
- 0- -1
--, .0
�- .
o+ 0-- 1.r
0 --2.0
1
li11oxygen 1\ 180xygen I) 1Soxygen
1.0,------, 1.0,-------, 1.0,------,
M&
0.5 + M. WOOINO
0.0 ---....-__.____________________________
MACIGNO

o. o-j
: J'
---------1
M. CERVAROLA 0.0 1---t--..---_._
_____,_,_.__
----+:;:-:;+c-�:---;;.--
-1.0 -1.0 �
..
·0.5 ••
� 1
- .0
-2.0 •
-2.0
+
•
� -1.5 •
-3.0 -3.0
� -2.0
• 8.0
•Bod
-2.5 oiled -".0 -".0
"'
• Shale
-3.0 • +
·5.0
Calcite vein
+
6.0
Shale Vein
-3.5 + Vein • Shale

-".O .f.o+--.--,.-.--.--,---.---.-�-.-----1 -e.o-f-�,.---.--.:.---.--.---,----,--1

-
16.0 6.0
·1-4.0 ·12.0 ·10.0 -8.0 - -4.0 -2.0 0.0 - 6.0 1
1 - 2.0 ·10.0
- -4.0 1 - .0
8 - .0 ·-4.0
6 -2.0 0.0 -11.0 -1-4. 0 -12.0 ·10.0 -8.0 -6.0 --4.0 - .0
2 0.0
18
B 1)180xygon I\ 0xygen
1)1 Qxyggn

Oxygen and ca rbon isotopic data for the stra ti graphical units e xa mined in this stu dy. All da ta , e xce pti ng
Fig. 7.
Antognola a nd Loiano, a re for whole rocks (u ncorrecte d for de trital versu s au thigenic conte nt).

and 81 80. The more deeply buried formations have and authigenic calcite (usually detrital grains with
lower 81 80 values, although the shallowly buried authigenic overgrowths). Thus, oxygen values as
Antognola samples overlap the more deeply buried 1 80 depleted as - 1 5o/oo in the Macigno may reflect a
and 1 80-enriched M. Modino samples). The more component of even more 1 80-depleted authigenic
deeply buried formations lack concretions and the calcite, assuming that some of the relatively 1 80-
8 1 80 values are from beds that contain both detrital enriched detrital calcite remains unreplaced. Such
222 K.L. Milliken et a/.

• Borello
.a.
0.0
lA Pliocene
+ Marnoso-a Upper

!
..,
0
-

·10.0
5
.0 •

c
Marnoso-a Lower
Bismantova

-c.o o Antognola
4 Ranzano
·1 .0
5
0 Loiano
ALL CONCRETIONS & BEDS
-20.0 + M. Cervarola
n-211
• Macigno
-�.0+-�
���-r�,-���r-��-r� e M. Modino
·16.0 -14.0 ·12.0 -10.0 -8.0 -6.0 -4.0 ·2.0 0.0 2.0

2.0

0.0
8
··················

+ 0
• �
�· ·············· ···li. \tlt
. .
�:
---····
•

•
Borello
Bismantova
+ + Antognola
• c:
c: ·2.0 •
• • Ranzano
c
�as 0 c Loiano
-4.0
Shallowly
0
.., Deeply buried buried 0 Marnoso-a Upper
�
-6.0 1:;. Marnoso·a Lower

ALL SHALES 0 M. Cervarola

-8.0 n - 37 ..: + Macigno
• M. Modino
-10.0
·14.0 ·12.0 -10.0 -8.0 -6.0 -4.0 ·2.0 0.0 2.0

• Bismantova
• Ranzano
c: Deeply
� -4.0 buried Shallowly buried + Loiano

0 + • Marnoso·a. Lower
.., -6.0
+
+ c M. Modino
c.Q •
- .0
8 0 Macigno
·10.0 1:;. M. Cervarola

-12.0
CALCITE VEINS
n -32 + Fig. 8. Isotopic data (whole rock,
-14.0 -l--.2.0..---,-...1,-..,--.--r---r--T---t uncorrected). (A) All concretions
·16.0 ·14.0 ·12.0 -10.0 -8.0 -6.0 -4.0 ·2.0 0.0
and be ds. (B) Mudrocks. (C)
B180xygen Calcite veins.

low o 180 values necessarily reflect elevated precip­
itation temperatures (> 700C for o'80water < - 5o/oo;
> l20"C for positive values of o'80water).
Among the shallowly buried formations, the
Loiano is distinctive because it has three relatively
early-formed concretions (IVG 32%) that have
distinctively high 0180 values and more 13C-
depleted carbon isotopic composition than later
concretions. In all formations except the Bis­
mantova, carbon isotopic values are more 13C
depleted in concretions than in detrital grains. This
reflects the greater contribution of organic carbon in
= the calcite cement component of the whole-rock
data. Oxygen isotopic values for host samples (pri-
 Calcite precipitation in Tertiary sandstones 223

marily detrital carbonate) tend to be somewhat
more 1 80 depleted (about 2%o) than whole-rock
values of concretions and beds for several forma­
tions; these differences are statistically significant
for the Antognola (significance level oft-test > 0.95)
and Loiano formations (significance level > 0.98).
The more 1 80-enriched oxygen in calcite cement
must be the result of precipitation of calcite from
marine formation water whose oxygen isotopic
content was enriched in 1 80 during silicate diagen­
esis (see below). Although detrital dolomite in the
Loiano Formation may contribute somewhat to its
relatively 1 80-enriched whole-rock average oxygen
value, the oxygen values for calculated pure calcite
cement are also among the highest of all the forma­
tions.
We found no difference in isotopic values be­
tween concretions and completely cemented beds
where both are present (Ranzano and Bismantova
formations).
Concretions and beds of all formations contain
mixtures of detrital and authigenic calcite that
cannot be separated for isotopic analysis. Detrital
rock fragments and skeletal grains are mostly cal­
cite, but a few samples of the Marnoso-arenacea
and Bismantova formations have more than 10%
detrital dolomite (among total carbonate), which
shifts whole-rock oxygen isotopic values to more
1 80-enriched values. Isotopic values for host sand­
stone with minor amounts of cement provide an
estimate of the average isotopic composition of
detrital rock fragments and bioclasts. Using the
average isotopic values for host samples as the
composition of detrital calcite grains, and determin­
ing the relative amounts of detrital and authigenic
calcite of representative samples from point counts,
the isotopic value for oxygen of authigenic calcite
was calculated from the following formula:
8180wr = X(8180detri tal) + ( l - X)(8180authigenic)
where X is the fraction of detrital calcite in total
calcite and 8180wr and 8180authigenic are the oxygen
isotopic values for the whole rock and authigenic
calcite, respectively. Carbon isotopic values were
treated similarly. These calculations were made for
representative samples of all formations with con­
cretions except the Borello Formation, from which
no host-rock samples were taken. Table 4 shows
that this calculation reveals differences between
813Cwr and 813Cauthi enic of l -2%o for only four
g
formations; corrections of this magnitude for
8180wr are required for only two formations. Cor­
rected values from Table 4 are used to make
interpretations of isotopic data.
The calculated isotopic values for authigenic
calcite for each of these units have somewhat
greater ranges than for whole-rock-samples. As ex­
amples, values for whole rock and authigenic calcite
for the Antognola and Loiano formations are shown
in Fig. 7. The uniformity of whole-rock isotopic
values for the stratigraphical units in general, and
the Ranzano in particular, are the result of the
uniformity of the isotopic values of detrital calcite
grains and their damping effect on the whole-rock
samples. The accuracy of our calculated values is of
course uncertain, because all host samples contain

Table 4. Summary of isotopi c data and i nterpre tation of te mpe rature and de pth of ce me ntati on

Corre ction to WR Calculate d values Calculate d

T ("C)* from Calculate d
o l3C 1)1 80 o 1 3C 1) 1 80
Stratigraphical unit burial data• 0 1 80water

lA Plioce ne 0 +I -3 to - I - 5 to -2 7-1 1 - 6 to-I

Bismantov a +I to + 2 + 2 - 1 to + 2 - 2 to 0 9- 1 9 - 3 to + 2
Borello 0 0 -I to 0 - 3 to - I 1 9- 25 - 1 to + 2
Loiano 0 0 - 20 to + 2 - 4 to 0 30-37 0 to + 5
Mamoso-are nacea (upper) - 2 0 - 1 1 to-I - 5 to- 3 55 + 3 to + 5
Ranzano -I 0 0 to- 2 - 4. 5 to- 3 1 7- 1 9 -4 to- 2
Antognola 0 0 - 7 to- 2 - 7 to- 6 30- 36 - 4 to- 2
Mamoso-are nacea (lowe r) - 2 0 - 3 to-I - 7 to- 4 85 + 5 to + 8
M. Cervarola 0 0 - 1 to +I - 1 1 to- 8 75- 1 1 5 0 to + 3
M. Modino 0 0 -2 to 0 - 1 2 to- 7 65- 1 1 9 - 2 to + 3
Macigno 0 0 - 3 to 0 - 1 5 to- 1 3 6 5- 1 49 - 5 to + 3

lA, Intra-Apennine ; WR, whole rock.

*Assumes sea-floor te mperature of 5"C and a ge othe rmal gradient of 20"C/km. Buri al depths from Table l. Isotopic
values are rounde d off.
224 KL. Milliken et a!.

B�
PLIOCENE
A
�m®�
�a.yaMt•�
r--:::
- : --1 37cm
36cm

S''c 0111 0 S13c 01'o

A • 2.0 • 3.7 A -1.6 . 3.4
B . 2.0 . 3.7 B -1.4 . 3.2

o"o c -2.0 . 3.6 c -1.5 • 3.5

I
c 813C s "o

50 em
� D
A
B
c
D
·1.6
. 1.3
. 1.5
. 1.8
. 3.7
. 3.0
• 3.3
-3.1

l
80cm

0
D
S ''c
P1 -1.1
1

0
P2 -1.3

BISMANTOVA FORMATION
E 60cm

s 'b s "o
�� :�� :�� RANZANO FORMATION
17 -1 8 -3.8
H

0
s'b s 'b
•
50 em
41 -1.9 3.5
42 ·0.9 . 3.2

F ANTOGNOLA FORMATION 40cm

�
� 28
29
� 'b
. 6.5
-4.4
� 'b
. 5.9
• 6.9
c
__ ___;:,;.
_....
54• ss-_
..;;:;:... :::> 54
29
•
s ,3c
0.6
. 0.6
o "o
. 3.0
. 3.7

30 -8.2 . 0.2 400cm

Shaleclast

55 em

JG
013c o''o
G 149A -1.2 -4.1
1498 -1.3 -4.3

0
149C . 1.6 . 4.1
513C 6110 22cm
34 . 1.9 . 7.3
35 35 -4.7 . 6.5
K

g
B13C s"o
150A • 0.9 . 3.8
1508 ·0.9 . 3.9
150C . 1.3 . 4.3

65cm 20cm

RANZANO FORMATION RANZANO FORMATION

N
-0�6 s1 3 c
!o.3

-0�5
-0:2 -0� 1

M -1�8 0
�2.4

-1:0 -:i.o
Fig. 9. Spatial vari ati on of carbon
and oxygen i sotopes wi thi n
concretions and beds.
 Calcite precipitation in Tertiary sandstones 225

p LOIANO FORMATION some cement and some samples contain anoma­

lously large and irregularly distributed detrital car­
bonate clasts_
o 13c o"o
1578 -6. 0 -3.5
1578 1579 -4.4 -3.4 Comparisons of isotopes within and
85 em 15808 -5.0 -2.9
-4.8 among concretions
158 1 -3.4
a
Val Savena In order to determine whether isotopic/temperature

� o 13 c o"o conditions were uniform during concretion growth,

157 0 -2.3 -2.3 multiple samples (2-30) were collected from 22
1571 -3.6 -2.3
concretions (Fig_ 9), analysed and compared_ In
220 em·
most formations the differences between adjacent
Vado
R
samples differ by less than I %o in 8180 and less than
o"'c o"o 2%o in 813C (Fig_ 9). Oxygen isotopic variations
1565 -2.3 -1.7 among concretions in the same stratigraphical unit
1566 -2.3 -1.6
1567 -2.6 -1.6 are greater than within concretions. Adjacent sam­
ples within the same concretions comonly differ by
100cm less than 1 %o in 8180, whereas different concretions
of the same unit commonly differ by 4%o. The

. -I�
s centre of concretions is not consistently more 180
8elluno
o13C o '• o depleted or more 180 enriched than concretion
250 -0.52 -0.89 margins. 813C values differ by about 2%o within and
251 -0.33 -2.36 between concretions. Multiple samples for com­
pletely cemented beds show similar variations to
50 em
concretions.
BORELLO FORMATION One spherical IA Pliocene concretion was sam­
pled at nine locations in a plane perpendicular to
o13c o'•o bedding. This showed a concentric pattern of both
245A -0.25 -1.9 0 813C and 8180 values, although the variations are
2458 -0.20 -0.54 small (Fig. 9A). From the nucleus outward carbon
245C -0.39 -1.58 values become 13C enriched by I %o and oxygen
246 -0.42
values become 180 depleted by 1 %o. These trends
-2. 0 0
are the most common found in isotopes in concre­
247 -0.3 0 -2.10
tions in marine rocks (Mozley & Bums, 1 993), and
60cm 249 -0.69 -0.95 are compatible with a decrease in organic carbon
contribution with time, and either an increase in
meteoric water component or increase in tempera­
u MARNOSQ-ARENACEA (Inner Belt) ture with time. We assume that concretions grew
from their centre outward beqmse we find no
hollow ones (but see Coleman, 1 993). An elongate
o'3 c o"o lA Pliocene concretion ( 1 00 em x 20 em, not de­
12 +0.4 -2.1
11 +0.6 -2.9 picted in Fig. 9) from a different locality was sam­
10 - 0.3 -5.2
9 - 0.4 -4.7 pled at 30 places on a grid. Carbon values differ by
8 -3.1 -5.8 0.9%o and show one 20 em diameter concentric
7 - 0.3 -5.4
pattern, whereas oxygen values differ by 1 . 1 %o and
show no pattern.
A spherical Ranzano concretion was sampled at
013C o'•o six places on a grid and at two adjacent places in the
- 0.3 -3.3
+0.4 -3.3 host sand. Carbon values become 13C enriched out­
+0.3
+0.4 -3.7 ward, like the spherical concretion in the IA Pliocene
+0. 1 -4.6 above, but, contrary to the latter concretion, so do
-0.3 -4.6
oxygen values become enriched (Fig. 9E). One
Fig_ 9- (Continued).
226 K.L.
spherical concretion from the Borello Formation
was sampled in three places on a vertical plane and
three on a horizontal plane. There is no trend in
carbon or oxygen isotopes on either face (Fig. 9T).
Concretions aligned along faults occur in the
Loiano and Bismantova formations. In the Loiano
at Locality 8, i5 1 80 values for bed-parallel calcite
cement average -2.9%o, whereas those values for
fault-parallel calcite average -1.3%o (Fig. 6). The
differences are significant at the 95% level using the
t-test. Field relations do not indicate the relative
ages of the two types of concretions (faults do not
intersect bed-parallel concretions). If faulting oc­
curred later than cementation of the bed-parallel
concretions, the slightly more 1 80-enriched oxygen
values of the younger fault-parallel concretions are
the result of cementation by waters that were either
cooler or more evolved from silicate reactions than
that which cemented the bed-parallel concretions.
Carbon isotopic values do not differ between the
two types of concretions.
We compared the isotopic values of calcite from
the Marnoso-arenacea Formation from its upper
and lower parts. Only weak differences (significance
level :::::0 .90) exist in both carbon and oxygen values.
Calcite veins
Fractures filled by calcite occur in all beds and
many concretions. Calcite veins formed later than
calcite cement as shown by cross-cutting relation­
ships. The i5 1 80 signatures of the veins (Figs 7 and
8) range from - 1 5 to +1%o, but in general are from
2 to 4%o more 1 80-depleted than calcite in beds and
concretions of the same formations. For example,
in the Ranzano and Marnoso-arenacea formations,
il 1 80 values are 2%o more 1 80 depleted than the
lightest whole-rock concretion value, and about 4%o
more 1 80 depleted than authigenic calcite. These
lower values indicate that vein calcite precipitated
from either hotter water or water with a greater
meteoric component, or both, than most calcite
cement. The salinity of fluid inclusions in veins in
the Ranzano (0%o) and the lower part of the
Marnoso-arenacea Formation (10%o, Table 2) fa­
vours, respectively, precipitation from meteoric
water and saline formation water greatly diluted by
meteoric water as the explanation for the 1 80-
depleted oxygen isotopic values in these forma­
tions. A similar explanation seems feasible for the
other shallowly buried formations.
The deeply buried formations have veins with
Milliken et a!.
oxygen isotopic values of -1O%o or less. Such light
values are commensurate with elevated tempera­
tures. Fluid inclusions in the M. Cervarola and
Macigno, based on salinity estimates from freezing
point data, have a meteoric water component. Both
primary and secondary inclusions in the M. Modino
Formation have salinities greater than sea water,
and second inclusions are high-salinity brines typi­
cal of some hydrothermal rocks (Scratch et a!.,
1984).
Mudrocks
Whole-rock samples of mudrock from both shal­
lowly and deeply buried rocks have similar carbon
isotopic ratios, but distinctly different oxygen isoto­
pic ratios (Fig. 8). il1 3C values for most samples are
from -2 to + 1%o, values typical of marine biogenic
carbonate or detrital clasts of marine limestone.
One anomalously light sample (-8%o) in the Antog­
nola Formation, and other samples with il1 3 C
values less than -2%o, contain some authigenic
carbon derived from organic material.
i5 1 80 values from shallow-buried mudrocks range
from about 0 to 4%o; the deeper buried formations
have values from -5 to -13%o. The slightly 1 80-
depleted values for some samples from the shal­
lowly buried rocks can be attributed to the presence
of mixtures of calcite from detrital limestone, indig­
enous marine skeletal debris, and authigenic calcite.
The strongly depleted oxygen isotopic values for the
deeply buried foreland basin samples can be attrib­
uted to a similar admixture of calcite types with a
greater proportion of calcite formed at higher tem­
perature.
Mg, Fe and Mn in calcite
Assessment of intraconcretion v�riations in minor
and trace element concentrations of calcite was
undertaken in nine vein samples, three host rocks
(carbonate content dominantly carbonate rock frag­
ments, CRFs), and 29 other samples of concretions
and generalized cements, including seven concre­
tions for which multiple analyses were performed at
different places relative to the centre of the concre­
tion (Table 5).
As with stable isotopic data, the interpretation of
minor and trace element data in authigenic calcite
is complicated by the presence of detrital carbonate.
In the three data sets from host rock/concretion
pairs, CRFs display a range of trace element content
Table 5. Summary of e leme ntal composition of authige nic and de trital carbonate s

Sample Formation n Ca (mole %) Mg (mole %) Fe (mole %) Mn (mole %) Mg ppm Fe ppm Mn ppm o 1 3C (%o PDB) 8 1 80 (%0 PDB)

Foreland basin samples

B0250 Borello (core) 10 94.92 2.42 2.38 0.28 5844 13131 1 543 - 0.52 - 0.89
B025 1* Borello (intermed.) 10 9 5 . 62 2. 1 3 1 .84 0. 5 1 5 1 83 96 1 8 28 1 1 -0.33 - 2.36
CE7 1 Ce rvarola 10 98.68 0.47 0.23 0.44 1 1 27 1 250 2396
CE76* Ce rvarola 9 98.07 0.69 0.68 0.48 1 648 3761 2592 - 0.57 -8.48
CE79* Ce rvarola 2 98.33 0. 7 1 0.30 0.6 1 1 696 1 627 3285 - 0. 1 8 - 1 0.49
CE87* Ce rvarola 10 98.08 0.8 1 0.30 0. 8 1 1 95 8 1 658 4429
F88-B4 Marnosa-aren. 12 94.89 2.57 1 .95 0.59 6045 1 0520 3 1 74 - 0.98 -3.4 1
MA I 97 Marnosa-aren. 10 97.74 0.84 0.98 0.24 2036 5455 1317 1.10 - 4.23
MA58 Marnosa-aren. 5 96.82 1 .23 1 . 39 0.39 2990 7789 2 1 50 - 1 .64 - 3.38
MA66W Marnosa-aren. 5 96.30 1 . 55 1 . 77 0.38 3682 9665 2034
MC- 1 24 Macigno (grai n re pl.?) 4 98.79 0.4 1 0.23 0.57 1 005 1 269 3 1 47 - !I
.I - 1 5.03 Q
MCI 35 Macigno (grai n re pl.?) 1 0 97.92 0.60 0.49 0.95 1 464 2724 5208 - 0.69 - 1 1 .27 �
MOI OO Modi no 2 98.2 1 0.73 0.33 0.68 1 706 1 779 360 1 �
'1::3
M098 Modi no 7 96.48 0.96 1 .32 1 .05 2296 7234 5659 0.55 - 7.58 ....
""
CE76 Ce rvarola (vien) I0 98.48 0.73 0.35 0. 1 5 1 769 1 957 795 '"'

CE79 Cervaro1a (vei n) 10 98 . 3 1 0. 57 0.44 0.40 1 375 2393 2229 - 0.30 - 1 0. 4 1 .;:; ·
CE8 1 Ce rvarola (vein) II 98.93 0.53 0.29 0.24 1310 1612 1 329 0.47 - 1 1 .05 §'
M089 Modi no (vein) 10 99. 1 5 0.22 0. 1 0 0.53 533 564 29 1 1 0.46 - 1 3.83 g.
;:,;
MOI OO Modi no (vei n) 12 98.63 0.37 0.20 0.44 871 1 089 2306 0. 1 0 - 1 3.73
s·
Piggy-back basin samples �
....
1 56 5 Loiano (margin) 9 97.60 1 .70 0.53 0. 1 4 4 1 46 2960 774 - 2.29 - 1 .68
1 566 Loi ano (core) 14 97.79 1 .66 0.45 0. 1 1 4063 25 1 2 582 -2.33 - 1.55 �·
1 569 Loi ano 15 98. 1 0 1 .78 0.02 0.09 4303 93 518 �
"'
1 5 74 Loi ano 18 97.87 1 . 50 0.42 0. 1 2 365 1 23 1 8 666 0. 1 9 - 1 .71 �
;:,;
1 5 78 Loiano 16 97. 1 9 1 .7 1 0.70 0.28 4085 3833 1 495 - 6.02 -3.48 �
A-28 Antognola 10 97.87 0.79 0.27 1 .07 1 888 1 469 5733 - 6.45 - 5 .86 0
A-29 Antognola 10 98.55 0.63 0.22 0.60 1512 1 1 87 3272 -4.38 ;:,;
- 6.86
M-1 1 0 Bismantova 14 96. 7 1 2. 1 4 0.99 0. 1 7 5 1 43 5448 920 -0.06 - 2.82 �
M- 1 5 Bismantova (core ) 12 96. 6 1 2.2 1 1 .04 0. 1 3 5283 5730 699 - 0. 7 1 - 2.02
M- 1 6* Bismantova (margi n) 12 97. 6 1 2.07 0.09 0.23 5 1 22 524 1 270 - 3. 5 5 - 2.23
M- 1 7 Bismantova (CRF) 16 98.92 0.67 0.23 0. 1 8 1 640 1 276 1 008 - 1 .84 - 3. 8 1
MC-B- 1 * Pli oce ne (core ) 14 94.24 3.07 2.02 0.66 7494 1 1 362 3670 - 1 .58 - 3.42
MC-B-2* Pliocene (intermed.) 16 94. 4 1 3.32 1 .73 0.53 794 1 9486 2886 - 1 .42 - 3.22
MC-B-3* Pli oce ne (margi n) 15 93.53 3.54 2.25 0.68 85 1 1 1 2426 366 1 - 1 .49 - 3.46
-
0-4 1 Ranzano (core ) 12 98. 1 6 1 .06 0.29 0.49 2556 1 607 2662 - 1 .91 - 3. 5 2
0-42* Ranzano (margi n) 14 98.49 0.84 0.21 0.4 1 2 1 33 1 1 70 225 1 - 0.89 -3.24
0-54* Ranzano (core) 22 96. 8 1 1 .87 0.60 0.72 447 1 3279 3866 - 0.64 -2.98
0-55 Ranzano (margin) 14 98. 1 3 0.83 0.23 0.80 20 1 7 1 307 4405 - 0.62 - 3.64
0-56 Ranzano (CRF) 18 97.86 0.85 0.92 0.37 2076 5094 2027 - 0.08 - 3.58
P- 1 2 Pli oce ne (CRF) 15 97.65 0.72 0.87 0.64 1 740 5493 3521 - 0. 2 1 - 3.42
N
N
Continued -..1
228
'V N 00 0'\ \0 00 0 \0 - f"") V
O '<t t-- N a. "<t - t-- - '0 0
r..) .....: N N _; N N N � � v)
I I I I I I I I I
0 \0 N t- N N t- N 00 0\ \0
0 00 0\ t- 0\ M OO C"'i \O rr'l -
_; \Ci c) _: _: _: c) _; r-: _: o
I '( I I I I '7 I I I
E
0.
0.
= rocks. Many of the trace elemental values reported
� for concretionary calcite cements in Table 4 repre­
E sent averages from which anaiyses that probably
0.
0.
� The reported trace element values have also been
E modified by averaging. This was deemed necessary
0.
0.
00
�
O M N � O\ \O � t- O f"""l t- 0
'V l.ri \O V"'I 'V Ir:l V"'I f"") - VI \0 00
ci ci ci ci ci ci ci ci ci ci ci ci
t- N O\ N \0 0\ 0\ V) - - - V)
\O O f"") N O\ M N 0 0\ 0 - N
.....: o N � N N o _: o o o o
M \O M N V1 0\ \0 0\ N ("f") 0 1o0
0 0\ 0'\ \0 \0 \0 ("f") M O\ - V) \0
r-i <'l r-i -<i <'l r-i <'l .....: o o o o
O V \0 - 0 \0 0\ 00 t- f"") N OO
a. "<t O >O a. "<t t-- - O M t-- N
� � � ;; � � � � � � � �
(!)
0.
E (PL)) show covariations between trace elements
0:
{/'J
KL. Milliken et a!.
that is somewhat larger, with less covariation
among elements, than that present in the associated
authigenic calcite. Despite frequent reference to
BSE images and reflected light views during elec­
tron microprobe work, a number of analyses in­
tended to represent authigenic carbonate fall within
the compositional field of the CRFs (Fig. 10).
Trace elements in CRFs were used as reference
values in order to assess the trace element values of
authigenic carbonate in the same way that isotopic
analyses of CRFs in host beds were used to assess
isotopic values determined from associated whole
represent CRFs have been discarded.
to accomplish a comparison between trace element
values and 0180catcite · As previously mentioned,
isotopic values necessarily represent averages
among inseparable detrital and authigenic compo­
nents (including possible multiple generations).
Trace elemental data obtained with the electron
microprobe have a spatial resolution that is orders
of magnitude greater (though still imprecise with
regard to some authigenic zoning; see below) than
that represented by isotopic analysis of bulk whole­
rock powders. At small scales, for example, within
the confines of a single pore, the superior spatial
resolution of probe analysis is valuable for docu­
menting historical trends in calcite precipitation.
However, when plotted against bulk analyses, high­
resolution data add unnecessary 'noise' to the re­
sults. Averaging the trace elemental data for
individual samples allows for more sensible com­
parison with 0180catcite ·
The total range for Mg, Fe and Mn content of
authigenic calcites observed in the concretions is
small relative to the potential ra�ge for authigenic
calcite documented in the literature (e.g. Veizer,
1 983). Mg ranges from below detection to around
5 mole % MgC0 ; Fe ranges from below detection
"' 3
1:: to around 3 mole % FeC0 ; Mn ranges from below
(!) 3
E detection to around 0.8 mole o/o MnC0 , ranging
00 3
0: much higher in CRFs.
<!::::
"" As with isotopic values, the content ofMg, Fe and
u
e Mn and the nature of covariati6ns among these
(!)
til elements differ markedly among concretions. Three
=
0 concretions (two Ranzano and one lA Pliocene
..0
8
* that are consistent with the presence of zoning
 Calcite precipitation in Tertiary sandstones 229

1 2000 X
lA Pliocene
X •
• •
1 0000 • •
•X ••
8000 • • -
e xx •
a. X
.e 6000
of
4000 X X X
X
2000
X

(A) 1 000 2000 3000 4000 5000 6000

1 0000
Ranzano Formation
9000 X X
8000
7000
X
e 6ooo • •
a.
.e 5000
•
X
• •
of 4000 • I •
• X •
• •
• X •
3000
2000
•
1 000
x __r
o t-�� x----�------+------r----�x�----+------r----�
(B) 0 1 000 2000 3000 4000 5000 6000 7000 8000
7000
Bismantova Formation
•
6000
Ill , : •

•• • •
5000

'[ 4000 X
X
.e X
� 3000
X X
2000
X X
1 000 X
X oo
0

Fig. 10. Comparison of Mg versus 0 ��x--��----+------+------�0---AY-�x·��--�x��

Fe in CRFs and authigenic calcite (C) o 1 ooo 2000 3000 4000 5000 6000 7000 8000
in three samples. Mg (ppm)

(Fig. l l A,D,E). In these three concretions, later­
formed calcite from the concretion margin is lower
in both Mg and Fe and slightly higher in Mn than
calcite in the concretion core. This represents the
only clear temporal trend in the trace elemental
composition of calcite. Two other concretions (Bis­
mantova and IA Pliocene, Fig. l l B,C) show clear
difierences between concretion core and margin,
but no systematic covariation among trace ele­
ments. Concretions from the Loiano and Borello
(Fig. l l F,G) show neither clear differences between
calcite from core and margin nor convincing cova­
riation among trace elements. Two Loiano concre-
tions stand out from all other calcites analysed in
terms of their very high Mg content in the presence
of very low amounts of Fe. These concretions also
have very 180-enriched oxygen and 1 3C-depleted
carbon, possibly consistent with precipitation near
the sea floor (described above).
Using the averaged values for trace elemental
composition, the between-concretion variations in
the content of Mg, Fe and Mn and the nature of
covariations among these elements are shown in
Fig. 12. The most consistent trend is a positive
covariation between Mg and Fe. This mimics that
observed within some of the individual concretions
230 K.L. Milliken et a!.

20000
+
lA Pl iocene--Concretion A � +
1 8000 + +

1 6000

1 4 000
o PL31 8-core
E'
c.
1 2 000
+ PL31 3-intermediate
.3: 1 0 000
• PL321 -margin1
.,
u. 8000 o PL3 1 1 -margin2

6000

4000

2000

1 4 000
(A) 2000 4000 6000 8000 1 0 000 1 2 000

1 4 000
lA Pliocene--Concretion B 2
•
1 2 000
0 0 0 • • � 0' 0�"'0
0
o MCBl -core 0 0 0 + +
1 0 000 * +
+ MCB2-intermediate ++ _.,_
+
+
E'
c.
8000 • MCB3-margin

.3:
Q) 6000
u.

4000

2000

(B) 1 0 00 2000 3000 4000 5000 6000 7000 8000 9000 1 0 000

7000
Bismantova Fm .--Concretion E
0
6000
s
o 0
00 0
5000

E'
c.
4000

.3:
Q) 3 0 0 0
u.

2000

• •
1 000
••
• •
•

(C) 1 0 00 2000 3000 4000 5000 6000 7000

8000

Ranzano Fm .--Concretion H •
7000

6000

5000
E'
c. •
.3: 4000
Q)
u.
3000

2000

•
1 0 00

Fig. I L Mg and Fe vari ations

withi n concre tions. Le tte r
500 1 000 1 5 00 2000 2500 3000 3500
(D) de signations are keyed t o the
Mg (ppm) concre tions depicted in Fig. 9.
 Calcite precipitation in Tertiary sandstones 231

1 0 000
•
Ranzano Fm.--Concretion I
9000

8000

7000

e
a.
6000
0 0
.3: 5000
., 0
u. 4000 0
0 0 8
3000
0
o o 0
2000 0
0
0
� o••
1 000

(E) 1 000 2000 3000 4000 5000 6000 7000 8000

7000
Loiano Fm.-Concretion R •

6000
•
5000

'[ 4000 0

.3: 0
0 0
•
., 3 0 0 0 • 0 0
u. •
0
0
•
2000
0
0
• 0
0
1 0 00
•
•

(F) 1 000 2000 3000 4000 5000 6000 7000

30000
Borello Fm.-Concretion T

25000

o 80250-core
20000
+ 80251 -intermediate
e
a.
.3: 1 5 000 o O <e 0
.,
0
u.
+ +
1 0 000
....,_ :j: +

5000

(G) 1 0 00 2000 3000 4000 5000 6000 7000 8000

Fig. 11 (Continued). Mg (ppm)

described above, allowing a general, but somewhat
speculative, temporal history to be imposed on the
data set as a whole, as indicated by the arrows in
Fig. 1 2 and and considered further in the discus­
sion.
There is no systematic difference in elemental
concentrations or covariation, either within or be­
tween concretions, that can be correlated with
either age or tectonic setting of the sandstone. In
Fig. 12 both foreland and piggy-back basin samples
occur across the compositional ranges identified
with relatively early and relatively late calcite pre-
cipitation. Samples in the Macigno, M. Cervarola,
M. Modino and lower Marnosa-arenacea (foreland
basins) and in the Antognola and Ran-zano (piggy­
back basins) have, in general, lower overall Mg and
Fe contents, a less erratic covariation between Mg
and Fe, and higher Mn contents than calcites in
other formations (see Table 4). The Loiano, Bis­
mantova and IA Pliocene (piggy-back basins) and
the Borello and upper Marnosa-arenacea (foreland
basins) have higher Mg and Fe contents and low to
moderate amounts of Mn (see Table 4). Veins tend
to have overall lower concentrations for all trace
232 KL. Milliken e t a!.

1 80 0 0 D
• foreland basin
1 60 0 0 & vein in foreland basin
0 piggy-back basin
1 40 0 0 6 vein in piggy-back basin
•
D
1 2000

'[ 1 00 0 0
•
.e 0
.. 8000 •
LL •
6000
cP
4000 • 0
" l ate" l' 0
o B "early"
D
2000
118".. j� 0
.. 0

0 2000 4000 6000 8000 1 0000 1 2 000

(A ) Mg (ppm)

6000

" l ate" •

5000

D
4000
0 0
'E
Q.
0
•
.e 3000 0
c:
• 0
:::;; "early"
0
2000 .

•
Fig. 12. Trace element
1 000
covariations for averaged,
between-concreti on data. Arrows
indicate the inferred temporal
2ooo 4ooo 6ooo 8ooo 1 oooo 1 2ooo 1 4ooo 1 6 ooo 1 80oo trends in calcite composition. (A)
(B) Fe (ppm) Mg vs. Fe. (B) Fe vs. Mn.

elements than do concretions, although composi­ from dull to bright orange. There are differences in
tions for veins fall within the range observed for CL between concretions in different formations, but
concretions. generally it is uniform within the same concretion,
A plot of average Mg concentration versus among concretions, and between beds and concre­
o 180calcite shows a wide scatter of Mg values across tions within the same formation. Exceptions show
the more 1 80-enriched range of 0 180calcite composi­ clear evidence of zoning. As expected, electron
tions, with a trend toward more 1 80-depleted oxygen probe data indicate that zon�s with high Fe/Mn
values for some samples with very low Mg contents ratios have dull CL and zones with low Fe/Mn
(Fig. 1 3A). A similar trend is observed between Fe ratios have bright CL. Variations in Mg/Ca and Sr
and 0180calcite (Fig. 13B). Notably, both piggy-back do not affect CL.
and foreland samples occur across nearly the whole
range of trace element values, whereas only foreland
and vein samples are characterized by highly nega­ DISCUSSION
tive oxygen values. The correlation between Mn or
Mn/Fe and 0180calcite is less systematic.
Time, depth and temperature of cementation
The cathodoluminescence (CL) response for 30
samples shows that calcite is for the most part The low IGV values of all but a couple of samples
homogeneous and unzoned. However, some con­ indicate that the sandstones underwent moderate to
cretions contain calcite cement that is zoned on the strong compaction before cementation. In the lA
micrometre scale. CL of authigenic calcite ranges Pliocene and Bismantova formations cementation
 Calcite precipitation in Tertiary sandstones 233

4
2 0 0
0
•
O l!IJ 0 Do o 0 0 0
If 0 . o 0 D O�
-2
!!. •
� -4
!!.
D t;
� -6
D •
-oo -8 • • foreland sandstone
-10 A foreland vein

-12
.. .
0 piggy-back sandstone
!!. piggy-back vein
-1 4 • •
•
-1 6

0 2000 4000 6000 8000 1 0000 1 20 0 0

(A) Mg (ppm)

4
2 0
0
•
0 �0 0 D
0 oo o
-2
� lfp
Do cB • � •
u -4 •
!!.
0 !!.
-6
D • • foreland sandstone
�
-oo -8 •
A foreland vein
-10
I •
. o piggy-back sandstone
-1 2
!!. piggy-back vein
-14 ••
•
-16

0 5000 1 0000 1 5000 20000

Fig. 13. (A) Mg and (B) Fe i n
calcite ceme nt versus 0 1 80calcite· (B) F e (ppm)

took place at their maximum burial depths, as can
be seen from the fact that the IGVs of host sand­
stones are the same as those of concretions. We
have no comparable data for the Borello Forma­
tion. In the Ranzano, upper Mamoso-arenacea,
Antognola and Loiano formations cementation
took place slightly shallower than their maximum
burial depths, the difference in IGV between host
sandstones and concretions being less than 5%
(Cibin et a!., 1993). A few concretions that were
cemented at burial depths considerably shallower
than most samples have higher IGV values than
concretions that formed at near-maximum burial
depths, and also have different isotopic signatures
(see below). The extremely low IGV values of the
M. Modino, M. Cervarola and Macigno formations
indicate that they were cemented at depths just
shallower than the depth at which all pores were lost
by compaction. In general, the low IGV values of all
the units suggest that the greater burial depth
estimates given in Table 1 are the more realistic
ones.
We estimated the temperature of cementation for
the base of each stratigraphical unit from burial
depths (assuming a thermal gradient of 20. C/km
and a sea-floor temperature of 5 ·q. Data from
fluid inclusions, vitrinite reflectance and apatite
fission tracks provided limits on maximum temper­
atures reached for most units (Tables 2 and 5)
(Boettcher & McBride, 1993). There is a range of
values for the estimated tempera,tures for most
stratigraphical units because of uncertainty about
burial depths. Estimated temperatures for forma­
tions in piggy-back basins (burial depths from 100
to 1600 m) range from around 10·c for the lA
Pliocene to about 4o ·c for the Loiano Formation.
Even though, as noted above, some formations were
cemented at less than maximum burial depth, and
therefore at less than maximum burial temperature,
we believe the higher temperatures are more realis­
tic values because we may have underestimated
burial depths and possibly the geothermal gradient.
Estimates of burial depths and hence tempera­
tures for the foreland basin units are more problem-
234 KL. Milliken et a!.
atic, except for the shallowly buried (400 m) Borello
Formation. Estimates of burial depths for the three
most deeply buried formations range from 2.5 to
7.3 km, which corresponds to burial temperatures
of from 55 to 150"C (Tables 2 and 5). For the
reasons mentioned above, we may have underesti­
mated the temperature of cementation. Fluid inclu­
sion data in veins indicate that the M. Cervarola
and Macigno reached temperatures of 225 o C and
265 c respectively. However, cementation was
o ,
complete before the hot water that precipitated the
veins circulated through these formations.
Oxygen isotopic composition of cementing fluids
The range of D180caicite in the Borello, Bismantova,
upper Marnoso-arenacea, Ramzano and Antognola
is 2%o or less. Such values indicate precipitation by
water that varied little in temperature and compo­
sition with time. The other formations have il180
ranges of 3-5%o. Variations in il180 of 4-6%o are
common in calcite cements in other shallow marine
40
(A) Piggy·back Basins
__ ..
35
30
25
'{: 20
15
10
sandstones (Wilkinson, 1991; Bjerkum & Walder­
haug, 1993), and a range of 8%o occurs in siltstones
(Lawrence, 1991). However, in the shallowly buried
formations a 4%o range in oxygen equates to ap­
proximately a 30o C range in temperature. If the
geothermal gradient was 20o C/km, the 30° range
equates to a 1. 5 km difference in burial depth. Such
a range in the depth of cementation is greater than
that deduced from our other data for shallowly
buried units. However, a 4%o range in D180sMow
for water at a temperature of so c ; for example, o
equates to nearly a 4%o il180p08 range in calcite.
Such a range in water composition is even more
difficult to explain from our other data. The range
in oxygen values reflects both differences in temper­
ature and water composition plus noise, as cited
above.
Figure 14 summarizes the ranges of il 180authigenic
observed for the various stratigraphical units in this
study. The maximum burial temperatures esti­
mated for the different units place certain con­
straints on the range of D180water that can be
- - - - - - - - - - - - - - - - - - - - - - - - ·.:-··
Bismantova
Fig. 14. Summary diagram for
temperature and 0 1 80water
conditions possible for calcite
precipitated in (A) piggy-back
basins and (B) foreland basins.
The range of o' 80caicite for each
formation is that shown in Table
4. Calculated from equation for
6 10
calcite-water fractionation in
Friedman & O'Neil ( 1 977).
 Calcite precipitation in Tertiary sandstones
postulated for calcite precipitation. I n general, cal­
cite in piggy-back basins has precipitated from
fluids depleted in 1 80 relative to sea water. Because
of their limited extent of burial, calcite in the
Antognola, Ranzano, and IA Pliocene is necessarily
restricted to values of 1 80water more negative than
-2o/oo. In the absence of evidence for subaerial
exposure in the units cemented by meteoric water,
we propose that meteoric water was introduced into
these basins (at some unspecified time) through
deep circulation of surficial waters from nearby
fold/thrustbelt mountains (e.g. Bethke et al., 1 988).
The o 1 80sMow of meteoric water for the latitude of
the northern Apennines today is -6 to -8o/oo at sea
level and - 1 1 o/oo at elevations between 500 and
1 500 m above sea level (Yurtsever & Gat, 1 98 1 ) . through water-rock interaction. XRD data from
Thus, it is necessary to postulate that the 1 80-
depleted fluids that precipitated calcite in the IA
Pliocene, Ranzano and Antognola had some contri­
bution of 1 80 from residual sea water or water-rock
interaction, or, more simply, were less depleted in
1 80 than modem meteoric waters in this area.
Bismantova and Borello (a foreland unit) calcites
have a less well constrained field of possible
o 1 80water-temperature combinations and fluids be­
tween 0 and -4o/oo are permitted. Loiano calcites are
permissive of a wide range of o 1 80waten although
very 1 80-enriched values are deemed unlikely be­
cause of the limited degree ofwater-rock interaction
that is manifested by the Loiano grain assemblage.
The modest level of clay diagenesis, lack of signifi­
cant albitization of feldspars and low burial temper­
atures of the Loiano suggest that if 1 80-enriched
fluids were involved in calcite precipitation those
fluids must have been derived from more-altered
underlying units. In general this limited degree of
rock alteration is confirmed by XRD data from
mudrock samples from the shallowly buried forma­
tions, showing that, although all mudrocks are dom­
inated by illite and chlorite, all contain mixed-layer
liS above trace amounts, and that randomly ordered,
highly smectitic liS is dominant among the mixed­
layer clays.
Because of their greater maximum burial, most of
the foreland units (Borello excepted) are not at all
well constrained with regard to o 1 80water-tempera­
ture relationships. IGVs suggest that calcite precip­
itation postdates considerable burial and compac­
tion, thus probably ruling out precipitation at highly
negative values of o 1 80water (< -3o/oo?). Values for
o 1 80water extending from slightly negative to very
positive cannot, however, be excluded.
235
The most likely source of 1 80-enriched water
during burial of marine rocks is water modified by
reaction with silicate minerals, such as clays under­
going alteration to illite and chlorite (e.g. Suchecki
& Land, 1 98 3 ; Land & Fisher, 1 98 7 ; Lundegard,
1 989). Water with oxygen isotopic values of +5%o
and greater resulting from silicate diagenesis appar­
ently is not generated at burial depths less than 5 or
6 km (Suchecki & Land, 1983). Such 1 80-enriched
water is entirely compatible with the burial depths
reached by the three deepest formations. The degree
of grain alteration (dissolution + replacement) ob­
served in the most altered units (Macigno, Modino,
Cervarola and the lower Mamosa-arenacea) is suf­
ficient to have produced 1 80-enriched fluids
mudrocks in these deeply buried formations also
indicate that illite and chlorite make up nearly all of
the clay species. Little or no liS remains, and that
which does is well ordered. This clay suite is typical
of thermally evolved minerals and an environment
capable of generating 1 80-enriched oxygen values
as described above.
Sources of carbon
The o 1 3 C values of whole-rock samples, as well as
for calculated calcite cement, in all the formations
fall predominantly between -2 and + 1 o/oo (Figs 6
and 7; Table 4). This points to carbonate rock
fragments (CRFs) and fossil skeletal grains as the
major source of carbon in calcite cement. Both are
present to various degrees of abundance in the
sandstones and mudrocks. Probe data show that
CRFs are low Mg-calcite. Any original skeletal
aragonite or high Mg-calcite in sandstones has been
lost by dissolution or replacement. If any such
unstable grains survived burial, they would have
been the first to dissolve. The petrog{aphic evidence
of pressure dissolution attests to local mobilization
of carbonate during compaction. However, CRFs
are several per mil more 1 3 C-enriched than most
cements in all formations except the Bismantova.
Further, the carbonate grains in mudrocks are
several per mil more 1 3 C-enriched than cements in
all formations for which we have isotopic data
(except the Bismantova). Even though carbonate
grains within the sandstones or mudrocks were the
prime source of carbon for cement, some compo­
nent of isotopically light organic carbon was added
to produce the more 1 80-depleted values of our
samples. Based on isotopic data, only the Bis-
236 K.L.
mantova Formation could have had all its carbon
recycled from CRFs within the formation.
Reaction paths by which organic carbon could be
sequestered in cements include bacterial sulphate
reduction or direct microbial oxidation of organic
matter, oxidation of methane, and/or the thermal
degradation of organic matter (see Curtis, 1 977;
Irwin et a!., 1 97 7). In our samples the depth of
cementation was at least several hundred metres,
which is deeper than the depth to which marine
sulphate survives (Hesse, 1 990). Therefore, the
organic carbon must have been derived from the
oxidation of methane or, more likely, from the
thermal degradation of organic matter.
Several concretions in the Pliocene, Mamoso­
arenacea and Loiano formations possess calculated
or whole-rock o 1 3 C values more 1 3 C depleted than
- 1 0%o. These have higher IGV values than other
cemented samples (e.g. 3 1 % for more 1 3 C-depleted
carbon samples vs. 1 4% for more 1 3 C-enriched
carbon samples in the Loiano), which indicates that
they were cemented at shallower depths than most
samples. But these earlier-phase concretions still
formed after significant compaction had occurred,
which would also be below the depth of sulphate
survival. Mass balance calculations indicate that
from 20 to 30% of the carbon in these more
1 3 C-depleted cements must have been derived from
the oxidation of methane, assuming that methanic
carbon has a o 1 3 C composition of at least -40%o
(Curtis, 1 97 7).
Samples of each formation were etched briefly in
weak HCI and examined with the scanning electron
microscope (SEM). Rod-shaped particles of possi­
ble nannobacteria (see Folk, 1 993) or biofilms of
microbially formed polymers (see Westall & Rince,
1 994) occur entombed in calcite cement in both
Pliocene units and in the upper part of the
Mamoso-arenacea Formation. The role played by
microbes is uncertain, but the possibility of micro­
bially mediated precipitation of calcite must be
considered even at the depths at which these rocks
were cemented (Folk, 1 99 3).
Sources of Ca, Mg, Fe and Mn in calcite
Calcium in calcite cement was probably also de­
rived from the large reservoir of calcium in CRFs in
the sandstones and, where present, interbedded
mudrocks. Release of substantial quantities of Ca
through dissolution of detrital carbonate is docu­
mented in other basins (e.g. Milliken & Land,
Milliken et al.
1 993). Albitization is another possible source of Ca
(e.g. Land et a!., 1 98 7), but albitization is not
uniform across the formations examined (Cibin et
a!., 1 993; Milliken & McBride, unpublished data).
Isotopically evolved formation water from deeper
in the basins, interpreted from oxygen isotopic data
from calcite cement, probably contributed the bi­
carbonate component of cement (see below); the
same water may have contained calcium released by
albitization of detrital feldspar assemblages in
deeper, more altered sandstones. The amount of
calcium introduced from deeper in the basins,
however, was probably miniscule compared with
locally derived calcium.
Shell-rich layers in the Pliocene sandstones are
not preferentially cemented. This suggests that the
source of calcium ions was so uniformly distributed
that there was no tendency to preferentially cement
the shell layers. This contrasts with cemented shell­
rich layers noted by other workers (e.g. Davies
1 969; Fiirsich, 1 982; Kantorowicz et a!., 1 98 7).
Large amounts of carbonate grains remain in the
host sandstones we studied; cementation did not
cease because of a lack of available calcium.
Trace elements in calcite give some clues to the
nature of reactions that accompanied the mobiliza­
tion of Ca. Unfortunately, documenting temporal
trends for elemental sources is greatly complicated
by the overriding heterogeneity of between-con­
cretion variability. Given this limited temporal
constraint on calcite precipitation it is difficult to
reconcile the oxygen and carbon isotopic evidence
with the trace element data. Very low Sr (mostly
below detection), relatively low Mg contents, and
the relatively enriched amounts of Fe and Mn in the
calcites support the evidence from IGV and stable
isotopes in ruling out unmodified sea water as a
cement source. Thus, sources for both Ca and trace
elements in the calcite mus\ be ones that are
plausible in later diagenesis.
A few individual Loiano concretions (PU C and
PL3C) have a combination of 1 80-enriched oxygen,
1 3 C-depleted carbon and Mg-rich, Fe-poor, Mn­
moderate trace element contents that are compati­
ble with relatively early precipitation from sea
water modified by bacterial oxidation of organic
matter and Mn mobilization from the sediment.
Mg- and Fe-enriched calcites in the lA Pliocene
precipitated from 1 80-depleted fluids. In these
cases, the sources of all trace elements are plausibly
construed to be materials mobilized from the sedi­
ments during diagenesis at low temperature, be-
 Calcite precipitation in Tertiary sandstones
cause higher-temperature sources are not an option.
Similar, though less pronounced, Mg- and Fe­
enriched calcites in the Bismantova, Loiano,
Borello and upper Mamosa-arenacea could have
precipitated from fluids characterized by a wide
range in l5 1 s0water· Covariations between Mg and
Fe (especially within concretions) suggest that sup­
plies of these two elements progressively declined
during diagenesis, being highest in the centres of a
few concretions and lowest in the veins. The pre­
calcite timing of the minor dolomite precipitation
lends further credence to a temporal sequence in
which Mg-enriched precipitates are relatively early.
The 1 s0-depleted nature of the fluids responsible
for these early Mg- and Fe-enriched calcite cements
suggests that the source ofMg, as well as Fe and Mn,
was material mobilized from the rocks as opposed
to residual sea water. Sources dominantly mobi­
lized relatively early in diagenesis-though later
than the near-seafloor alteration of the early Loiano
concretions-possibly include the dissolution of
very unstable heavy minerals and amorphous oxy­
hydroxides and material weakly adsorbed onto clay
surfaces.
It is paradoxical that Mg and Fe contents are
lowest in the rocks that have the greatest docu­
mented degree of grain alteration. Lower Mg and Fe
contents in the later-formed cements, both those
necessarily precipitated from 1 s0-depleted fluids
(Ranzano and Antognola) and those permissive of
1 s0-enriched fluids (lower Mamosa-arenacea, M.
Cervarola, M. Modino and Macigno) suggest that,
whatever the dominant sources of Mg and Fe, the
supply of these elements was exhausted prior to the
onset of substantial grain alteration. It is also
interesting that some formations with highly unsta­
ble grain assemblages-for example the Ranzano,
which has substantial quantities of serpentinitic
debris-contain calcite with very low Mg content.
Clearly, mobilization during grain alteration was
either insufficient to raise Mg contents in the fluids,
or meteoric fluid volumes were sufficiently high to
maintain low Mg concentrations despite mobiliza­
tion of this element from altered grains.
The temporal distribution of Mn sources is ap­
parently more complicated. Mn is relatively en­
riched in the early precipitates, depleted in some of
the later ones, and enriched in others (e.g. the
Ranzano and Antognola). This trend hints at the
possiblity of multiple sources for Mn, some mobi­
lized relatively early, whereas others were possibly
affiliated with grain alteration later in diagenesis.
237
CONCLUSIONS
I G V values indicate that, with few exceptions,
calcite cementation in the rocks studied occurred
close to maximum burial depth. Possibly four strati­
graphical units were cemented at < 1 km depth;
three formations were cemented deeper than 5 km.
Carbon isotopes indicate that 70-80% of the
carbon in calcite cement in concretions and beds
was derived from carbonate grains in the sandstone
and, where present, interbedded mudrocks. The
grains include CRFs in the sandstones and
mudrocks plus coeval intrabasinal bioclasts in
mudrocks. Evidence of pressure dissolution of
CRFs in the sandstones is ubiquitous. All samples
contain some carbon derived from organic sources,
and the few concretions that have 15 1 3 C values more
1 3 C depleted than -1O%o have a significant propor­
tion of organic carbon. The abnormally high IGV
values of the latter samples indicate that they were
cemented at shallower depths than the norm for
their respective formations, probably within the
zone of methanogenesis.
The IA Pliocene, Ranzano and Antognola forma­
tions were cemented by meteoric water; the Bis­
mantova Formation was cemented in part by water
with a meteoric component; the Loiano and the
Borello formations were cemented by slightly mod­
ified marine pore water; and all the foreland basin
units (except the Borello) were cemented by water
variably enriched in 1 s0 (15 1 s0 -2 to +8) gener­
=
ated from silicate reactions. The most 1 s0-enriched
values for l5 1 s0water are compatible with depths and
temperatures of cementation of the three deepest
formations, but not for the less deeply buried
Loiano and upper part of the Mamoso-arenacea
formations. I SO-enriched fluids in these latter for­
mations were more probably derived from underly­
ing, more deeply buried rocks and expelled by
'
compaction. Possibly, the calcite in the deepest
buried formations re-equilibrated with hot water
after precipitation.
The calcium in calcite cement was also derived
chiefly from the large reservoir of calcium in CRFs
and skeletal grains in the sandstones and, where
present, interbedded mudrocks. Some calcium in
cement in the deepest buried formations may be
derived from albitized plagioclase, but this source
was probably minor compared with CRFs and
carbonate skeletal grains.
Shell-rich layers in the Pliocene sandstones are
not preferentially cemented. This indicates that the
238 KL. Milliken et a!.
source of calcium was so uniformly distributed in
these rocks that there was no tendency to preferen­
tially cement the shell-rich layers.
Very low Sr, relatively low Mg content, and
relatively enriched amounts of Fe and Mn in authi­
genic calcite supports evidence from IGV and stable
isotopes, and we can rule out unmodified sea water
as a source of cement. Sources of trace elements
must be ones efficacious in later diagenesis. Covari­
ations between Mg and Fe suggest that supplies of
these two elements progressively declined during
diagenesis, being highest in the centres of a few
concretions and lowest in the veins. Temporal
distribution of Mn sources are less clear. Docu­
menting temporal trends for elemental sources is
complicated by the large variability between con­
cretions.
In the Loiano Formation, bed-parallel concre­
tions formed at a different time than fault-parallel
concretions.
In some shallowly buried units, variations in
oxygen isotopes (up to 6%o) are greater than can be
explained by temperature differences alone.
Burial depth and its attendant temperature­
induced chemical reactions played a more impor­
tant role in diagenesis than whether a formation
was deposited in a piggy-back or a foreland basin.
ACKN O WLEDGE M E NTS
Financial support was provided by NSF grant EAR-
9 1 03985 (McBride, Milliken), J. Nalle Gregory
Chair in Sedimentary Geology (McBride), and
CNR grants 92.08 74/05 and 9 3 .0 1 03 1 105 (D. Fon­
tana, G.G. Zuffa). Isotopic data were provided by
Lynton Land, Guoqiu Gao and Rachel Eustice.
Analyses and interpretation of clay minerals were
provided by F. Leo Lynch. Luigi Folk provided
advice on bacteria. Stefan Boettcher provided data
from apatite fission tracks. Editorial reviews by Jim
Hendry, Christoph Spot! and Sadoon Morad im­
proved the manuscnpt.
REFERENCES
AMOROSI, A . ( 1 990) Analisi di facies e stratigrafia sequen­
ziale della Formazione di Bismantova ad est del Fiume
Panaro ('placca' di Zocca-Montese, Appennino setten­
trionale). Giorn. Geol. , 52, 1 59- 1 77.
ANDREOZZI, M. ( 1 99 1 ) Stratigrafia fisice delle Arenarie di
M. Cervarola nel settore nord-occidentale deli'Appen-
nino settentrionale tra Ia Val Secchia (R.E.) e Ia Val
Panaro (MO). Mem. Descrillive Carta d'Italia, 46, 269-
285.
BETHKE, C.M., HARRISON, W.J., UPSON, C. & ALTANER, S.
( 1 988) Supercomputer analysis of sedimentary basins.
Science, 239, 26 1 -267.
BETTELLI, G., BONAZZI, U., FAZZINI, P. & PANINI, F. ( 1 987)
Schema introduttivo alia geologia delle Epiliguirdi del­
l'Appennino modenese e delle aree limtrofe. Mem. Soc.
Geol. Ita!., 39, 2 1 5-244.
BJORKUM, P. A. & WALDERHAUG, 0. ( 1 993) Isotopic com­
position of a calcite-cemented layer in the Lower Juras­
sic Bridport Sands, southern England: implications for
formation of laterally extensive calcite-cemented layers.
J. sediment. Petrol. , 63, 678-682.
BOETTCHER, S.S. & McBRIDE, E.F. ( 1 993) Thermal histo­
ries of piggy-back and foreland basins in the northern
Apennines, Italy, derived from apatite fission track
thermochronology. EOS, 74, 547 (Abstract).
BRUNI, P., CiPRIANI, N. & PANDELI, E. ( 1 994) Sedimento­
logical and petrographical features of the Macigno and
the Monte Modino sandstone, in the Abetone area
(Northern Apennines). Mem. Soc. Geol. It., 48, 3 3 1 -
34 1 .
CAVANNA, F., DIGUILIO, A., BALBIATI, B. et a!. ( 1 989) Carta
Geologica dell'estremit orientale del Bacino Terziario
Ligure-Piemontese. Alli Ticinesi di Scienze della Terra,
32 (map).
C!BIN, U., CAVAZZA, W., FONTANA, D., MILLIKEN, K.L. &
McBRIDE, E.F. ( 1 993) Comparison of composition and
texture of calcite-cemented concretions and host sand­
stones, northern Apennines, Italy. J. sediment. Petrol. ,
63, 945-954.
COLEMAN, M. ( 1 993) Microbial processes: controls on the
shape and composition of carbonate concretions. Mar.
Geol. , 1 13, 1 27- 1 40.
CuRTIS, C. D. ( 1 977) Sedimentary geochemistry: environ­
ments and processes dominated by involvement of an
aqueous phase. Philos. Trans., Roy. Soc. Lond. 286A,
3 5 3-372.
DAVIES, D.K. ( 1 969) Shelf sedimentation: an example
from the Jurassic of Britain. J. sediment. Petrol. , 39,
1 344- 1 370.
DE NARDO, M.T., IACCARINO, S., MARTELLI, L. et a{. ( 1 992)
Osservazioni sui bacino satellite epiligure Vetto­
Carpineti-Canossa (Appennino settentrionale). Mem.
Descrillive Carta Geol. ltalia, 46, 209-220.
FAILLA , A. ( 1 987) Evoluzione Diagenetica dei Minerali
Agrillosi e della Sostanza Organica Vegetate in Succes­
sioni Terziarie dell'Appennino Sellentrionale. Dip. di
Scienze Mineralogiche, Univ. degli Studi di Bologna,
1 08 pp.
FAILLA, A. & MEZZETTI, R. ( 1 987) Grado diagenetico di
successioni terziarie deli'Appenino Settentrionale sulla
base di parametri mineralogici. Mem. Soc. Geol. /tal. ,
39, 325-3 3 5 .
FOLK, R.L. ( 1 993) SEM imaging o f bacteria and nanno­
bacteria in carbonate sediments and rocks. J. sediment.
Petrol. , 63, 990-999.
FRIEDMAN, I. & O'NEIL, J.R. ( 1 977) Compilation of stable
isotope fractionation factors of geochemical interest. In:
Data of Geochemistry, 6th edn (Ed. Fleischer, M.). US
Geol. Surv. Prof. Paper 440K, Chapter KK, 12 pp.
 Calcite precipitation in Tertiary sandstones
FOCHTBAUER, H. ( 1 967) Influence of different types of
diagenesis on sandstone porosity. Seventh World Petro­
leum Congress Proceedings, 2. Elsevier, New York,
pp. 3 5 3-369.
FORSICH, F.T. ( 1 982) Rhythmic bedding and shell bed
formation in the Upper Jurassic of East Greenland. In:
Cyclic and Event Stratification (Eds Einsele, G. &
Seilacher, A.), pp. 208-222. Springer-Verlag, Berlin.
GHIBAUDO, G. ( 1 980) Deep-sea fan in the Macigno Forma­
tion (Middle-Upper Oligocene) of the Gordana Valley,
northern Apennines, Italy. J sediment. Petrol. , 50,
723-742.
GHIBAUDO, G. & MuTTI, E. ( 1 973) Facies ed interpretazi­
one paleoambientale delle Arenarie di Ranzano nei
dintorni di Specchio (Val Pessola, Appennino par­
mense). Mem. Soc. Geo!. Ita!. , 1 2 , 2 5 1 -265.
GRATON, L.C. & FRASER, H.J. ( 1 935) Systematic packing of
spheres with particular relation to porosity and perme­
ability. J. Geol. , 43, 785-900.
HESSE, R. ( 1 990) Early diagenetic pore water/sediment
interactions: modern offshore basins. In: Diagenesis
(Eds Macllreath, LA. & Morrow, D.W.). Geosci. Can.
Reprint Series 4, 277-3 1 6.
IRWIN, H., CuRTIS, C. & COLEMAN, M. ( 1 977) Isotopic
evidence for source of diagenetic carbonates formed
during burial of organic-rich sediments. Nature, 269,
209-2 1 3.
l<ANTOROWICZ, J.D., BRYANT, l.D. & DAWANS, J.M. ( 1 987)
Controls on the permeability and distribution of car­
bonate cements in Jurassic sandstones: Bridport Sands,
southern England and Viking Group, Troll Field, Nor­
way. In: Diagenesis of Sedimentary Sequences (Ed.
Marshall, J.D.). Spec. Publ. Geol. Soc. London, 36,
1 03- 1 1 8.
LAND, L.S. & FISHER, R.S. ( 1 987) Wilcox sandstone dia­
genesis, Texas Gulf Coast: a regional isotopic compari­
son with the Frio Formation. In: Diagenesis of
Sedimentary Sequences (Ed. Marshall, J.D.). Spec. Publ.
Geol. Soc. London, 6, 2 1 9-235.
LAND, L.S., MILLIKEN, K.L. & McBRIDE, E.F. ( 1 987)
Diagenetic evolution of Cenozoic sandstones, Gulf of
Mexico sedimentary basin. Sediment. Geol. , 50, 1 95-
225.
LAWRENCE, M.J.F. ( 1 99 1 ) Early diagenetic dolomite con­
cretions in the late Cretaceous Herring Formation,
eastern Marlborough, New Zealand. Sediment. Geo!. ,
75, 1 25- 1 40.
LUNARDINI, J. ( 1 993) Meccanismi di Indizione delle Paras­
ita nelle Areniti dell ' Appennine Settentrionale: Confr­
onto Ire le Formazione dell' Arafone e di Sezione Sottili.
Unpublished thesis, Univ. Bologna, 50 pp.
LuNDEGARD, P.D. ( 1 989) Temporal reconstruction of
sandstone diagenetic histories. In: Burial Diagenesis
(Ed. Hutcheon, I.E.). Miner. Ass. Canada Short Course
Handbook 15, 1 6 1 -200.
LYNCH, L.F. ( 1 997) Frio shale mineralogy and the stoi­
chiometry of the smectite to illite reaction-the most
important reaction in sedimentary diagenesis. Clays
Clay Miner. , 45, 6 1 8-63 1 .
McBRIDE, E.F., MILLIKEN, K.L., CAVAZZA, W. et a/. ( 1 995)
Inhomogeneous distribution of calcite cement at the
outcrop scale in Tertiary sandstones, northern Apenni-
239
nes, Italy. Bull. Am. Assoc. petrol. Geo!. , 79, 1 044- 1 063.
MILLIKEN, K.L. & LAND, L.S. ( 1 993) The origin and fate of
silt-sized carbonate in subsurface Miocene-Oligocene
mudstones, south Texas Gulf Coast. Sedimentology, 40,
1 07- 1 24.
MozLEY, P.S. & BuRNS, S.J. ( 1 993) Oxygen and carbon
isotopic composition of marine carbonate concretions:
an overview. J sediment. Petrol. , 63, 73-83.
NAESER, C.W. ( 1 979) Fission track dating and annealing of
fission tracks. In: Lectures in Isotope Geology (Eds Jager,
E. & Hunziker, J.C.), pp. 1 54- 1 69. Springer Verlag,
Berlin.
REUTTER, K.J., TEICHMOLLER, M., TEICHMOLLER, R. &
ZANZUCCHI, G. ( 1 983) The coalification pattern in the
Northern Apennines and its paleogeothermic and tec­
tonic significance. Geo!. Rundsch. , 72, 86 1 -894.
RICCI LuCCHI, F. ( 1 9 8 1 ) The Marnoso-arenacea turbidites,
Romagna and Umbria Apennines. In: Excursion Guide­
book, 2nd lAS Regional Meeting, Bologna (Ed. Ricci
Lucchi, F.), pp. 229-303.
RICCI LucCHI, F. ( 1 986) The Oligocene to Recent foreland
basins of the northern Apennines. In: Foreland Basins
(Eds Allen, P.A. & Homewood, P.). Spec. Publ. Int. Ass.
Sediment., 8, 1 05-1 39.
RICCI LuccHI, F., COLELLA, A., ORI, G.G., OGLIANI, F. &
CoLALONGO, M.L. ( 1 98 1 ) Pliocene fan deltas of the
Intra-apenninic Basin, Bologna. In: Excursion Guide­
book, 2nd lAS Regional Meeting, Bologna (Ed. Ricci
Lucchi, F.), pp. 79- 1 62.
SCRATCH, R.B., WATSON, G.P., KERRICH, R. & HUTCHIN­
SON, R.W. ( 1 984) Fracture controlled antimony-quartz
mineralization, Lake George Deposit, New Brunswick;
mineralogy, geochemistry, alteration, and hydrothermal
regimes. Econ. Geol. Bull. Soc. Econ. Geol. , 79, 1 1 59-
1 1 86.
SUCHECKI, R. & LAND, L.S. ( 1 983) Isotopic geochemistry of
burial-metamorphosed volcanogenic sediments, Great
Valley sequence, Northern California. Geochim. Cos­
mochim. Acta, 47, 1 487-1 499.
VALLONI, R., LAZZARI, D. & CALZOLARI, M. ( 1 99 1 ) Selec­
tive alteration of arkose framework in Oligo-Miocene
turbidites of the Northern Apennines foreland. In:
Developments in Sedimentary Provenance Studies (Eds
Morton, A.C., Todd, S.P. & Houghton, P.O.W.). Spec.
Publ. Geol. Soc. London, 57, 1 25- 1 36.
VEIZER, J. ( 1 983) Trace elements and isotopes in sedimen­
tary carbonates. In: Carbonates: Mineralogy and Chem­
istry (Ed. Reeder, R.J.). Reveral., Mineral. Soc. Am., 1 1 ,
265-299.
WESTALL, F. & RINCE, Y. ( 1 994) Biofilms, microbial mats,
and microbe-particle interactions: electron microscope
observations from diatomaceous sediments. Sedimen­
tology, 4 1 , 1 47-1 62.
WILKINSON, M. ( 1 99 1 ) The concretions of the Bearreraig
Sandstone Formation: geometry and geochemistry. Sed­
imentology, 38, 899-9 1 2.
YuRTSEVER, Y. & GAT, J.R. ( 1 98 1 ) Atmospheric waters. In:
Stable Isotope Hydrology: Deuterium and Oxygen-18 in
the Water Cycle (Eds Gat, J.R. & Gonfianti, R.). Inter­
ational Atomic Energy Agency, Vienna, Tech. Rept.
Series No. 2 1 0, 103- 1 42.
Spec. Pubis int. Ass. Sediment. (1998) 26, 241-260

Diagenetic evolution of synorogenic hybrid and lithic arenites

(Miocene), northern Apennines, Italy

E. SPADAFORA*1, L.F. DE ROSt2, G.G. ZUFFA*, S. MORADt a n d I.S. AL-AASM:j:

*Dipartimento di Scienze della Terra e Geologico-Ambientali, University ofBologna,
via Zamboni, 67, 40127, Bologna, Italy, e-mail zuffa@geomin.unibo.it;
tSedimentary Geology Research Group, Institute of Earth Sciences.
Uppsala University, S-752 36 Uppsala. Sweden. e-mail sadoon.morad@geo.uu.se; and
:j:Department of Earth Sciences, University of Windsor, Windsor,
Ontario N9B 3P4, Canada, e-mail alaasm@delta.uwindsor.ca

ABSTRACT

The Bismantova-Termina (Miocene) succession was deposited in satellite basins generated within the
collisional orogenic framt; of the northern Apennines. The succession is divided into four major
sequences separated by regional unconformities. Sequences S l and S2 are composed of hybrid arenites
rich in carbonate bioclasts and deposited in a shallow-marine shelf environment. Sequence S3 contains
outer shelf/slope arkosic turbidites interbedded with marls, and sequence S4 is composed of turbiditic
arenites rich in carbonate rock fragments, shales and marls deposited in slope and basin settings. Calcite
cementation in the shelf arenites started with marine rims and syntaxial overgrowths on echinoderms,
and proceeded towards blocky pore-filling cements. The loose packing of the arenites and the isotopic
values of these cements (8180p08 from -3.6o/oo to Oo/oo and 813Cp08 from -4.5o/oo to +0.5o/oo) indicate
precipitation at shallow depth below the sea floor from marine pore waters influenced by bioclast
dissolution. Similar isotopic values in the arkosic slope arenites suggest potential additional derivation
of ions for carbonate cementation from the interbedded marls. Small amounts of dolomite, heulandite.
chlorite and K-feldspar are related to the· early alteration of volcanic rock fragments, heavy minerals and
detrital dolomite grains. The isotopic values of calcite cement (8180p08 from -5.8o/oo to -1.7o/oo; 813Cp08
from -2.8o/oo to +0.1o/oo) and the tighter packing in the S4 turbiditic arenites indicate cementation under
progressive burial, related mostly to the pervasive pressure dissolution of extrabasinal carbonate rock
fragments. Maximum burial depth is, however, estimated to be less than 1 km.

INTRODUCTION

Sandstones with abundant carbonate grains consti­ 1968), mixed with terrigenous quartz, feldspars and
tute an important petrofacies in many sedimentary rock fragments. These sandstones are more pro­
sequences (Zuffa, 1987). One class of such sand­ perly termed hybrid arenites (sensu Zuffa, 1980). A
stones has abundant contemporaneous intrabasinal voluminous literature has been published during
carbonate particles which include bioclasts, ooids, the past decade on the deposition and provenance
peloids and intraclasts (allochems, sensu Folk, of hybrid arenites (Mount, 1984; Doyle & Roberts,
1988; Fontana et a!., 1989; Budd & Harris, 1990;
1 Present address: AGIP Servizi ELSI, Via Fabiani 1, Ctr Loman do & Harris, 1991; Critelli & Le Pera, 1994;
Studi S.D. Milanese, 20097 Milano, Italy. Zuffa et a!., 1995). Hybrid carbonate sand-sized
2Present address: Universidade Federal do Rio Grande sediments are normally deposited: (i) in low­
do Sui, lnstituto de Geociencias, Departamento de
latitude shelf environments by, for example, punc­
Mineralogia e Petrologia, Av. Bento Goncalves, 9500,
CEP 91501-970 Porto Alegre, RS, Brazil, e-mail tuated mixing by storms or in situ allochem
lfderos@if. ufrgs. br. generation (e.g. carbonate fauna in siliciclastic set-

Carbonate Cementation in Sandstones Edited by Sadoon Morad 241

© 1998 The International Association of Sedimentologists
ISBN: 978-0-632-04777-2
242 E. Spadafora et al.

tings); (ii) at medium and high latitudes (foramol
types, Lees & Buller, 1972), where siliciclastic
sedimentation is commonly considered as virtually
exclusive; or (iii) are derived from episodic mixing
of siliciclastic and carbonate sand or along bound­
aries between siliciclastic and carbonate facies (see
Mount, 1984). Despite their common occurrence,
studies on the diagenesis of hybrid arenites are still
relatively scarce (Hudson & Andrews, 1987; Kan­
torowicz et al., 1987; Molenaar et al., 1988; Mo­
lenaar & Martinius, 1990; Cavazza & Gandolfi,
1992; James, 1992; Searl, 1994).
Another class of carbonate-rich sandstones has
abundant extrabasinal limestone and dolostone
fragments (including the calclithites of Folk, 1968).
These rocks occur mostly in orogenic settings, as the
survival of abundant chemically unstable carbonate
rock fragments depends on rapid erosion, transpor­
tation and burial. Studies on the deposition and
provenance of arenites rich in terrigenous carbonate
grains are likewise numerous in the recent literature
(e.g. Mack, 1984; Valloni & Zuffa, 1984; Lawton,
1986; Massari et al. 1986; Ingersoll et al., 1987;
Zuffa, 198 7; Fontana et al., 1989; Fontana, 1991),
whereas work on their diagenesis is very scarce
(Fiichtbauer, 1967; Kukal & Al-Jassim, 1971; Dick­
inson, 1988).
This study aims to decipher and compare the dia­
genetic evolution of hybrid arenites and arenites
rich in carbonate rock fragments belonging to the
Bismantova-Termina succession, a synorogenic se­
quence of the northern Apennines (Fig. 1). Special
emphasis is given to the roles of the complex detrital
composition, provenance and facies organization of
the arenites in their carbonate cementation.
GEOLOGICAL SETTING
Tectonic evolution
The northern Apennines formed by collision be­
tween the European plate (Corsica-Sardinia) and
the African plate (Adria spur, or microplate) after
consumption of the westernmost arm of the Tethys
Sea (Ligurian Ocean). Subduction was initiated in
the Late Cretaceous with an east-dipping (Boccaletti
et al., 197 1; Durand-Delga, 1984) or a west-dipping
subduction zone (Abbate & Bortolotti, 1984; Treves,
1984). Sedimentation in the Ligurian Ocean ceased
in the middle Eocene owing to the continental colli­
sion (mesoalpine phase). Post-collisional orogeny
continued through the late Oligocene along the east­
ern margin of Corsica-Sardinia, probably by ensialic

FOREDEEP UNITS BISMANTOVA· TERMINA PLIO-QUATERNARY UNITS

� EPILIGURIAN SUCCESSION �
LIGURIAN UNITS MONTE PIANO- ANTOGNOLA TORTONIAN-MESSINIAN UNITS
- EPILIGURIAN SUCCESSION ..
Fig. I. Simplified geological map of the northern Apennines showing location of the study areas: A, Bologna area; B,
Yetto-Carpineti area. Modified from Gasperi et a/. (1986).
 Synorogenic hybridand lithic arenites
subduction of the thinned Adria continental litho­
sphere, with progressive eastward migration of the
thrust belt (ophiolite fragments, mudstones and
deep-water turbidites). Continuous convergence re­
sulted in the development of progressively younger
foreland basins, which were filled by thick clastic
sequences (the well known Apenninic flysch). Coeval
deposition of the Epi-Ligurian succession (Ricci
Lucchi, 1987) occurred in smaller basins located on
top of the advancing thrust system.
A generally accepted evolutionary scheme of the
northern Apennines (e.g. Scandone, 1980; Lavec­
chia et a!., 1984; Patacca & Scandone, 1989; Cas­
tellarin et a!., 1992) is as follows: (i) the main
structural development of the northern Apenninic
belt occurred in the late Oligocene-lower Burdi­
galian, synchronous with back-arc extension in the
western Mediterranean (Ligurian-Balearic Sea) and
counter-clockwise rotation of the Corsica-Sardinia
block; (ii) late Tortonian extension caused the
opening of the Tyrrhenian Sea (Sartori, 1989), with
progressive propagation to the east of the contrac­
tional front.
In a recent revision of the stratigraphical and
structural data, Carmignani et a!. (1995) proposed
that both the Balearic basin and the northern
Tyrrhenian Sea developed contemporaneously after
the Oligocene-early Miocene northern Apennines
collision. According to these authors, in Burdigalian
time the tectonic regime switched from contrac­
tional to extensional. Extension caused the opening
of the Balearic basin and northern Tyrrhenian Sea,
(A)
Fig. 2. Schematic stratigraphy of
the northern Apennines Bpi­
Ligurian sequences. (A) Upper
Eocene-Lower Miocene
succession. (B) Bismantova­
Termina succession. S 1-S4,
depositional sequences; B l -B3,
arenite petrofacies. B I, hybrid
arenites; B2, arkosic arenites; B3,
feldspathic lithic arenites; I, marls;
2, resedimented arkosic and lithic
arenites; 3, hybrid arenites; 4, silty
marls. Modified from Amorosi &
Spadafora (1995).
243
detachment of the Corsica-Sardinia block from the
European plate, development of the Alpi-Apuane
core complex in Tuscany, various types of magma­
tism and a widespread transgression. The northern
Apennine thrust system continued its migration to
the east, with contemporaneous shortening at the
front and extension at the back of the orogenic belt.
The Bismantova-Termina succession, the target
of this study (Fig. 2), was deposited from late
Burdigalian to early Tortonian with an angular
unconformity on the Epi-Ligurian succession (mid­
dle Eocene-early Burdigalian) or directly on top of
the Ligurian thrust units. The overall stratigraphy
of the Bismantova-Termina succession exhibits a
transgressive trend and is referred to sedimentation
taking place in 'floating' basins moving on top of an
accretionary wedge driven by active SW-directed
subduction. The succession was strongly dismem­
bered into separate fragments by late Neogene
tectonics, and unfortunately we lack field evidence
to support its deposition in a contractional regime
in 'piggy-back' basins (see Or & Friend, 1984) or in
an extensional regime which may have occurred at
the top of the accretionary wedge (see Platt, 1986).
Stratigraphy, depositional and burial history
of the Epi-Ligurian succession
After the Middle Eocene continental collision, sedi­
mentation was characterized by olistostromes un­
conformably overlying the deformed thrust system
of the Ligurian oceanic units (Fig. 2a). Pelagic and
(B)
..
�c
\ ..
e
"'
\ iiS
\
Antognola
244 E.
hemipelagic reddish deep-sea marls (Monte Piano
Marls) overlie both the olistostromes and the Lig­
urian units. Turbiditic arkoses (Loiano Sandstones)
I km thick are interbedded within the Monte Piano
Marls in the east. A complex turbidite unit (Ran­
zano succession) comprising several composition­
ally different subunits (Cibin, 1993; Zuffa et a/.,
1995) overlies the Monte Piano Marls in the west­
em part of the northern Apennines (Mutti et a/.,
1995). Hemipelagic marls (Antognola Marls; late
Oligocene to early Burdigalian) with interbedded
bodies of turbiditic sandstone of various composi­
tions (Cibin, personal communication) blanketed
all the northern Apennines area.
The Bpi-Ligurian Miocene succession (Bisman­
tova and Termina Formations) occurs as isolated
outcrops in the northern Apennines and is separated
by a regional angular unconformity from the under­
lying Bpi-Ligurian and Ligurian units. Four major
sequences were distinguished by Amorosi (1992).
Stratigraphy in the study area is strongly controlled
by the irregular morphology inherited by the Eocene
and the lower Burdigalian tectonic phases. There­
fore, the column represented in Fig. 2B does not
fully account for local vari(\tions (Bettelli etal., 1987;
Papani eta/., 1987; De Nardo eta/., 199 1; Amorosi,
1992).
w E
� Inner [l]J
T shelf
deposits
Resedimented
shelf deposits
----
--
--
---
.;::
Fig. 3. Simplified stratigraphical and lithofacies schemes of the Bismantova-Termina succession in the Bologna area
B, Burdigalian; L, Langhian; S, Serravallian; T, Tortonian. Modified from Amorosi (1992).
Spadafora et a/.
The base of sequence I (S I) is bounded by an
angular unconformity marked by a glaucony�rich
horizon, and consists of hybrid arenites of marginal
marine facies with shelf silty-marly deposits which
locally interfinger with res�dimented hybrid aren­
ites (Fig. 3). An inversion of the sedimentation
trend from coarsening upwards to fining upwards is
associated with an increase in glaucony content,
and has been chosen to define the boundary be­
tween S l and sequence 2.
Sequence 2 (S2) consists of storm- and tide­
influenced inner-shelf deposits passing to outer­
shelf pervasively bioturbated, fine-grained arenites
and siltstones.
Sequence 3 (S3) starts with a 50 m thick lenticu­
lar body of resedimented arenites which overlies S2
with an erosional contact and passes upward to
fossiliferous silty and clayey marls. S3 arenites are
interpreted as channel-fill and shelf/slope deposits.
The boundary surface between S3 and Sequence
4 (S4; late Serravallian-early Tortonian) is an angu­
lar or paraconformable unconformity of regional
extent, associated with significant depositional hia­
tuses and overlain by glaucony-rich horizons (Am­
orosi & Spadafora, 1995). S4 is constituted of
turbidite bodies overlain by clayey marls, inter­
preted as base of slope/basin deposits.
0 Outer shelf
E-=-=]
Slope
deposits -- deposits
Facies boundary
200 j Sequence boundary
(m)
S3 Depositional sequenc·e
t
Angular unconformity
0
G Glaucony
·.::: .,:::. :: ::· : :::·
 Synorogenic hybridand lithic arenites
The maximum burial depths of the Miocene suc­
cession were estimated to be between 200 and 700 m
(Milliken et a/., this volume), based on variations of
the present thickness range of the succession in the
study area. Two apatite fission track analyse� indi­
cate a burial temperature of 50-55 ·c in the Bologna
area and 70-75 ·c in the Vetto-Carpineti turbidites
(Boettcher & McBride, 1993). The latter sample was
buried slightly deeper than the first one, but neither
is reset in terms of the fission-track data. Fission­
track results are not conclusive, considering the
poorly defined palaeogeothermal gradients, and can
involve some underestimation of the maximum
burial depths. Nevertheless, based on tectonic and
stratigraphical evidence it is unlikely that the succes­
sion reached more than I km of maximum burial
depth.
SAMPLES AND
ANALYTICAL METHODS
One hundred fresh outcrop samples were collected
in the Bologna and Vetto-Carpineti areas (Fig. 1).
Thin sections were prepared from blue epoxy­
impregnated samples, stained with alizarin red plus
K-ferrocyanide solution for carbonate, and with
cobalt nitrite for K-feldspar identification, and ex­
amined with a petrographic microscope. The modal
compositions were obtained by counting 500 points
in each thin section and particular care was taken to
discriminate between non-coeval extrabasinal and
coeval intrabasinal carbonate grains using the crite­
ria of Zuffa (1980, 198 7). The modal point counts
were performed using the Gazzi technique (Gazzi,
1966; Zuffa, 1985). A second point count was
performed to identify a minimum of I 00 fine­
grained rock fragments. Packing proximity index
(Pp) (Kahn, 19 56) was quantified in the transverses
of I 00 grain interfaces, in order to evaluate grain
compaction and the timing of cementation.
Polished thin sections were examined with a
CITL 8200 cathodoluminoscope (CL) at an acceler­
ation voltage of 15- 18 kV and a beam current of
400-500 JlA in order to detect replacement fea­
tures, zonations and different generations of calcite
and dolomite cements.
The chemical composition of minerals was deter­
mined in a total of I 7 polished, carbon-coated thin
sections. A Cameca Camebax BX50 microprobe
equipped with three spectrometers and a back­
scattered electron detector (BSE) was used for quan-
245
titative determination. Operating conditions were:
20 kV acceleration voltage, 8 nA (for carbonates and
clay minerals) to 12 nA (for feldspars) measured
beam current, and a 1-10 Jlm beam diameter (de­
pending on the extent of homogeneous areas). Stan­
dards and count times were: wollastonite (Ca, 10 s),
orthoclase (K, 5 s), albite (Na, Si, 5 and 10 s, respec­
tively), corundum (AI, 20 s), MgO (Mg1 I 0 s),
MnTi03 (Mn, 10 s) and hematite (Fe, 10 s). Preci­
sion during analysis was better than 0.1 mol%. Ad­
ditional semiquantitative examinations were per­
formed with a Philips XL30 scanning electron
microscope (SEM) equipped with BSE and an
EDAX energy-dispersive X-ray analyser (EDS) with
an average 8 kV acceleration voltage.
For carbon and oxygen isotope analyses, precision
microdrilling of the carbonate cements and grains
was carried out on 19 I 00 Jlm thick polished thin
sections following the Dettman & Lohmann ( 199 5)
technique. The 29 separated fractions obtained were
reacted with 100% phosphoric acid at 25·c and the
evolved gas for each carbonate fraction was analysed
using a SIRA- 12 mass spectrometer. The calcite­
phosphoric acid fractionation factor used was
1.0 I 025 (Friedman & O'Neil, 1977). Carbon and
oxygen isotope data are presented in the normal o
notation relative to PDB (Craig, 1957). Precision
(1 cr) was monitored through daily analysis of the
NBS-20 calcite standard and was better than
± 0.05%o for both o13C and o180.
PETROGRAPHY AND PROVENANCE
The average framework compositions of the four
depositional sequences are reported in Table I. The
triangular plots of Fig. 4 show the compo�ition of
the total framework (Fig. 4A), the terrigenous
framework (Fig. 4B) and the fine-grained
' rock frag-
ments (Fig. 4C).
Sequences S1 and S2 comprise hybrid arenites
(sensu Zuffa, 1980) which can be compositionally
differentiated and have been characterized as Petro­
facies B I (Fig. 4A). These arenites contain large
amounts of carbonate intrabasinal bioclasts, such as
echinoderms, bryozoans, algae, corals, and benthic
and planktonic foraminifers of Miocene age (Fig.
5A,B). These bioclasts are commonly micritized
along their margins and skeletal pores, which delin­
eated the original shell shapes and textures in
recrystallized bioclasts. The terrigenous fraction is
composed of quartz, K-feldspar and plagioclase, as
246 E. Spadafora et a!.
Fig. 5. (A) Optical photomicrograph showing the characteristic aspect of petrofacies B I with abundant benthic
bioclasts, including echinoderms with syntaxial overgrowths (e), molluscs (m), red algae (r) and intraclasts (i);
uncrossed polarizers. (B) Optical aspect of petrofacies B I in levels of sequence S2 characterized by large amounts of
planktonic foraminifers and marly matrix; uncrossed polarizers. (C) Photomicrograph of arkosic petrofacies 82 in the
turbidite sequence S3: abundant feldspar grains cemented by blocky calcite; crossed polarizers. (D) Optical aspect of
the lithic arenites of the petrofacies 83 in S4 sequence turbidites; abundant sedimentary lithic fragments: limestone (Is)
and mudstone (m); crossed polarizers. (E) BSE image of rims of calcite prismatic crystals within intraparticle pores of
a bioclast. (F) Cathodoluminescence (CL) photo of a hybrid arenite (shown in A) cemented by rims of non-luminescing
prismatic crystals around bioclasts (arrows), some of which show bright luminescence, and blocky pore-filling
bright-luminescing calcite.
248 E. Spadafora et a!.
well as fragments of phyllite, muscovite-schist, ser­
pentine-schist, subordinate serpentinite, volcanic
rocks and rare dolostone. Intrabasinal glaucony and
phosphate grains are locally abundant. The compo­
sitional and textural characteristics of the terrige­
nous fraction indicate that the detritus was derived
from recycling of the underlying arenites of the
Antognola and Ranzano Formations (Fontana &
Spadafora, 1994; Spadafora, 1996). The intra­
basinal origin of the glaucony grains is indicated by
large rip-up glaucony intraclasts and the irregular
shape of the grains. Based on a study that related
sequence stratigraphy and glaucony distribution in
the Eocene of the Isle of Wight and the Miocene
Bismantova succession, Amorosi (1995) distin­
guished two populations of grains in sequence S2:
(i) para-autochthonous glaucony (intrasequential,
transported), associated with high-energy deposits in
the basal part of the sequence which represents the
TST (transgressive systems tract), and (ii) autochth­
onous glaucony (intrasequential, in situ) concen­
trated at the boundary between the TST and the HST
(highstand systems tract), which is interpreted as an
MFS (maximum flooding surface). The two kinds of
grains are characterized by different degrees of evo­
lution based on their potassium content and para­
magnetic susceptibility.
The turbiditic arenites of S3 (Petrofacies B2)
contain fewer carbonate intrabasinal grains (silici­
clastic arkosic arenites) (Figs 4A,B and 5C). The
siliciclastic fraction is qualitatively similar to that of
sequences S1 and S2 (Fig. 4B,C) and came from the
same source.
The turbiditic arenites of S4 (Petrofacies B3) are
siliciclastic feldspathic litharenites (Fig. 4B) and are
characterized by a sharp change in the types of fine­
grained rock fragments (Figs 4C and 50), which con­
sist mostly of micritic limestones, shales and silt­
stones. This compositional variation is attributed to
an important tectonic change in the source/basin
palaeogeography which delivered a new terrigenous
material to the basin from the Ligurian units (Amo­
rosi & Spadafora, 1995).
The heavy mineral composition in the Vetto­
Carpineti area (Fig. I) (Zuffa, 1969; Fontana &
Spadafora, 1994) shows that metamorphic minerals
such as allanite, pistacite, chloritoid, glaucophane,
hornblende, orthopyroxene and augite are abundant
in shelf arenites from the basal members, whereas in
Vetto turbidites they are very minor constituents.
Ultrastable minerals such as tourmaline, zircon and
rutile are ubiquitous, but the greater amounts occur
in the Vetto arenites. Picotite grains, which are indi­
cators of ophiolitic rocks, are present in all samples
in variable percentages.
CARBONATE CEMENTS
Carbonate cements in the Bismantova-Termina
succession include calcite and subordinate dolomite.
Calcite cement is by far the most abundant
diagenetic constituent, averaging � 18% bulk rock
volume and reaching up to 30% in massively ce­
mented arenites (Table 1). The distribution of cal­
cite at the outcrop scale is commonly homogeneous
and pervasive. Layers interbedded with marls are
usually massively cemented. In some cases, how­
ever, cementation is heterogeneous, taking the
shape of oblate spheroidal and irregular concre­
tions, as well as tabular areas along fractures and
small faults (McBride eta!., 1995). Approximately
15% of the arenites are massively cemented,
whereas the remainder are partially cemented. Cal­
cite cement distribution also varies on thin-section
scale, although in a few samples only it is scarce.
There are important textural and compositional
differences in calcite cements between the shelf
hybrid arenites and the turbiditic arkosic and feld­
spathic lithic arenites. In the shelf hybrid arenites,
calcite cement was precipitated initially as rims of
prismatic crystals (2-20 Jlm across) around mic­
ritized bioclasts and within the skeletal voids
(Fig. 5E). These rims are thinner or absent on the
terrigenous grains. CL imaging revealed that the
prismatic crystals of the rims have non-luminescent
cores covered with orange-yellow zones (Fig. SF).
Calcite cement also occurs as syntaxial overgrowths
on echinoderms (Fig. 6A). Coarse pore-filling cal­
cite, which is volumetrically the main cement in the
hybrid arenites, is characteriz�d by interlocking
euhedral blocky crystals and anhedral drusiform
mosaics, which almost totally occlude the pores
(average remaining macroporosity �0.2% inter­
granular and �0.1% intraparticle) (Fig. 6B).
In the graded storm layers with abundant plank­
tonic foraminifers microcrystalline, moderately fer­
roan calcite replaced clayey/marly matrix (Fig. 6C).
Some bioclasts display their original composition or
show a well-preserved original shell texture, owing
to pseudomorphic neomorphism to low-Mg calcite,
but others are totally recrystallized/replaced by
drusiform or equant mosaic calcite. Blocky calcite
and syntaxial overgrowths display non-luminescent
 Synorogenic hybrid and lithic arenites 249

Fig. 6. (A) CL photograph of syntaxial overgrowths (arrows) on echinoderm bioclasts; overgrowths are initially
non-luminescent and became bright luminescent towards the pore centre; bioclasts luminesce in red and orange. (B)
BSE image of coarsely crystalline, interlocking replacive blocky calcite in hybrid arenite. (C) BSE image of finely
crystalline calcite replacing marly matrix in a hybrid arenite rich in planktonic Foraminifera (f). (D) CL photograph of
blocky calcite cement in a hybrid arenite showing non-luminescing crystal cores (c) followed by bright orange zones
filling the pores. (E) BSE image of coarse, postcompactional calcite cement (bright) replacing lithic fragments and
pseudomatrix after the compaction of micaceous metamorphic fragments. (F) BSE image of coarse calcite rich in Sr
replacing a detrital plagioclase (p).
250 E. Spadafora et a!.
cores or inner zones which are covered by bright
yellowish orange zones toward the centre of the pores
(Fig. 6A,D). Many bioclasts display reddish brown
to orange luminescence, which is additional evi­
dence of their diagenetic recrystallization/re­
placement (Figs SF and 6A). The rims, overgrowths
and pore-filling calcite cements are precompac­
tional, as suggested by the low packing of the ce­
mented areas (commonly <25% Pp) (Table 1).
Calcite cement occurs in the turbiditic arenites
from S3 and S4 mostly as pore-filling blocky or mo­
saic aggregates, with pervasive to patchy distribu­
tion. Calcite replaces silicate grains and pseudoma­
trix (Fig. 6E,F). Lenticular turbidite deposits
enclosed in marls are pervasively cemented. In some
turbidite samples intergranular calcite cement was
partially dissolved. Coarse-crystalline pore-filling
blocky or mosaic cements are luminescent from
brown to orange, in places with wide or irregular
zoning (Fig. 7A). Detrital carbonate rock fragments
show bright orange to yellow luminescence (Fig. 7A).
Generally the calcite cements are characterized by
relatively low average mol% of Fe ( 1.0%), Mg (0.6%)
and Mn (0.2%) (Table 2). Cements in the turbiditic
arenites have somewhat more Fe than those in the
shelf hybrid arenites (av. 1.4% and 0.6%, respec­
tively) (Table 2). Bioclasts such as echinoderms,
originally composed of high-Mg calcite, show low
Mg contents, which confirms their pseudomorphic
replacement. Calcite which replaced feldspar grains
is occasionally Sr rich (up to 2.1 o/o SrC03) (Fig. 6F;
Table 2).
The o180p08 values of calcite cement range from
3
-5.8o/oo to -0.3o/oo, and the 01 CPDB values from
-4.5o/oo to +0.5o/oo (Table 3; Fig. 8). Bioclasts in S1
and S2 show 0180pos values ranging from -1.3 to
3
O.Oo/oo and 01 CPDB values varying from -3.9 to
+0.7o/oo.
Dolomite cement is disseminated and averages
0.2% of the bulk volume. However, in places it forms
up to 1.6% of the hybrid arenites and 0.8% of the
turbidites. Diagenetic dolomite occurs mostly as
syntaxial overgrowths (up to 40 Jlm thick) on detrital
dolomite grains (Fig. 7B). Monocrystalline dolomite
grains are detrital, as revealed by their abraded out­
lines, size equivalence an punctual contacts with
adjacent grains (see Young & Doig, 1986). Polycrys­
talline dolostone fragments developed only small
rhombohedral outgrowths (Fig. 7C). Commonly,
discrete small (4-25 Jlm) dolomite rhombohedra
also occur adjacent to dolomite grains (Fig. 7C).
Both the dolomite overgrowths and the discrete do-
lomite crystals are covered and engulfed by, and thus
pre-date, the calcite cements (Fig. 7B). Detrital and
diagenetic dolomites reveal evidence of partial dis­
solution (Fig. 7D)
Dolomite overgrowths commonly show concen­
tric and oscillatory Fe zonation from less than 5 to
12 mol% FeC03 (Fig. 7E). Discrete crystals have
usually less than 5% FeC03. Manganese values are
low, averaging 0.2% (Table 2), but dolomite over­
growths usually show bright orange luminescence
(Fig. 7F). Because of the small amounts and size of
the overgrowths it was not possible to separate dia­
genetic from detrital dolomite for isotopic analysis.
OTHER CEMENTS
K-feldspar occurs as overgrowths and fracture­
healing of detrital K-feldspar, and as discrete
K-feldspar crystals (Fig. 9A) disseminated in both
shelf and turbidite arenites in trace amounts.
K-feldspar overgrowths and discrete crystals are
covered and engulfed by, and therefore pre-date,
calcite cement (Fig. 9A). The diagenetic K-feldspar
shows a near stoichiometric KA1Si30 end-member
8
composition.
Zeolite occurs in trace amounts as small (5-30 Jlm
long) prismatic crystals is in the intergranular space
and within Foraminifera chambers in some of the
hybrid arenites that are rich in volcanic rock frag­
ments. These zeolite crystals are engulfed by, and
thus pre-date, the pore-filling calcite cements
(Fig. 9B). The crystal habit and electron microprobe
analyses indicate that the zeolite is a relatively
Ba-rich heulandite (average formula (M&J.4Ba0_3
Na0.2K0.1 Fe0.1 )Ca3.5(Al9Si27072).24H20). Similar
Ba enrichment occurs in hydrothermal heulandite
derived from the alteration of basic volcanic rocks
(Gottardi & Galli, 1985). ,
Authigenic clay minerals are generally scarce in
the succession. Thin chlorite rims surround some
heavy minerals and volcanic rock fragments. Chlo­
rite usually occurs in trace amounts but in places it
forms up to 3.2% in S4 turbidites. The rims are
covered by, and thus pre-date, calcite cement. Glau­
cony, as discussed above, is intrabasinal and par­
tially reworked. The higher potassium contents of
the autochthonous glaucony reflect its more ad­
vanced degree of diagenetic evolution.
Chalcedony replaces bioclasts, particularly echin­
oderms, and calcite cement in some samples of shelf
arenites near the base of S I (up to IOo/o). It occurs as
 Synorogenic hybrid and lithic arenites 251

Fig. 7. (A) CL photograph of luminescing carbonate rock fragments (cr) pressure-dissolved along the contacts with
adjacent siliciclastic grains (arrows) in a lithic arenite cemented by postcompactional zoned calcite cement (em). (B)
BSE image of detrital dolomite grain (dd) covered by syntaxial dolomite overgrowths (arrows), followed by blocky
calcite (bright). (C) BSE image of polycrystalline detrital dolomite surrounded by rhombohedral dolomite outgrowths
and discrete crystals (arrows), covered by blocky calcite (be). (D) BSE image of detrital dolomite surrounded by a
rhombohedral overgrowth, both partially dissolved; later blocky calcite cement. (E) BSE image of detrital dolomite
surrounded by overgrowth with oscillatory Fe zonation. (F) CL photograph of dull-luminescing detrital dolomite
involved by a bright overgrowth (partially dissolved); intergranular blocky zoned calcite cement.
252 E. Spadafora et al.

Table 2. Re pre sentative microprobe analyse s of de trital and diagenetic carbonate s in the Bismantova-Term ina
succe ssion

Se quence and
Sample pe trofacie s MgC03 SrC03 CaC03 MnC03 Fe C03 Constituent

48 P4 S3 B2 1.76 0.21 97.82 0.00 0.21 Partially dissolved foram bioclast

68 P3 Sl Bl 0.94 0.25 98.60 0.06 0.15 E chinoid bioclast with glaucony in pore s
89 P4 S4 83 0.16 0.66 96.07 1.05 2.07 Coarse ly crystalline lime stone fragment
1745 P I S4 B3 0.00 0.00 99.65 0.15 0.20 Microcrystalline lime stone grain
48 P2 S3 B2 0.00 0.00 100.00 0.00 0.00 Detrital monocrystalline calcite grain
DF 1772 P5 S3 B3 0.85 0.16 96.57 0.27 2.15 Calcite prismatic rim on bioclast
68 P5 Sl Bl 0.14 0.38 98.87 0.45 0.16 Calcite rim ce ment on quartz grain
68 P6 S l Bl 0.10 0.51 98.86 0.06 0.48 Calcite rim ce ment on bioclast
31 VC P7 Sl Bl 1.89 0.19 97.65 0.01 0.26 Intergranular coarse calcite ce ment
31 VC P9 Sl Bl 0.24 0.08 98.23 0.12 1.34 Intergranular mosaic calcite ce ment
34 VC P3 S2 Bl 1.27 0.31 96.82 0.21 1.40 Intergranular blocky calcite ce ment
34 VC P5 S2 Bl 0.00 0.14 99.86 0.00 0.00 Intergranular blocky calcite ce ment
34 VC P7 S2 Bl 0.43 0.00 99.57 0.00 0.00 Intergranular blocky calcite ce ment
34 VC P8 S2 Bl 0.18 0.41 96.54 1.00 1.87 Bright zone in patchy blocky calcite
34 VC P9 S2 Bl 0.92 0.17 98.80 0.06 0.05 Dark zone in blocky patchy calcite
34 VC P!O S2 Bl 0.00 0.40 99.61 0.00 0.00 Drusiform calcite within bioclast
48 P I S3 B2 0.00 1.05 98.31 0.12 0.52 Intergranular blocky calcite
48 P3 S3 B2 0.14 0.75 98.19 0.17 0.75 Intergranular blocky calcite ce ment
48 P5 S3 B2 2.99 0.13 96.56 0.00 0.32 Intergranular blocky calcite ce ment
68 P I Sl Bl 1.87 0.08 97.89 0.16 0.00 Intergranular blocky calcite ce ment
68 P2 Sl Bl 0.98 0.16 98.78 0.08 0.00 Intergranular blocky calcite ce ment
89 P I S4 B3 0.82 0.00 96.15 0.55 2.49 Intergranular blocky calcite ce ment
89 P3 S4 B3 0.79 0.00 96.63 0.21 2.37 Intergranular blocky calcite ce ment
89 P5 S4 B3 2.02 0.00 97.98 0.00 0.00 Intergranular blocky calcite ce ment
DF 1772 P I S3 B3 0.86 0.06 96.80 0.27 1.99 Intergranular blocky calcite ce ment
DF 1772 P4 S3 B3 0.93 0.32 95.63 0.27 2.85 Intergranular blocky calcite ce ment
1740 P3 S2 Bl 0.52 0.21 97.26 0.01 2.00 Calcite mosaic within foram
1745 P2 S4 B3 0.45 0.11 97.01 0.20 2.23 Intergranular blocky calcite with pyrite
1745 P3 S4 B3 0.61 0.23 96.70 0.08 2.39 Intergranular blocky calcite ce ment
1749 P4 S4 B3 0.00 0.25 99.63 0.12 0.00 Intergranular mosaic calcite
1749 P5 S4 B3 0.54 0.00 96.81 0.13 2.52 Intergranular mosaic calcite
1749 P6 S4 B3 0.00 0.34 98.48 0.25 0.93 Intergranular mosaic calcite
1749 P7 S4 B3 1.32 0.00 96.91 0.08 1.69 Intergranular blocky calcite ce ment
1749 P8 S4 B3 1.48 0.06 96.47 0.04 1.95 Intergranular blocky calcite ce ment
89 P6 S4 B3 0.25 0.19 99.55 0.00 0.00 Calcite mosaic within foram with pyrite
31 VC P8 Sl Bl 0.59 0.22 99.03 0.00 0.16 Syntaxial overgrowth on echinoid
68 P8 Sl Bl 0.12 0.89 98.69 0.03 0.27 Syntaxial overgrowth on echinoid
89 P2 S4 B3 0.63 0.07 95.37 1.65 2.28 Coarse blocky calcite replacing quartz
DF 1772 P2 S3 B3 0.00 2.10 96.80 0.25 0.86 Calcite replacing de trital albite
DF 1772 P3 S3 B3 0.00 0.66 98.26 0.55 0.53 Calcite replacing de!rital albite
1740 P I S2 Bl 0.43 0.03 97.74 0.11 1.69 Microcrystalline calcite replacing matrix
1740 P4 S2 Bl 0.50 0.17 97.63 0.01 1.69 Microcrystalline calcite replacing matrix
1740 P2 S2 Bl 0.23 0.13 97.99 0.02 1.64 Microcrystalline calcite replacing matrix
1745 P4 S4 B3 0.38 0.30 99.17 0.00 0.15 Calcite replacing clay pseudomatrix
34 VC P4 S2 Bl 0.11 0.06 99.44 0.00 0.40 Calcite replacing clay intraclast
34 VC P6 S2 Bl 0.00 0.24 99.61 0.07 0.08 Calcite replacing clay intraclast
48 P6 S3 B2 27.87 0.05 60.03 0.33 11.72 Overgrowth in detrital dolomite
48 P8 S3 B2 29.72 0.00 62.56 0.40 7.33 Overgrowth in de trital dolomite
48 P7 S3 B2 33.87 0.18 61.46 0.03 4.45 Small discre te dolomite crystal
48 P9 S3 B2 32.20 0.05 63.75 0.00 4.00 Small discrete dolomite crystal
68 P4 Sl Bl 41.86 0.00 56.86 0.16 1.12 Discrete dolomite, partially dissolve d
68 P7 Sl Bl 35.38 0.13 64.24 0.06 0.18 Discrete dolomite, partially dissolve d
89 P7 S4 B3 36.56 0.26 62.93 0.18 0.07 Small discrete dolomite crystal

Continued
 Synorogenic hybridand lithic arenites 253

Table 2. (Continued)
Sequence and
Sample petrofacies MgC03 SrC03 caco, MnC03 FeC03 Constituent

0.62 0.28 97.87 0.19 1.04 Diagenetic calcite average-general

2.99 2.10 99.86 1.65 2.85 Diagenetic calcite maximum-general
0.00 0.00 95.37 0.00 0.00 Diagenetic calcite minimum-general
0.53 0.24 98.51 0.13 0.60 Diagenetic calcite average-S! + S2
1.89 0.89 99.86 1.00 1.87 Diagenetic calcite maximum-S!+ S2
0.00 0.00 96.54 0.00 0.00 Diagenetic calcite minimum-S!+ S2
0.72 0.32 97.33 0.25 1.38 Diagenetic calcite average-53 + S4
2.99 2.10 99.63 1.65 2.85 Diagenetic calcite maximum-53+ S4
0.00 0.00 95.37 0.00 0.00 Diagenetic calcite minimum-53 + S4
33.92 0.10 61.69 0.17 4.12 Diagenetic dolomite average-general
41.86 0.26 64.24 0.40 11.72 Diagenetic dolomite maximum-general
27.87 0.00 56.86 0.00 0.07 Diagenetic dolomite minimum-general

Table 3. Isotopic values of representative diagenetic and detrital calcite in the 8ismantova-Term ina succession

Sample Original number Sequence Petrofacies Constituent 013CPDB ()IBQPDB

42 65 Sl 81 Echinoid 0.72 -0.17

42 65 Sl 81 Intergranular cement 0.15 -1.00
42 65 Sl 81 Foraminifer -0.23 -1.07
6 68 Sl 81 Intergranular cement -1.21 -2.61
7 71 Sl 81 Syntaxial overgrowth -0.69 -1.45
7 71 Sl 81 Cavity of bioclast -0.56 -1.58
59 Jive Sl 81 Intergranular cement 0.35 -1.05
59 Jive Sl 81 Intergranular cement 0.46 -0.95
63 8vc S2 81 Intergranular cement -0.39 -3.32
67 34vc S2 81 Intergranular cement -2.50 -0.31
II 81 S2 81 Echinoid -0.62 -1.34
II 81 S2 81 8ioclast -1.05 -1.00
12 1738 S2 81 Echinoid -3.90 -0.16
12 1738 S2 81 Syntaxial overgrowth -4.47 -3.64
51 36 S2 81 Cavity of foraminifer 0.39 -0.92
51 36 S2 81 Intergranular cement 0.37 -1.02
51 36 S2 81 Echinoid 0.74 -0.01
54 40 S3 82 Intergranular cement 0.30 -1.47
41 63 S3 82 Intergranular cement -0.43 -2.63
21 48 S3 82 Intergranular cement 0.47 -1.08
23 54 S3 82 Intergranular cement 0.43 -1.31
28 61 S3 82 Intergranular cement -0.�4 -3.17
28 61 S3 82 8ioclast -0.86 -0.61
36 87 S4 83 Intergranular cement -1.14 -5.50
37 89 S4 83 Intergranular cement -0.28 -5.15
37 89 S4 83 Detrital calcite 0.07 -4.33
18 1749 S4 83 Echinoid -0.51 -1.72
44 27 S4 83 Intergranular cement -2.80 -5.29
46 30 S4 83 Intergranular cement -0.98 -5.75

small globular and spherulitic aggregates (Fig. 9C). framboids in trace and up to 1.5% within chambers
Similar silicification was not observed in the other of foraminifers, and partially replacing bioclasts
sequences. and mud pseudomatrix. Framboids are engulfed by
Pyrite is widespread in all facies, occurring as blocky and mosaic calcite cements (Fig. 90).
254 E. Spadafora et a/.

- 5 -4 - 3 I'� -2 - 1 0 _ __§13C PDB

,. ' ./:0
' ', () <¥ •�/1 2
', �� ,J&._-z � I
........... . . //
�--.!_I---,..- ----,
..-�
(.- ------ -4 0 <> bioclasts

___________ 6 o intraparticle cement

Fig. 8. o13CPDB versus o'80PDB
......._

t. syntaxial overgrowths (51 +52) plot of representative diagenetic

� -8 and detrital calcite in the various
• lntergranular cement (51 +52+53)
0 depositional sequences of the
"' -1 0 o carbonate rock fragment (54)
;o Bismantova-Termina succession.
• intergranular cement (54) Arrow denotes the probable
-1 2
derivation of S4 cements from
marine carbonate rock fragments.

DISCUSSION sors) rims associated with these fragments.

Despite the relatively shallow maximum burial
depths reached, the combination of cementation
Paragenetic evolution and porosity destruction
with mechanical and chemical compaction resulted
The arenites in this study display evidence of dia­ in a near total porosity destruction. Mechanical
genetic modifications that started on the sea floor compaction in the arenites is evidenced by moder­
and continued during progressive burial (Fig. 10). ate fracturing of bioclasts, quartz and feldspar
The earliest diagenetic processes include micritiza­ grains, and by ductile deformation of grains such as
tion and precipitation of calcite rims around the bio­ glaucony peloids, mud intraclasts, shale, schist and
clastic fragments in the shelf hybrid arenites. The altered volcanic rock fragments (Figs 6E and 9E).
precipitation of minor amounts of pyrite, K-feldspar Chemical compaction is evidenced by pressure
overgrowths and chlorite rims occurred at shallow dissolution along contacts between bioclasts and
depths below the sediment-water interface, in both between silicate grains in the shelf arenites. Con­
the shelf and turbiditic arenites. The chloritic rims spicuous pressure dissolution is observed between
probably evolved from a berthierine or trioctahedral extrabasinal carbonate grains and detrital silicates
smectite precursor derived from the alteration of in the S4 turbidites (Figs 7A and 9F).
volcanic and heavy mineral grains, as the formation The packing proximity values provide further
of chlorite requires higher temperatures under burial clues to the timing of cementation, being low in the
diagenetic, metamorphic or hydrothermal condi­ shelf S1 and S2 ar�nites (av. 28.9% and 34.5%, re­
tions (see Velde, 1985). Trace amounts of zeolite spectively), intermediate in the S3 arkoses (av.
formed in the shelf arenites in close association with 35.3%) and higher in the S4 turbidites (av. 40.2%
altered volcanic rock fragments, which probably (Table 1). This suggests that cementation was largely
supplied the needed ions. These early cements were pre-compactional in the shelf ar;enites and syncom­
later engulfed by coarse pore-filling calcite that pre­ pactional in the turbidites. Surprisingly, the similar
cipitated during progressive sediment burial. average intergranular volume (IGV) of these se­
The formation of conspicuous early dolomite quences (21.6% in S l to 25.3% in S4) does not
overgrowths pre-dating the blocky calcite cements indicate the variations in cementation timing sug­
indicates elevated Mg/Ca ratios of pore waters at gested by the packing proximity values (Table 1).
very shallow burial depths below the sea floor. The Furthermore, Milliken eta!.(this volume) found that
formation of these overgrowths was probably fa­ the IGV values of Bismantova arenites cemented by
voured because of the dissolution of high-Mg bio­ concretionary calcite cement were similar to those of
clasts and their neomorphism to low-Mg calcite. A non-cemented areas (av. IGV 19. 7%), and con­
=

possible additional source of Mg was the alteration cluded that cementation occurred at near maximum
of unstable basic volcanic rock fragments. This is burial depths. These discrepancies are probably due
supported by the occurrence of heulandite and the to two factors: first, that the sampling of Milliken et
early chloritic (Mg-smectite or berthierine precur- a!. was restricted to the turbidite facies, where ce-
 Synorogenic hybridand lithic arenites 255

Fig. 9. (A) BSE image of a K-feldspar grain with overgrowths and fractures healed by darker authigenic K-feldspar;
note the small surrounding discrete K-feldspar crystals (arrows). (B) BSE image of prismatic crystals of heulandite
(arrows) engulfed by blocky calcite cement in a hybrid arenite. (C) BSE image of silicified echinoderm plates (ep),
echinoderm spines (es) and intergranular calcite cement (cc) in a hybrid arenite. (D) BSE image of framboidal pyrite
(white) and calcite (light grey) replacing pseudomatrix after clay intraclasts. (E) BSE image of a glaucony grain
compacted and penetrated by adjacent quartz and feldspar grains. (F) BSE image of foraminifer bioclasts
pressure-dissolved along the contacts with adjacent quartz and perthitic feldspar grains.
256 E. Spadafora et al.

HYBRID SHELF ARENITES TURBIDITE FELDSPATHIC LITHARENITES

Vl
Vi
UJ
z
UJ
"'
<1:
0
0
UJ

Vl
Vi
UJ
z
UJ
"'
<1:
0
0
Vl
UJ
::;:
- - - - -

Fig. 10. Generalized paragenetic

<'-·

0
<1: sequence in the shelf hybrid
0
0 arenites and in turbidite
...J
UJ
f- feldspathic litharenites.

mentation is clearly later and essentially postcom­

pactional, and secondly, to the conspicuous presence
of thin, precompactional carbonate rims around the
grains of S1 and S2 arenites. These rims, although
not in quantities sufficient to sustain compaction of
the mechanically weak framework, effectively sepa­
rate the grains from each other, promoting low pack­
ing indices.
A plot of IGV versus cement% for samples with
less than 10% of matrix (see Houseknecht, 1987)
shows that compaction and cementation were gen­
eral equally important in destroying the porosity
Sources and processes of carbonate diagenesis
The abundant and recurrent calcite cementation is
related to the internal source and nucleus for car­
bonate precipitation provided by the bioclasts and
the associated early marine rims and overgrowths in
the shelf petrofacies, and by the carbonate rock
fragments in S4 turbidites. The influence of abun­
dant carbonate grains on calcite cementation is
illustrated by Fig. 12, which shows a positive corre­
lation between the amounts of bioclasts and cement
in S1 and S2 arenites (R2 0.43), and between
=

(Fig. 11A). It also appears that cementation was carbonate fragments and cement in S4 turbidites
more important than compaction in shelf arenites, (R2 0.71). This suggests that the carbonate grains
=

whereas the opposite is true for the turbidites.
However, by plotting indices which take into con­
sideration the reduction in bulk rock volume due to
compaction (see Lundegard, 1992), a more realistic
evaluation of the relative roles of compaction and
cementation is obtained (Fig. 11B). The same sam­
ples with less than 10% of matrix plotted in Lunde­
gard's (1992) diagram reveal that compaction was
actually more important than cementation in reduc­
ing porosity. Remaining interparticle porosity is
very low in the shelf arenites (av. 0.2%). Slightly
provided preferential nucleation sites for the pre­
cipitation of cements, and that their partial dissolu­
tion constituted an important source for calcite
cementation. Hybrid arenite layers are known for
being commonly cemented by massive or concre­
tionary stratabound calcite in many shelf and tur­
bidite sequences (e.g. Kantorowicz et al., 1987;
Molenaar et al., 1988; Carvalho et al., 1995). The
S3 turbidites, interbedded with thick marls, show
no correlation between the amounts of carbonate
grains and cement (R2 0.13) (Fig. 12). This rela­
=

higher average intergranular porosity in the tur­ tionship, and the pervasive cementation of the thin
bidites (2.3%) is partly of secondary origin and arenite bodies interbedded in marls, suggests that
formed by slight dissolution of dolomite and calcite the marls were an additional source of carbonate for
cements. the cements. Marls interbedded with the turbidite
 Synorogenic hybrid and lithic arenites 257

ORIGINAL POROSITY DESTROYED B Y CEMENTATION (%) sequences show a bulk isotopic composition vary­
A --'
"
u ing from -2.2 to l %o o 1 3C and -2 to O%o o 1 80 (see
z
" Milliken et a!., this volume), which is similar to that
I
35
::cl t
�

.� :>: of S3 cements.
� 30 0
z
,_

"
� "
6 Many of the bioclast types were originally arago­
5 25 >- a.
w
> 0 :>: nitic or high-Mg calcite in composition, and thus
IGV � 20 50 � �" chemically labile even during initial burial. As there
"
>- ""
a

�
a:
15 are no mouldic pores preserved, however, it is more
Cl � �
0 I
ffi 10 "' u likely that most of the ions for the precipitation of
....
;;
2 �
;/. " early calcite rims in the hybrid arenites were de­
0
�-------'--' 1 00 � rived from sea water, and not from the dissolution of
0 50 1 00 § these bioclasts. The conspicuous oversized patches
CEMENT (")

B .-------, filled by blocky calcite cement were probably formed

111 Sequence S 1·1
50 Petrofacies 8 1 by the dissolution of bioclastic fragments and early
o Sequence S2 rims during progressive burial. An additional inter­
45

40
• Sequence S3 Petrofacies 82 nal source of relatively late carbonate cementation
35 Sequence S4 Petrofacies 83
was the widespread pressure-dissolution which af­
fected the bioclasts during compaction. The o 1 3C
30
...J
values of calcite cement in the hybrid arenites indeed
g; 25
indicate a dominantly marine source probably re­
u 20
lated to marine pore waters, bioclasts and marls
15
(Table 3; Fig. 8).
10
The absence of kaolinite cement and kaolinized
grains, as well as the high o 1 80PoB values of calcite
10 15 20 25 30 35 40 45 50 cement in the hybrid and arkosic arenites (-3.6 to
CEPL
O%o) (Table 3; Fig. 8), indicates that meteoric fluids
Fig. 1 1 . (A) Plot of intergranular vol% versus cement% did not play an important role in carbonate cemen­
for arenites with less than I 0% of matrix (see tation. Considering that the succession remained
Houseknecht, 1987). (B) Plot of compactional porosity buried at shallow depths in a region of rugged relief,
loss (COPL) versus cementation porosity loss (CEPL) for
arenites with less than I 0% of matrix (see Lundegard,
the lack of meteoric influence is surprising and
1992). probably related to the early and pervasive destruc­
tion of the porosity and permeability of the arenites
caused by intense cementation and compaction.

30 0
0 0
� 0

.... cfl
c 25 0 • 0
Q) 0
DO e 0
oo
E oo
Q) •
u • 0
20 •
• Oo
• '
Q)
.... D q. 0 0
0 0
0
0
:.!;! 15 'ill 0
0 5 1 +52
ro •
u • 0 0 o o
0 0 53
0
.... 0 0 0
ro 10 0
• 54
Fig. 12. Plot of carbonate grains% 'S
c
versus intergranular calcite ro
.... 5 0
Cl
cement% showing the positive ....
Q)
correlation between amounts of .... 0
c
bioclasts and cement in S l and S2 0
arenites and between carbonate
0 5 10 15 20 25 30 35 40
fragments and cement in S4
turbidites. bioclasts + carbonate rock fragments %
258 E. Spadafora et a/.
Consequently, carbonate precipitation is assumed to
have occurred from marine pore waters which may
have been slightly modified by the dissolution of
carbonate grains and early cements. Oxygen isotopic
values close to normal marine indicate a fully open
diagenetic system in relation to the overlying sea
water, with no sensible influence of silicate interac­
tions, such as the alteration of volcanic grains. As­
suming original sea water with a o 1 80sMow value of
- l .2%o (Shackleton & Kennett, 1975) and the oxy­
gen isotopic range of calcite cements, the calculated
precipitation temperatures (see O'Neil & Epstein,
1966) range from �10 to 3o · c. The high 0 1 80PDB
values of diagenetically altered bioclasts (-1.3 to
O.O%o) indicate that their neomorphism occurred at
shallow depths below the sea bottom and at low tem­
peratures (:::::1 2-17• C). Cements in S4 turbidites, on
the other hand, show consistently lower o 1 80p08
values (from -5.8 to -1.7%o) (Table 3, Fig. 8) than
the bioclastic and arkosic arenites. This is in agree­
ment with the petrographic observation of a later,
syncompactional cementation in the S4 arenites. As­
suming precipitation from unmodified marine pore
waters, the cementation of S4 turbidites occurred at
�20-40 " C.
Calculated precipitation temperatures for the cal­
cite cements in both the shelf and turbidite arenites
are far below the temperatures estimated during
maximum burial (50-55 · c in the Bologna area and
70-75· c in the Vetto-Carpineti area). This indi­
cates that cementation did not occur at maximum
burial depths, and is supported by the relatively
high IGV and low packing proximity values shown
by all sequences. Incipient dissolution of calcite and
dolomite cements and grains, and local silicification
were probably caused by limited telogenetic circu­
lation of meteoric waters. Meteoric infiltration in
these tight arenites is expected to have an influence
only in the vicinity of fractures developed during
orogenic uplift and deformation.
CONCLUSIONS
Calcite cementation in bioclastic hybrid and lithic
arenites of the Bismantova-Termina succession is
pervasive along layers and concretionary horizons.
Cementation in the hybrid shelf arenites was mostly
precompactional and began with marine calcite
rims, syntaxial overgrowths on echinoderms,
K-feldspar and dolomite overgrowths, chloritic clay
rims, framboidal pyrite and heulandite, followed by
blocky pore-filling calcite. Loose packing, high IGV
values, and the oxygen and carbon isotopic values
of calcite cement indicate that precipitation oc­
curred close to or at shallow depths below the sea
floor, derived from marine pore waters and bio­
clasts. Marls interbedded with the slope arkosic
arenites probably also supplied the abundant calcite
cement of this sequence, which contains lower
amounts of carbonate grains. The tighter packing
and lower oxygen isotopic values of calcite cements
in the feldspathic litharenite turbidites indicate that
precipitation occurred relatively later, during pro­
gressive burial, derived from the pressure-dis­
solution of abundant carbonate rock fragments. The
hybrid and lithic arenites were subjected to a rapid
and shallow porosity destruction by pressure­
dissolution and cementation. Calcite cement was
derived and nucleated on bioclasts and carbonate
rock fragments. The rapid and intense destruction
of porosity and permeability prevented any major
influence of meteoric fluids on carbonate cementa­
tion or dissolution.
ACKNOWLEDGEMENTS
We thank S. Boettcher for information on the
thermal history, and D. Fontana for thoughtful
discussions on the Miocene succession. We also
thank L. Martire (CL), K.C. Lohman (isotopes) and
H. Harryson (microprobe analyses) for analytical
assistance, and B Gios for photographic work.
Comments by reviewers K.L. Milliken and W.
Dickinson helped to improve the manuscript. The
financial support by the Italian National Council of
Research (CNR grant 95.00324.CT05), the Brazil­
ian National Council of Research (CNPq grant
200465/92.9-GL to L.F.D.R.), the Natural Sciences
and Engineering Research Council of Canada (to
I.S.A.) and by the Swedish Natural Science Re­
search Council (to S.M.) is gratefully acknowledged.
REFERENCES
ABBATE, E. & BORTOLOTTI, V. ( 1 984) Le unita Liguri
dell'Appennino settentrionale: sintesi dei dati e delle
interpretazioni dal 1 970 in poi. Mem. Soc. Geol. /tal. ,
Suppl. E, 2 1 1 -240.
AMOROSI, A. ( 1 992) Correlazioni stratigrafiche e sequenze
deposizionali nel Miocene epiligure delle Formazioni di
Bismantova, S. Marino e M. Fu maiolo (Appennino
settentrionale). Giorn. Geol. , 54, 95-105.
 Synorogenic hybrid and lithic arenites
AMOROSI, A (1995) Glaucony and sequence stratigraphy: a
conceptual framework of distribution in siliciclastic
sequences. J. sediment. Res., B65, 419-425.
AMOROSI, A. & SPADAFORA, E. (1995) The Upper Serraval­
lian unconformity in the Epi-Ligurian units of the
Bologna Apennines. In: Rapporti Alpi-Appennino (Eds
Polino, R. & Sacchi, R.), pp. 69-86. Accademia Nazio­
nale delle Scienze, Peveragno, Italy.
BETTELLI, G., BONAZZI, U., FAZZINI, P. & PANINI, F. (1987)
Schema introduttivo alia geologia delle Epiliguridi
dell'Appennino modenese e delle aree limitrofe. Mem.
Soc. Geol. !tal. , 39, 215-244.
BOCCALETTI, M., ELTER, P. & GUAZZONE, G. (1971) Plate
tectonics model for the development of the Western
Alps and Northern Apennines. Nature, 234, 108-111.
BOETTCHER, S.S. & McBRIDE, E.F. (1993) Thermal histo­
ries of piggy-back and foreland basins in the northern
Apennines, Italy, derived from apatite fission track
thermochronology. EOS, 74, 547.
Buoo, D.A. & HARRIS, P.M. (Eds) (1990) Carbonate­
Siliciclastic Mixtures. Soc. econ . Paleont. Miner. Re­
print Series 14, 272 pp.
CARMIGNANI, L., DECANDIA, F.A., DISPERATI, L. et a/.
(1995) Relationships between the Tertiary structural
evolution of the Sardinia-Corsica-Provencal Domain
and the northern Apennines. Terra Nova, 7, 128-137.
CARVALHO, M.V.F., DE Ros, L.F. & GOMES, N.S. (1995)
Carbonate cementation patterns and diagenetic reser­
voir facies in the Campos Basin Cretaceous turbidites,
offshore eastern Brazil. Mar. Petrol. Geol. , 12, 741-758.
CASTELLARIN, A, CANTELLI, L., FESCE, A.M. et a!. (1992)
Alpine compressional tectonics in the Southern Alps:
relationships with the N-Apennines. Ann. Tecton., VI,
62-94.
CAVAZZA, W. & GANDOLFI, G. (1992) Diagenetic processes
along a basin-wide marker bed as function of burial
depth. J. sediment. Petrol. , 62, 261-272.
CIBIN, U. (1993) Evoluzione composizionale delle areniti
nella successione epiligure eo-oligocenica (Appennino
settentrionale). Giorn. Geol. , 55, 69-92.
CRAIG, H. (1957) Isotopic standards for carbon and
oxygen correction factors for mass spectrometric analy­
sis of carbon dioxide. Geochim. Cosmochim. Acta, 12,
133-149.
CRITELLI, S. & LE PERA, E (1994) Detrital modes and
provenance of Miocene sandstones and modem sands
of the southern Apennines thrust-top basins (Italy). J.
sediment. Res. , A64, 824-835.
DE NARDO, M.T., IACCARINO, S., MARTELLI, L. et a/. (1991)
Osservazioni del bacino satellite epiligure Vetto­
Carpineti-Canossa (Appennino Settentrionale). Mem.
Descr. Carta Geol. !tal. , 46, 209-220.
DETTMAN, D.L. & LoHMANN, K.C. (1995) Microsampling
carbonates for stable isotope and minor element analy­
sis: physical separation of samples on a 20 micrometer
scale. J. sediment. Res. , A65, 566-569.
DICKINSON, W.W. (1988) Isotopic and petrographic evi­
dence for carbonate diagenesis in nonmarine sand­
stones, Green River Basin, Wyoming. J. sediment.
Petrol. , 58, 327-338.
DOYLE, L.J. & ROBERTS, H.H. (Eo.) (1988) CARBONATE­
CLASTIC TRANSITIONS. DEVELOPMENTS IN SEDIMENTOLOGY
42. ELSEVIER, AMSTERDAM, 304 PP.
259
DURAND-DELGA, M. (1984) Principaux traits de Ia Corse
alpine et com!lations avec les Apes Ligures. Mem. Soc.
Geol. !tal., 28 , 285-329.
FOLK, R.L. (1968) Petrology ofSedimentary Rocks. Hemp­
hills, Austin, TX, I 07 pp.
FoNTANA, D. ( 1991) Detrital carbonate grains as prove­
nance indicators in the Upper Cretaceous Pietraforte
Formation (northern Apennines). Sedimentology, 38 ,
1085-1095.
FoNTANA, D. & SPADAFORA, E. (1994) The Miocene Bis­
mantova Formation in the Vetto-Carpineti syncline
(Parma-Reggio Emilia Apennines): evolution of sedi­
ment provenance. Giorn. Geol. , 56, 119-132.
FONTANA, D., ZUFFA, G.G. & GARZANTI, E. (1989) The
interaction of eustacy and tectonism from provenance
studies of the Eocene Hecho Group Turbidite Complex
(Eocene-Central Pyrenees, Spain). Basin Res., 2, 223-
237.
FRIEDMAN, I. & O'NEIL, J.R. (1977) Compilation ofStable
Isotopic Fractionation Factors of Geochemical Interest.
US Geol. Surv. Prof. Paper 440-KK, 12 pp.
FOCHTBAUER, H. (1967) Influence of different types of
diagenesis on sandstone porosity. 7th World Petroleum
Congress, Mexico, Proceedings, 2, 353-369.
GASPER!, G., GELATI, R. & PAPANI, G. (1986) Neogene
evolution of the northern Apennines on the Po Valley
side. Giorn. Geol. , 48, 187-195.
GAZZI, P. (1966) Le arenarie del flysch sopracretaceo
dell'Appennino modenese; correlazione con il flysch di
Monghidoro. Miner. Petrogr. Acta, 1 2, 69-97.
GoTTARDI, G. & GALLI, E. (1985) Natural Zeolites. Miner­
als and Rocks No. 18. Springer-Verlag, Berlin, 409 pp.
HousEKNECHT, D.W. (1987) Assessing the relative impor­
tance of compaction processes and cementation to
reduction of porosity in sandstones. Bull. Am. Ass.
Petrol. Geol. , 7 1 , 633-642.
HUDSON, J.D. & ANDREWS, J.E. (1987) The diagenesis of
the Great Estuarine Group, Middle Jurassic, Inner
Hebrides, Scotland. In: Diagenesis of Sedimentary Se­
quences (Ed. Marshall, J. D.). Spec. Pub!. Geol. Soc.
Lond., 36, 259-276.
INGERSOLL, R.V., CAVAZZA, W., GRAHAM, S.A. & IUFS
Participants (1987) Provenance of impure calclithites in
the Laramide Foreland of southwest Montana. J. sedi­
ment. Petrol. , 57, 995-1003.
JAMES, W.C. (1992) Sandstone diagenesis in mixed
siliciclastic-carbonate sequences: Quadrant and Ten­
sleep formations (Pennsylvan ian), 'northern Rocky
Mountains. J. sediment. Petrol. , 62, 810-824.
KAHN, J.S. (1956) The analysis and distribution of the
properties of packing in sand-size sediments: I . On the
measurement of packing in sandstones. J. Geol. , 64,
385-395.
KANTOROWICZ, J.D., BRYANT, I.D. & DAWANS, J.M. (1987)
Controls on the geometry and distribution of carbonate
cements in Jurassic sandstcnes: Bfidport sands, south­
em England and Viking Group, Troll Field, Norway. In:
Diagenesis of Sedimentary Sequences (Ed. Marshall,
J.D.). Spec. Pub!. Geol. Soc. Lond., 36, 103-118.
KuKAL, Z. & AL-JASSIM, J. (1971) Sedimentology of
Pliocene molasse sediments of the Mesopotamian geo­
syncline. Sediment. Geol. , 5, 57-81.
LAVECCHIA, G., MINELLI, G. & PIALLI, G. (1984) L'Apen-
260 E. Spadafora et a/.
nino umbro-marchigiano: tettonica distensiva e ipotesi
di sismogenesi. Boll. Soc. Geol. !tal. , 103, 467-476.
LAWTON, T.F. (1986) Compositional trends within a clas­
tic wedge adjacent to a fold-thrust belt: Indianola
Group, central Utah, USA. In: Foreland Basins (Eds
Allen, P.A. & Homewood, P.). Spec. Publ. Int. Ass.
Sediment., 8, 411-423.
LEES, A. & BULLER, AT. (1972) Modern temperate-water
and warm-water shelf carbonate sediments contrasted.
Mar. Geol. , 13, M67-M73.
LOMANDO, A.J. & HARRIS, P.M. (Eds) (1991) Mixed
Carbonate-Siliciclastic Sequences. Soc. econ. Paleont.
Miner. Core Workshop 15, 560 pp.
LUNDEGARD, P.D. (1992) Sandstone porosity loss-a 'big
picture' view of the importance of compaction. J
sediment. Petrol. , 62, 250-260.
MACK, G.H. (1984) Exceptions to the relationship between
plate tectonics and sandstone compositions. J sedi­
ment. Petrol., 54, 212-220.
MASSARI, F., GRANDESSO, P., STEFANI, C. & JOBSTRAIBIZER,
P.G. (1986) A small polyhistory foreland basin evolving
in a context of oblique convergence: the Venetian basin
(Chattian to Recent, Southern Alps, Italy). In: Foreland
Basins (Eds Allen, P.A. & Homewood, P.). Spec. Publ.
Int. Ass. Sediment., 8, 141-168.
McBRIDE, E.F., MILLIKEN, K.L., CAvAZZA, W. et a/., (1995)
Heterogeneous distribution of calcite cement at the
outcrop scale in Tertiary sandstones, northern Apenni­
nes, Italy. Bull. Am. Ass. Petrol. Geol. , 79 , 1044-1063.
MOLENAAR, N. & MARTINIUS, A.W. (1990) Origin of
nodules in mixed siliciclastic-carbonate sandstones, the
Lower Eocene Roda Sandstone Member, Southern
Pyrenees, Spain). Sediment. Geol. , 66, 277-293.
MOLENAAR, N., VAN DER BILT, G.P., 0STENDE, V.D.H. &
Ni o, S.D. (1988) Early diagenetic alteration of shallow­
marine mixed sandstones: an example from the Lower
Eocene Roda Sandstone Member, Tremp-Graus Basin,
Spain. Sediment. Geol. , 55, 295-318.
MoUNT, J.F. (1984) Mixing of siliciclastic and carbonate
sediments in shallow shelf environments. Geology, 12,
432-435.
MuTT!, E., PAPANI, L., DI BlASE, D. et a/. (1995) II Bacino
Terziario Epimesoalpino e le sue implicazioni sui rap­
porti tra Alpi ed Appennino. Mem. Sci. Geol. Univ.
Padova, 47, 217-244.
O'NEIL, J.R. & EPSTEIN, S. (1966) Oxygen isotope fraction­
ation in the system dolomite-calcite-carbon dioxide.
Science, 152, 198-20 I .
ORI, G.G. & FRIEND, P.F. (1984) Sedimentary basins
formed and carried piggyback on active thrust sheets.
Geology, 12, 475-478.
PAPANI, G., TELLINI, C., TORELLI, L., VERNIA, L. & IACCAR­
INO, S. (1987) Nuovi dati stratigrafici e strutturali sulla
Formazione di Bismantova nella 'sinclinale' Vetto­
Carpineti (Appennino Reggiano-Parmese). Mem. Soc.
Geol. /tal. , 39, 245-275.
PATACCA, E. & SCANDONE, P. (1989) Post-Tortonian moun­
tain building in the Apennines. The role of the passive
sinking of a relict lithospheric slab. In: The Lithosphere
in Italy (Ed. Boriani, A. et a/.), pp. 157-176. Atti dei
Convegniolei Lincei 80. Roma Accademia Nazionale
dei Lincei, Rome.
PLATT, J.P. ( 1986) Dynamics of orogenic wedges and uplift
of high-pressure metamorphic rocks. Geol. Soc. Am.
Bull. , 97, 1037-1053.
RICCI LucCHI, F. (1987) Semi-allochthonous sedimenta­
tion in the apenninic thrust belt. Sediment. Geol. , 50,
119-134.
SARTORI, R. (1989) Evoluzione neogenico-recente del Ba­
cino Tirrenico e i suoi rapporti con Ia geologia delle aree
circostanti. Giorn. Geol. , 5 1 , 1-39.
SCANDONE, P. (1980) Origin of the Tyrrhenian Sea and
Calabrian Arc. Boll. Soc. Geol. Ita/. , 98, 27-34.
SEARL, A (1994) Diagenetic destruction of reservoir po­
tential in shallow marine sandstones of the Broadford
Beds (Lower Jurassic), north-west Scotland: deposi­
tional versus burial and thermal history controls on
porosity destruction. Mar. Petrol. Geol. , 1 1 , 131-147.
SHACKLETON, N.J. & KENNETT, J.P. (1975) Paleotempera­
ture history of the Cenozoic and the initiation of
Antarctic glaciation; oxygen and carbon isotope analy­
ses in DSDP sites 277, 279, and 281. In: Initial Reports
of the Deep Sea Drilling Project, 29, 743-755. US
Government Printing Office, Washington, DC.
SPADAFORA, E. ( 1996) Compasizione e diagenesi delle
successioni epiliguri mioceniche dell'Appennino setten­
trionale. Unpublished PhD thesis, Universita di Bolo­
gna, 165 pp.
TREVES, B. (1984) Orogenic belts as accretionary prisms:
the example of the Northern Apennines. Ofioliti, 9,
577-618.
VALLONI, R. & ZuFFA, G.G. (1984) Provenance changes for
arenaceous formations of the northern Apennines, Italy.
Geol. Soc. Am. Bull. , 95 , 1035-1039.
VELDE, B. (1985) Clay Minerals: a Physico-Chemical Ex­
planation of their Occurrence. Developments in Sedi­
mentology, 40. Elsevier, Amsterdam, 427 pp.
YouNG, H.R. & DOIG, D.J. (1986) Petrography and prov­
enance of the Glauconitic Sandstone, south-central
Alberta, with comments on the occurrence of detrital
dolomite. Bull. Can. Petrol. Geol. , 34, 408-425.
ZuFFA, G.G. ( 1969) Arenarie e calcari arenacei miocenici di
Vetto-Carpineti (formazione di Bismantova, Appenino
settentrionale). Miner. Petrogr. Acta, 15, 191-219.
ZuFFA, G.G. (1980) Hybrid arenites: their composition
and classification. J sediment. Petrol. , 50, 21-29.
ZuFFA, G.G. (1985) Optical analysis of arenites: influence
of methodology on compositional results. In: Prove­
nance of Arenites (Ed. Zuffa, G.G.). NATO-AS! Series
C: Mathematical and Physical Sciences, 148, 165-189.
D. Reidel, Dordrecht.
ZuFFA, G.G. (1987) Unravelling hinterland and offshore
palaeogeography from deep-water arenites. In: Marine
Clastic Sedimentology-Concepts and Case Studies (a
volume in memory of C. Tarquin Teale) (Eds Leggett,
J.K. & Zuffa, G.G.). pp. 39-61. Graham & Trotman,
London.
ZUFFA, G.G., CIBIN, U. & D1 GIULIO, A (1995) Arenite
petrography in sequence stratigraphy. J. Geol. , 103,
451-459.
Spec. Pubis int. Ass. Sediment. (1998) 26, 26 1 -283

Carbonate cementation in Tertiary sandstones,

San Joaquin basin, California

J.R. BOLES

De partment o f Geolo gical Sciences , Uni versity o f Cali fo rnia ,

Santa Barbara , C A 93106, USA,
e -mail boles @ma gic .geol. ucsb. ed u

ABSTRACT

Carbonate-cemented sandstones occur throughout the San Joaquin basin. New isotopic data from nine
additional areas combined with published papers allow comparison of cement compositions through­
out the basin and a quantitative model of cement timing.
In marine turbidite sandstones of the central basin, following minor siderite precipitation,
dolomites formed early in the zone of methanogenesis. These have Ca-rich compositions similar to
dolomites reported from contemporaneous fine-grained rocks of the Monterey Formation, coastal
California. The dolomites are an example of young ( < 6 Ma) dolomite formation at shallow burial
depth in marine pore water, and they may have undergone some recrystallization during shallow burial
without resetting their initial 87 Sr/86Sr values.
Calcite cements in the central basin formed between burial depths of about 1 . 5 km and> 4 km. The
calcites show significantly lower 8 7 Sr/86Sr values than the depositional marine water, and have progres­
sively lower ratios with increasing burial depth. The latest cements have 87 Sr/86Sr ratios lower than any
possible marine pore water, and the ratios indicate that Sr and Ca are sourced from plagioclase feldspar.
Calcite cements formed at intermediate burial depths have carbon isotopic compositions sourced in part
from thermogenic-derived carbon but during deep burial, carbon isotopic values near zero (PDB) suggest
carbon derived from unknown reactions, possibly related to the organic acids in the oil reservoir. Sr
isotopic values preclude dissolution of shell tests as the primary carbon source in these late calcites. Late
cements appear to have formed in a relatively closed system, on the reservoir scale, during the dissolution
of detrital plagioclase within the reservoirs, possibly during hydrocarbon emplacement.
The basin margins are characterized by calcite and minor dolomite cements, many of which which
formed in isotopically light brackish or meteoric water at low temperature. In general, calcites did not
form near the sediment-water interface, but during shallow burial. On the east side of the basin these
cements are characterized by widely varying o13Cp08 values (+20 to -30) compared with central basin
cements (+5 to - 1 0). Sr isotopic ratios in cements are lower than the marine depositional waters on the
east side of the basin, but are higher than expected for depositional waters on the west side.
Although the San Joaquin basin has evidence for cross-formational fluid flow, in many cases each
reservoir has carbonate cements with distinctive compositions. This indicates that the flow that has
occurred has not been at a rate or magnitude sufficient to homogenize the pore fluids within closely
spaced reservoirs.

INTRODUCTION

Carbonate cements occur in small amounts in many
sandstone hydrocarbon reservoirs of the San Joa­
quin basin. The cements formed throughout much of
the burial history of the basin, and thus provide an
extensive record of organic-inorganic diagenesis.
Moreover, the cements record the nature and mag-
nitude of diagenetic mass transfer during burial.
Textural, isotopic and trace element data from car­
bonate cements help constrain the timing of fluid
movement, including meteoric incursions into the
basin.
The San Joaquin basin has several attributes that

Carbonate Cementation in Sandstones Edited by Sadoon Morad 261

© 1998 The International Association of Sedimentologists
ISBN: 978-0-632-04777-2
262 J.R. Boles

'' ''
\ \
\
\ \
\ \
\
\ �
\ 00'<
\
''
\
\ I
I
I I

i
I
I
I Fig. 1. Location of oilfields in the
I
I San Joaquin basin with carbonate
N I
I
cement studies. Codes are: North
\ Coles Levee (NCL), South Coles
I
I Levee (SCL), Canal (C), Paloma
\
\ (P), Landslide (L), Yowlumne (Y),
'. Rio Viejo (RV), San Emidio Nose
RR •
(SN), Rosedale Ranch (RR),
Fruitvale (F), Mountain View
(MV), Edison (E), Mount Poso
(MP), Round Mountain (RM),
Kern River (KR), Poso Creek
(PC), North Belridge (NB) and
Kettleman North Dome (KND).
Dashed lines are thickness of
sedimentary rocks (metres), after
KILOMETERS Callaway ( 1 97 1 ). Cross-section line
X-X' is shown in Fig. 2.

make it an exceptional locality to study cementa­ several diagenetic processes, including the smectite/
tion processes. Exploration and the development of illite transition (Ramseyer & Boles, 1986) and pla­
hydrocarbon resources have produced an abun­ gioclase alteration (Boles & Ramseyer, 1988).
dance of subsurface data from fields as shallow as Finally, carbonate cements reveal the sources of
500 m to deeper than 4 km (Figs 1 and 2). The basin dissolved carbon in the evolving pore waters of the
contains more than 7 km of Cenozoic sediment, San Joaquin basin. The clastic-rich basin is free of
most of which is at maximum burial depth and carbonate rocks but contains a considerable amount
temperature. At depths greater than 4 km in the of organic matter, both in fine-grained sediment
central basin, in sit u temperatures exceed 14o·c in and as relatively recent hydrocarbon accumulations.
hydrocarbon-producing reservoirs. Therefore, ce­ Potential carbon sources for the carbonate cements
mentation can be studied from the surface up to are marine shell tests, thermogenesis and, possibly,
relatively high temperatures. The very young depo­ organic reactions related to the presence of the oil.
sitional age of the basin (much of the section is less This paper includes 58 carbon-oxygen and 25
than 15 Ma) and its relatively simple subsidence strontium isotopic analyses from my research group
history allow the construction of accurate time­ and synthesizes these data with previous published
temperature burial paths. When combined with work in the basin. The new a,nalyses extend the
estimates of cementation temperatures, the timing coverage of the basin to nine additional areas,
of cementation can be constrained to a higher including the southern and eastern parts, where no
degree than is possible in most other basins. similar data have been published. This paper intro­
The carbonate cements of the San Joaquin basin duces new comparisons between cementation in the
have been useful for reconstructing the diagenetic central basin and that in basin margins.
history of non-carbonate reactions in the basin.
Where rock is completely cemented, carbonate ce­
mentation can be used to deduce the reaction SAN JO AQUIN BASIN
progress before and after cement sealing by compar­
ing diagenesis both within and outside the cement
Geological history
zone. The low permeability of extensively cemented
sandstone has effectively prevented further diagene­ Callaway ( 1971) provides an excellent review of the
sis within that rock. This has been demonstrated for geological framework of the San Joaquin basin
 Carbonate cementation in Tertia ry sandstones 263

Sierra '
X Nevadas X
"

:;
0
u.
"'
0
..
�
"0
c
<{

c
0
(/)

20km IOkm

..
l500m

?
IOOOm

Fig. 2. West to east cross-section across San Joaquin basin. See Fig. I for location of the cross-section line. Most of
basin-fill is marine, including the Stevens sandstone, and is at maximum burial depth. Non-marine strata are Chanac
and Kern River Formations. Low lateral continuity of beds and abundant shales have prevented meteoric water from
entering the deep central basin. Cross-section from California Division of Oil and Gas.

(Fig. 1). According to this summary, more than 80%
of the basin fill is of Miocene and younger age
(Fig. 2). The oldest sediment in the basin is Pale­
ocene to Early Miocene strata deposited in a marine
basin that opened to the southwest (Graham, 1987).
Non-marine to marginal marine sands on the flanks
of the basin grade into deep marine shales towards
the basin centre. Shallow marine to non-marine
facies were deposited during the Oligocene, when
the sea withdrew from most of the basin. During the
Middle to Late Miocene a sequence of marine
shales and deep-sea fan sands, the Stevens sand­
stone, were deposited in the central basin. The
shallow marine to non-marine equivalents of the
Stevens are preserved on the east basin flank as the
Santa Margarita and Chanac Formations. The basin
began to shoal in the Pliocene, resulting in deposi­
tion of the non-marine Kern River Formation
(Fig. 2).
The basin centre or central basin and the basin
flank areas are described separately, because of their
different geological histories. The central region of
the southern San Joaquin basin includes North
Coles Levee, South Coles Levee, Paloma, Canal,
Landslide, Yowlumne, San Emidio Nose and Rio
Viejo fields (Figs 1 and 2). The sedimentary se­
quence consists of up to 7 km of Miocene and
younger arkosic sediments deposited in deep-sea
fan environments (Callaway, 1971), and many of
the data come from Upper Miocene Stevens and
equivalent-age sandstones (Webb, 1981 ). Sedi­
ments of the central basin have undergone simple
burial to their present depth (Fig. 3). In the central
basin pore fluid temperatures have increased along
the prevailing geothermal gradient, which Wood &
Boles (1991) estimate to be about 30-36.C/km.
Reservoir temperatures at the time of field discov­
ery for the samples range from a low of about 100 • C
at 2.4 km (8000 ft) in the Canal field to a high of
about 14o·c at 4.3 km (14 000 ft) in the Rio Viejo
field (California Division of Oil and Gas, 1985).
The eastern margin of the San Joaquin basin is in
depositional contact with Sierra Nevada crystalline
rocks, and includes shallow marine and non-marine
264
EOCENE OUGOCENE MIOCENE PLIO-PLEISTOCENE
0.
a:
UJ
1-­
UJ
::2
0
...J
S2
2.25
J:
Well ClA 67-29
li:
UJ
Cl
5.5
50 40 30 20 10
AGE [Ma]
Fig. 3. Time-depth-temperature burial history plot at
North Coles Levee. Modified from Wood & Boles
( 1 99 1 ). Note the rapid burial of the Stevens sandstone
(shaded) at North Coles Levee, which is also typical of
Stevens sandstone's burial history throughout the central
basin. Carbonate cements formed throughout the burial
history of the Stevens.
facies of Eocene to Pleistocene age (Callaway, 1971;
Dunwoody, 1986; Goodman & Malin, 1 992). The
contact is faulted in many places. Fields that have
been studied on the eastern area include Edison,
Fruitvale, Kern River, Mountain View, Mount
Poso, Poso Creek and Round Mountain. The east­
ern margin has undergone modest uplift (less than
300 m; Olsen, 1988) and burial depths are generally
less than a few kilometres. Maximum burial tem­
peratures can therefore be accurately estimated.
The deepest studied samples from the eastern flank
are at 1770 m in the Kern River field, where present
temperatures are at 67 ·c; the shallowest sample is
at 317 m in the same field, at a temperature of
24·c.
The western margin of the San Joaquin basin,
including Kettleman North Dome and North Bel­
ridge fields, is in contact with the San Andreas fault
and has undergone significant uplift and erosion.
The Oligocene depositional facies studied at North
Belridge is interpreted as a submarine fan (Taylor &
Soule, 1993) and the early Miocene strata at Kettle­
man North Dome include non-marine facies to
submarine fan deposits (Kuespert, 1985). Strati-
J.R. Boles
graphical data in the Kettleman Hills and South
Belridge areas indicate up to several kilometres of
T[•C] uplift and erosion (Bloch et a!. , 1993, and Schwartz,
1988, respectively) and this magnitude of uplift is
confirmed by diagenetic studies (Boles & Ramseyer,
1988; Taylor & Soule, 1993). Westerly sediment
sources are no longer in contact with the sediment
package owing to right-lateral movement along the
San Andreas fault. The west basin flank is the most
geologically complex area to interpret in the basin
because of the poor constraints on uplift history and
the uncertain pore fluid evolution. Present temper­
atures are as high as 110·c at 2.6 km in the North
Belridge field samples (Taylor & Soule, 1993) and
1oo·c at 2.3 km in the Kettleman North Dome
samples (Lee & Boles, 1996), but maximum tem­
0 peratures could have been greater than 1so·c in
some areas (see Taylor & Soule, 1993).
Basin pore water
The relatively uniform arkosic composition of the
basin sediment and the marine depositional setting
in much of the basin, particularly the central area,
combined with the absence of underlying salt,
suggests that the evolution of the pore water is
relatively simple and is largely the result of marine
water-arkosic rock interaction. In the central basin,
the only water involved in diagenesis was original
marine pore water, and possibly that derived from
dehydration reactions in fine-grained sediment
(smectite/illite or perhaps opal/quartz alteration).
At present much of the basin pore water has a
salinity similar to sea water, but in the deeper parts
of the basin has been modified by reactions with
plagioclase and organic matter (Fisher & Boles,
1990; Feldman et a! ., 1993). The extent of palaeo­
meteoric water incursion into the basin margins is
deduced from the presence of relatively fresh water
in some marine sediments (Fisher & Boles, 1990)
and the recognition of meteoric cements in marine
strata (Hayes & Boles, 1993; Taylor & Soule, 1993;
Lee & Boles, 1 996).
In summary, reconstructed geological histories
and interpretations of cement timing are relatively
accurate for the central basin, are somewhat less so
for the shallow-buried and modestly uplifted east­
ern margin, and are much less well constrained for
the structurally complex western margin. Because
of the different geological histories, the discussion
of carbonate cements is divided into basin centre
and flank regions.
 Carbonate cementation in Tertia ry san dstones
SAMPLES AND METHODS
Sample location
Figure 1 shows the location of the hydrocarbon
reservoirs discussed in this paper and Table 1
references the data sources. Table 2 contains the
new isotopic data from the basin. Numerous studies
of the North Coles Levee have been published
(Boles & Ramseyer, 1987; Schultz et al. , 1989;
Wood & Boles, 1991), but only sparse data have
been published from other fields in the area (e.g.
Paloma and Lakeside fields; see Fischer & Surdam,
1988). The new data in Table 2 add seven new areas
to the central basin database, including South Coles
Levee, Canal, Paloma, Landslide, Rio Viejo, San
Emidio Nose and Yowlumne fields.
On the basin flanks there have been several
papers on the eastern margin, including geochemi­
cal data on cements in the Mountain View and
Edison fields (Fischer & Surdam, 1988) and Kern
River, Poso Creek, Rosedale Ranch, Fruitvale and
Mount Poso fields (Boles & Ramseyer, 1987; Hayes
& Boles, 1993). On the western flank, studies of
carbonate cementation include the North Belridge
field (Taylor & Soule, 1993) and Kettleman North
Dome (Merino, 1975; Lee & Boles, 1996). The new
analyses from the basin margin include data from the
Mount Poso field and a wildcat well Ohio KCLG-1,
about 5 km southwest of the Fruitvale field.
Sample selection
The geochemical data described here are based on
sampling of diamond drill cores. They are believed
to represent a random sampling of the basin ce­
ments, because the sampling process was guided
only by finding zones that appeared cemented
without any previous knowledge of what the cement
composition might be. It is important to note that
these samples are spot samples, and that consider­
able heterogeneity can exist over the scale of a few
metres of core. In most cases, samples represent a
randomly selected sample of a cement zone. De­
tailed systematic sampling of cement zones on a
scale of centimetres has generally not been done,
but is the thrust of our present investigation.
Isotopic analyses are measured on samples with a
preponderance of either calcite or dolomite, based
on X-ray diffraction. The samples were further
high-graded by excluding samples with obvious
mixtures of shell tests, cements, grain replacements
265
and fracture-fill. Because of the relatively high iron
content of most San Joaquin carbonates (see later),
cathodoluminescence is generally not effective in
recognizing fine-scale cement zones within crystals,
a successful method in carbonate rocks (see Meyers,
1974). Typically, calcites and dolomites show either
no luminescence or a dull orange colour. In some
cases calcite samples were selected on the basis of
relatively uniform iron, magnesium and manganese
trace element composition, based on spot micro­
probe analysis (Boles .& Ramseyer, 1987).
TIMING OF CEMENTATION
One of the most challenging aspects of diagenesis is
to quantify the time and duration of cementation.
In the San Joaquin basin we used several methods
to estimate when carbonate cements formed. One
method, which has also been applied by many
previous workers in other basins (e.g. Galloway,
1979), is to infer the porosity at the time of
cementation from the volume of pore-filling ce­
ment. If the compaction history for uncemented
sands is known, the cement volume in fully ce­
mented sandstones implies the burial depth at
which cementation occurred. In the case of the San
Joaquin basin, most areas underwent simple sub­
sidence to their present depth, and the relation
between depth of burial and porosity for unce­
mented sandstones is known from abundant
porosity-depth data (e.g. Ziegler & Spotts, 1976).
Thus the depth of cementation can be inferred and,
combined with a time-depth burial curve, so can
the timing of cementation.
Another useful guide to the timing of cementa­
tion is estimation of the temperature of cementa­
tion from oxygen isotopic analysis and an assumed
oxygen isotopic composition of. the pore water.
From the temperature, cementation timing can be
inferred from a time-temperature burial history.
The present burial temperature defines the upper
temperature limit of precipitation in much of the
basin, where the sediment is currently at its maxi­
mum burial temperature. In the central basin, the
pore water oxygen isotopic composition can be well
constrained because only marine or evolved marine
pore waters exist, and present-day values are known
(Carothers & Kharaka, 1978; Fisher & Boles, 1990).
For the central basin the pore water evolved from
the initial Miocene marine value near zero (SMOW
scale) to its present-day value near +4 (see Boles &
 N
a­
a-

Table I. Carbonate cement studies in the San Joaquin basin

Depositional
Field Formation Age environment Data Reference

Basin centre
North Coles Levee Stevens Upper Miocene Deep marine V, OC, SR, TR Boles, 1 987; Boles &
Ramseyer, 1 987; Mozley,
1 989; Schultz et al.,
1 989; Wood & Boles,
1 99 1
South Coles Levee Stevens Upper Miocene Deep marine V, OC, SR, TR This chapter
Paloma Stevens Upper Miocene Deep marine oc Fischer & Surdam, 1 988
Lakeside Stevens Upper Miocene Deep marine oc Fischer & Surdam, 1 988
Yowlumne Stevens Upper Miocene Deep marine oc This chapter
Rio Viejo Stevens Upper Miocene Deep marine oc This chapter
San Emidio Nose Stevens Upper Miocene Deep marine oc This chapter �
�
Basin margin (east) l:l;l
a
Rosedale ranch Kern River Late Miocene-Pleistocene Non-marine OC, TR Boles & Ramseyer, 1 987
Fruitvale Chanac Late Miocene Non-marine OC, TR Boles & Ramseyer, 1 987 �
Mountain View Santa Margarita-Fruitvale Late Miocene Shallow marine OC, TR Fischer & Surdam, 1 988
Edison Santa Margarita-Fruitvale Late Miocene Shallow marine OC, TR Fischer & Surdam, 1 988
Mount Poso Vedder Oligocene Shallow marine V, OC, TR Hayes & Boles, 1 993; this
chapter
Round Mountain Vedder Oligocene Shallow marine V, OC, TR Hayes & Boles, 1 993
Kern River Kern River Late Miocene-Pleistocene Non-marine V, OC, TR Hayes & Boles, 1 993
Vedder Oligocene Shallow marine V, OC, TR Hayes & Boles, 1 993
Famosa Eocene Shallow marine V, OC, TR Hayes & Boles, 1 993
Poso Creek Chanac Late Miocene Non-marine V, OC, TR Hayes & Boles, 1 993

Basin margin (west)

North Belridge 64-Zone Sandstone Oligocene Deep marine V, OC, SR, TR Taylor & Soule, 1 993
Kettleman North Temblor Lower to Middle Miocene Non-marine to V, OC, TR Lee & Boles, 1 996
Dome shallow marine

V, point count cement volume; OC, oxygen-carbon isotopic analyses; SR, strontium isotopic analyses; TR, trace element analyses.
Table 2. Carbonate cement data, San Joaquin basin

Field Well Depth core ( ft) Depth core (m) Depo age 01 80PDB 013 CPDB 87 Sr/s6sr Sr ppm Analyst

Calcite
S. Coles Levee 25- 1 2 9857. 3-9857.8 3004.60 Upper Miocene -5.90 -5.08 0. 707544 388 Unocal
S. Coles Levee 32-9 9 1 77 2797 . 1 5 Upper Miocene -9.48 1 .0 1 0.707 1 84 nd Unocal
S. Coles Levee KCL 67-1 1 1 0040.3 3060.28 Upper Miocene -5.00 -4. 1 5 0.7080 1 6 279 Unocal
S. Coles Levee KCL 67-1 1 1 0043 306 1 . 1 1 Upper Miocene -7 . 9 1 -6. 8 1 0. 707655 414 Unocal
S. Coles Levee KCL 67-1 1 1 0537.8 32 1 1 . 92 Upper Miocene - 1 0.67 3.52 0.707348 938 Unocal
S. Coles Levee KCL 67-1 1 1 0562.2 32 1 9.36 Upper Miocene - 1 0. 3 8 4. 1 2 0.7074 1 5 916 Unocal
S. Coles Levee KCL 67- 1 1 1 0586.9 3226.89 Upper Miocene -9.73 0.60 0.707446 1 047 Unocal
S. Coles Levee KCL 67- 1 1 1 0587.9 3227 . 1 9 Upper Miocene -8.67 0.39 0.707550 862 Unocal
S. Coles Levee KCL 67-1 1 1 0588.9a 3227.50 Upper Miocene -8.76 -0.24 Unocal
S. Coles Levee KCL 6 7- 1 1 1 0588.9b 3227.50 Upper Miocene -8.78 -0.20 Unocal
S. Coles Levee KCL 67-1 1 1 0588.9c 3227.50 Upper Miocene -8.93 - 1 .48 Unocal Q
....
S. Coles Levee KCL 67- 1 1 1 0589.9 3227.80 Upper Miocene -9. 1 0 -0.63 0.707852 367 Unocal <:l-
0
S. Coles Levee KCL 67- 1 1 1 0600 3230.88 Upper Miocene -8.46 1 . 20 0.707835 700 Unocal ;-s
�
S. Coles Levee KCLB 67-4 9 1 02-92 1 4 2776 . 1 0 Upper Miocene -7.2 1 -0.63 0.707 7 1 3 763 Unecal ;:;;
S. Coles Levee KCLB 67-4 92 1 2-9224 2809.60 Upper Miocene -7.80 - 1 .46 Unocal '"'
""
Canal KCL E-23 8090.5 2465.98 Upper Miocene -6.82 -3.82 0.70820 1 290 Unocal 2i
""
Canal SHE A-2 1 8300 2529.84 Upper Miocene -6.52 -9.95 0.7080 1 6 1 1 05 Unocal ;:::s
Canal SHE A-2 1 8436 257 1 .29 Upper Miocene -5. 3 3 -3.08 0. 708446 1 063 Unocal §
<::;·
Landslide 27X- 1 9 1 2555.6 3826.95 Upper Miocene - 1 2.35 -7.47 UCSB ;:::s
Landslide 27X- 1 9 1 2596.5 3839.4 1 Upper Miocene - 1 1 . 23 -3.41 Unocal �·
Landslide 27X- 1 9 1 27 0 1 387 1 .26 Upper Miocene - 1 1 . 72 - 1 .84 UCSB
Landslide Transco 83X-23 1 2 1 65.8a 3708. 1 0 Upper Miocene - 1 3.98 -3.70 Unocal �
;::t
Landslide Transco 83X-23 1 2 1 65.8b 3708. 1 0 Upper Miocene - 1 4. 4 1 5. 1 5 Unocal iS"
Landslide Transco 83X-23 1 2 1 65.8c 3708. 1 0 Upper Miocene - 1 4. 1 1 3.51 Unocal �
"'
Paloma Sup. And. 1 8-35 1 10 1 0 3355.85 Upper Miocene -8. 1 1 -7. 1 3 0.707792 357 Unocal �
;:::s
74-2 1 0982- 1 1 03 1 Upper Miocene -7.5 1 -9.45 Marathon
Paloma
�
Paloma KCLA A72-4 1 6790 5 1 1 7.59 Middle Miocene - 1 5.62 4. 1 1 Marathon 0
;:::s
Rio Viejo Tennaco 22X-34 1 5 1 24- 1 5 1 25 Upper Miocene - 1 5.66 -0.68 UCSB Cl
Rio Viejo Tennaco 22X-34 1 5 1 36 46 1 3.45 Upper Miocene - 1 5.62 -0.27 UCSB
Rio Viejo Tennaco 22X-34 1 5 1 3 7- 1 5 1 3 8 Upper Miocene - 1 5.57 -0.34 UCSB
Rio Viejo Tennaco 22X-34 1 5 1 44 46 1 5.89 Upper Miocene - 1 5.76 -1.31 UCSB
Yowlumne 22X-3 1 1 8 1 6.2 360 1 .58 Upper Miocene - 1 2. 8 1 -4.80 UCSB
Yowlumne 22X-3 1 1 86 1 . 2 36 1 5.29 Upper Miocene -7.45 -3. 7 1 UCSB
Yowlumne Tennaco 68-32 1 1 942 3639.92 Upper Miocene -7.94 - 1 .74 UCSB
Yow1umne Tennaco 68-32 1 1 948 364 1 .75 Upper Miocene -8. 1 8 - 1 .32 UCSB
Yowlumne Tennace 68-32 1 1 972 3649.07 Upper Miocene -8.70 -2.55 UCSB
Yowlumne Tennaco 68-32 1 1 97 7 3650.59 Upper Miocene - 1 0. 79 -0.39 UCSB
San Emidio Nose KCLH 1 3- 1 5 1 3280 4047.74 Upper Miocene -7.97 -2.88 Marathon
San Emidio Nose KCLH 1 3- 1 5 13319 4059 .63 Upper Miocene -6. 3 7 -0.96 Marathon
Roco KCLM 87-3 1 4 1 2 1 - 1 4 1 43 Upper Miocene - 1 0. 74 -2. 1 4 UCSB N
San Emidio Nose 0..
--- --.1
Continued
 N
a..
00

Table 2. (Continued)

Field Well Depth core (ft) Depth core (m) Depo age o180p os 1>13Cpos 87Sr/B6sr Sr ppm Analyst

Mount Poso Vedder I 2483.7 7 5 7.03 Oligocene -5.62 - 1 5 .06 0.706779 63 Unocal
Mount Poso Vedder Rail 352 1 95 6 . 5 596.34 Oligocene -2.29 -4.27 0.70698 1 42 Unocal
Mount Poso Vedder Rall 43 1 1 785.5 544.22 Oligocene - 1 0. 1 7 - 1 2.66 0.706344 44 Unocal
Mount Poso Vedder Rall 43 1 20 1 8 . 1 6 1 5. 1 2 Oligocene -2.40 - 1 5.20 0.707864 1 82 Unocal
Mount Poso Vedder Rall 43 1 203 3 . 5 6 1 9. 8 L Oligocene -8.93 5.55 0.706832 72 Unocal
Mount Poso Vedder Rall 43 1 2040.1 6 2 1 .82 Oligocene -7.65 -32.83 0.7064 1 6 64 Unocal
Wildcat Ohio KCLG- 1 6063-6067 1 848.60 Upper Miocene -7.36 6.97 0.70760 1 nd Unocal
Wildcat Ohio KCLG- 1 6236-6238 1 90 1 .00 Upper Miocene 1 . 36 - 1 4.42 Unocal
Wildcat Ohio KCLG-2 6236-6238 1 90 1 .00 Upper Miocene -4.20 2.94 0.7087 1 7 1 79 Unocal �
Wildcat Ohio KCLG- 1 6257-62 6 1 1 907.70 Upper Miocene -4.93 3.95 0. 7083 3 3 934 Unocal ?:l
\::J::l
Dolomite
0

�
-...

N. Coles Levee NCL 487-29 9082-9083 2768.30 Upper Miocene -3.43 4.63 ARCO
N. Coles Levee NCL 488-29 8947 2727.05 Upper Miocene -2.76 4.96 ARCO
N. Coles Levee NCL 488-29 9037 2754.48 Upper Miocene -2.66 3.69 ARCO
N. Coles Levee NCL 488-29 9037.2 2754.54 Upper Miocene -2.35 5 . 59 ARCO
N. Coles Levee NCL 48 8-29 9040. 3 2 7 5 5 .48 Upper Miocene -2.40 4.85 ARCO
N. Coles Levee NCL 48 8-29 9047 2757.53 Upper Miocene -3. 5 1 4.07 ARCO
San Emidio Nose KCLM 87-3 1 4 1 32.2 4307.50 Upper Miocene -2.45 8.65 ARCO
Paloma KCLA 72-4 1 7090 5209.03 Middle Miocene -3.22 9.54 0.708682 nd Marathon

Depo, depositional.
Carbon oxygen isotope analysts are M. DeNiro (UCSB), G. Thyne (ARCO), P. Dobson (Unocal), D. Wallwey (Marathon).
Sr isotopic analyses by J. Schultz, UCSB. nd, not determined.
 Carbonate cementation in Tertiary sandsto nes
Ramseyer, 1987; Schultz et a/ ., 1989). By assuming
that the water has evolved linearly between the
initial and present-day values, a best estimate for
the water composition is used to refine the estimate
of precipitation temperature (Boles & Ramseyer,
1987).
On the basin margin, where either the isotopically
light meteoric waters may have been exchanged for
the original marine pore water or the original
depositional water composition is unknown, it is
difficult to estimate the evolutionary path of the
water's oxygen isotopic composition. Constraints
on possible values are present Kern River water
draining into the eastern basin margin with
o180sMow -13.5 (see Boles & Ramseyer, 1987) and
present-day shallow meteoric groundwaters ranging
from about -6 to -12 (Coplen et a! ., 1985). Fischer
(1 986, p. 1 23) reports subsurface waters in the
Edison field ranging from -8.1 to -10.6, suggesting
a modified meteoric water.
The final method for determining carbonate ce­
ment timing, particularly in the central basin, is
87 Sr/86 Sr composition. Based
from the carbonate
on studies at North Coles Levee, Schultz et a!.
(1989) found that Sr isotopic ratios decrease with
increasing crystallization temperatures (as inferred
from the oxygen isotopic ratios). Thus Sr isotopic
ratios are a useful guide to cementation timing.
New data presented here for South Coles Levee
confirm this relationship.
DIAGENETIC HISTORY OF THE BASIN
The sandstones of the San Joaquin basin are com­
posed largely of quartz and subequal proportions of
plagioclase and K-feldspar, with less than 25%
igneous and metamorphic rock fragments. The
chief mafic mineral is biotite. The diagenesis of
these sediments includes compaction, cementation
and mineral dissolution (Boles, 1987; Fischer &
Surdam, 1988; Hayes & Boles, 1993; Taylor &
Soule, 1993). Much of the present reduced porosity
in the basin is due to compaction rather than
cementation. Biotite, for example, shows increasing
deformation with burial depth. Average porosities
in Pleistocene sands of the Kern River field are
30-40% at 1 20-400 m depths, but Upper Miocene
sands average 15% at 4300 m in the Rio Viejo field
(California Division of Oil and Gas data). Some
sands underwent early cementation by sparse pyrite
and siderite and, more commonly, dolomite. Our-
269
ing moderate to deep burial calcite cementation was
common, up to and at maximum burial depths.
Detrital plagioclase feldspar is the most impor­
tant source of silicate diagenesis in sandstones.
Feldspar dissolution occurred relatively late in t}le
burial history of the central basin, as indicated by
its absence in sandstones with an early carbonate
cement; in addition, active albitization occurs
where burial temperatures exceed about l.20"C
(Boles & Ramseyer, 1988). Kaolinite and calcite are
byproducts of the plagioclase alteration. Plagioclase
dissolution is also recognized as a weathering prQd­
uct (Bloch & Franks, 1993), apparently as a result of
early diagenesis from meteoric water at the basin
margin (Hayes & Boles, 1992). Diagenesis Qf smec­
tite to illite is occurring at deeper levels in the basin,
but overall this reaction appears to be less, advanced
than found at comparable temperatures in the older
strata of the Texas Gulf Coast (Ramseyer & Boles,
1986).
In addition to inorganic reactions, the basin has
abundant hydrocarbon accumulations-chiefly oil-"
that have resulted from organic diagenesis. The oil&
are found throughout the stratigraphical section,
indicating there has been considerable cross­
formational flow from deeper levels in the basin.
The central basin pore waters are notable in that
they are known to contain an abundance of organic
acids (Carothers. & Kharaka, 1978; MacGowan &
Surdam, 1988; Fisher & Boles, 1990). As is dis­
cussed later, kerogen maturation or organic acids
from oils may have been a source of carbon in
cements precipitating during moderate to de�p
burial.
BASIN CENTRE CEMENTS
Early fluid pathways
Sandstones in the central basin, with the highest
initial porosities and permeabilities, were the first to
experience extensive cementation. Presumal;>ly
these were flow pathways and, considering the pore
fluid volumes required for extensive cementation, it
is not surprising that this relationship occurs. The
basis for this observation is that sandstones with
early high-volume cements are notably clay free
(relatively low elemental AI content) compared with
adjacent uncernented sandstones (Boles, 1989). The
clay is a smectite-rich mixed-layer smectite/illite
clay, believed to be detrital on the basis of its
270
relatively high Sr isotopic composition, which is
consistent with its being a weathering product of
Sierran K-feldspar rather than a product of Mio­
cene sea water or later diagenesis (Schultz et a/ .,
1989). The smectite is believed to have reduced the
permeability of the sands so that flow was focused
through other, more permeable zones, where ce­
mentation began. Interestingly, cementation con­
tinued in these relatively clay-free sands to the point
that permeability was below 1 mD, much lower
than for adjacent clayey sands. Presumably, favour­
able nucleation kinetics promoted continued ce­
mentation of the sand beyond the point
unfavourable mass transfer properties.
Cement petrology
Pore-filling cements in the central basin show a
characteristic paragenetic sequence of early siderite
to dolomite to calcite. In terms of abundance,
calcite is by far the most abundant and siderite is
the least common. Photomicrographs of cement
types are shown in Fig. 4.
Siderite occurs as scattered, small (10-15 J.lm),
yellowish euhedral crystals attached to detrital
grains or enclosed by later carbonate cements
(Fig. 4A). The siderite has a relatively Mg-rich
composition, up to 40 mol% relative to Fe (Boles,
1987), probably reflecting the Mg-rich composition
of marine water (Mozley, 1989). Extensive cement
zones or concretions of siderite have not been
found, indicating that sideritization is not as com­
mon in this deep marine environment as in shallow
marine and non-marine environments (e.g. see Mo­
zley, 1989).
Dolomite, forming after siderite, is the most im­
portant early cement in the central basin (Fig. 4B). It
accounts for 30-40% of the sandstone volume in
cemented zones, which are up to 100 em thick.
These can be correlated laterally for at least 120 m
in some cases at North Coles Levee, where closely
spaced well data are available (Boles & Ramseyer,
1987). More commonly, dolomite occurs as scat­
tered euhedral rhombs (up to 200 J.lm) enclosed in
later calcite cements. Evidence for an early origin of
dolomite includes high intergranular cement vol­
ume, euhedral rhomb-shaped crystals indicating
crystallization into open pore spaces, and the align­
ment of cement zones parallel to bedding even in
deformed areas, suggesting early crystallization prior
to compaction and folding (see Boles & Ramseyer,
1987). Detrital biotite in extensively cemented rock
J.R. Boles
shows little compaction deformation, and in some
cases has undergone expansion from carbonate
growing between the cleavage flakes (Fig. 4B).
Calcite cement is the dominant cement type in
the central basin. Cemented zones can be visually
recognized in cores and are from I 0 em to, in a few
cases, more than 1 m thick (Boles & Ramseyer,
1987). Cement zones cannot be easily traced be­
tween wells spaced as close as 100 m, suggesting
that the intensely cemented zones are relatively
isolated and discontinuous, certainly on a basin
scale and in most cases on a reservoir scale. Most
of cement zones have not been studied in sufficient
detail to establish growth patterns. A few detailed
analyses of individual zones show that some have a
composite history (i.e. variable isotopic composi­
tions) on a scale of less than 0.5 m (e.g. cement zone
at North Coles Levee, well NCL 488-29, 2621 m
depth), whereas others show little variation (Schultz
et a/ ., 1989). Systematic growth patterns, such as
are typical for concretions in shales (e.g. Raiswell,
1971; Boles et a/ ., 1985) or in concretions that
coalesce to form continuous cemented beds (Bjm­
kum & Walderhaug, 1990), have not been recog­
nized in the zones studied to date. Apart from
extensively cemented zones, calcite occurs as scat­
tered crystals in many samples.
In thin section, calcite appears most commonly as
scattered euhedral crystals from I 0 J.lm up to
500 J.lm (Fig. 4C,D). Completely cemented samples
have an interlocking mosaic of crystals that fill pore
space but typically do not replace detrital grains.
Some sandstones with abundant calcite cements are
relatively uncompacted, with uncrushed biotites,
and these generally do not have partially dissolved
detrital plagioclase grains, whereas adjacent com­
pacted and uncemented sandstones may exhibit up
to 5% secondary porosity after plagioclase. These
cements are inferred to be relatively early.
In other sandstones with a relatively compacted
fabric, biotites are deformed, plagioclase is partially
dissolved and calcite cements partially replace dis­
solved plagioclase or, more commonly, enclose
kaolinite in intergranular pore space (Fig. 4D). This
kaolinite has been demonstrated to be a byproduct
of plagioclase dissolution (Boles, 1984). Fracture-fill
calcite is minor but tends to be more common in the
relatively compacted sands, where grains have been
fractured and micas crushed (Fig. 4D). These cal­
cite cement occurrences in altered sandstones are
interpreted to be a later generation than those
occurring in less altered rock.
 Carbonate cementation in Tertiary sandstones 271

Fig. 4. Photomicrographs of carbonate cements in sandstones of the San Joaquin basin. (A) Siderite rhombs (arrows)
in pore space (dark areas). Well NCL 8 8-29, 2746 . 5 m (90 1 0. 8 ft). White bar is 0.25 mm. (B) Dolomite pore-filling
sandstone from the central basin. Note high cement volume and undeformed detrital biotite (dark grains). Detrital
grains are chiefly quartz and feldspar. Well NCL 88-29, 27 1 7 m (89 1 3 ft). White bar is 0.5 mm. (C) Calcite pore-filling
cement from central basin. Note relatively high cement volume and partially crushed biotite. Well NCL 487-29,
272 1 m (8927.3 ft). White bar is 0.5 mm. Estimated precipitation temperature is 52"C (see Schultz et a!., 1989).
(D) Calcite pore-filling and grain replacement in the central basin. Note crushed biotite (B) and plagioclase replaced by
calcite (arrows). Most of the light-coloured mineral is pore-filling calcite. Well NCL 48 7-29, 2722.8 m (8933. 1 -0.3 ft).
White bar is 0.5 mm. Estimated precipitation temperature 88"C (see Schultz et a/., 1989). (E) Calcite pore-filling
cement from basin margin. Well Fruitvale No. I , 1 3 61.3 m ( 4466 ft). Estimated water composition is o180sMow = - I 0,
suggesting a meteoric pore water (see Hayes & Boles, 1 993). White bar is 0.5 mm. (A, B, C, Di) Plane-polarized light;
(Dii, E) crossed nicols.
272
Although some samples on a thin-section scale
undoubtedly represent multiple generations of ce­
ment, many cements appear to be a single genera­
tion, based on spot analysis with the electron
microprobe, which indicates less than a few percent
variation in trace element content (Boles & Ram­
seyer, 1987). Cathodoluminescence is not particu­
larly useful for recognizing zonation within crystals
or differences between crystals, presumably owing
to the relatively high Fe content of most crystals,
which causes very low luminescence.
Dolomi te composition
Evidence supporting an early precipitation origin
of dolomite is 87 Sr/86 Sr ratios similar to sea water.
Dolomites in Stevens sandstones at North Coles
87 86
Levee have Sr/ Sr ratios of about 0.70860-
0.70865. This is slightly less than the 0. 7090 value
expected for uppermost Miocene-age sea water
during deposition of the Stevens at 5-6 Ma (Hess et
a! ., 1986). Subsequent calcite cements in the basin
have distinctly lower Sr isotopic ratios
than the dolomites (Schultz et a! ., 1989).
The oxygen and carbon isotopic compositions of
early dolomite in the Central basin have 018QPDB
13
between about 0 and -3.5o/oo and 8 Cp08 between
+4 and + 1Oo/oo (Fig. 5). Virtually all samples have
positive carbon values, and in this respect they are
distinctive from almost all calcite cements in the
basin. The positive carbon value indicates that they
formed in a zone of methanogenesis, where meth­
ane preferentially fractionates light carbon, leaving
10 r------,
..
•
8 •
iii" 6
c • •
e,.
0 •
• •
..,
4 •
co •
2
0 dolomite cements in sandstones of
·4 ·3
0 18
JR. Boles
the pore water rich in 13C (Curtis, 1978). The zone
of methanogenesis is believed to be at a depth
ranging from tens of metres to about 1 km of burial
(Irwin et a!., 1977), consistent with the other evi­
dence indicating an early origin, such as high
pore-filling cement volumes and Sr isotopic values
near sea water.
Based on the oxygen isotopic fractionation equa­
tion of Fritz & Smith ( 1970) for water and dolo­
mite, the oxygen values indicate crystallization at
about 30-4Q"C from a water with 0180sMOW of
Oo/oo, or at an even higher temperature if the pore
water evolved to a more positive oxygen isotopic
composition (Boles & Ramseyer, 1987). The value
indicates that the minimum depth at which the
dolomites would have formed is 500 m below the
sediment-water interface (assuming a 15 ·c bottom
temperature and 30-36.C/km geothermal gradi­
ent). The bottom water temperature could have
been as low as 5 • C, indicating crystallization at
burial depths of up to 1 km. In the light of other
evidence it is surprising that isotopic analysis does
( < 0. 7083) not indicate crystallization nearer to the sediment­
water interface. There is no evidence to suggest that
the initial waters in the central basin may have had
a meteoric component and hence were isotopically
lighter than sea water.
Trace element substitution in the early marine
dolomite from the central basin is relatively Fe rich,
with slight excess calcium substitution from the
ideal formula Ca/(Mg + Fe+ Mn) = 1, and very little
Mn substitution (Boles & Ramseyer, 1987). Excess
Ca is about 1-4 mol%. Typical compositions have
•
•
.
I I
N. COL£8 LEVEE '
• SAN EMIDIO NOSE
.oPALOMA Fig. 5. Oxygen and carbon isotopic
composition of early-formed
·2 ·1 0 the central San Joaquin basin.
Data from Boles & Ramseyer
Q (PDB) ( 1 987) and Table 2 .
 Carbonate cementation in Tertiary sandstones
5-10mol% Fe (Boles & Ramseyer, 1987). Late­
forming dolomites grown between the cleavages of
crushed biotite grains (Boles & Johnson, 1983;
Boles & Ramseyer, 1987) are slightly more Ca and
Fe rich than the dolomite cements. Compared with
dolomites occurring on the flanks of the basin (Lee
& Boles, in press), those in the central basin have
higher Fe contents. High Fe content is also a
characteristic of many of the calcite cements in the
central basin (see below).
Calcite composition
Basin centre calcite cements typically have 8180p08
between about -4%o and -1 2%o and 813Cp08 be­
tween O%o and -1 0%o (Boles & Ramseyer, 1987;
Schultz et a! ., 1989) (Table 2). Based on pre-cement
pore space volume and these oxygen isotopic val­
ues, calcite cements are inferred to have formed
over a broad depth range, including at or near
present-day maximum burial depth (Boles & Ram­
seyer, 1987; Schultz et a! ., 1989). Intergranular
cement volumes in fully cemented sandstones are
between 32 and 1 5% (Fig. 2 in Schultz et a! ., 1989),
which, if compared with San Joaquin basin com­
paction curves (Ziegler & Spotts, 1976), suggests
cementation at burial depths ranging between less
than I km to over 3 km. Calcite cementation tem­
peratures at North Coles Levee are calculated to
range from about 40 to 9o·c , based on oxygen
isotopic analysis (Schultz et a! ., 1989). Based on a
time-temperature burial curve for North Coles
Levee (Fig. 3), precipitation of calcite at 9o·c
implies that cementation occurred as recently as a
few million years ago, close to the present in sit u
reservoir temperature at North Coles Levee of
1o5·c .
In addition to North Coles Levee, several other
reservoirs in the basin also show evidence of cemen­
tation near present-day maximum burial depths. In
the Yowlumne and Rio Viejo fields, calcites have
particularly light oxygen isotopic compositions
compared with other central basin cements (Table
2). These calcites appear to have formed at deep
burial levels, judging from the low intergranular
cement volume in compacted sandstone (less than
20%). In sit u temperatures in these two reservoirs
are 125·c and l4o·c and present pore waters have
8180sMow of O%o and +i. lo/oo, respectively (un­
published data). Assuming that the calcites with the
lightest oxygen precipitated from these waters, the
calcite is calculated to have formed at a maximum
273
temperature of 95•c at Yowlumne and 1 40"C at
Rio Viejo. Thus, these central basin cements
formed relatively late in the burial history and, in
the case of the Rio Viejo, could be forming today in
equilibrium with present reservoir conditions.
Sr isotope studies at North Coles Levee (Schultz
et a! ., 1989) indicate that a major source of Ca and
Sr in calcite cements is plagioclase feldspar. Sr
isotopic values of the calcite cements are consis­
tently lower than that expected for Miocene sea
water, and show a systematic decrease with decreas­
ing oxygen isotopic composition (Fig. 5). Decreas­
ing oxygen values are interpreted to result from
precipitation at higher temperatures as the pore
water 8180sMow value increased from O%o to +4%o.
The Sr isotopic value of the present-day pore waters
87
( Sr/86 Sr = 0. 7072) is similar to that of the last
calcite cements to form (0.7072-0.7073), which
also have the lowest oxygen isotopes. In addition,
the Sr content of the cements systematically in­
creases with increasing burial temperature (Schultz
et a! ., 1989).
Based on observations at North Coles Levee,
Schultz et a! . (1989) proposed a model in which
Stevens sands receive Ca and Sr from plagioclase
alteration in underlying formations, either by albi­
tization or by dissolution of plagioclase. Both pla­
gioclase alteration studies (Boles & Ramseyer,
1988; Ramseyer et a! ., 1992) and pore water com­
positional studies (Feldman et a! ., 1993) have doc­
umented the importance of plagioclase alteration in
the basin. Sr isotopes indicate that during the
precipitation of calcite cements at intermediate
burial depths, plagioclase feldspar was a partial
source of Ca to the pore water, yet no evidence of
plagioclase alteration is observed within these ce­
mented sandstones at North Coles Levee. In the late
history of central basin sands, calcite cementation
and alteration of plagioclase is occvrring locally in a
more closed system within the reservoir. Evidence
for this is that late calcites have Sr isotopic values
approaching that of present-day pore water, plagio­
clase dissolution has occurred within these reservoir
sands, and some calcite infills partially dissolved
plagioclase. Thus, late-stage dissolution of plagio­
clase at North Coles Levee is contributing calcium
directly to late-forming calcite cements, and this is
also reflected in a marked increase in their Sr
content (Schultz et a! ., 1989).
Data from South Coles Levee and Canal fields
(Table 2) confirm the positive correlation between
oxygen and Sr isotopes, as noted at North Coles
274
Levee (Fig. 6). However, the South Coles Levee
data show a lower slope than the North Coles Levee
data. The scatter in Fig. 6 may be due to real
variations among the bulk sample analysis, or to
unrecognized multiple generations of cement. Al­
though samples were selected for their uniform
mineralogy and thin-section
amounts of early dolomite, for example, would raise
the Sr and oxygen isotopic ratio, causing scatter
(Schultz et a! ., 1989). The scatter may, however, be
real, and reflect local variations in the degree of
plagioclase alteration. Schultz et a! . (1989) have
shown that less than a few volume percent plagio­
clase alteration is required to markedly lower the Sr
ratio of the pore water from a Miocene marine
value.
The new data from South Coles Levee are inter­
esting in that a number of samples with o 1 80p06
between -5%o and -9%o have lower Sr ratios than
the equivalent interval at North
(Fig. 6). This might suggest that upwelling pore
waters at South Coles Levee have experienced more
extensive water-rock (feldspar) interaction and/or
arrived earlier than those at North Coles Levee. The
South Coles Levee reservoir is 300 m deeper and
down dip from North Coles Levee, and it is proba­
ble that fluids transporting mass up from deeper
levels in the basin would have arrived at South
Coles Levee before reaching North Coles Levee. In
any case, the data suggest that at any given time
these two closely spaced reservoirs did not have
identical pore fluid compositions.
0.7090 r------,
0.7085
...
en
CD
• . ..
� 0.7080
t-
co • •• •
• •
0.7075
• •
• • •
• calculated composition of calcite
• •
0.7070
IIIIIllllmi •
· 12 · 10 -8
J.R. Boles
Schultz et a! . (1989)report that Sr/Ca ratios are a
factor of 3 higher in late calcite cements with low
87 Sr/86 Sr ratios t·han in early cements with high
87 Sr/86Sr ratios. Data from South Coles Levee show
generally lower Sr contents than do the data from
North Coles Levee, but also show a similar pattern
appearance, small 87
of higher Sr contents associated with lower Sr/
86
Sr values (Fig. 7 ). The Sr content of South Coles
Levee calcites is surprisingly low considering the
relatively low Sr isotopic ratio of many of these
samples, indicating extensive plagioclase-water in­
teraction. One possibility is that another phase, not
present at North Coles Levee, has competed for Sr
at South Coles Levee. Both reservoirs have Sr
contents of about 53-63 Mg/1 in their current pore
waters (Fisher & Boles, 1990), suggesting that
present conditions do not reflect the differences in
the cement compositions.
The last calcite cements to form at North Coles
Coles Levee Levee precipitated during the dissolution of plagio­
clase, and were presumably formed during emplace­
ment of the hydrocarbons (Boles, 1987 , 1992).
Although systematic patterns of carbonate precipi­
tation have not been recognized within individual
cement zones, it is interesting to note that on a res­
ervoir scale, late calcite cements are restricted to the
zone below the gas cap (upper 50 m of the reservoir
sand in well NCL 4 88- 29), high on the North Coles
Levee structure. In fact, re-examination of the Table
1 data of Schultz et a!. (1989) indicates that late­
stage, relatively high-temperature carbonate ce­
ments are found in the lower parts of the reservoir,
Fig. 6. Isotopic composition of
calcite cements from North Coles
Levee, South Coles Levee and
Canal fields. North Coles Levee
X X
X X dolomite cements shown for
comparison. N.orth Coles Levee
.. N. COLES LEVEE
DOLCMITE data from Schultz et al. ( 1 989);
• South Coles Levee and Canal data
• • ..
• from Table 2. Increasingly negative
• S 1 80p08 and lower Sr ratios are
.
.•
correlated with higher
temperatures of crystallization.
•
• Strontium ratio decrease is
attributed to Sr from plagioclase
alteration. Shaded box shows
in equilibrium with present pore
water at Coles Levee fields based
on fluid temperature and
-6 -4 -2
composition in Fisher & Boles
5 18 ( 1 990) and Sr isotopic data in
Q(PDB) Feldman et al. ( 1 993).
 Carbonate cementation in Tertiary sandstones 275

• ••
�------,

i •N�S�E�
2500

••
•
2000
. s. �S �EE ·

• • • •• • • •
E' 1 500
Fig. 7. Sr content (ppm) and c.
8 7 Sr/86 Sr ratio in calcite cements .!?;
...

• •
from Coles Levee reservoirs. North en 1 00 0
•

Coles Levee data are from Schultz

• • . ,
•
•
et a/. ( 1 989) and South Coles •
Levee data are unpublished 500
analyses of J. Schultz (Table 2). • •
•
•
Data show general increase of Sr
content with decreasing Sr isotopic 0
ratio, interpreted to be a result of 0. 7070 0.7072 0.7074 0 . 7076 0.7078 0.7080 0.7082 0.7084
plagioclase dissolution with
increasing burial depth (see
87/86
Schultz et a/., 1 989). Sr

rather than throughout the section. It has also been
noted that the gas cap at North Coles Levee is
essentially void of plagioclase dissolution, even
though water saturation (45 % ) is similar to that in
the underlying strata (Boles, 1 992). Calcite with Sr
ratios similar to present-day pore water, and thus
forming from plagioclase sources, appears to be
restricted in distribution to the lower part of the
reservoir. During the late diagenetic history, the
reservoir was apparently a relatively closed system
in which neither diffusion nor fluid circulation was
able to move calcium to the top of the structure for
carbonate cementation. Calcium, however, was able
to move from the dissolving plagioclase and precip­
itate in adjacent pore spaces.
Carbon isotopic compositions of the central basin
calcite cements are interpreted to be a mixture of
biogenic (shell tests) and thermogenic carbon reser­
voirs (Wood & Boles, 1 991 ). Biogenic carbonate
from adjacent marine shales WOUld have 0 1 3CPDB
near zero, and thermally derived C02 would result in
an HC0 - in the pore water with a 0 1 3C P D B value of
3
-18o/oo to - 23o/oo (Wood & Boles, 1991 ). A frequency
histogram of carbon isotopes for central basin calcite
cements shows the dominance of values around Oo/oo
to -4o/oo , suggesting that biogenic sources (shell tests)
could be the major sources of carbon in these calcites
(Fig. 8). The lightest o 1 3C values in the cements are
around -!Oo/oo to - 1 2o/oo , indicating a subequal con­
tribution of biogenic and thermogenic sources.

25

Fig. 8. Frequency histogram > 20

(n 83) of the distribution of
= 0
carbon isotopes (PDB) in calcite z
w
cements from the central basin. :::1 15
Data from Boles & Ramseyer 0
w
( 1 987), Wood & Boles ( 1 99 1 ), a:
Table 2 (this chapter) and u.. 10

additional unpublished data of the

author. Data indicate that carbon
sources near zero are the most
important in moderate to deep
burial calcites of the San Joaquin
basin. This carbon may in part be · 10 ·8 -4 ·2 0 6 8

of non-biogenic origin (see text for

discussion). I) 13 c (PDB)
276
At North Coles Levee, carbon isotope values
fluctuate between zero and - 1 2o/oo in cements esti­
mated to have formed at intermediate burial depths
between 4 and 5 Ma (Wood & Boles, 1 99 1 ). These
fluctuations are interpreted as being the result of
pulses of flu id from deep basin levels mixing with
local fluids (Wood & Boles, 1 9 9 1 ). Burial history
reconstruction for the basin indicates that shale
source rocks, such as the Eocene Kreyenhagen,
would have been at an appropriate depth to be a
C02 source (Fig. 3).
Trace amounts of Fe, Mg and Mn are present in
virtually all calcite cements that have been mea­
sured in the central basin. Samples are character­
ized by relatively high Fe and Mg, but relatively low
Mn contents. Compositions have a characteristic
(Fe + Mn)/Mg ratio of about 3/1 (Fig. 9). Late ce­
ments tend to be more enriched in Ca (Boles &
Ramseyer, 1 98 7), which may be due to late precip­
itation of calcite during the dissolution of Ca­
plagioclase at burial temperatures greater than
about 70-80 ' C. Fe and Mn contributions to the
pore water from this source would presumably be
negligible. Marine pore waters at both North and
South Coles Levees have up to 1 800 mg/1 Ca (see
analysis in Fisher & Boles, 1 990), due to Ca enrich­
ment from feldspar dissolution (these values are
considerably higher than originally reported by
Boles & Ramseyer ( 1 987)).
The source of the Fe, Mg and Mn in these
carbonates has not been identified. Diagenesis of
Fe+Mn
CENTRAL .BASIN
•
0 BASIN MARGIN
Ca Mg
Fig. 9. Mg-Fe-Mn content of San Joaquin calcite
cements. Data from Boles & Ramseyer ( 1 987), Fischer
& Surdam (1988) and Hayes & Boles (1993). Note that
central basin cements are characterized by relatively
high total Fe + Mg + Mn with an (Fe + Mn)/Mg ratio of
3/ I compared with basin margin cements.
J.R. Boles
smectite to illite does not appear to be prevalent in
the young strata of the San Joaquin basin (Ram­
seyer & Boles, 1 986), and thus cannot be considered
as a likely source of the components (cf. Gulf Coast
Tertiary; Boles & Franks, 1 9 79). Accessory miner­
als, such as hornblende, are possible sources but
dissolution textures or remnant grains are not ob­
served. It is interesting to note that hornblende is
abundant in the Sierran plutons and is almost
completely absent in the subsurface. It is generally
not seen even in early cemented sediment, where
unstable minerals are often preserved. It was prob­
ably removed during weathering, even though pris­
tine biotite is often seen in these sandstones.
Nevertheless, hornblende may be an unrecognized
subsurface source of trace metals.
BASIN MARGIN CEMENTS
Cement petrology
Dolomite-cemented zones are on a scale of less than
30 em thick, and in thin section dolomite occurs as
euhedral to subeuhedral pore-filling cement. Dolo­
mite also occurs in partially cemented sandstones,
forming small euhedral rhombohedral crystals at­
tached to detrital grains. A relatively high-volume
dolomite cement in a wildcat well, Ohio KCL G- 1
at 1 9 1 1 m (6270 ft), has interlocking rhomb-shaped
crystals up to 1 50 �m across, each exhibiting more
than 1 0 zones of orange to brown luminescence.
This striking multiple zonation has not been recog­
nized in other carbonate samples from the basin.
Calcite occurs as cement zones 1 0- 1 50 em thick.
Pore-filling calcite crystals are anhedral blocky spar
crystals (0. 1 -0.5 mm). In general, the petrographic
appearance of basin margin calcite crystals is no
different from calcite cements oc;curring in the basin
centre. Some samples exhibit moderate to strong
luminescence, whereas others are uniformly non­
luminescent.
High cement volumes characterize many calcite
cements from the basin margin sediments, and
these presumably formed at shallow burial depths
before significant compaction (Fig. 4E). Intergranu­
lar cement volumes are of the order of 30% to more
than 40% in these samples. However, most cements
from the basin margins have volumes indicating
that they formed after some compaction and, like
the central basin, cementation is a process that
extends through much of the burial history (a!-
 Carbonate cementation in Tertiary sandstones 277

though not usually in the same sample). Examples o 1 3CPDB +5.9%o) (Fischer & Surdam, 1 988) and
=

of relatively low-volume cements forming in marine from the Mount Poso field (o1 80p08 -2.29%o, =

or mixed marine-meteoric waters are recognized o 1 3Cp08 -4.27%o) (Hayes & Boles, 1 993), and
=

from Mount Poso and Kern River fields (Hayes & from the west side in the North Be1ridge field, where
Boles, 1 993), North Belridge field (Taylor & Soule, high-volume early marine cements have o1 80p08 =

1 993) and Kettleman North Dome (Lee & Boles, -2.69%o and o 1 3CPDB - 1 9. 56%o (Taylor & Soule, =

1 996). 1 99 3). Overall, however, calcites with o1 80p08 near

Cement composi tion
Early marine cements are recognized in a few cases
where high-volume cements have o1 80p08 values
near zero, indicating low-temperature crystallization
near the sediment-water interface in marine pore
water (Fig. 1 OA ). For example, on the east side of
the basin early marine calcite cements are reported
for the Mountain View field (o1 80p08 +0.3%o,
= dolomite. Based on their occurrence in shallow
zero, irrespective of their volume, are relatively rare
at the basin margins, indicating that most of these
cements did not form in equilibrium with sea water.
A meteoric origin is suggested for some of basin
margin cements that have relatively low oxygen
isotopic compositions yet could not have been
exposed to high burial temperatures. These cements
have relatively high interstitial volumes (> 30%)
and are usually calcite, but in some cases are

·16 ·14 ·12 ·10 ·8 ·6 ·4 ·2 0 2

30

0 18 0 (PDB)
•
20
•
• •
•

10
•
• •
• • • iii'
• c
• • • • • !!;.
• 0
0
• •
• -
"'

• • • •
•
00
..>• .:: • -10
• ..
• •
•
• -20

•
-30
•

-40

(A)
·16 ·14 ·12 ·10 ·8 ·6 -4 -2 0 2

30

0 18 0 (PDB)
20

10

•
• • •• iii'
Fig. 10. (A) Oxygen and carbon
, • ::.-.'!.·
.
• c
!!;.
:r.. ' "' . .
••
• • 0
• • •
stable isotopic composition of •
•
• .gl 0
calcite cements from the San • .. • . . .
• • • -1 0
Joaquin basin margins. Data • • • 00
sources from Table I . (B) Oxygen
and carbon stable isotopic -20
composition of calcite cements
from the San Joaquin basin centre.
-30
Data sources from Table I . Note
the restricted range of carbon
isotopes compared with basin -40

margin cell!ents (cf. with (A)). (B)

27 8 J.R. Boles
sediment, they must have formed at low tempera­
tures from isotopically light meteoric water
(Fig. 4E). Examples of calcite cements are described
for the Edison field (Fischer & Surdam, 1988) and
for the Mount Poso and Fruitvale fields (Hayes &
Boles, 1993). Examples of early dolomites with a
clear meteoric signature come from the west side of
the basin in the North Belridge (dolomite volume
26% ) (Taylor & Soule, 1993) and Kettleman North
Dome fields (dolomite volume 30% ) (Lee & Boles,
1996). In all cases cited here, the cements have
oxygen isotopic compositions recording waters sig­
nificantly lighter than sea water. For example,
o1 80p08 values of -9%o and -12%o are estimated
for the parent water of the calcite cements in the
Fruitvale and Round Mountain fields, respectively
(Hayes & Boles, 1993), and these values are similar
to modern meteoric water in the basin. Burial
temperatures have never been sufficiently high to
account for the relatively low o 1 80pos values found
in many of these cements. In many cases where
sandstones have been buried less than I km it is not
possible to determine the depth at which cementa­
tion occurred, nor to constrain the cementation
temperature, other than limits placed by the
present-day burial depth.
The early marine cements on both basin margins
have relatively wide variations in carbon isotopic
composition compared with the basin centre (Fig.
I 0). Carbon isotopes in the basin centre have
0 1 3 Cp os values between +5o/oo and -10o/oo , whereas
basin margin cements commonly have values be­
tween + 1O %o and -20%o; values as low as -30%o and
as high as + 20%o also occur. Presumably, the overall
shallower depositional conditions at the basin
margin and/or greater fluid mobility compared with
the basin centre have resulted in different reactions,
producing dissolved carbon within tens of metres
of the sediment-water interface. Negative 0 1 3CPDB
values presumably represent bacterial oxidation
or sulphate reduction, and positive values are from
bacterial fermentation (Irwin et a/ ., 197 7 ).
The carbon isotopic values of the early meteoric
calcite cements are negative, generally much more
so than early marine calcite cement. o 13Cpos of
-11.33%o , - 14 .91%o , and - 32. 83%o are reported for
the Fruitvale and Mount Poso fields, with one
exceptionally positive value ( + 21.89%o ) reported
for the Round Mountain field (Hayes & Boles,
1993). Fischer & Surdam ( 1988) also report
negative carbon values, generally associated with
cements of meteoric or mixed meteoric-marine
origin ( -9.3%o , -12.5 %o , - 28.0%o ). These generally
strongly negative values are consistent with a signif­
icant carbon contribution from shallow processes of
either bacterial oxidation or sulphate reduction,
both of which result in carbon values as low as
- 25 %o (Irwin et a! ., 197 7 ).
The calcite cements that formed at moderate to
deep burial levels show a more restricted range of
carbon isotopic values than the early cements.
0 1 3C
PDB
for cements forming during moderate
burial on the eastern side of the basin varies
between -15 %o and + 16%o (Hayes & Boles, 1993),
and values reported for the western side vary from
-9%o to -12%o at North Belridge (Taylor & Soule,
1993)to -4 %o to -16%o at Kettleman North Dome.
Carbon values from late cements of the basin
margin include markedly positive values that do not
occur in late calcite cements from the central basin.
Presumably, bacterial fermentation producing iso­
topically heavy dissolved carbon during methane
production was shut off during deep burial in the
central basin, and carbon sources there are derived
from a combination of thermogenic and lighter
sources near zero (PDB) carbon.
There are only a few published 87Sr/86Sr analyses
of basin margin cements. Early cements of calcite
and dolomite at North Belridge field have 87Sr/86Sr
ratios of about 0.7 080, similar to that expected for
an Oligocene marine depositional water (Taylor &
Soule, 1993). Later calcite cements have similar to
slightly higher values, suggesting the influence of
clay or feldspar (K-feldspar?) diagenesis on the pore
water. This contrasts with carbonate cements else­
where in the basin, which have Sr isotopic ratios
equal to or less than that expected for their respec­
tive marine depositional ages (Schultz et a! ., 1989)
(Table 2).
On the east side of the basin, Sr isotopic ratios
from calcite-cemented marine �trata in the Mount
Poso field and from the wildcat well Ohio KCL G- 1
are low relative to that expected for unaltered sea
water (Table 2). The cements from Mount Poso
were buried only a few kilometres, and cement
volumes suggest that they formed during the inter­
mediate to late burial history. Sr isotopic values are
especially low in the cements of the Mount Poso
field, with 87Sr/86Sr ratios being as low as 0.7 063 in
Oligocene marine sediment. These ratios are com­
parable to those measured in Sierran plagioclase of
0.7 080-0.7 05 0 (mean 0.7 068) (Schultz et a!.,
1989), suggesting that plagioclase has made a signif­
icant contribution to the pore water of the east
 Carbonate cementation in Tertiary sandstones
basin flank. Hayes & Boles ( 1992) report up to 3%
plagioclase dissolution in the basin margin sand­
stones.
One notable difference between samples from the
eastern basin margin and the central basin is that
although both areas show low 87Sr/86Sr ratios rela­
tive to the values expected from the marine pore
water in which they were deposited, the Sr content
of cements from the basin margin is generally less
than a few hundred ppm, whereas the Sr content of
the central basin is from 1000 to 2000 ppm. This
difference may reflect a higher temperature of crys­
tallization or significantly greater water-rock inter­
action in a more closed system in the central basin
than in the basin margin. Modern pore waters from
the central basin (e.g. North and South Coles
Levees) have Sr isotopic ratios of 0.7070 and Sr
contents of 5 0- 60 mg/1, whereas modern waters
from Mount Poso have ratios of 0.7061- 0.7068
and 0.06- 0.1 mg/1 Sr (Feldman et a!. , 1993). Thus,
there is a general correlation between the Sr content
of the calcite cements in each area and the modern
pore fluids.
The meteoric and marine cements of the basin
margin are relatively pure, containing less than
5 mol% total Mg + Mn + Fe (Boles & Ramseyer,
198 7; Fischer & Surdam, 1988; Hayes & Boles,
1993; Lee & Boles, 1996). In contrast, cements in
the central basin, particularly those forming at
moderate burial depths and temperatures (about
4 0- 7o·q , commonly have 5 -10 mol% of
Mg + Mn + Fe (Fig. 9).
There are some differences to be noted in the
proportion of trace elements in the basin margin
cements compared with those of the central basin.
High proportions of Mn and/or Fe are found only in
meteoric cements (Boles & Ramseyer, 198 7; Fis­
cher & Surdam, 198 8 ). Further studies have shown
that meteoric cements may also have Mg as the
dominant trace element (Hayes & Boles, 1993).
This latter work also showed that the trace element
distribution of basin margin calcites appears to vary
on a reservoir scale. Thus, the proportion of Mg­
Mn-Fe is distinctly different in calcite cements of
the Vedder sandstone from Mount Poso and Round
Mountain fields, even though they are interpreted
to have formed at a similar time (i.e. burial temper­
ature). As concluded by Hayes & Boles ( 1993), local
reactions in the reservoir are occurring at a faster
rate than potential mixing processes between these
reservoirs.
279
DISCUSSION
Early marine Tertiary dolomite
The central basin dolomites have similar isotopic
compositions to the early dolomite concretions and
cemented layers from the upper Miocene Monterey
Formation exposed along coastal California (Pis­
ciotto, 198 1; Isaacs, 1984; Hennessy & Knauth,
1 98 5 ; Burns & Baker, 198 7; Malone et a! ., 1 994 ).
The Monterey Formation differs from the coarse­
grained Stevens sandstones in that it contains dom­
inantly fine-grained rocks, including organic-rich
shales, altered diatomite, dolomites and minor phos­
phates. The Monterey dolomites are believed to have
formed near the marine sediment-water interface,
but recent studies suggest that they have recrystal­
lized during burial (Malone et a!. , 1994 ). Evidence
cited for recrystallization includes a covarying trend
of lighter oxygen and carbon isotopic compositions,
and an increase in ordering and stoichiometric com­
position in areas with higher burial temperatures. It
is possible that the San Joaquin dolomites under­
went recrystallization during burial, as they show a
similar trend to Monterey dolomites of covarying
oxygen and carbon isotopic composition (Fig. 5 ).
The 4 %o spread of the Stevens dolomite oxygen val­
ues (Fig. 5 ) would represent an approximate differ­
ence of about 22·c in precipitation temperature at
low temperature and at constant isotopic water com­
position. Thus recrystallization, if it has occurred,
would possibly be over a 1 km interval of burial. If
Stevens dolomites are recrystallized they have re­
tained a distinct carbon and Sr isotopic signature
relative to calcites that are calculated to be forming at
comparable depths.
Another similarity between Monterey and Stevens
dolomites is their excess Ca, typical of Tertiary
dolomite compositions. Apparently these non­
stoichiometric compositions are �etained in San
Joaquin dolomites at temperatures at least as high
as 10o·c (their present burial condition). Stevens
dolomites are generally much richer in Fe than
Monterey dolomites, possibly owing to the alter­
ation of Fe-bearing minerals in the Stevens sand­
stones.
Many of the Monterey dolomites have carbon
isotopic values more positive (> 10%o PDB) than
Stevens central basin dolomites. Carbon isotopes in
Monterey dolomites have been related to sedimen­
tation rates (Pi sciotto & Mahoney, 198 1). When
sedimentation rates are low dolomites have nega-
2 80 J.R. Boles
tive carbon values, owing to precipitation in the
sulphate-reducing zone during extensive reaction of
organic matter near the marine sediment-water
interface. When sedimentation rates are high, rapid
burial allows organic matter to be buried into the
zone of methanogenesis, resulting in dolomites with
positive carbon isotopic values. Sedimentation
rates for Stevens turbidite sandstones are undoubt­
edly much higher than for much of the Monterey
Formation, but the initial amount of organic carbon
in the sands at the time of deposition was probably
much less than in the fine-grained Monterey sedi­
ments. Thus the positive carbon values of Stevens
dolomites are attributed to precipitation in the zone
of methanogenesis within a sediment having mod­
erate amounts of organic matter.
Early marine calcite
Only a few early calcite cements from the basin
margin have a 8180p08 of zero (Fig. l OA), and
calcite cements from the basin centre also rarely
show values near zero (Fig. 1OB). For example,
Fischer & Surdam ( 1988) noted that calcite crystal­
lized at shallow depths in meteoric water and at
deeper burial levels in marine water in some east
basin margin reservoirs. Apparently calcite cements
rarely precipitate from sea water near the sediment­
water interface in the San Joaquin basin. This is
somewhat surprising, considering the abundance of
marine sediment deposited in a range of water
depths within the basin. The marine shales within
the central basin contain abundant forams (Lagoie,
1987) and these, as well as microfossils which occur
in shallow marine facies on the basin margin
(Loomis, 1988), are potential sources of cement
material. Early calcite cementation was not com­
mon in marine pore water on a basin scale in the
San Joaquin basin, and it appears that the bulk of
the cementation in marine pore waters occurred
either after significant burial or, in some cases, at
shallow burial depths during mixing with meteoric
water.
Evi d ence for magnitude of mass transfer
Dolomitization in the San Joaquin basin, as indi­
cated by precipitation at or near the sediment­
water interface, apparently did not require long
transport distances for components, which is con­
sistent with a very shallow burial process. However,
repeated calcite precipitation events during burial
in which cements have variable chemistry suggests a
relatively open system, in which considerable mass
transport has occurred. A relatively open system
during burial is suggested by the movement of
reaction products from deeper levels to shallower
levels, where they are incorporated into carbonate
cements. For example, at North and South Coles
Levees Sr isotopic studies indicate that calcium is
derived from the alteration of Sierran plagioclase
(Schultz et a!., 1989). This calcium is transferred
upward for up to several kilometres, perhaps by
compaction-driven cross-formational flow. Carbon
isotope studies also support the upward mobility of
fluids, as is suggested by pulses of isotopically light
thermogenic C02 incorporated in relatively shallow
cements (Wood & Boles, 1991). Although upward
flow is suggested by cement compositions, pore
-fluids within individual reservoirs developed dis­
tinctive compositions indicating that the flow rate
was not sufficiently fast to homogenize the differ­
ences between reservoirs.
Evidence for repeated calcite cementation events
during intermediate to deep burial, based largely on
isotopic studies of cements, suggests multiple peri­
ods of pore fluid flux rather than a single period of
calcite precipitation resulting from thermodynamic
equilibrium· bet�een sediment and pore fluid dur­
ing burial. Episodic cementation may have been
due to repeated pulses of C02-charged water from
deeper levels in the basin (Wood & Boles, 1991).
Late calcite cements are interpreted to have
formed under relatively closed conditions, with
cement components being derived from within the
scale of the reservoir. In the central basin many of
these calcites precipitated at near present-day reser­
voir conditions (> 9o·q and have Sr isotopic
values consistent with current reservoir pore water
values.
Carbon source in late calcite cement
One would expect thermogenic carbon sources to
become increasingly important as the sands are
buried. However, the late carbonate cements at
North Coles Levee (after about 4 Ma) have carbon
isotopic values in a very restricted range (- 2%o to
-4 %o ), as if they were buffered by some process
taking place in the reservoir (Fig. 9 in Wood &
Boles, 1991). Similarly, late calcite cements forming
in the Rio Viejo, Yowlumne and San Emidio Nose
fields (Table 2) also have relatively heavy carbon
values, between -5 and 0. The timing of the 'buff-
 Carbonate cementation in Tertiary sandstones
ered' carbon value may reflect a process related to
the presence of oil in the reservoir. In the case of
North Coles Levee, oil emplacement partly coin­
cides with the timing of late carbonate cementation
(Boles, 1 98 7), and the presence of this oil is proba­
bly the source of organic acids in the present pore
waters (Boles, 1 992). The carboxyl carbon has a
variable o 1 3Cp08 (-0.8 to 1 2. 8%o) (S. Franks, per­
sonal communication), and as yet unknown reac­
tions involving these acids may influence the o 1 3C
of the precipitating carbonate. The point is that
carbon isotope values near zero may be due to other
processes in addition to or iri lieu of dissolution of
biogenic carbonate.
Sr isotopes provide further evidence that carbon
isotope values near zero are not simply due to
dissolving marine shell tests. Sr isotopes clearly
show that the late cements are significantly different
from the Sr ratios expected for Tertiary sea water.
The source of Ca and Sr is mostly from plagioclase.
Marine shell tests such as forams have a molar
Sr/Ca ratio of about 0.002 (Milliman, 1 974),
whereas Sierran plagioclase has a molar Sr/Ca ratio
nearly an order of magnitude higher (0.0 1 7-0.0 1 8)
(Schultz et a!., 1 989). Thus a plagioclase source can
easily dominate the Sr isotopic signature in a pore
water. However, for every mole of carbon released
to the pore water from the dissolution of a marine
shell, a corresponding mole of Ca would be re­
leased. It is difficult to imagine how such a dissolu­
tion process would result in a carbonate cement
with a nearly zero carbon value yet having such a
distinct Sr signature from plagioclase.
Origin of cementation episodes
The question of how long subsurface cementation
continues can be constrained in the San Joaquin
basin as well as anywhere, owing to the young age of
the strata. However, in spite of the brief and simple
burial history of the San Joaquin basin, the rela­
tively well constrained fluid composition history
and the relatively well documented geochemistry of
its carbonate cements, we still cannot determine the
growth time of a given cement zone to any better
' 0 . :!han .perhaps � oo .ooo years or less.
., At North Coles Levee,. where basin cements have
been studied in the .greatest detail, the total time
available for cementation is less than 6 Myr. Within
the time span of about 1 millennium between 4 and
5 Myr ago, we can recognize about 1 3 cementation
periods (Wood & Boles, 1 99 1 ) (Fig. 7), implying
28 1
that, on average, individual cementation events
occur every 1 00 000 years. Nur & Booker ( 1 97 1 )
believe that the arrangement and decay of after­
shocks after large earthquakes is evidence of pore
fluid movement in sedimentary basins, and Sibson
( 1 989) has also suggested that episodic flow in
basins is a result of seismic events. Sieh ( 1 98 1 ) has
shown that rupture frequency along the San An­
dreas fault is of the order of hundreds of years for
ruptures producing large earthquakes. It is not
known to what extent major earthquakes affect deep
basin fluids at a considerable distance from a
rupture, but if the cementation described in this
paper is related to seismic events, it would have to
represent infrequent substantial earthquakes. Alter­
natively, the cementation may represent other cy­
clic phenomena occurring during burial of the
sediment, unrelated to earthquakes, such as the
breaking of basin seals during periodic pressure
build-up (Walder & Nur, 1 984; Hunt, 1 990).
ACKNOWLEDGEMENTS
Karl Ramseyer, University of Bern, contributed
much to the early studies of the basin cements.
Former graduate student Jan Schultz contributed
greatly to our understanding ofthe strontium isotope
systematics in the basin, including the Sr data pre­
sented in this paper. Mike Hayes, former graduate
student currently at EXXON research, contributed
much to the knowledge of eastern basin margin ce­
ments. Many groups contributed data and/or sam­
ples to our research, including ARCO, California
Core Repository, Chevron, EXXON, Marathon,
Mobil, Shell, Texaco and Unocal and Vintage Petro­
leum. Dr Stacey Zeck and Peter Eichhubl reviewed
earlier drafts and considerably improved the manu­
script. This research is currently supported by ACS­
PRF Grant No. 27046-AC, NSF Grant EAR-93-
045 8 1 and DOE grant DE-FG03096EB 1 9620.
REFERENCES
BJORKUM, P.A. & WALDERHAUG, 0. ( 1 990) Geometrical
arrangement of calcite cementation within shallow ma­
rine sandstones. Earth-Sci. Rev. , 29, 1 45- 1 6 1 .
BLOCH, R.B., VAN HUENE, R., HART, P.E. & WENTWORTH,
C.M. ( 1 993) Style and magnitude of tectonic shortening
normal to the San Andreas fault across Pyramid Hills
and Kettleman Hills, South Dome, California. Geol.
Soc. Am. Bull., 105, 464-478.
BLOCH, S. & FRANKS, S.G. ( 1 993) Preservation of shallow
282 J.R. Boles
plagioclase dissolution porosity during burial: implica­
tions for porosity prediction and aluminum mass bal­
ance. Bull. Am. Ass. Petrol. Geol. , 77, 1 488- 1 50 1 .
BOLES, J.R. ( 1 984) Secondary porosity reactions i n Stevens
sandstone, San Joaquin Valley, California. In: Clastic
Diagenesis (Eds McDonald, D.A. & Surdam, R.C.).
Mem. Am. Ass. Petrol. Geol., Tulsa, 37, 2 1 7-224.
BoLES, J.R. ( 1 987) Six million year diagenetic history,
North Coles Levee, San Joaquin Basin, California. In:
Diagenesis of Sedimentary Sequences (Ed. Marshall,
J.D.). Spec. Pub!. Geol. Soc. London, 36, 1 9 1 -200.
BOLES, J.R. ( 1 989) Chemical analyses as indicators of
sandstone petrology: an example from the North Coles
Levee reservoir, California. In: International Spectros­
copy and Geochemistry Symposium Transactions,
Schlumberger. Doll Research, Ridgefield, CT, Appendix
AA, pp. 1 -5 .
BOLES, J.R. ( 1 992) Evidence for oil-derived diagenesis. In:
Proceedings of the 7th International Symposium on
Water-Rock Interaction, Park City, Utah (Eds Kharaka,
Y.K & Maest, A.S.), pp. 3 1 1 -3 1 4. A.A. Balkema Pub­
lishers, Brookfield, UT.
BOLES, J.R., LANDIS, C.A. & DALE, P. ( 1 985) The Moeraki
boulders-anatomy of some septarian concretions. J.
sediment. Petrol., 55, 398-406.
BOLES, J.R. & FRANKS, S.G. ( 1 979) Clay diagenesis in
Wilcox sandstones of southwest Texas: implications of
smectite diagenesis on sandstone cementation. J. sedi­
ment. Petrol., 49, 5 5-70.
BOLES, J.R. & JoHNSON, K.S. ( 1 983) Influence of mica
surface charge on pore-water pH. Chern. Geol. , 43,
303-3 1 7.
BOLES, J.R. & RAMSEYER, K. ( 1 987) Diagenetic carbonate
in Miocene sandstone reservoir, San Joaquin Basin,
California. Bull. Am. Ass. Petrol. Geol. , 7 1 , 1 475- 1 487.
BOLES, J.R. & RAMSEYER, K. ( 1 988) Albitization of plagio­
clase and vitrinite reflectance as paleothermal indica­
tors, San Joaquin Basin. In: Studies ofthe Geology ofthe
San Joaquin Basin (Ed. Graham, S.A.). Soc. Econ.
Paleont. Miner. Pacific Sec. Pub!., 60, 1 29- 1 39.
BuRNS, S.J. & BAKER, P.A. ( 1 987) A geochemical study of
dolomite in the Monterey Formation, California. J.
sediment. Petrol., 57, 1 28- 1 3 9.
CALIFORNIA DIVISION OF OIL AND GAS ( 1 985) California Oil
and Gas fields-Central California. 3rd edn, Vol. I .
California Division of Oil and Gas, Sacramento, CA.
CALLAWAY, D.C. ( 1 97 1 ) Petroleum potential of the San
Joaquin basin, California. In: Future Petroleum Prov­
inces of the United States-Their Geology and Potential
(Ed. Cram, I.H.). Mem. Am. Ass. Petrol. Geol., Tulsa,
15, 239-25 3.
CAROTHERS, W.W. & KHARAKA, Y.K. ( 1 978) Aliphatic acid
anions in oil field waters-implications for origin of
natural gas. Bull. Am. Ass. Petrol. Geol. , 62, 244 1 -2453.
CoPLEN, T.B. et a/. ( 1 985) Oxygen and Hydrogen Stable
Isotope Measurements of Ground Waters of the Central­
West Side of the San Joaquin Valley, California. US
Geol. Surv., Open File Report, 85-490, 20 pp.
CuRTIS, C. D. ( 1 978) Possible links between sandstone
diagenesis and depth related geochemical reactions
occurring in enclosing mudstones. J. Geol. Soc. Land. ,
135, 107- 1 1 7 .
DuNWOODY, J.A. ( 1 986) Correlation Section No. 8 (Re-
vised) across the South San Joaquin Valley from the San
Andreas · Fault to the Sierra Nevada Foothills: San
Joaquin Valley Sub-committee on the Cenozoic. Am.
Ass. Petrol. Geol. Pacific Sec. Pub!.
FELDMAN, M.D. et a/. ( 1 993) Diagenetic mass transport in
the southern San Joaquin Basin, California: implica-·
tions from the strontium isotopic composition of mod­
em pore fluids. Chern. Geol. , l lO, 329-343.
FISCHER, K.S. ( 1 986) Diagenesis and mass transfer in · an
Upper Miocene source/reservoir system, southern San
Joaquin basin, California. MS thesis, University of
Wyoming, 237 pp.
FISCHER, K.S. & SURDAM, R.C. ( 1 988) Contrasting diage..
netic styles in a shelf turbidite sandstone sequence: the
Santa Margarita and Stevens sandstones, San Joaquin
Basin, California, USA. In: Studies ofthe Geology ofthe
San Joaquin Basin (Ed. Graham, S.A.). Soc. Econ.
Paleont. Miner. Pacific Sec. Pub!., 60, 233-247.
FISHER, J.B. & BOLES, J.R. ( 1 990) Water-rock interaction
in Tertiary sandstones, San Joaquin basin, California,
USA: diagenetic controls on water composition. Chern.
Geol. , 82, 8 3- 1 0 1 .
FRITZ, P. & SMITH, D.C. ( 1 970) The isotopic composition
of secondary dolomites. Geochim. Cosmochim. Acta,
34, 1 1 6 1 - 1 1 7 3.
GALLOWAY, W.E. ( 1 979) Diagenetic Control of Reservoir
Quality in Arc-Derived Sandstones: Implications for Pe··
troleum Exploration (Eds Scholle, P.A. & Schluger,
P.R.). Spec. Pub!. Soc. Econ. Paleont. Miner., Tulsa, 36,
2 5 1 -262.
GooDMAN, E.D. & MALIN, P.E. ( 1 992) Evolution of the
Southern San Joaquin basin and mid-Tertiary 'transi-­
tional' tectonics, central California. Tectonics, I l , 47 8--
498.
GRAHAM, S.A. ( 1 987) Tectonic controls on petroleum
occurrence in central California. In: Cenozoic Basin
Development of Coastal California. Rubey Vol. VI (Eds
Ingersoll, R.V. & Ernst, W.G.), pp. 47-63. Prentice
Hall, Englewood Cliffs, NJ.
HAYES, M.J. & BOLES, J.R. ( 1 992) Volumetric relations
between dissolved plagioclase and kaolinite in sand­
stones: implications for mass transfer in the San
Joaquin basin, California. In: Origin, Diagenesis, and
Petrophysics of Clay Minerals in Sandstones (Eds
Houseknocht, D.W. & Pittman, E.D.). Spec. Pub!. Soc.
Econ. Paleont. Miner., Tulsa, 47, 1 1 1 - 1 2 3.
HAYES, M.J. & BOLES, J.R. ( 1 993) Evidence for meteoric
recharge in the San Joaquin basin, California, provided
by isotope and trace element geochemistry. Mar. Petrol.
Geol. , 10, 1 3 5- 1 44 .
HENNESSY, J. & KNAUTH, L.P. ( 1 985) Isotopic variations i n
dolomite concretions from the Monterey Formation,
California. J. sediment. Petrol. , 55, 1 20- 1 30.
HESS, J. et a/. ( 1 986) Evolution of the ratio of a Sr-87 to
strontium-86 in seawater from Cretaceous to present..
Science, 231, 979-984.
HuNT, J.M. ( 1 990) Generation and migration of petro­
leum from abnormally pressured fluid compartments.
Bull. Am. Ass. Petrol. Geol. , 74, 1 - 1 2.
IRWIN, H., COLEMAN, M.L. & CURTIS, C. D. ( 1 977) Isotopic
evidence for source of diagenetic carbonates formed
during burial of organic-rich sediments. Nature, 269,
209-2 1 3 .
 Carbonate cementation in Tertiary sandstones
ISAACS, C. ( 1 984) Disseminated dolomite in the Monterey
Formation, Santa Maria and Santa Barbara areas,
California. In: Dolomites of the Monterey Formation
and Other Organic Rich Units (Eds Garrison, R.E. &
Zenger, D.H.). Soc. Econ. Paleont. Miner. Pacific Sec.
Publ., 4 1 , 1 1 9- 1 40.
KuESPERT, J. ( 1 985) Depositional environments and sedi­
mentary history of the Miocene Temblor Formation
and associated Oligo-Miocene units in the vicinity of
Kettleman North Dome, San Joaquin Valley, Califor­
nia. In: Geology of the Temblor Formations, Western
San Joaquin Basin, California (Ed. Graham, S.A.). Soc.
Econ. Paleont. Miner. Pacific Sec. Publ., 44, 53-67.
LAGOE, M.B. ( 1 9 87) The stratigraphic record of sea level
and climatic flucuations in an active margin basin: the
Stevens Sandstones, Coles Levee area, California.
Palaios, 2, 48-68.
LEE, Y.I. & BoLES, J.R. ( 1 996) Depositional controls on
carbonate cement in the San Joaquin basin. In: Silici­
clastic Diagenesis and Fluid Flow: Concepts and Appli­
cations (Eds Crossey, L., Loucks, B. & Totten, M.). Spec.
Publ. Soc. Econ. Paleont. Miner., Tulsa, 55, 1 3-22.
LooMIS, K.B. ( 1 988) Paleoenvironmental and paleocli­
matic interpretation of upper Miocene-Pliocene litho­
facies and macrobiota of the Etchegoin Group, Jacalitos
Canyon, San Joaquin Valley, California. In: Studies of
the Geology of the San Joaquin Basin (Ed. Graham,
S.A.). Soc. Econ. Paleont. Miner. Pacific Sec. Publ., 60,
303-3 1 8 .
MALONE, M.J., BARKER, P.A. & BuRNS, S.J. ( 1 994) Recrys­
tallization of dolomite: evidence from the Monterey
Formation (Miocene), California. Sedimentology, 4 1 ,
1 223- 1 239.
MAcGOWAN, D.B. & SuRDAM, R.C. ( 1 988) Difunctional
carboxylic acid anions in oil field waters. Org.
Geochem. , 1 2 , 245-259.
MERINO, E. ( 1 9 7 5) Diagenesis in Tertiary sandstones from
Kettleman North Dome, California-!. Diagenetic min­
eralogy. J. sediment. Petrol., 45, 320-336.
MEYERS, W.J. ( 1 97 4) Carbonate cement stratigraphy of the
Lake Valley Formation (Mississippian), Sacrement
Mountains, New Mexico. J. sediment. Petrol. , 44, 837-
86 1 .
MILLIMAN, J.D. ( 1 974) Marine Carbonates. Springer­
Verlag, New York, 375 pp.
MozLEY, P.S. ( 1 989) Relation between depositional envi­
ronment and the elemental composition of early diage­
netic siderite. Geology, 17, 704-706.
NuR, A. & BooKER, J.R. ( 1 97 1 ) Aftershocks caused by pore
fluid flow? Science, 175, 885-887.
OLSEN, H.C. ( 1 988) Middle Tertiary stratigraphy, deposi­
tional environments, paleoecology, and tectonic history
ofthe southeastern San Joaquin basin. PhD dissertation,
Stanford University, Palo Alto, CA, 305 pp.
PISCIOTTO, K.A. ( 1 98 1 ) Review of secondary carbonates in
the Monterey Formation, California. In: The Monterey
283
Formation and Related Siliceous Rocks of California
(Eds Garrison, R.E., Douglas, R.G., Pisciotto, K.E.,
Isaacs, C.M. & Ingle, J.C.). Soc Econ. Paleont. Miner.
Pacific Sec. Publ., 15, 273-284.
PISCIOTTO, K.A. & MAHONEY, J.J. ( 1 98 1 ) Isotopic survey of
diagenetic carbonates. DSDP Leg 63. In: Initial Reports
ofthe Deep Sea Drilling Project (Eds Yeats, R. et al.), 63,
595-609. US Government Printing Office, Washington,
DC.
RAISWELL, R. ( 1 97 1 ) The growth of Cambrian and Liassic
concretions. Sedimentology, 17, 1 4 7- 1 7 1 .
RAMSEYER, K. & BOLES, J.R. ( 1 986) Mixed-layer illite/
smectite minerals in Tertiary sandstones and shales, San
Joaquin basin, California. Clays Clay Miner. , 34, 1 1 5-
1 24.
RAMSEYER, K., BOLES, J.R. & LICHTNER, P.C. ( 1 992) Mech­
anism of plagioclase albitization. J. sediment. Petrol. ,
6 2 , 349-356.
SCHULTZ, J., BOLES, J.R. & TITTON, G.R. ( 1 989) Tracking
calcium in the San Joaquin basin, California: a stron­
tium isotopic study of carbonate cements at North
Coles Levee. Geochim. Cosmochim. Acta, 53, 1 9 9 1 -
1 999.
SCHWARTZ, D.E. ( 1 988) Characterizing the lithology,
petrophysical properties and depositional setting of the
Belridge Diatomite, South Belridge field, Kern County,
California. In: Studies of the Geology ofthe San Joaquin
Basin (Ed. Graham, S.A.). Soc. Econ. Paleont. Miner.
Pacific Sec. Publ., 60, 28 1 -30 1 .
SIBSON, R.H. ( 1 989) Earthquake faulting as a structural
process. J. Struct. Geol. , 1 1 , 1 - 1 4.
SIEH, K. ( 1 98 1 ) A review of geological evidence for
recurrence times of large earthquakes. In: Earthquake
Prediction, an International Review (Eds Ewing, M.,
Simpson, D. & Richards, P.), pp. 209-2 1 6. Am. Geo­
phys. Union, Washington, DC.
TAYLOR, T.R. & SouLE, C.H. ( 1 993) Reservoir character­
ization and diagenesis of the Oligocene 64-Zone sand­
stone, North Belridge field, Kern County, California.
Bull. Am. Ass. Petrol. Geol. , 77, 1 549- 1 566.
WALDER, J. & NuR, A. ( 1 984) Porosity reduction and
crustal pore pressure development. J. Geophys. Res., 89,
1 1 539- 1 1 548.
WEBB, G.W. ( 1 9 8 1 ) Stevens and earlier Miocene turbidite
sandstones, southern San Joaquin Valley. Bull. Am. Ass.
Petrol. Geol. , 65, 438-465.
WooD, J.R. & BOLES, J.R. ( 1 99 1 ) Evidence for episodic
cementation and diagenetic recording of seismic pump­
ing events, North Coles Levee, California, USA. Appl.
Geochem. , 6, 50 1 -5 2 1 .
ZIEGLER, D.L. & SPOTTS, J.H. ( 1 976) Reservoir and source
bed history in the Great Valley, California. In: Tomor­
row 's Oil from Today 's Provinces (Ed. Jantzen, R.E.).
Am. Ass. Petrol. Geol. Pacific Sec. Misc. Publ ., 24,
1 9-38.
Spec. Pubis int. Ass. Sediment. ( 1 99 8) 26, 285-307

Carbonate cementation in the Middle Jurassic Oseberg

reservoir sandstone, Oseberg field, Norway: a case of
deep burial-high temperature poikilotopic calcite

J. - P . G I RA R D
BRGM, B P 6009, 45060 Orleans Cedex 2 , France,
e-mail jp.girard@brgm.fr

A BSTRACT

Diagenetic carbonate cement in reservoir sandstones of the Oseberg Formation (Brent Group) in the
Oseberg field, Norwegian North Sea, occurs as disseminated siderite and ankerite, and as massively
calcite-cemented intervals. Other diagenetic features include extensive feldspar dissolution and
K-feldspar, quartz, kaolinite and dickite cements. Conditions of carbonate cementation are constrained
on the basis of textural, geochemical and fluid inclusion evidence.
Siderite formed early in the diagenetic history at low temperature (20-40"C) from mixed
marine/meteoric waters. Ankerite formed at a higher temperature (70-80"C) in the Latest Cretaceous­
Early Tertiary from waters of meteoric and marine origin with a significant influence of shale-derived
fluids. Siderite and ankerite are sporadically distributed and affect only minor volumes of sediments.
Conversely, calcite cement is volumetrically significant at field scale ("='I%). The calcite is ferroan and
occurs as inter- and intragranular poikilotopic cement. Textural relationships indicate that it postdates
all other diagenetic cements but dickite. Aqueous and hydrocarbon fluid inclusions yield formation
temperatures (90- 1 1 O"C) close to present-day reservoir temperatures. Combined with the recon­
structed thermal and oil-filling history of the reservoir, this implies that calcite cementation occurred in
the past 40 Myr and simultaneously with oil emplacement. Calcite is characterized by consistently low
3180 ( "='- 1 5%o PDB) and 3 1 3C ( "='- 1 2%o PDB) values, reflecting primarily a high formation temperature
and organic sources of carbon, and by elevated 87Sr/86Sr ratios (av. 0. 7 1 1 9). 3180 and 87Sr/86Sr values
indicate that calcite-forming fluids were isotopically similar to present-day formation fluids. They
acquired their isotopic signature following pervasive dissolution of feldspars prior to calcitization and,
possibly, as a result of the introduction of oil-accompanying basinal brines in the reservoir.
Calcite-cemented intervals are heterogeneously distributed within the field. They are a few metres
thick, with lateral extents no greater than a few kilometres. The amount of Ca necessary to form the
<I % calcite present in the reservoir is small and compatible with either internal or external sources.
The exact source of Ca could not be pinpointed, nor could the mechanism of reconcentration in specific
intervals. REE compositions of calcite samples, however, suggest that trace elements originated from
local and variable sources, and support the prevalence of relatively closed conditions during calcite
cementation.

I NTRODUCTION

Carbonate cement o f variable mineralogy (siderite,
calcite, dolomite and ankerite) is a common diage­
netic feature in North Sea Jurassic reservoir sand­
stones (e.g. Saigal & Bj0rlykke, 1 987; Walderhaug et
al., 1 989; Bj0rlykke et al., 1 992; Giles et al., 1 992;
Haszeldine et al., 1 992; Macaulay et al., 1 99 3). In a
number of studies particular attention has been
paid to extensive calcite-cemented intervals present
in many oilfields of the northern North Sea, as they
may constitute permeability barriers to fluid flow.

Carbonate Cementation in Sandstones Edited by Sadoon Morad 285

© 1998 The International Association of Sedimentologists
ISBN: 978-0-632-04777-2
286 J.-P. Girard
In most instances this type of calcite cement was
argued to have formed relatively early in the diage­
netic history, at temperatures lower than 7o·c
(Saigal & Bj0rlykke, 1 987; Glasmann et a!., 1 989;
Walderhaug & Bj0rkum, 1 992; Prosser et a!., 1 993;
Lundegard, 1 994). In only a few cases has it been
interpreted to have a late diagenetic, possibly high­
temperature (>90"C), origin (L0n0y et a!., 1 986;
Walderhaug & Bj0rkum, 1 9 92). In addition, it has
often been argued that calcite-cemented zones
formed by extensive flushing of the reservoir by
meteoric water (Walderhaug & Bj0rkum, 1 992;
Prosser et a/., 1 993; Lundegard, 1 994) and that
calcium was mainly derived from pre-existing car­
bonates, either diagenetic or detrital (Saigal &
Bj0rlykke, 1 987; Walderhaug et a!., 1 989; Bj0rkum
& Walderhaug, 1 99 3).
This paper reports petrological, geochemical and
fluid inclusion data gathered for calcite-cemented
intervals occurring in the Oseberg Formation in the
Oseberg field, as well as for coexisting siderite and
ankerite cements. The conditions of carbonate for­
mation are reconstructed and the origin of fluids
involved and the source of cements are discussed.
Particular emphasis is placed on providing evi­
dence documenting the late, deep-burial-high­
temperature origin of calcite. The work presented
herein is part of a larger integrated study on the
diagenesis of the Oseberg Formation (Girard, 1 995;
NORWAY
100 km
Girard, et a!., 1 995; Sanjuan et a!., 1 995), the
results of which will be published elsewhere.
G E O LO G I C A L S E T T I N G
The Oseberg field is located in the Norwegian part
of the North Sea on the eastern flank of the Viking
Graben (Fig. I). It spreads along a north-south
trend over the western halves of Norwegian blocks
30/6 and 30/9, about 1 40 km west of Bergen (Figs I
and 2). The Oseberg structure is one of the major
Mesozoic tilted fault blocks forming the transition
between the Horda platform to the east and the
Viking graben rift to the west (Badley et a!., 1 984;
Yielding et a!., 1 992). The fault block topography
generally developed during Late Jurassic and Early
Cretaceous times, as a result of the Cimmerian phase
of the extensional rift tectonics in the North Sea, and
is characterized by a gentle eastward dip of individ­
ual blocks separated by planar normal faults (Fig. 3).
In the Oseberg field fiv e main structural blocks have
been identified, namely Alpha, Alpha North, Alpha
South, Gamma and Gamma South (Fig. 2). The fault
blocks are essentially composed of clastic sediments
ranging in age from Permian-Triassic to Tertiary
(Fig. 4). The most significant hydrocarbon-bearing
reservoirs are within the Brent Group, of Middle
Jurassic age.
\
j
\
\
Fig. 1. Location map of the
Oseberg field in the northern
North Sea area. · · ·
 Carbonate cementation in the Middle Jurassic 287

Oseberg Fm Group and is overlain by sandstones, siltstones and

Oseberg Field shales of the Brent deltaic succession (Rannoch,
Core samples
Etive, Ness and Tarbert Formations) and mudrocks
• Water filled
e Oil/Gas filled
of the Viking Group (Heather and Draupne Forma­
tions). In crestal areas, the top part of the Brent
(Tarbert, Ness) is truncated and unconformably
overlain by marine shales of the Upper Cretaceous
Shetland Group (Fig. 3), as a result of substantial
erosion during Late Jurassic and Early Cretaceous
times.
The Oseberg Formation is interpreted as a prox­
imal, marine-dominated fan delta deposit brack­
eted by alluvial deposits. It is 20-60 m thick and
Limit of proven oil
laterally continuous throughout the main structures
of the Oseberg field, with few or no changes in
lithology. In contrast, the overlying reservoir units
(Rannoch/Etive, Ness and Tarbert; Fig. 3) are thin­
ner and often discontinuous as a result of lateral
8 changes of lithofacies and/or erosion (Helland­
• Hansen et al., 1 992). The Oseberg Formation is
composed primarily of coarse- to medium-grained,
poorly sorted, immature lithic sandstones with per­
meabilities ranging from 1 to 1 0 D and porosities
30/6 between 1 5 and 25%. In the Oseberg field it occurs
30/9 at present-day depths of 2200-3200 m below the
sea floor (water depth � 1 1 0 m) and temperatures of
95- 1 2 5 · c.
The burial/thermal history of the Brent Group
in the 30/6 quadrant has been reconstructed in
GAMMA
SOUTH ---f-ll- a detailed integrated basin modelling study by
Dahl et al. ( 1 987) and Dahl & Yiikler ( 1 99 1 ) A
.

time-temperature curve, taking into ·account the

variation in burial rate, temperature at the water­
Fault
sediment interface, palaeobathymetry and heat flow
--

= Major fault
Cimmerian unconformity
as a function of time, established by Dahl & Yiikler
"""""""'

4 6Km
0
Fig. 2. Simplified map of the study area showing major
structural features of the Oseberg field and location of
wells sampled. The dashed line A-B corresponds to the
position of the geological cross-section shown in Fig. 3.
This study concerns the reservoir sandstone of
the Oseberg Formation, which constitutes the low.­
ermost formation of the Brent Group (Figs 3 and 4).
The Oseberg Formation represents early lateral
infill of the basin during Aalenian ( 1 87- 1 83 Ma) as
a result of tectonic uplift of the Horda Platform and
the Norwegian mainland (Graue et al., 1 987). It lies
above the Lower Jurassic shales of the Dunlin
( 1 99 1 ) for the northernmost part of the Oseberg
area, is shown in Fig. 5. A similar thermal history
curve was provided by Walderhaug ( 1 994) for the
Oseberg reservoir in well 30/6- 1 0 (Figs 2 and 5).
Temperature in the Oseberg Formation remained
low ( <50-6o·q during the Jurassic and most of the
Cretaceous, then increased steadily to l 00-:- 120·c
from Upper Cretaceous to the present time as a
result of a sustained rate of burial. Present-day
depths and temperatures represent maximum
burial conditions. Lundegard ( 1 994) reported ·a
similar burial/thermal history for' the Brent Group
in the nearby field of Veslefrikk, located i 0- 1 5 km
to the northeast of the Oseberg field.
According to Dahl & Yiikler ( 1 99 1 ) the earliest
phase of oil migration in the Oseberg structures
occurred in the Late Maastrichtian. By the late
288 J.-P. Girard
ssw
Depth 8-04
•
mMSL
'
2 400 Shetland
2 500
2 700
2 800 Gamma structure
Fig. 3. Geological cross-section through the Alpha and Gamma structures (see location on Fig. 2), showing the
eastward tilt of Brent Grot,1p sedime11ts and the Late Jurassic-Early Cretaceous Cimmerian erosional unconfonnitY
(between Shetland and Viking Groups). Note how the top part of the Brent Group can b� affected by the Cimmerian
erosion in crestal areas.
Eocene a significant oil column and gas cap were
established, at least in the upper units of the Brent
Group. The main pulse of oil charging occurred in
the Oligocene and Miocene, with maximum oil
filling (to spill point) ip. the latest Miocene. Later
(Pliocene-present) the oil-water contact moved up
( 1 0- 1 5 m) owing to gas leakage and pressure in­
crease. Pre-production oil-water contact was at
2 7 1 0 m relative to mean sea level (MSL) (2600 m
relative to sea floor) over most of the field.
S A M PLE S A N D M ETHO D S
S amples
Analytical data from 2 1 samples of carbonate­
cemented sandstones of the Oseberg Formation are
examined in this paper. These were selected from a
set of 1 3 cores ( 1 57 thin sections) investigated as
part of an integrated study of the diagenesis in the
Oseberg Formation (Girard, 1 99 5; Girard et a/.,
1 995). The carbonate-cemented samples described
here come from nine wells located in the Alpha
NNE
Dun lin
Alpha structure
0 Gas zone
GJ Oil zone
Owaterzone
North, Alpha and Gamma structures (Fig. 2). Well
30/9-B04 is situated within a major fault zone
between Alpha and Gamma (Figs 2 and 3), and was
selected to investigate the influence of faults in the
carbonate cementation. All but one of the samples
studied lie at present-day depths of 2500-2700 m
below the sea floor (�2600-2800 m MSL) at tem­
peratures of 1 00- l l O • C. The singled-out sample
comes from well 30/6-8 in the eastern part of the
block (Fig. 2), where the Oseberg reservoir is
slightly deeper and hotter (3000.m below sea floor
and l200C). Samples from wells 30/6-7, 30/6-8 and
30/9- l come from the water zone, whereas other
samples were located l 0-40 m within the oil zone.
Petrography and mineralogy
Standard petrographical and mineralogical tech­
niques, including optical and cathodoluminescence
(CL) microscopy, X-ray diffraction (XRD) and
electron microprobe analysis (EPMA), were used to
characterize detrital and diagenetic minerals and
textural relationships. Thin sections were half
stained with K-Fe cyanide for rapid identification
 Carbonate cementation in the Middle Jurassic 289

140
PLEISTOCENE
NORDLAND
120 --+--
Dahl and Yukte; (1991)
>Uts"�
GP. PLIOCENE

- Wa/derhaug (1994)
I Fm
MIOCENE

OLIGOCENE
E
E
100

.a 60

� �
HORDALAND
GP. ..
c.
E
vvvvvvvvvv .. 60
1-
ROGALAND PALEOCENE
=175 Ma
40
GP
.

oil filling

I
TARBERT FM.
SHETLAND UPPER TRANSGRESSIVE

20
GP CRETACEOUS SHOREFACE

�
CROMER NOLL LOWER
� E
NESS FM. -200 -160 -120 ·80 -40 0
GP. � C� RE
� T AC � O�US
�--� � � �
Age (my)
� � DELTA PLAIN
VIKING
GP.
�����·�e
� UPPER
JURASSIC
ETIVE FM. Fig. 5. Reconstructed time-temperature curves for the
UPPER SHOREFACE.
MIDDLE
JURASSIC DELTA FRONT
Brent Group in the Oseberg field, and timing of oil
emplacement. After Dahl & Yiikler ( 1 99 1 ) and

\
RANNOCH FM. Walderhaug ( 1 994).
LOWER SHOREFACE

:�:���;;;'
����·�.·��
:;� .
,:o::,__,
______

"J< AND OLDER

OSEBERG FM
SUBAQUEOUS
DELTAS
FAN

=187 Ma
Fig. 4. General stratigraphy of the northern North Sea
sediments in the Norwegian sector and depositional
environments of successive formations in the Brent
Group.
of carbonates. Mineral abundances were deter­
mined by point counting 600 points per thin sec­
tion, resulting in an uncertainty of about ± 2-4%
(Van der Plas & Tobi, 1 965). EPMA and CL
microscopy were performed on a few selected
samples only.
F luid inclusions
All occurrences of carbonate cement were systemat­
ically investigated for fluid inclusions. Micro­
thermometric measurements were conducted on
1 00 Jlm thick doubly polished sections using a
modified Chaixmeca heating-freezing stage, pro­
viding standard errors on temperature measure­
ments of ± 2 • C. Care was taken to ensure that
overheating and stretching of inclusions were
avoided during temperature measurements by sys­
tematically collecting lower to higher homogeniza­
tion temperatures. Rare instances of non­
reproducible results were discarded. Heating runs
were carried out at a rate of about 5-6 • Cfmin up to
so·c, then at 2·c1min for most inclusions and
<I ·cfmin for small inclusions. Because of the
extreme difficulty in actually seeing the ice crystals
in the fluid inclusions studied, melting tempera­
tures were generally not measurable. All inclusions
were also examined under blue light and by Raman
spectrometry (Dilor spectrometer equipped with Ar
and Kr lasers) to document the presence of hydro­
carbons.
Isotopes and rare earth elements
Diagenetic carbonates were analysed for strontium,
oxygen and carbon isotope ratios and for rare earth
elements (REEs). Pure siderite was obtained by a
combination of heavy liquid segregation and mag­
netic separation, whereas pure ankerite was hand­
picked under a binocular microscope. Calcite was
dissolved from whole-rock powders using dilute
acid solutions (see below).
Strontium ratios were analysed on a multicollec­
tor Finnigan MAT262 mass spectrometer in static
mode using W filaments. Analytical uncertainty
(2cr), based on the reproducibility of analyses of the
international standard NBS987, is ± 0.00002. Pure
separates of siderite and ankerite were totally dis­
solved in I N HCL Calcite was dissolved from
wbole-rock powders by leaching in 0.2 N HCL This
procedure was verified to not leach strontium from
silicates. Pure calcite extracted (by heavy liquids)
290 J.-P. Girard

from sample 30/6-C08 3985 yielded a 87Sr/86Sr
ratio of 0. 7 1 1 8 1 1 , whereas acid leaching of the
whole-rock powder gave a ratio of 0. 7 1 1 846 using
0.2 N HCI and 0. 7 1 1 847 using 1 N HCI. All of these
ratios are within analytical uncertainty of each
other, indicating no significant contribution from
non-carbonate minerals.
Carbon and oxygen isotope ratios were deter­
mined using a triple collector Finnigan MAT252
mass spectrometer, after extraction using conven­
tional methods (McCrea, 1 950). Siderite and anker­
ite were reacted with 1 00% anhydrous phosphoric
acid at 2 5 • c for 2 weeks and 5 days, respectively,
following the recommendations of Al-Aasm et a!.
( 1 990). Calcite was analysed by reacting whole-rock
powders for 1 8 hours. Unless otherwise specified
o 1 3C and o180 values are reported relative to PDB.
Repeated analyses of standards (NBS 1 8, NBS 1 9)
indicate an analytical reproducibility of ± 0.2%o.
REEs were analysed by inductively coupled
plasma mass spectrometry (ICP-MS). Siderite was
dissolved in 1 N HCl and ankerite in HF + HN0 .
3
Analysis of calcite was performed on acid (0.2 N
HCl) leachates of whole-rock powders used for Sr
determinations.
PE T R O G R A PHY A N D
SPAT I A L D I S T R I BU T I O N O F
C A R BO NATE C E M E N T S
The modal compositions of the Oseberg Formation
are typical of first-cycle lithic arenites. The detrital
mineral assemblage is very uniform throughout the
study area, and consists of polycrystalline quartz
(23-39%), monocrystalline quartz ( 1 9-30%), K­
feldspar (4- 1 1 %), lithic clayey clasts ( <4%), rock
fragments ( <4%), muscovite ( <3%), plagioclase
( <2%) and biotite ( < 1 %). Lithic clayey clasts are
most commonly composed of micaceous or kaoli­
nitic material, whereas rock fragments are almost
exclusively quartzo-feldspathic mixed grains. Heavy
minerals, present in trace amounts, include garnet,
zircon and sphene.
The chronological sequence of diagenetic phases
established from petrographic observations in­
cludes: siderite and pyrite, K-feldspar overgrowths,
ankerite, feldspar dissolution, vermiform kaolinite
and quartz overgrowths, poikilotopic calcite and
dickite (Girard, 1 995). The most common and per­
vasively distributed diagenetic features at field scale
are feldspar dissolution, and quartz and kaolinite
cementation. The next most abundant diagenetic
phases are feldspar overgrowths and poikilotopic
calcite. Other diagenetic components affect ex­
tremely small volumes of rocks within the field.
S id erite
Siderite is present in wells 30/6-7 , 30/6-6 and
30/9-B26 in small amounts ( 1 -5%). It occurs as
intergranular rhombs or spherules disseminated
throughout the sandstone, and often closely associ­
ated with detrital clays or micas (Fig. 6A). Siderite
is an early diagenetic phase. It is never found
postdating any of the other diagenetic phases, and is
systematically engulfed by calcite in pores where
both carbonate cements occur (Fig. 6A).

Fig. 6. (Opposite) Thin-section photomicrographs of textural occurrences of diagenetic carbonates in the Oseberg
Formation. (A) Spherulitic precipitates of early diagenetic siderite (S) embedded in late poikilotoP.ic calcite (C).
'
Siderite commonly occurs in association with (replacement of?) detrital biotites, as shown in the !ower right comer of
the view (arrow). Sample 30/9-826 4286.8; crossed polars; x 100. (B) Rhombs of diagenetic ankerite precipitated
between the layers of an exfoliated detrital muscovite (A) and on detrital quartz (white arrow). Note that the ankerite
rhombs are embedded in a late pore-filling poikilotopic calcite (C). Sample 30/6-7 2794.1; crossed polars; x I 00.
(C) Poikilotopic calcite cement filling intergranular space and intragrain secondary porosity after feldspar. The
homogeneous royal blue colour which occurs after K-Fe cyanide staining indicates an iron-rich composition. Sample
30/6-CIO 3169.9; plane light; x 100. (D) Poikilotopic calcite cement filling secondary porosity developed at the
expense of diagenetic feldspar overgrowth (arrows). Sample 30/6-7 2784.5; plane light; x I 00. (E) Poikilotopic calcite
cement (C) filling residual intergranular space after quartz cementation (white arrows). Sample 30/6-7 2794.1; crossed
polars, x 100. (F) Yermicules of diagenetic kaolinite embedded within the poikilotopic calcite cement (darker areas).
Sample 30/6-7 2784.4; plane light; x 160. (G) Brecciated texture of detrital framework grains typical of samples from
well 30/9-804 (fault zone). The intergranular space is completely cemented by poikilotopic ferroan calcite stained by
K-Fe cyanide. Sample 30/9-804 2865.3; plane light; x 40. (H) Isolated inclusion typically encountered in poikilotopic
calcite cement. The dark linear feature inside the inclusion is not a daughter crystal but is due to the internal
morphology of the inclusion wall. This inclusion yields a Th of 99.8 C and does not fluoresce under blue light. Sample
•

30/6-C08 3983.3; plane light, x 500.

Carbonate cementation in the Middle Jurassic 291
 292 J-P. Girard

EPMA data (Table 1 ) typically show small "'

�
�
-o-aoooV)�oo-, r-
amounts of Ca and Mg, and traces of Mn in siderite.
� 66N � 666606 �
oo oo-oooo-voN

Chemical composition is relatively constant from

v5
one spherule to another within the same thin section. N

� ONNr-o---ooo N
Individual rhombs and spherules show a systematic ....,, 00 00\0 000\0 0 -
o oo_:Moooooo -.o
("!")

peripheral enrichment in Ca and Mg relative to the � � �

core (Table 1 ). 0V000000lr"'MO\

00NI.(')000r<iON
r-
V
oo__:Noooooo ..¢-
� �

Ankerite � 88N�080��� �
r- oo.....:Moooooo V)

Ankerite � �
N
occurs as disseminated intergranular ,_·
rhombi> in wells 30/6-7, 30/6-8 and 30/9- l . It was ..a 88��088�=N �
o oo....�
.: oooooo \0
identified by XRD and EPMA (Table 1 ). Ankerite � � �

rhombs are often precipitated within exfoliated 0:

mic a flakes or oil detrital surfaces of adjacent quartz
N 00\NN-.:t'OOOOOO \0
oo
� oooooo �
o-vl.l")ooo(""')ON r-
� oo.....:
grains (Fig . 6B). Ankerite abundance does not ex­
00
ceed a few percent, except in sample 30/9- 1 278 1 .0, 0

where it reaches 33% of the total rock volume. This ':J..

_.
O\Ooo-ooor--N
O-Nt-OOOVOM 0\
o oo.....:Noooooo -.i-
is the only occurrence of abundant ankerite cement � � �

fb'tmd in the samples investigated in this study. OOr"'"'l-lr'lOO-M\0 0'\

V"'
OOVV"'I000("1")0N
In samples where calcite and ankerite coexist, M oo....;Moooooo V)

M V)
1.(')

ankerite rhombs are systematically embedded "'

-t-V)000("<")0'\00"1 M
within the calcite cement (Fig. 6B). Because of the 0 oo.....: � oooooo
M 00\0t-OOOtr�ON M
�
scarcity of ankerite, textural relationships with other
diagenetic cements are rarely observable. However, ("f')NVt--00(""')0t-
OOVOOOOO\OON
\0
N
oo-:�oooooo r---:
�tis notable that in the ankerite-rich sample (30/9- 1 � �

27 8 1 .0) feldspar grains are essentially unaffected by (""')N00000Ntr�ON 0

dissolution, although they are extensively leached in � dd��ddd::d�

_....,
�
�
ankerite-free samples of the same well. In addition, -.i
in the single instance of partial leaching of a feldspar � 08�3888��� �
oo.....:Nooooooo
N
�
grain found in sample 30/9-1 278 1.0, ankerite was r----" -� lr)

.J:,
not observed to fill the intragrain porosity. These 0 ""d"00'\0'\0NOO'\O""d"
OONN0000'\-0'\
\0
\0
ool"""i_:ooooooo ..¢-
�
observations indicate that ankerite pre-dates feld­
� -� �

spar dissolution and calcite cementation. ·c

� vovvo o ooooo
00lr)0000000000
.... OO_;NOOOOOoO ..¢-
-<( -� �

Calcite

Calcite strictly occurs as poikilotopic cement filling

intergranular porosity and intragranular porosity
developed at the expense. of feldspar grains and
overgrowths (Fig. 6C, D).' It typically makes up
21-25% of the total rock volume, with intergranular
calcite ranging from 1 3 to 22% and intragranular
calcite from 5 to 8%. It is non-undulose and fluid
inclusion poor.. NQ. coexisting finely crystalline or
mqsaic calcite cement of the types. described in
· othe r Jurassic rese}o/oirs of the nearby Norwegian
fields (Saig�l & �jerlykJce, 1 987; Walderhaug et a!.,
1 989; Lundegard, 1 994) was found in the Oseberg
Formation. When coexisting with diagenetic sider­
ite, ankerite or quartz overgrowths in a same pore,
poikilotopic calcite 'atways fills the residual primary
 Carbonate cementation in the Middle Jurassic
porosity in the centre of the pore (Fig. 6A,B,E). In
addition, diagenetic kaolinite vermicules are com­
monly embedded within the poikilotopic calcite
(Fig. 6F). These textural relationships unambigu­
ously indicate that calcite postdates K-feldspar
overgrowths, siderite, ankerite, quartz and vermic­
ular kaolinite.
Point-counting analysis indicates that minus­
cement porosity in calcite-cemented sandstones
ranges from 1 8 to 25% (Girard, 1 995). This is
similar to the range of 1 4-25% minus-cement po­
rosity exhibited by surrounding calcite-free samples
in the same or nearby wells. Predicted residual
primary porosity at depths of 2.5-3.0 km is around
2 1 -25% for mature sandstones (Baldwin & Butler,
1 985) and as low as 1 5- 1 8% for immature sand­
stones such as the Oseberg sandstones (Bond et a!.,
1 98 3). Porosity loss due to mechanical compaction
in the Oseberg Formation is thus essentially normal
for this type of sediment, given the uncertainty of
± 3% associated with point counting. Point­
counting results further indicate that calcite­
cemented ancl calcite-free sandstones underwent
approximately the same degree of compaction,
compatible with their present-day burial depth.
This precludes an early, precompactional precipita­
tion of the calcite cement at shallow depth. The
latter situation would resl!lt in higher minus-cement
porosities in calcite-cemented samples (in the range
of 30-40%) (Glasmann et a!., 1 989; Lundegard,
1 994) than in calcite-free samples. Thus, all petro­
graphic evidence supports a late-burial timing of
poikilotopic calcite precipitation in the Oseberg
Formation.
K-Fe cyanide staining and XRD patterns indi­
cate that the calcite cement is iron rich. According
to EPMA data it contains 1 -2 wto/o FeO along with
small amounts (0.2-0.6 wto/o) of MnO and MgO
(Table 1 ), and is rather uniform compositionally at
sample, well and field scales. The ferroan composi­
tion, a common feature of 'burial' carbonate ce­
ments (Scholle & Halley, 1 985; Giles et a!, 1 992), is
consistent with the late diagenetic origin indicated
by petrography. CL microscopy indicates that the
calcite cement exhibits orange to red luminescence,
with abundant sector zoning that occurs as non­
concentric alternating bands of bright and dull
luminescence in a random orientation with respect
to growth features within crystals. No systematic
CL zonation pattern could be found from crystal to
crystal within a sample, or from sample to sample.
In well 30/9-B04, located in a fault zone (Figs 2
293
and 3), ferroan calcite with similar optical and
petrographic characteristics as in the other wells
constitutes the most abundant diagenetic phase
throughout the Oseberg Formation (Fig. 6G). It is
encountered, locally in association with ankerite,
over a 1 5 m depth interval in the lower half of the
40 m thick Oseberg Formation. In samples from
well 30/9-B04, the framework detrital grains are
pervasively microfractured into � breccia t(;!d facies
(Fig. 60), presumably as a result of faulting. The
calcite cement, which fills intergranular primary
porosity and secondary porosity after feld�par
grains, also fills the microfractures, affecting the
detrital grains. This indicates that calcite cementa.
tion in this well postdates microfracturing.
S patial d istribution
There is no discernible pattern to the spatial distri­
bution of siderite other than its sporadic occurrence
within wells and within the field. Ankerite is also
sporadically distributed and restricted to samples
located at or close to the boundary with the under·
lying shales of the Drake Formation (Dunlin
Group; Fig. 4). Systematic distribution of diage­
netic anlcerite in the vicinity of mudstone units h,as
been noted by Macaulay et a/. ( 1 993) in Jurassic
sandstones of the Magnus field, North Sea. Siderite
and ankerite represent minor c(;!ments in terms of
total volume precipitated a,t field scale.
The distribution of ferroan calcite in the Osebt;rg
Formation is very heterogeneous at field and well
scales. Calcite occurs in one well out of four and is
typically restricted to particular depth intervals
about 1 -6 m thick. Calcite-cemented intervals are
readily detected on wireline logs as porosity drops
abruptly, from 20-30% to less than �%. The upper
and lower boundaries are distinctly abrupt and
sharp. Detailed examination of witeline logs indi­
cates that calcitized intervals occur at different
levels within the Oseberg Formation, and cannot be
correlated between nearby wells. Therefore, they
are not related to any continuous depositional bed.
Most cemented intervals are found in the lower part
of the formation, within I 0 m of the lower bound­
ary, and occur in the oil zone as well as in the water
zone.
Petrographically and compositionally similar
poikilotopic calcite cement, distributed as subcon­
tinuous intervals within the Oseberg Formation,
was described by Walderhaug & Bj0rkum ( 1 992)
and by Lundegard ( 1 994) in the nearby field of
294 J.-P. Girard

Veslefrikk. Lundegard ( 1 994) indicated that most 10 ,------,

calcite-cemented intervals in Veslefrikk had lateral Ankerite 30/9-1 2780

extents of less than 1 km, and many less than 8 n = 6

1 00 m. Walderhaug & Bj0rkum ( 1 992) proposed avg.=72°C

"
lateral extents no greater than a few hundred me­ g 6
"
"'
tres. In the Oseberg field lateral extents of calcitized 0"

intervals are not precisely known, but are not � 4

expected to exceed a few kilometres, as calcitized 2
intervals cannot be correlated between wells. De­
pending on the actual size, cemented intervals may o ���n�n�--����
constitute permeability barriers to vertical flow but 10
they may not dramatically impair lateral hydrolog­ Calcite 30/6-C08 3983
30/6-C10 3192

�
ical connectivity in the reservoir. Indeed, the n = 27
oo
30/6-10 2568
8
++
chemical/isotopic composition of present-day for­ °
avg.=99 C -
0
mation water is remarkably uniform at field scale g
"
()
6 r-
'+
0 0
0 +
(Ziegler & Coleman, 1 998). c
" 8++
0
"
++
0 0 salinity :

� 4
0
g +
0 +
� ++
.... 0 1.8-3.5 wgt %

2 g t +t NaCI eq.

I
F LU I D I N C LU SI O N S gt
0
g g �I
50 70 90 110 130 150
M icrothermometry and R aman spectrometry
Homogenization temperature (°C)
Rare two-phase fluid inclusions are found in anker­
ite and in calcite (Fig. 6H). They occur as isolated Fig. 7. Histograms of homogenization temperatures of
inclusions or in groups of a few inclusions with a fluid inclusions in ankerite and calcite in the Oseberg
Formation. Salinities derived from limited Tm data are
random distribution. Owing to the scarcity of inclu­ indicated. Vertical black arrows represent present-day
sions and to the lack of visible intracrystalline reservoir temperature.
growth features in the carbonate cements, relation­
ships between fluid inclusions and carbonate
growth could not be evaluated. As a result, the crystal, and uniform apparent L/V ratios of 8/ 1 .
primary or secondary origin of the inclusions stud­ They are about 1 0 11m in size and do not fluoresce
ied cannot be ascertained. Microthermometric under blue light. Homogenization temperatures
measurements are summarized in Table 2 and ( Th) range from 66 to 88·c, averaging 72 ± 8·c
Fig. 7. (Fig. 7; Table 2). No melting temperatures could be
Only six inclusions were found in the ankerite measured.
cement of one sample (Table 2). These exhibit Inclusions in calcite were studied in three sam­
prismatic shapes, sometimes approaching negative ples from three of the northern wells (Table 2). They

Table 2. Summary of fluid inclusion data for diagenetic carbonates in the Oseberg Formation

Th Th Salinity
Depth* Temp.t range mean Tm (% NaCI
Mineral Well Sample (m) ("C) Type n ( "C) ("C) lcr ("C) equiv.)

Ankerite 30/9-1 2780.7 2649 1 08- 1 09 nf 6 66-88 72 8

Calcite 30/6-C08 3983.3 2559 1 06- 1 07 nf 6 88- 1 00 92 4
30/6-CIO 3 1 92.5 2555 1 05- 1 06 nf II 9 1 - 1 07 97 5 - 1 . 1 ; -2.2 1 .8; 3.5
30/6- 1 0 2568.2 2430 1 0 1 - 1 02 nf + f 10 1 0 1-1 12 1 06 3

f, fluorescent under blue light; n, number of individual fluid inclusions measured for Th or Tm; nf, non-fluorescent
under blue light.
*Present depth below sea floor (add 1 1 0 m for depths MSL).
tPresent reservoir temperature.
 Carbonate cementation in the Middle Jurassic
have variable shapes and sizes but are most com­
monly round or prismatic, and measure 1 5-30 IJ.m.
LIV ratios are of the order of 6/ 1 to 7/ 1 . Locally,
abundant small-sized ( <2-3 �J.m) inclusions were
observed, but these could not be used for microther­
mometry. Examination under blue light indicates
that no inclusions in samples 3 0/6-C08 3983.3 and
30/6-C 1 0 3 1 92.5 are fluorescent (Table 2). In sam­
ple 30/6- 1 0 2568.2 most inclusions exhibit a bright
yellow fluorescence under blue light, indicating the
presence of hydrocarbons, and coexist with aqueous
(non-fluorescent) inclusions. Homogenization tem­
peratures measured for a total of 27 inclusions
range from 88 to 1 1 2°C (Table 2), averaging 99
± ?"C. Th values for each individual sample are
very consistent and range between 1 1 and 1 6o C,
with 1 cr standard deviation typically around ± 4o C.
In sample 30/6- 1 0 2568.2, nine hydrocarbon­
bearing inclusions yielded an average Th of 1 06. 1
± 3° C, and one aqueous inclusion yielded a Th of
1 08 . 3oC. On the basis of this good agreement (and
in spite of a rather limited data set), we postulate
that coexisting aqueous and hydrocarbon inclusions
in calcite have indistinguishable homogenization
temperatures. Melting temperatures, approximated
as the last sudden movement of the bubble across
the inclusion, were only measured for two large
aqueous inclusions in sample 30/6-C1 0 3 1 92 . 5 . The
results suggest a low salinity, i.e. 1 . 8-3.5 wt% NaCl
equivalent (Table 2).
In Raman spectrometry the intense fluorescence
emitted by most inclusions, regardless of the type of
laser radiation, precluded any chemical character­
ization. The fact that non-fluorescent inclusions in
ankerite exhibit a strong fluorescence under laser
light suggests that those may contain hydrocarbons
of a type that does not fluoresce under blue light.
R epresentativeness of the data
If heated to temperatures significantly higher than
entrapment temperature, fluid inclusions in carbon­
ates may stretch or leak, resulting in partial or
complete resetting of Th values (Goldstein & Rey­
nolds, 1 995). In the Oseberg Formation there is
reasonable evidence that inclusions in ankerite and
calcite did not r!'s�t.
Evidence fo ankerite cement is that homogeni­
zation temperatur�s are consistent and 20-40oc
lower than present reservoir temperature (Fig. 7,
Table 2). As ankeriie f ormed in the middle part of
the diagencrtic sequence, it ha�.
�
Wflb
ably bee,n sub-
295
mitted to high temperatures (80- 1 00°C) for at least
several million years, and possibly several tens of
millions. However, the gap between homogeniza­
tion temperatures and reservoir temperature is still
significant. In contrast, fluid inclusions in calcite
have Th values equal or close to present reservoir
temperatures (Fig. 7; Table 2). This is generally
considered a hint that inclusions might have reset
(Goldstein & Reynolds, 1 99 5). However, the re­
markable consistency of Th values within a sample,
and from well to well, argues against resetting.
Variation in Th values in any sample is no greater
than 1 6°C, regardless of size (5-30 �J.m), location
and content (aqueous versus hydrocarbons).-
Additional evidence against resetting comes from
the study of primary fluid inclusions in the quartz
cement of the Oseberg Formation. These two-phase
inclusions have homogenization temperatures of
85- 1 05°C and contain appreciable amounts of
dissolved methane (Walderhaug, 1 994; Girard,
1 995). Walderhaug ( 1 994) argued that there is no
evidence to suspect that fluid inclusions in the
quartz cement in the Oseberg Formation might
have reset, and that measured Th values constitute
reliable indicators of formation temperatures of the
diagenetic quartz. Because calcite postdates quartz
in the Oseberg reservoir, the temperature of calcite
precipitation is expected to range between 85°C
and present-day temperature. The T h values of
90- 1 1 0°C obtained for calcite fluid inclusions are, ·
therefore, perfectly coherent, and strongly suggest
that inclusions are primary and not reset.
Pressure correction and formation temperatures
Three lines of evidence suggest that no significant
pressure correction needs to be applied to homoge­
nization temperatures measured in calcite. First, Th
values are equal or close to prese.nt-day tempera­
tures, which represent maximum burial tempera-·
tures experienced oy the reservoir (Dahl & Yiikler,
1 9 9 1 ). Secondly, the good agreement between ho­
mogenization temperatures of aqueous and hydro­
carbon inclusions suggests that Th values are close
to trapping temperatures. Thirdly, the presence of
CH4-rich fluid inclusions in the quartz cement
(Girard, 1 995), and the occurrence of hydrocarbon
inclusions in diagenetic quartz and calcite, implies
that these cements formed after hydrocarbon mi­
gration in the reservoir had begun. Because calcite
postdates quartz, and because the reservoir contains
a significant gas cap (Fig. 3), it is likely that dis-
296 J-P. Girard

solved CH4 is significant in calcite fluid inclusions.
Consequently, pressure correction would be negligi­
ble (Hanor, 1 980) and Th values of fluid inclusions
in calcite are close approximations of trapping
temperatures, hence the formation temperatures of
calcite.
If it is assumed that fluid inclusions in ankerite
are primary, measured Th values would represent
minimum trapping temperatures (the presence of
dissolved CH4 or hydrocarbons in these inclusions
could not be ascertained). However, as ankerite
pre-dates calcite and quartz in the diagenetic se­
quence, the lowest Th yielded by fluid inclusions in
calcite and quartz, i.e. 85-9o·c, may represent a
maximum formation temperature for ankerite. This
implies that diagenetic ankerite in the Oseberg must
have formed at temperatures between 65 and 8 5·c,
i.e. essentially identical to measured Th values
(Table 2).
The formation temperatures of ankerite and cal­
cite in the Oseberg reservoir derived from fluid
inclusions are in good agreement with petrographic
observations indicating a late diagenetic precipita­
tion. Combined with the thermal history of the
reservoir, these temperatures imply that ankerite
formed between about 65 and 35 Ma whereas cal­
cite could not have formed prior to 40 Ma (Fig. 5).
The timing of calcite cementation overlaps with the
timing of oil emplacement (Fig. 5). The occurrence
of calcitized intervals in the oil zone and the
presence of hydrocarbon inclusions in calcite sug­
gest that at least some calcite precipitated during oil
emplacement and prior to maximum oil fill (latest
Miocene).
I S O T OPE A N D R E E G E O CHE M I S T R Y
Sr content, 87Sr/86Sr ratios, o180 and o13C values
The 81 80, 813C, 87Sr/86Sr and Sr contents of repre­
sentative samples of diagenetic carbonates from the
structural blocks and from the fault are compiled in
Table 3 and plotted in conventional coordinates in
Fig. 8 and 9, along with relevant compositional
fields. The 8 1 80 values range from -1 5 . 5 to -3. 5%o,
813C values from - 1 8 . 1 to -4.6%o, 87Sr/86Sr ratios
from 0. 7 1 09 1 to 0. 7 1 3 8 5, and Sr contents from 1 7 3
to 1 8 1 3 ppm. The different types of diagenetic
carbonates have distinct 81 80-813C values, none of

Table 3. 1i 1 3C, 1> 1 80, Sr content and 87Sr/86Sr of diagenetic carbonates in the Oseberg Formation, Oseberg field

Depth* 813c 1)1 80 1)1 80 Sr

Mineral Well Sample (m) (%o PDB) (%o PDB) (%o SMOW) (ppm) 87Sr/86Sr

Block am�
Siderite 30/6-7 2778.2 2635 - 1 7.95 -3.69 27.06 1 73 0.7 1 1 5 1
30/9-B26 4288.3 2587 - 1 8. 1 3 -3.50 27.27 191 0.71 1 5 1
Ankerite 30/6-8 3 1 57 . 1 3007 - 1 2.37 - 1 1 .34 1 9. 1 8
30/9-1 2780.7 2649 - 1 2.32 -1 1 .82 1 8.68 270 0.7 1 38 5
Calcite 30/6-7 2784.3 264 1 - 1 0. 1 0 - 1 5.5 1 1 4.87 1 65 8 0.7 1 245
2794.2 265 1 - 1 0.8 1 - 1 5.2 1 1 5. 1 8 1 762
30/6-C08 3983.3 2559 - 1 2.92 - 1 4.48 1 5.94 1 65 8 0.7 1 1 69
3985. 1 2560 - 1 4. 5 9 - 1 3.74 1 6.70 1 495 0.7 1 1 83t
3987.4 2562 - 1 1 .96 - 1 4.70 1 5.70 18 1 3 0.7 1 1 88
40 1 1 .6 2573 - 1 2.6 1 - 1 4.28 1 6. 1 5 1 55 6 0.7 1 1 7 1
40 1 3.3 2574 - 1 4.55 - 1 3.83 1 6. 6 1 1 474
'
30/6-1 0 2568.2 2430 -9.90 - 1 5.53 1 4. 8 5 1 032 0.7 1 220
30/6-C!O 3 1 92.5 2555 - 1 2.26 - 1 4. 1 5 1 6.28 1 367 0.7 1 1 78
3 1 96.9 2560 - 1 3. 1 1 - 1 3.54 1 6.9 1 1 1 59 0.7 1 1 84
30/9-B26 4286.8 2559 -7.29 - 1 5. 1 5 1 5.24 1 936 0.7 1 200
4288.3 2560 -7. 1 3 - 1 5.15 1 5.24 1 470
3019-2 27 1 5. 8 25 8 1 - 1 2. 1 5 - 1 4. 1 2 1 6.3 1 1514 0.7 1 1 6 1

Fault wne
Ankerite 30/9-B04 2864.4 2557 -9. 3 5 - 1 1 .60 0.7 1 2 1 2
Calcite 30/9-B04 2868.7 256 1 -4.64 -7.25 0.7 1 09 1

*Depth below sea floor (add I I 0 m for depths MSL). -�- ·t'\_,.
tMean of three determinations (see text: Samples and methods). ·;.}�i_.·�-�-r·\;�;. ·�:.�<�� :;

�i2Y�:�;,- :.
'· . ' ...

..'• :-;t l·:!�/�.;·

i:�·J��:: :- : �i�r����r: ·
 Carbonate cementation in the Middle Jurassic 297

S18 o (0/oo PDB)

-20 -15 -10 -5 0
���--����-� 0

!::..
•
...
I
pristine primary
-5
01

4) marine carbonates -10 "'

<X>
0

-15
�
0
lOJ 0
-20 -u
0
�
-25

A
-30
0.715

Fm. water
<>
®
0.713 I
Fig. 8. (A) o180- 013C and (B)
87Sr/86Sr-l!Sr plots for diagenetic
'"-
(/)
<D
ca
0.711
� ...
•
0 0
carbonates occurring in the 2-
(/)
Oseberg Formation. Expected I'
0 Siderite
o180- 013C values for primary ca
0 Ankerite
marine carbonates, 87Sr/86Sr ratios 0.709
+ Ankerite fault
of present-day formation waters Mid. J u ra s sic t::. Calcite
(Girard, 1 995) and Middle / s e a water .A Calcite fault
Jurassic sea water (Jones et a/.,
0.707
1 994) are shown for comparison 0 10 20 30 40 50 60
(the Sr content of waters is not
-4
taken into account in the plot). 1/Sr X 10

which is compatible with a primary marine origin
(Fig. 8A) . In addition, all diagenetic carbonates
show 87Sr/86Sr ratios (Fig. 8B) significantly higher
than that of Middle Jurassic or younger sea water
(Burke et al., 1 982; Jones et al., 1 994). In agreement
with petrographic and fluid inclusion data, this
indicates that carbonate cements in the Oseberg
Formation did not form from pristine sea water on
the sea floor and that a significant proportion of 0,
C and Sr from sources other than sea water was
involved. Although calcite o 1 80 and o13C values are
linearly correlated, there is rio distinct linear trend
in the two plots suggestive of two-component mix­
ing between the three different carbonate minerals.
Siderite
The two siderite samples investigated have very sim­
ilar o180-o13C values, 87Sr/86Sr ratios and Sr con­
tents. The o 1 80-o13C values are close to, but not
within, the range of values compiled by Mozley &
Bums ( 1 992) for marine siderite concretions. The
very low o 13C (- 18%o) reflects a carbon sourced by
organic matter and not by sea water or dissolution of
marine carbonate fossils. The high o 1 80 value
(-3.6%o) is consistent with siderite formation from
Jurassic seawater (o1 80=- 1 %o sMow) during shal­
low burial at about 4o·c or lower in the presence of
significant meteoric water (Carothers et al., 1 988).
Assuming a formation temperature of 2o·c yields a
0 1 80 value of -6%o SMOW for the siderite-forming
water, which is essentially similar'to values inferred
for rainfall over the North Sea in the Jurassic (Has­
zeldine et at., 1 992). Considered together, o 180-
o1 3C values suggest that siderite precipitated (in
association with pyrite) in the sulphate reduction
zone at shallow depth (Irwin et al., 1 977).
Siderite exhibits the lowest 87Sr/86Sr ratio
(0. 7 1 1 5 1 ) of all carbonates from the structural
blocks (Table 3), still significantly higher than that
of Jurassic seawater (Fig. 8B). Because siderite is
often precipitated in association with detrital micas
and clays, it may have incorporated some of the
highly radiogenic Sr from the old detrital phyllosil-
298 J.-P. Girard

61s 0 ( 0/oo P D B )

-20 -15 -10 -5 0

0

O s e b . Fm.
V e s l efrikk -10 en

VI
0

-20
�

J u r . rese rv.
�
0
0
N o rway ""()
0
-30
�

® -40
0.715 Fig. 9. (A) 8 1 80-8 1 3C and (B)
Fm. water 87Sr/86Sr- 1 /Sr plots for diagenetic
poikilotopic calcite in the Oseberg
0.713
/ Formation (this chapter;
Wolderhoug

�
Welhaven, 1 985) and comparison
1...
(/) & 1Bjorkum with published data for
<D
IX) 1 99 2 poikilotopic. calcite in the Oseberg
0.711

IO
":::- Formation in the Veslefrikk field
(/) (Glasmann et a/., 1 989;
s e b . Fm Ve s l efrikk
r-...
IX) Walderhaug & Bj0rkum, 1 992;
Lundegord 1 99 4 !'::. struc. b l o c k Lundegard, 1 994) and in other
0.709
.A. f a u l t z o n e Jurassic reservoirs in the
0 W e l haven 1 9 85 Norwegian North Sea (Saigal &
0 . 707
® Bj0rlykke, 1 987). Welhaven ( 1 985)
did not report Sr content. 87Sr/86Sr
0 10 20 30 40 50 60
ratios of present formation fluids
1 /S r X 10
-4 (Girard, 1 995) are shown.

icates. Alternatively, the elevated 87Sr/86Sr ratio
may reflect the influence of meteoric water early in
the diagenetic history of the reservoir. The Oseberg
sediments were deposited in very shallow (<50 m
water depth) nearshore marine environments and
are overlain by shoreface and fluviatile deposits, i.e.
a setting favourable to the influx of continental
waters (Bjerlykke, 1 994). The fact that siderites
from wells 20 km apart yield identical 87Sr/86Sr
ratios (in addition to identical o 1 80-o 13C values)
strongly supports a regional control of the isotopic
composition of early diagenetic fluids through sig­
nificant meteoric water contribution, rather than a
local control due to dissolution of detrital micas.
Moreover, a significant contribution of meteoric
fluids in the early diagenesis is consistent with the
low Sr content ( 1 73- 1 93 ppm) of siderite.
Ankerite
Ankerite is characterized by moderately low o 1 80
(- 1 1 .8 io - 1 1 . 3%o) and o 13C (- 1 2.4 to -9.4%o)
values, consistent with precipitation at intermedi­
ate temperature in the thermal decarboxylation
zone of organic matter (Irwin et a/., 1 977). Using
fluid inclusion temperatures and the oxygen frac­
tionation equation of Fisher & Land ( 1 986) yields
o 1 80 values of -5 to -2%o SMOW for the parent fluid,
indicating a significant contribution of meteoric
water.
Ankerites from the structural block and from the
the fault have similar o 1 80-o 13C values, suggesting
formation under similar conditions. However,
ankerite from the block exhibits a significantly
higher 87Sr/86Sr (0. 7 1 385) than ankerite from the
fault (0. 7 1 2 1 2). In fact, the 87Sr/86Sr ratio of anker­
ite in sample 30/9-1 2780.7 (Table 3) is higher than
that of all other diagenetic carbonates in the Ose­
berg Formation, and is much higher than that of
present-day formation waters (Fig. 8B). Because
sample 30/9- 1 2780.7 is located within a few metres
of the underlying shales of the Dunlin Group
(Fig. 4), ankerite in this sample may have incorpo­
rated highly radiogenic Sr leached from the shales.
 Carbonate cementation in the Middle Jurassic
In contrast, ankerite sampled in the fault zone
comes from the middle part of the Oseberg Forma­
tion, 1 3 m above the shales, and originated from
different fluids of unknown sources circulating via
the fault system.
Calcite
Calcites from wells located within structural blocks
have very similar 15 1 80 values (- 1 5. 5 to - 1 3. 5%o,
av. - 1 4.6 ± 0. 7%o) and moderately variable 1513C
values (- 1 4.6 to -7. 1 %o, av. - 1 1.5 ± 2. 3%o). Simi­
larly, the 87Sr/86Sr ratio (0. 7 1 1 6 1 -0. 7 1 220, av.
0. 7 1 1 9 ± 0.0003) and Sr content ( 1 032- 1 936 ppm,
av. 1 5 30 ± 240) of calcite display little variation at
field scale (Fig. 8). Isotopic variations recorded
within a single calcite-cemented interval or within a
single well are small, of the order of 1 %o for 15180
and 2.5%o for 1513C and show no systematic pattern
from one interval to another. The 15180-15 13C values
and 87Sr/86Sr ratios obtained in this study for the
calcitized interval 27 1 5-27 1 8 m in well 30/9-2 are
identical to those previously reported by Welhaven
( 1 985) for the same interval (Fig. 9).
The uniformity in calcite isotopic composition at
field scale suggests a great similarity in the forma­
tion conditions (temperature, fluid, etc.) of calcite
in different calcitized intervals. This consistency
contrasts significantly with the very large scatter in
the 0, C and Sr isotope data reported in studies of
the poikilotopic calcite in the Oseberg Formation in
Veslefrikk, or in other Jurassic reservoirs of the
Norwegian North Sea (Fig. 9). Poikilotopic calcite
in the Oseberg Formation in the Oseberg field
typically exhibits very low 15180 values, equal to or
lower than the lowest values reported for Jurassic
sandstones of the Norwegian North Sea (Fig. 9A;
Bjerkum, 1 984; Saigal & Bjerlykke, 1 987). Petro­
graphic and fluid inclusion evidence indicates that
poikilotopic calcite in the Oseberg Formation in the
Oseberg field is a late diagenetic, high-temperature
cement. The low 15180 values may therefore reflect
an elevated temperature of formation. Late diage­
netic ferroan calcite is known to exist in reservoirs
of the Brent Group (Giles et al., 1 992). Walderhaug
& Bjerkum ( 1 992) and Lundegard ( 1 994) reported
the presence of late burial calcite cement, petro­
graphically postdating quartz cement, in Veslefrikk.
This late calcite is characterized by low 15 1 80 values
(- 1 5.9 to - 1 4. 7%o in Lundegard, 1 994; - 1 4. 3 to
- 1 1 . 1 %o in Walderhaug & Bjerkum, 1 992), and
thus very similar to calcite in the Oseberg field.
299
Isotopically light poikilotopic calcite was also re­
ported by Walderhaug et al. ( 1 989) in the Jurassic
sandstones of the Fensfjord Formation in the
nearby Brage field. Late diagenetic calcite, which
seems to be volumetrically minor in Veslefrikk,
constitutes the main, and apparently only, calcite
generation in the Oseberg field.
Using fluid inclusion temperatures (90- 1 1 0•C)
and the oxygen fractionation equation of Friedman
& O'Neil ( 1 977), 15 1 80 values of -3.6 to +0.6%o
SMOW are calculated for the diagenetic fluids from
which calcite grew in the Oseberg. This range
brackets the 15180 values of present-day formation
waters in the reservoir (-1.9 to -0.8%o SMOW)
(Ziegler & Coleman, 1 998) and is consistent with
the late diagenetic origin of calcite. The interstitial
water from which calcite formed during late burial
partially retained the isotopic signature of earlier
meteorically influenced diagenetic fluids, but under­
went a significant 1 80 enrichment with respect to
siderite- and ankerite-forming waters. This 1 80
enrichment is most likely related to massive feld­
spar dissolution (see discussion of Sr data below).
The 15 13C values of poikilotopic calcite in the
Oseberg field are in the upper range of those
reported for Veslefrikk (Fig. 9A). Extremely 13C­
depleted (1513C <-2%o) calcites found elsewhere in
Norwegian Jurassic reservoirs were not encoun­
tered in the Oseberg field. However, measured 15 1 3C
values point to a significant contribution of carbon
from organic sources. The presence of hydrocarbon
fluid inclusions in the Oseberg calcite indicates that
some oil was already present in the reservoir when
calcite formed. Thermal dehydroxylation of source­
rock kerogen or maturation/degradation of hydro­
carbons within the reservoir (Dahl & Yiikler, 1 99 1 )
are likely origins for light carbon.
The 87Sr/86Sr ratios of the Oseberg field calcite
(0. 7 1 1 6-0. 7 1 25) are very close to those of present­
day formation fluids (0. 7 1 25 ..:.0. 7 1 29; Fig. 9B)
(Girard, 1 995), consistent with a late diagenetic
formation. Petrographic evidence indicates unam­
biguously that calcite postdates feldspar dissolution
(Fig. 6C,D). Massive dissolution of feldspar is most
likely responsible for the elevated 87Sr/86Sr ratios
and Sr contents of calcite and present-day fluids.
Indeed, detrital K-feldspars from offshore Norway
reservoirs are expected to have 87Sr/86Sr ratios
around 0. 7 322 (Egeberg & Aagaard, 1 989). Walder­
haug & Bjerkum ( 1 992) found similar elevated Sr
ratios and contents for late burial calcite in the
Oseberg Formation in Veslefrikk.
300 J.-P. Girard
Calcite from the fault (30/9-B04) has much
higher � 1 80-� 1 3C values, but a significantly lower
Sr content and 87Sr/86Sr ratio than calcite from the
structural blocks (Fig. 9). This suggests that calcites
from the two locations formed under different
conditions and are not genetically linked. Sr-rich
fluids from which structural block calcite precipi­
tated were probably not routed via the fault system
between Alpha and Gamma.
Rare earth elements
The REE compositions of diagenetic carbonates in
the Oseberg are given in Table 4 and chondrite­
normalized patterns are shown in Fig. 1 0. The REE
patterns of all samples investigated are very dif­
ferent from that of sea water (Hogdahl et a!., 1 968),
indicating that REEs in Oseberg carbonate cements
were not primarily sourced by sea water or were
strongly fractionated during precipitation (Zhong &
Mucci, 1 994).
The REE patterns of the two siderites (Nos 1 5
and 1 6 in Fig. 1 OA) are alike. Along with the 0, C
and Sr isotope data (see above), this confirms that
the chemical and isotopic compositions of siderite
were controlled by fluid compqsition at a regional
scale.
The REE patterns of ankerite from the main field
and from the fault are markedly different, especially
with regard to heavy REEs (Fig. 1 OA). The REE
pattern of ankerite from the main block (No. 1 7 in
Fig. 1 OA) resembles typical patterns of shales
(McLennan, 1 989), suggesting, as postulated above
on the basis of 87Sr/86Sr ratios, that fluids from the
nearby shales were involved in the formation of
ankerite. The V-shaped REE pattern of ankerite
from the fault is unlike any other pattern (Fig. 1 OA).
In a similar way, calcite from the fault zone
exhibits a unique REE pattern, with a marked
positive Ce anomaly and a marked depletion in all
heavy REEs (No. 7 in Fig. 1 OB), which does not
resemble the patterns of other calcite samples. In
agreement with 0, C and Sr data, this confirms that
the calcites in the fault and in the main blocks are
not genetically linked. The REE patterns of calcite
from the structural blocks are extremely ·variable,
with any of the following features: marked deple­
tion in heavy REEs; extreme depletion in light
REEs; enrichment in middle REEs; marked posi­
tive Ce anomaly; and small positive Gd anomaly
(Fig. 1 OB,C). Different calcite-cemented intervals
in the same well or in different wells generally
exhibit extremely different REE patterns. In con­
strast, calcite samples from a single calcitized inter­
val usually have rather consistent REE patterns
(one exception to this is sample 30/6-C08 3985)
(Fig. l OC).
Zhong & Mucci ( 1 994) r;eported that partitioning
of REEs between calcite and sea water at 25·c is
significantly affected by concentrations of individ­
ual REEs and by the REE/Ca ratio of the solution.
This effect is most notable for light REEs. Partition­
ing during calcite precipitation at 1 0o·c is not
known, but might explain part of the variation
observed in the Oseberg field. Indeed, the total REE
content (LREE in Table 4) of calcite samples is very
variable ( 1 8-200 ppm), and the variation is much
more pronounced for light than for heavy REEs
(Fig. 1 0). Whatever the process responsible for the
observed variation, individual calcite�cemented in­
tervals have distinct patterns, suggesting a local and
variable source of REEs in solution. This is consis­
tent with the low mobility of REEs and has several
implications. First, it supports the ·observation that
calcite-cemented intervals represent disconnected
calcitized 'lenses' or intervals of finite lateral exten­
sion. Secondly, it suggests the prevalence of rela­
tively closed conditions with respect to REEs
during calcite precipitation (late diagenesis), as
opposed to the open conditions prevailing during
siderite formation (early diagenesi�). Considering
the uniformity in � 1 80 and 87Sr/86Sr values of
Oseberg calcites, the previous statement implies
that field scale homogenization of the isotopic
composition of interstitial fluids was essentially
achieved before calcite formed and REB-supplying
minerals started to alter. It also implies that Ose­
berg calcite precipitation did not.occur in a system
undergoing pervasive gravity-driven meteoric fluid
flow, as described by Waldei:haug _ . & Bjerkum
( 1 992) and by Lundegard ( 1 994,) for_�� oikilotopic
calcite in Veslefrikk.
Covariation in calcite isotopic compositi ons
Ca}cite � 1 80 and � 1 3C values ar�. negatively corre­
lated (r = -0.80) (Fig. 8A). This is the inverse of
what would be expected if calcite continuously
·
formed (or recrystallized) with· 'increasing depth­
temperature in the thermal decarboxylation zone of
organic matter (Irwin et al., 1 977; Prosser et a!.,
1 993). Additional cross-plots (not shown) indicate
that 87Sr/86Sr is also negatively correlated with
� 1 80 (r = -0. 77) and, to a lesser extent, positively
Table 4. REE analyses (in ppm) of diagenetic calcite, siderite and ankerite cements in the Oseberg Formation, Oseberg field

Calcite Siderite Ankerite

30/6-7 30/6-C08 .. 30/6- 1 0 3016-C l 0 30/9-826 . 30/9-2 30/9-804 . 30/6-7 30/9-826 30/9- 1 30/9-804
-- -- -- --
2784.2 2794.2 3983.3 3985.0 3987.3 401 1 .6' 40 1 3.3 2568.2 3 1 92'.4 3 1 96.9 4286.8 4288 . 3 27 1 5. 8 2868.7 2778. 1 . 4288.3 2780.7 2864.4
19 113 . T
Pattern* l 4 3 ll .: 12 8 10 5 14 2 ; 6 16 15 17 18

}.94 5.85 0. 372 7.35 0.558 0.682 1 0,04 . 6 ..6 2 . 6.0 1 2.48 1 0. 5 6.55
La 0. 1 99 2.76 4.35 1 7. 34
I 3 . 84 6.36
Ce 0. 8 7 1 8.09 28.23 8 1 .79 25.86 20. 1 2 33�82 1 .89 37.63 39.92 3.86 5.7 1 40.66 37:3 2 1 .45 1 1 . 76 34.23 1 3. 6 1
1
Pr 0. 1 28 1 .07 3.58 8.56 3.4L 2.0 1 2.95 0.242 3.92 3.73 0.449 0.824 4.07 2.69 2.56 1.4 3.44 1 .03
Nd 0.697 4.673 20. 1 6 38.24 1 9.38 'I 9.64 1 2.95 1.3 1 8.86 1 6. 2 1 2.42 4.88 1 7.9 1 0.27 1 1 .79 6.99 1 2.64 3.52
Sm 0.549 1.19 1 3.23 1 0. 74 l l .:i l I 3.6 3.21
.'· 0.922 7.01 6.02 1.19 2:09 6.91 2.28 3.2 2.23 3.23 0.7 7 1
Eu 0.498 0.368 6.41 3. 1 4 5.38 . 1 1 .46 l .Q4 0. 749 2.69 2. 1 3 0.48 0.7 1 7 2.6 1 0.558 0.935 0.725 0.958 0.233
Gd 2.27 1 . 34 32.7 1 2.56 25.54 5.68 4.2 4.62 1 2.06 9. 1 5 2. 1 2. 7 1 1 0. 5 5 1 .84 3.95 3.05 3.69 1 .07
4.29 I 1 .0 1
.
Tb 0.599 0.234 6.02 2. 1 2 0.6 6 4 1 .25 1 . 97 1 . 54 0.443 0.499 1 . 53 0.295 0.82 0.645 0.739 0.207
Dy 4.7 1 .46 30.99 1 1 . 84 2 1 .46 I 6.07 4.36 7.05 9.73 7.84 3.38 3.32 7.59 1 .6 1 5.66 5. 1 5. 4.41 l . 79
Ho 1. 135 0.37 5 .06 2.33 3 . 5 7 I 1 . 32 1 . 1 2 ·,_ 1 . 33 1 .68 1.51 0.893 0.793 1 .43 O.J8 1 .22 1 .27 0.93 0.599
Er 3. 1 3 1 .23 1 0. 7 3 5.83 7.82 1 3.49 3.33 3. 1 5 3.78 3.58 3. 1 5 2.43 3.58 1 .039 3.8 4.52 2.6 2. 1 7
Tm 0.44 0.207 1 . 34 0. 8 1 3 0.949 1 0.476 0.44 7 . 0.406 0.48 1 0.468 0.55 0.425 0.48 0. 1 56 0.652 0.833 0.389 0.39
Yb 2.42 1.41 7.4 4.77 5 .25 2.7 1 2.37 2.26 2.64 2.63 3.77 2.76 2.8 1 0.968 4.65 6.49 2.28 2.73
1 0.32 1 0.5
Lu 0.337 0.2 1 5 I 0. 7 1 0.755 . ·0.377 0.35 0.3 1 8 0.382 0.395 0.598 0.423 0.436 0. 1 48 0.68 0.952
1
I 1 8.0 24.6 1 7 1 .2 200.8 1 38.9 6 1 .8 76.7 25.9 1 09.2 1 02.5 23.8' 28.3 1 1 0.6 66.2 67.4 48.5 80.4 35.2

Vertical dashed lines separate different calcite intervals from the same well.
*Corresponding REE pattern number in Fig. I 0.

VJ
0
..,...
302 J.-P. Girard

(A) SIDERITE - ANKERITE (B) CALCITE (C) CALCITE

� 12

--<>-- 1 0

- 16
-- ..
100
--<>-- 1 7

-+- 1 8

fault zone

Fig. 10. Chondrite-normalized REE patterns of (A) siderite, ankerite and (B,C) calcite in the Oseberg Formation. The
numbers in the key boxes refer to the analysis numbers in Table 4. In (C) a same symbol shape denotes different
samples coming from a single calcite-cemented layer in a same well.

correlated with o 1 3C (? +0.5 8). These correla­

= o 1 3C within individual calcitized intervals was
tions are not due to undetected contamination
among carbonates because the analytical procedure
used to react calcite does not affect siderite/
ankerite, and because siderite/ankerite composi­
tions do not fall on the calcite linear trend (Fig. 8A).
Physical mixing of two different generations of
calcite is also ruled out for the following reasons: (i)
calcite samples do not show any linear trend,
diagnostic of two-component mixing, in the Sr plot
(Fig. 8B); (ii) no correlation was recognized between
o 1 80 (or 87Sr/86Sr) and any of the major (Ca, Fe,
Mg, Mn) or trace (Sr, Ba, Pb, U, Rb, Ga; data in
Girard, 1 99 5) elements; (iii) examination under
optical and CL microscopes did not provide any
evidence for the presence of two distinct popula­
tions of calcite crystals. Finally, calcite isotopic
composition does not correlate with sample depth
or distance from the underlying shales, and shows
no significant geographical zonation over the field.
pointed out by Lundegard ( 1 994) in Veslefrikk. The
general trend of all calcites in Veslefrikk is also that
of decreasing o 1 3C with increasing o 1 80 (Fig. 9A),
and was interpreted by Lundegard ( 1 994) as result­
ing from precipitation in a mixing zone between
meteoric (low o 1 80-high o1 3C) and basinal (high
o1 80-low o 1 3C) waters at less than 5o•c. The linear
array formed by the Oseberg calcite in Fig. 9A is in
the continuation of the Veslefri)ck calcite trend but
covers a much smaller range of o values at the low
o 1 80-high o 1 3C end. Lundegard ( 1 994) indicated
that the lowest o 1 80 values of his data set corre­
spond to late burial (post-quartz) calcite formed at
elevated temperature (>50•C). Thus, part of the
trend depicted in Fig. 9A for Veslefrikk and Ose­
berg calcites ought to represent temperature varia­
tion. The three Oseberg calcite samples studied for
fluid inclusions (Table 2) do show a significant
correlation (? +0. 83) between o 1 80 and fluid
=

Consequently, an alternative explanation must be inclusion average temperature. The range in o 1 80 of

sought. Oseberg calcites therefore most likely reflects small
A strong inverse correlation between o 1 80 and differences in formation temperature. The 2%o total
 Carbonate cementation in the Middle Jurassic
range in o 1 80 corresponds to a 20oC temperature
range (assuming constant o 1 80 for the parent fluid)
and is consistent with the total range of 24 o C in
fluid inclusion Th values (Table 2).
If the variation in o 1 80 is related to temperature,
the increase in 87Sr/86Sr with decreasing o 1 80 among
calcites may reflect greater dissolution of detrital
feldspars and greater release of radiogenic Sr in the
pore fluid at higher temperature. The increase of
o 1 3C with decreasing o 1 80 cannot be due to an in­
creasing contribution of marine inorganic carbon in
the reservoir at the time ofcalcite cementation, given
the deep burial conditions, nor to progressive disso­
lution of earlier-formed carbonate cements, as
siderite/ankerite o 1 3C values are no greater than
that of calcite (Fig. 8A). An increase in o 1 3C is best
explained as resulting from the fermentation of
organic matter or hydrocarbons (Irwin et a!., 1 917;
Carothers & Kharaka, 1 980). Dimitrakopoulos &
Muehlenbachs ( 1 987) argued that negative o 1 3C­
o 1 80 correlation in post-oil carbonate cement of the
Lower Cretaceous Mannville Group (Alberta, Can­
ada) resulted from biodegradation of petroleum,
involving a bacterial fermentation step. Whether a
similar scenario can be invoked for the Oseberg
calcite is arguable, considering the temperature of
calcite formation (90- 1 00o C), which approaches
the upper limit for bacterial activity. It is, however,
consistent with the early biodegradation of petro­
leum in the Alpha structure documented by Dahl &
Yiikler ( 1 99 1 ).
Alternatively, the calcite linear trend depicted in
Fig. 8A may reflect mixing of two different waters.
One would be characterized by a low o 1 80 ('%-3%o
SMOW, using T= 90- 1 00 o C), a high o 1 3C (�-8%o)
and a high 87Sr/86Sr (�0. 7 1 25), and the other by a
high o 1 80 (�0%o SMOW), a low o 1 3C ('%- 1 6%o) and a
low 87Sr/86Sr ('%0. 7 1 1 5). The former could repre­
sent interstitial fluid present in the Oseberg follow­
ing feldspar dissolution and quartz-kaolinite
cementation, whereas the latter water could be an
oil-accompanying basinal fluid (Ziegler & Coleman,
1 998). Depending on its exact isotopic composi­
tion, the volume contribution of basinal fluid to the
mixture would not necessarily be large.
S O U R C E OF C A L C I U M
The source o f calcium involved in the formation of
heavily cemented poikilotopic calcite intervals in
marine sandstones of the North Sea has been a
303
matter of debate. L0n0y et a!. ( 1 986) and Saigal &
Bj0rlykke ( 1 987) suggested that the dissolution of
earlier-formed, diagenetic calcite was the source of
Ca. Walderhaug & Bj0rkum ( 1 992), Prosser et al.
( 1 993) and Bj0rkum & Walderhaug ( 1 993) pro­
posed biogenic carbonates as Ca sources. Morad
& De Ros ( 1 994) pointed out that volcaniclastic
detrital fragments could also provide significant
amounts of calcium. Regarding more specifically
the extensive calcite cementation occurring in the
Oseberg Formation in the Veslefrikk field, Walder­
haug & Bj0rkum ( 1 992) favoured biogenic carbon­
ates as the most likely source of Ca, whereas
Lundegard ( 1 994) stated that the origin of Ca was
unclear.
The total volume of diagenetic calcite in the
Oseberg Formation in the Oseberg field can be
estimated as follows. Out of 22 cores examined in a
systematic survey, six were found to contain calcite­
cemented intervals, the cumulated thickness of
which is 23 m. This corresponds to an average
thickness of � 1 m of calcite-cemented sandstone
per well. Given an average thickness of 40 m for the
Oseberg reservoir and a calcite abundance of 2 5%
(point-counting data) the total volume of calcite is
0.6% at field scale. This is a minimum estimate, as
some calcitized intervals may not have been cored
through. However, it is highly unlikely that the total
volume of calcite in the Oseberg Formation exceeds
1 %.
A volume of 1 % calcite represents a relatively
moderate quantity of Ca (� 1 0 1 3 g), corresponding
to about 1 00 pore volumes of present-day forma­
tion water in the Oseberg (Ca concentration is 0.8
gil) (Ziegler & Coleman, 1 998) or 2.8% pure anor­
thite (Ca content 1 4%) at field scale. The Oseberg
calcite formed during deep burial diagenesis in
semi-closed conditions which are not compatible
with extensive fluid flow. This is fvrther indicated
by present-day overpressuring (7-8 MPa) in the
field, and is in agreement with the observation that
large-scale fluid movements in the North Sea are
not supported by present patterns of salinity distri­
bution (Bj0rlykke & Gran, 1 994). Thus, it seems
difficult to invoke advective transport of Ca from
external sources at the time of calcite precipitation,
given the very high pore-water fluxes required
(Bj0rkum & Walderhaug, 1 990; Bj0rlykke, 1 994).
In contrast, it is conceivable that Ca could have
been introduced into the Oseberg reservoir earlier
in the diagenetic history. The formation of early
diagenetic K-feldspar overgrowths in the Oseberg
304 J.-P. Girard
Formation requires the import of potassium from
outside the formation, and is attributed to the intro­
duction of evaporative brine in the reservoir (Gi­
rard, 1 995; Sanjuan et al., 1 995). Such brine could be
rich in dissolved Ca, particularly if it was involved in
dolomitization processes. Also, deep basinal fluids
originating in the Viking Graben and accompanying
oil emplacement in the reservoir could contain sig­
nificant dissolved Ca. Present-day formation waters
from offshore Norway in Triassic-Jurassic sand­
stones of the North Sea have Ca contents as high as
1 0 g/1 (Egeberg & Aagaard, 1 989). Only a few (four or
five) pore volumes of such a fluid would be necessary
to provide the required amount of Ca.
Alternatively, calcium may have been internally
derived. No biogenic carbonates have been found in
the samples studied, and there is no report of their
presence in the Oseberg Fm:mation in the Oseberg
field. Thus, one can only speculate that biogenic
carbonates may have been present and were totally
dissolved, as advocated by Bj0rlykke et al. ( 1 992).
Assuming transport was not significant, i.e. calcite­
cemented intervals developed in zones where shell
debris was abundant, this interpretation is not
supported by the 87Sr/86Sr ratios of Oseberg calcite,
which are much higher than expected for marine
biogenic carbonates. If the reservoir was essentially
free of biogenic carbonates, it is quantitatively
plausible that detrital Ca-plagiodase could provide
enough calcium to form the calcite cement. Disso­
lution of a few volume% of detrital anorthite
throughout the field would be sufficient. Mineralog­
ical studies of the Brent sandstones suggest that
detrital plagioclase is mostly albite. However, the
extent to which rock fragments present in sand­
stones and interbedded shales in the Brent may be
of volcanic origin, and rich in anorthite, has not
been fully assessed (Morad & De Ros, 1 994). A
major limitation to this interpretation is that anor­
thite would be expected to dissolve rather early at
shallow depth. An alternative is albitization of
Ca-Na plagioclase at 75- l Oo·c, as described by
Morad et al. ( 1 990). However, this would require a
greater amount (possibly several volume%) of pla­
gioclase grains to react, depending on their Ca
content. Finally, scenarios involving plagioclases as
the source of Ca are also problematic in that calcite
precipitation would be expected to be more scat­
tered throughout the reservoir.
Although the required quantity of calcium is not
large, the exact source of it cannot be pointed out
from present data alone. Nor can the mechanism
by which calcium was reconcentrated in hetero­
geneously distributed caleitized intervals in the
reservoir.
CONCLUSIONS
The timing and conditions o f carbonate cementa­
tion in the Oseberg Formation in the Oseberg field
have been discussed on the basis of petrological,
fluid inclusion and geochemical evidence. Samples
from structural blocks and from a major fault zone
were investigated.
Sporadic occurrences of minor amounts of early
siderite formed at low temperature (20-4o·q,
frequently in association with detrital biotites. The
fluid from which siderite precipitated had a consis­
tent oxygen, carbon and strontium isotopic compo­
sition over great distances, which is best explained
as representing homogeneous mixing of Jurassic
seawater and meteoric water at field scale. Carbon
was predominantly supplied by organic sources.
Ankerite cement is very scarce in the Oseberg
reservoir, except in one sample located at the
boundary with the underlying shales. In this sam­
ple, fluid inclusion analysis indicates that ankerite
formed at temperatures of 70-so ·c, i.e. during the
latest Cretaceous or Early Tertiary. The 87Sr/86Sr
ratios and REE patterns of ankerite are consistent
with a significant influence of fluids and/or trace
elements derived from shales. The o 1 80 values
indicate a marked contribution of meteoric water
during ankerite formation. Ankerite from the fault
zone and from the structural blocks have similar
o 1 80-o 1 3C values, suggesting similar conditions of
formation.
Diagenetic calcite occurs as inter- and intragran­
ular poikilotopic cement. Textural relationships
with other diagenetic phases, in agreement with its
ferroan chemical composition, indicate that calcite
is a deep burial cement. Formation temperatures
derived from aqueous and hydrocarbon-bearing
two-phase fluid inclusions are in the range of 90-
l l O·c, i.e. similar to present reservoir tempera­
tures, implying precipitation during the past 30-
40 Myr. The absence of resetting is attested to by
the excellent consistency of Th values at reservoir
scale. Calcite o 1 80 values are among the lowest
reported for intergranular poikilotopic calcite in the
Brent sandstones, and reflect primarily an elevated
formation temperature. Consistent with a late dia­
genetic origin, the reconstructed o 1 80 values and
 Carbonate cementation in the Middle Jurassic
87Sr/86Sr ratios for parent waters indicate that calcite
formed from fluids with isotopic compositions
similar or close to present-day formation flu ids. The
elevated 87Sr/86Sr and o 1 80 of calcite-forming
waters resulted mainly from pervasive dissolution
of detrital and diagenetic feldspar prior to calcitiza­
tion, and may be in part r�lated to the introduction
of oil-accompanying basinal brines. Calcite o 1 3C
values reflect carbon sourced by organic matter and
possibly influenced by biodegradation-fermenta­
tion processes that affected petroleum emplaced in
the reservoir.
Calcite-cemented intervals are heterogeneously
distributed at field scale. They are typically a few
metres thick and have lateral extents no greater
than a few kilometres. The mechanism responsible
for spatial heterogeneity is unknown. The amount
of Ca necessary to form the total volume of diage­
netic calcite present in the Oseberg r�servoir in the
Oseberg field is small and compatible with internal
and/or external sources. Dissolution or albitization
of detrital plagioclase, dissolution of biogenic car­
bonates and introduction of small amounts of
Ca-rich basinal fluids in the reservoir are possibili­
ties.
The integration of textural, compositional, fluid
inclusion and isotopic evidence indicate that the
Oseberg calcite is a deep-burial-high-temperature
cement Except for minor occurrences of deep burial
poikilotopic calcite in Brent reservoirs reported in
Veslefrikk and in Hild fields, this is different from
much of the poikilotopic calcite encountered as
extensive cemented intervals in Jurassic reservoirs
of the North Sea. In the Oseberg field there is no
evidence of any early marine Mg-calcite precursor
or precompaction meteoric or mixing-zone calcite
cementation. Although residual meteoric water, in­
herited from earlier invasion of the reservoir by
meteoric flow, was present in the Oseberg Forma­
tion at the time of calcitization, the formation of
calcite-cemented intervals was not linked to perva­
sive meteoric flushing of the reservoir.
A C K N O WL E D G E M E N T S
This research was conducted as part of the Reser­
voir Engineering Project (CEU-DGXII) in the
framework of the JOULE II Programme. Financial
support was pr�vided by the Commission of the
European Communities (contract No. JOU2-CT92-
0 1 82) and by the Research Division of BRGM
305
(Scientific Project S06). Special thanks are ex­
pressed to Norsk Hydro (Bergen, Norway) for pro­
viding samples and access to pre-existing data, and
for granting permission to publish the results. Ana­
lytical work was carried out at BRGM, Orleans,
France. Assistance in the laboratories and in data
acquisition was provided by 0. Legendre (FI micro­
thermometry), C. Beny (Raman spectrometry), A.
Cocherie (trace elements), C. Guerrot (Sr isotopes),
F. Pillard and C. Gilles (EPMA). Assistance in
selecting samples and collecting geological informa­
tion was provided by H. Moe (Norsk Hydro).
Interpretations presented herein are the sole re­
sponsibility of the author.
REFERENCES
AL-AASM, I.S., TAYLOR, B .E. & SouTH, B . (1990) Stable
isotope analysis of multiple carbonate samples using
selective acid extraction. Chern. Geo/. (!sot. Geosci.
Sect.), 80, 119-125.
BADLEY, M.E., EGEBERG, T. & NIPEN, 0. (1984) Develop­
ment of rift basins illustrated by the structural evolution
of the Oseberg feature, Block 30/6, offshore Norway. J.
Geo/. Soc. London, 141, 639-649.
BALDWIN, B. & BUTLER, C.O. (1985) Compaction curves.
Bull. Am. Ass. Petrol. Geo/. , 69, 622-626.
BJ0RKUM, P.A. (1984) A study of Middle Jurassic sand­
stones in cored sections of the 3016- 7 well of the Oseberg
field, with emphasis on diagenesis. Cand. Sci. Thesis,
University of Bergen.
BJ0RKUM, P.A. & WALDERHAUG, 0. (1990) Geometrical
arrangement of calcite cementation within shallow ma­
rine sandstones. Earth Sci. Rev., 29, 145-161.
BJ0RKUM, P.A. & WALDERHAUG, 0. (1993) Isotopic com­
position of a calcite-cemented layer in the Lower Juras­
sic Bridport sands, Southern England: implications for
formation of laterally extensive calcite-cemented layers.
J. sediment. Petrol. , 63, 678-682.
BJ0RLYKKE, K. (1994) Pore-water flow and mass transfer
of solids in solution in sedimentary basins. In: Quanti­
tative Diagenesis: Recent Development£ and Application
to Reservoir Geology (Eds Parker, A & Sellwood, B.W.},
pp. 127-140. K.luwer Academic, Dordrecht.
BJ0RLYKKE, K. & G RAN, K. (1994) Salinity variations in
North Sea formation waters: implication for large-scale
fluid movements. Mar. Petrol. Geol., 1 1 , 5-9.
BJ0RLYKKE, K., NEDKVITNE, T., RAMM, M. & GIRISH, C.
(1992) Diagenetic processes in the Brent Group (Middle
Jurassic) reservoirs of the North Sea: an overview. In:
Geology of the Brent Group (Eds Morton, A. C., Haszel­
dine, R. S., Giles, M.R. & Brown, S.). Spec. Pub!. Geol.
Soc. London, 61, 263-287.
BOND, G.C., KoMINZ, M.A. & DEVLIN, W.J. (1983) Ther­
mal subsidence and eustasy in the Lower Paleozoic
miogeocline of western North America. Nature, 306,
775-779.
BURKE, W.H., DENISON, R.E., HETHERINGTON, E.A. et a/.
306 J.-P. Girard
( 1 982) Variation of seawater 87Sr/86Sr throughout
Phanerozoic time. Geology, 10, 5 1 6-5 1 9.
CAROTHERS, W.W. & KHARAKA, Y.K ( 1 980) Stable carbon
isotopes of HCO) in oil-field waters-implications for
the origin of C02• Geochim. Cosmochim. Acta, 44,
323-332.
CAROTHERS, W.W., ADAMI, L.H. & ROSENBAUER, R.J.
( 1 988) Experimental oxygen fractionation between
siderite-water and phosphoric acid liberated COr
siderite. Geochim. Cosmochim. Acta, 44, 323-332.
DAHL, B. & YOKLER, A. ( 1 99 1 ) The role of petroleum
geochemistry in basin modeling of the Oseberg area,
North Sea. In: Source and Migration Processes and
Evaluation Techniques (Ed. Merrill, R.K.). Am. Assoc.
petrol. Geol. Treatise of Petroleum Geology, Handbook
of Petroleum Geology, pp. 65-85.
DAHL, B., NYSAETHER, E., SPEERS, G.C. & YOKLER, A.
( 1 987) Oseberg area-integrated basin modelling. In:
Petroleum Geology ofNorth West Europe (Eds Brooks, J.
& Glennie, K.), pp. 1 029- 1 038. Graham & Trotman,
London.
DIMITRAKOPOULOS, R. & MUEHLENBACHS, K. ( 1 987) Bio­
degradation of petroleum as a source of 1 3C-enriched
carbon dioxide in the formation of carbonate cement.
Chern. Geol. (/sot. Geosci. Sect.), 65, 283-29 1 .
EGEBERG, P.K. & AAGAARD, P. ( 1 989) Origin and evolution
of formation waters from oil fields on the Norwegian
shelf. Appl. Geochem. , 4, 1 3 1 - 1 42.
FISHER, R.S. & LAND, L.S. ( 1 986) Diagenetic history of
Eocene Wilcox sandstones, south-central Texas.
Geochim. Cosmochim. Acta, 50, 5 5 1 -56 1 .
FRIEDMAN, I. & O N EIL, J.R. ( 1 977) Compilation of stable
' (Middle Jurassic), Veslefrikk field, Norway. Geochim.
isotope fractionation factors of geochemical interest. In:
Data of Geochemistry, 6th edn (Ed. Fleisher, M). US
Geol. Surv. Prof. Paper 440-KK, 1 2 pp.
GILES, M.R., STEVENSON, S., MARTIN, S.V. et a/. ( 1 992) The
reservoir properties and diagenesis of the Brent Group:
a regional perspective. In: Geology of the Brent Group
(Eds Morton, A.C., Haszeldine, R.S., Giles, M.R. &
Brown, S.). Spec. PubI. Geol. Soc. London, 6 1 , 289-327.
GIRARD, J.-P. ( 1 995) Diagenesis of the Oseberg Reservoir
Formation (Middle Jurassic), Oseberg Field, Norwegian
North Sea: a Core Study. BRGM Int. Rep. R38556,
88 pp.
GIRARD, J.-P., SANJUAN, B., FOUILLAC, C., ZIEGLER, K. &
CoLEMAN, M. ( 1 995) Reconstruction and geochemical
modelling of the diagenetic history of the Middle­
Jurassic Oseberg sandstone reservoir, Oseberg field,
Norwegian North Sea. Am. Assoc. petrol. Geol. Interna­
tional Conference, Abstract Book, p. 26 (Abstract).
GLASMANN, J.R., CLARK, R.A., LARTER, S., BRIEDIS, N.A. &
LUNDEGARD, P.D. ( 1 989) Diagenesis and hydrocarbon
accumulation, Brent sandstone (Jurassic), Bergen High
area, North Sea. Bull. Am. Ass. Petrol. Geol. , 73, 1 34 1 -
1 360.
GOLDSTEIN, R.H. & REYNOLDS, T.J. ( 1 995) Systematics of
Fluid Inclusions in Diagenetic Minerals. Soc. Econ.
Paleont. Miner., Short Course No. 3 1 .
GRAUE, E., HELLAND-HANSEN, W., JoHNSEN, J.R. et a!.
( 1 987) Advances and retreat of the Brent delta system,
Norwegian North Sea. In: Petroleum Geology of North
West Europe (Eds Brooks, J. & Glennie, K. ) , pp. 9 1 5-
937. Graham & Trotman, London.
HANOR, J.S. ( 1 980) Dissolved methane in sedimentary
brines: potential effects on the PVT properties of fluid
inclusions. Econ. Geol., 75, 603-6 1 7.
HASZELDINE, R.S., BRINT, J.F., FALLICK, A.E., HAMILTON,
P.J. & B ROWN , S. ( 1 992) Open and restricted hydrolo­
gies in Brent Group diagenesis: North Sea. In: Geology
of the Brent Group (Eds Morton, A. C., Haszeldine, R.S.,
Giles, M.R. & Brown, S.). Spec. Publ. Geol. Soc.
London, 61, 40 1 -4 1 9.
HELLAND-HANSEN, W., ASHTON, M., L0MO, L. & STEEL, R.
( 1 992) Advance and retreat of the Brent delta: recent
contributions to the depositional model. In: Geology of
the Brent Group (Eds Morton, A.C., Haszeldine, R.S.,
Giles, M.R. & Brown, S). Spec. Publ. Geol. Soc. Lon­
don, 6 1 , 1 09- 1 27.
HOGDAHL, O.T., MELSON, S. & BOWEN, V. ( 1 968) Neutron
activation analysis of lanthanide elements in seawater.
Adv. Chern. Ser., 73, 308-325.
IRWIN, H., CURTIS, C. & COLEMAN, M. ( 1 977) Isotopic
evidence for source of diagenetic carbonates formed
during burial of organic-rich sediments. Nature, 269,
209-2 1 3.
JoNES, C.E., JENKYNS, H.C., CoE, A.L. & HESSELBO, S.P.
( 1 994) Strontium isotopic variations in Jurassic and
Cretaceous seawater. Geochim. Cosmochim. Acta, 58,
306 1 -3074.
L0N0Y, A., AKSELSEN, J. & RONNING, K. ( 1 986) Diagenesis
of a deeply buried sandstone reservoir: Hild field,
Northern North Sea. Clay Miner. , 21, 497-5 1 1 .
LUNDEGARD, P.D. ( 1 9 94) Mixing zone origin of 1 3C­
depleted calcite cement: Oseberg Formation sandstones
Cosmochim. Acta, 58, 266 1 -2675.
MACAULAY, C.J., HASZELDINE, R.S. & FALLICK, A.E. ( 1 993)
Distribution, chemistry, isotopic composition and ori­
gin of diagenetic carbonates: Magnus sandstone, North
Sea. J. sediment. Petrol., 63, 33-43.
McCREA, J.M. ( 1 950) On the isotopic chemistry of carbon­
ate and paleotemperature scale. J. Chern. Phys. , 18,
849-857.
McLENNAN, S.M. ( 1 989) Rare earth elements in sedi­
mentary rocks: influence of provenance and sedi­
mentary processes. In: Geochemistry and Mineralogy of
Rare Earth Elements (Eds Lipin, B.R. & McKay, G.A.).
Rev. Mineral., Mineral. Soc. Am., 21, 1 69- 1 99.
MORAD S. & DE Ros, L.F. ( 1 994) Geochemistry and
diagenesis of stratabound calcite cement layers within
the Rannoch Formation of the B'rent Group, Murchison
field, North Viking Graben (northern North Sea)­
comment. Sediment. Geol. , 93, 1 35- 1 4 1 .
MORAD, S., BERGAN, M., KNARUD, R. & NYSTUEN, J.P.
( 1 990) Albitization of detrital plagioclase in Triassic:
reservoir sandstones from the Snorre field, Norwegian
North Sea. J. sediment. Petrol., 60, 4 1 1 -425.
MOZLEY, P.S. & BURNS, S.J. ( 1 992) Oxygen and carbon
isotopic composition of marine carbonate concretions:
an overview. J. sediment. Petrol. , 63, 73-83.
PROSSER, D.J., DAWS, J.A., FALLICK, A.E. & WILLIAMS,
B.P.J. ( 1 993) Geochemistry and diagenesis of stra··
tabound calcite cement layers within the Rannoch
Formation of the Brent Group, Murchison field, North
Viking Graben (northern North Sea). Sediment. Geol.,
87, 1 39- 1 64.
 Carbonate cementation in the Middle Jurassic
SAIGAL, G.C. & BJ0RLYKKE, K. ( 1 987) Carbonate cements
in clastic reservoir rocks from offshore Norway­
Relationships between isotopic composition, textural
development and burial depth. In: Diagenesis of Sedi­
mentary Sequences (Ed. Marshall, J.D.). Spec. Pub!.
Geol. Soc. London, 36, 3 1 3-324.
SANJUAN, B., GIRARD, J.-P. & CZERNICHOWSKJ-LAURIOL, I.
( 1 995) Geochemical Modelling of the Main Diagenetic
Processes in the Oseberg Sandstone Reservoir Oseberg
Field, Northern North Sea. BRGM Internal Report
R38599, 83 pp.
SCHOLLE, P.A. & HALLEY, R.B. ( 1 985) Burial diagenesis:
out of sight, out of mind. In: Carbonate Cements (Eds
Schneiderman, N. & Harris, P.M.). Spec. Pub!. Soc.
Econ. Paleont. Miner., 36, 309-334.
VAN DER PLAS, L. & TOBI, A. C. ( 1 965) A chart for judging
the reliability of point counting results. Am. J. Sci., 263,
87-90.
WALDERHAUG, 0. ( 1 994) Temperatures of quartz cemen­
tation in Jurassic sandstones from the Norwegian con­
tinental shelf-Evidence from fluid inclusions. J
sediment. Res., A64, 3 1 1 -323.
WALDERHAUG, 0., BJ0RKUM, P.A. & BOLAS, H.M.N. ( 1 989)
Correlation of calcite-cemented layers in shallow­
marine sandstones of the Fensfjord Formation in the
307
Brage Field. In: Correlation in Hydrocarbon Exploration
(Eds Collison, J.D. et a/.), pp. 367-375. Graham &
Trotman, London.
WALDERHAUG, 0. & BJ0RKUM, P.A. ( 1 992) Effect of mete­
oric water flow on calcite cementation in the Middle
Jurassic Oseberg Formation, well 30/3-2, Veslefrikk
field, Norwegian North Sea. Mar. Petrol. Geo/. , 9,
308-3 1 8.
WELHAVEN, E. ( 1 985) Formasjonvann og diagenese, Ose­
bergj/etet, Brenn 30/9-2. Cand. Sci. thesis, University of
Oslo.
YIELDING, G., BADLEY, M.E. & ROBERTS, A.M. ( 1 992) The
structural evolution of the Brent province. In: Geology
ofthe Brent Group (Eds Morton, A. C., Haszeldine, R.S.,
Giles, M.R. & Brown, S.). Spec. Pub!. Geol. Soc.
London, 61, 27-43.
ZHONG, S. & Mucci, A. ( 1 998) Partitioning of rare earth
elements (REEs) between calcite and seawater solutions
at 2 5 · c and I atm, and high dissolved REE concentra­
tions. Geochim. Cosmochim. Acta, 59, 443-453.
ZIEGLER, K. & COLEMAN, M. ( 1 998) Chemical and isotopic
variations of formation waters from the Oseberg field
(Brent Group, Norwegian North Sea): Implications for
paleohydrodynamics. App/. Geochem. , in revision.
Spec. Pubis int. Ass. Sediment. ( 1998) 26, 309-325

Origin and timing of carbonate cementation of the

Namorado Sandstone (Cretaceous), Albacora Field, Brazil:
implications for oil recovery

R.S. D E SOUZA a n d C.M. D E A S S I S SI LVA

Petrobras Research and Development Center, Cidade Universitdria, Quadra 7, Ilha do Fundiio,
CEP 21949-900, Rio de Janeiro, RJ, Brazil, e-mail rschi.ffer@mail.utexas.edu;
carlosmas@cenpes. petrobras. com. br

ABSTRACT

The N amorado S andstone is one of the m ain oil reservoirs of the C ampos basin, offshore Brazil, and
consists of a t hick t urbiditic sequence deposited d uring the Albian-Cenomanian. C alcite cement is
volumetrically t he most important diagenetic parameter controlling reservoir q uality in t he sandstones.
Three types of c alcite cement were detected: (i) blocky mosaic with an oscillatory cathodolumines­
cence (CL) zonation pattern, formed by the replacement of bioclasts in response to meteoric water
inv asion during eodiagenesis; (ii) very fine mosaic, showing t hin CL zonation, formed by nucleation
around bioclasts and precipitating relatively early d uring diagenesis at an estimated temperat ure of
about 25 ·c, prior to oil emplacement; (iii) coarse mosaic with a poikilotopic texture, formed after oil
charging into the reservoir at a precipitation temperature of approximately 40·c, in t he water­
sat urated sandstones. Cement supply was from t he underlying carbonate sediments (t he Albian­
Turoni an M ac ae Formation). Oil emplacement inhibited both calcite cement ation and mec hanical
compaction within t he reservoir, and t hus explain t he higher porosity and permeability in t he oil zone
t han in t he water zone.

INTRODUCTION

Feldspathic turbidite sandstones are the main oil
reservoirs in the Campos basin, offshore Brazil. The
Namorado Sandstone (Albian-Cenomanian) is one
of these reservoirs, consisting of a thick sequence of
arenites with calcite cement being the volumetri­
cally most abundant diagenetic constituent. The
calcite occurs as concretions, which strongly reduce
the porosity and permeability distribution in this
reservoir. However, the average gas porosity of 20%
(range 17-30%) and the mean gas permeability of
120 mD (range 0.1-1600 mD) are unusually high
for present reservoir depths (> 3200 m) compared
with other reservoirs at a similar depth in the
Campos basin.
Calcite concretions generate a complex porous
system which influences fluid flow (hydrocarbons
and water), oil productivity and recovery efficiency.
Unravelling the distribution, origin and timing of
these calcite-cemented zones is important in reser-
voir evaluation of the Namorado Sandstone.
The general sedimentological and diagenetic his­
tory of the Namorado Sandstone in the Campos
basin has been discussed by several authors. Car­
valho et a!. ( 1995) studied the carbonate cementa­
tion pattern, controls on calcite qistribution and
porosity evolution of the Namorado turbidites in
the Albacora Field. These authors distinguished
seven textural types of carbonate cement with
different origins. Abreu et a!. (1992) suggested two
main sources of calcite cement: marine water for
the eodiagenetic microcrystalline calcite, and com­
pactional water sourced from the underlying
carbonates of the Macae Formation for the meso­
diagenetic sparry calcite. Further, Carvalho (1990)
and Carvalho et a!. (199 5) concluded that oil
charging was an important factor in the inhibition
of calcite cementation in the Namorado Sandstone.
These authors noticed that below the oil-water

Carbonate Cementation in Sandstones Edited by Sadoon Morad 309

© 1998 The International Association of Sedimentologists
ISBN: 978-0-632-04777-2
310 R.S. de Souza and C.M. de Assis Silva
interface, porosity was almost completely occluded
by sparry calcite cement. However, the relationship
between the origin and timing of calcite cementa­
tion relative to oil emplacement remains poorly
understood. Such information is essential to an
understanding of the diagenetic history and of the
-
reservoir quality.
The objectives of the present study are (i) to
understand the timing of cementation and the
origin of the calcite cement; (ii) to investigate the
relationships between diagenesis of the sandstones
and oil emplacement in the reservoirs; and (iii) to
characterize the main cementation types, both
above and below the oil-water interface.
SAMPLES AND METHODS
This study is based on the integrated analysis of
211 m of cores taken from two oil wells (AB-1 and
AB-2) in the Albacora Field (Fig. 1). A total of 170
thin sections were selected for detailed petrographic
investigation of the carbonate cements; 54 of these
were selected for modal compositional analyses by
counting of 300 points per thin section. Sampling
was representative of the cored intervals. The car­
bonate cements were stained with a solution of
0.15% HCl, 1.0% potassium ferrocyanide and 0.1o/o
alizarin red-S in order to determine Fe distribution.
The cathodoluminescence of the cements was ex­
amined through a TECHNOSYN luminoscope,
model CCL 8200 MK2, under the following condi­
tions: air pressure between 0.1 and 0.05 Torr,
voltage about 15 kV and electric current about
2 50 �A. In order to obtain the best photomicro­
graph results the following conditions were routinely
used: (i) a colour negative film (Kodak Ektapress
Plus, 400 ASA, 35 mm); (ii) an exposure time of
about 2 min; (iii) a reciprocity factor of 3; and (iv)
the film was processed as if it were 800 ASA.
After describing the texture, mineralogy and
cathodoluminescence features of the main types of
carbonate cements, 12 polished thin sections, rep­
resenting the different textural types of calcite, were
carbon coated and analysed using a JEOL JXA-
840A electron microscope with a back-scattered
electron detector (BSE). In order to investigate the
carbonate chemical compositions, samples were
examined by a Tracor-Northem energy-dispersive
X-ray analyser (EDS).
Stable carbon and oxygen isotopes of the different
diagenetic carbonates were determined in 22 sand­
stone samples, which were selected after careful
optical and cathodoluminescence examination.
Samples in which contamination caused by allo­
chems or mixtures of different textural types of
calcite (both easily recognized by optical petrogra­
phy and cathodoluminescence) were not considered
for isotopic analysis. The 'Kiel' system for calcite
sample preparation was used, and calcite was re­
acted with anhydrous phosphoric acid at 2 5 oC
under vacuum for 2 h, the produced C02 was
Fig. 1. Location of t he deep-water
(25 0- 2000 m) Albacora Field,
offshore C ampos basin, Brazil,
where t he t urbiditic Namorado
Sandstone is an important oil
reservoir. Isocontours shown
represent water depth.
 Carbonate cementation in the Namorado Sandstone
analysed in a Delta Finnigan MAT 252 mass
spectrometer. o180 values were corrected for o1 7 0
using the method of Craig (1957). The phosphoric
acid fractionation factor used was 1.010225. Both
o13C and o180 are reported relative to PDB.
The nitrogen permeability and porosity (follow­
ing Boyle's Law method) of 193 plugs, 1. 5 inches in
diameter, were obtained from representative litho­
facies of the Namorado Sandstone, derived from
both below and above the oil-water interface. Sam­
ples were analysed in a CORELAB porosimeter and
permeabilimeter in the petrophysical laboratory of
the Petrobras Research Centre. Permeability values
were not corrected for Klinkenberg effects.
Fluid saturation, which defines oil-water contact,
was determined using wireline log interpretation
and RFT data. Shaliness was obtained through
density/neutron log data and the salinity of the
formation water was analysed from RFT sampled
water. Additional parameters used in the calcula­
tion, such as saturation and cementation exponents
(n and m, respectively}, and tortuosity coefficient
(a}, were measured in the laboratory.
The reconstruction of the burial and thermal
history of the Namorado Sandstone was performed
using the BaSS software (Basin Simulation System)
developed by Chang et al. (1991). The calibration of
the thermal history was carried out by measured
vitrinite data and smectite/illite conversion rate.
The amount of illite in the interlayered liS clays was
determined via XRD analysis. The organic matter
residue of the associated shales close to the reservoir
intervals was petrographically analysed to obtain
the vitrinite index. Depth in the diagrams and
tables is referred to datum level (driller-measured
depth).
GEOLOGICAL SETTING
The Campos basin is a typical passive margin basin
located in offshore Rio de Janeiro State, southeast­
em Brazil (Fig. 1). It represents the most prolific oil
province of the country. The tectonostratigraphical
evolution of the Brazilian margin basins has been
described by several authors (Ponte & Asmus,
1976; Ojeda & Oyeda, 1982). Three major distinct
tectonostratigraphical sequences are recognized
(Fig. 2). The continental sequence (Neocomian)
includes volcanic rocks and coquinas of the Lagoa
Feia Formation. The transitional sequence (Aptian)
is represented by evaporites of the upper Lagoa Feia
311
Formation, which characterizes a period of relative
tectonic quiescence. The passive marine margin
sequence (Albian to Recent) is characterized by the
development of a shallow-water carbonate platform
(Macae Formation}, which prograded into deep­
water siliciclastics (Campos Formation).
The Namorado Sandstone represents a thick
interval (aveq:tge of 100 m) of siliciclastic deep­
water turbidites. These were deposited above the
Macae Formation, which consists of marls and
calcilutites (Fig. 3). The turbiditic sequence consists
mostly of massive sandstone lobes, locally associ­
ated with channel-levee complexes (Moraes, 1985,
1989; Freitas, 1987; Bruhn & Moraes, 1988; Becker
et al., 1989}, representing a lowstand fan system
(Dias et al., 1987; Abreu et al., 1992).
The Namorado Sandstone is composed primarily
of massive fine- to very fine-grained arenites. Sedi­
mentary structures are rare, but occasionj!l normal
grading is observed. The fluidization process is
identified by unusual dish and pillar structures.
Rarely, individual beds present ripple cross­
laminated divisions at the top. Individual beds have
an average thickness of 1m. Several individual beds
may be amalgamated, resulting in sandy intervals
up to 10 m thick. Individual beds or cycles are
capped by thin beds of shales and calcilutites.
The oil present in these turbidites was generated
from the Lower Cretaceous lacustrine shales of the
Lagoa Feia Formation (Dias et al., 1987; Rizzo et
al., 1990; Soldan et al., 1990) and migrated into the
reservoirs along major faults (Fig. 4). Oil in the
reservoir ranges from 25 to 30" API. At present the
reservoir temperature is about 80oC.
The Albacora Field has two main fault systems:
one of them affects both the Cretaceous and the
Tertiary sequences, the other influences only the
Cretaceous rocks (Fig. 4). The two systems resulted
from halokinetic processes (Candido & Coni,
1991). Faults are mainly normal in style and run
approximately north-south.
SANDSTONE PETROGRAPHY
The Namorado Sandstone is composed of fine- to
medium-grained, moderately to well sorted arkoses
and subarkoses (Fig. 5). The main framework con­
stituents are quartz, potassium feldspar, plagioclase
and quartz-feldspar rich rock fragments, most of
plutonic origin. Although previous work indicated
the predominance of K-feldspars (Carvalho, 1990;
 MIOCENE

>
a:<r:
OLIGOCENE

a:0UJ

;a
�

1-
u..U):E
EOCENE
0a..
�
c("0
a:c(Cl
" zUJ
PALEOCENE UJ> a:c(
:E
U)
c(U)a..

CENOMANIAN

:E
oUJc(u..
ALBIAN c(0"
U):l
0 ..J
c(
0c(UJ
w
z
0i= z-c( a:
1- :E O:z
!;;:::>
a:UJ0 APTIAN c(u..iii zu;
:::>s �
u.. a:c(1-
C5Cl
:s
BARREMIAN

Fig. 2. Stratigraphical col umn of t he C ampos basin showing the position of t he N amorado S andstone. Modified from
R angel et a/. ( 1994 ) .
 Carbonate cementation in the Namorado Sandstone
DEEP WATER TURBIDITES
Fig. 3. Schematic depositional
environment of the Namorado
Sandstone turbidites. Modified
from Barros et al. ( 1982).
LEGEND
t OIL MIGRATION PATHWAYS
� OlL ACUMMULATION
Fig. 4. Schematic model of oil migration and
accumulation in t he C ampos basin. The oil present in
these turbidites was generated from the lower Cretaceous
lacustrine s hales of the L agoa Feia and migrated into t he
reservoirs along major faults. The seal is t he M acae
Formation, which consists primarily of m arls and
calcilutites. Modified from Figueiredo ( 1985).
Abreu et a!., 1992; Carvalho et a/., 1995), recent
studies based on cathodoluminescence show that
plagioclase amounts have been underestimated.
This is probably due to the difficulty in recognizing
plagioclase by optical petrography, because most
313
L
Fig. 5. Detrital composition of the Namorado sandstone
turbidites plotted on t he McBride ( 1963) classification
diagram. S andstones are mainly arkoses and subarkoses.
plagioclase grains are untwinned and unaltered in
the Namorado Sandstone.
Together with calcite, other diagenetic features of
the Namorado Sandstone include dolomite, opal,
kaolinite, K-feldspar overgrowths, chlorite, albite,
anatase, barite, pyrite and pseudomatrix generated
by squeezing of clay clasts. The whole diagenetic
evolution is detailed by Carvalho (1990), Abreu et
a/. (1992) and Carvalho et a/. (1995).
Calcite types
Calcite is the most abundant cement in the Namo­
rado Sandstone, but is patchy at many scales of
occurrence. Commonly, calcite forms discrete and
tightly cemented concretions embedded in weakly
314 R.S. de Souza and C.M de Assis Silva
25cm
Fig. 6. Typical lithofacies of the N amorado S andstone.
Fine- and very fine-grained sandstones arranged as
t urbidite cycles, e ach c apped by s h ale and siltstone.
Pervasively calcite-cemented beds resulted from
concretion coalescence. Both individual concretions and
calcite-cemented beds are discontinuous on the l ateral
scale, as indicated by core analysis of a recently drilled
horizontal well. Concretions show v ariable s h apes and
intensity of cementation, and a grad ual vertical
transition to partially cemented and porous intervals.
or uncemented host sandstones (Fig. 6). Completely
calcite-cemented beds found in the Namorado Sand­
stone are the product of concretion coalescence, as
described by Bj0rkum & Walderhaug (1990, 1993)
for shallow marine sandstones in northern Europe.
Both individual concretions and calcite-cemented
beds are discontinuous on the lateral scale, as indi­
cated by core analysis of a recently drilled horizontal
well. The main shapes of the concretions are tabular,
ellipsoidal and irregular, with the intensity of occur­
rence increasing from the top to the base of the res­
ervoir (Sombra et a!., 1995).
Fig. 7. (A) Blocky zoned meteoric c alcite cement (type I)
filling mouldic pore microcrystalline type II calcite
(white arrow). Note intense bioclast replacement and
l oose packing (packing proximity = 36%); uncrossed
polars. (B) S ame field of view observed with
c at hodol uminescence: type I calcite presents s h arP CL
zonation (black arrow), whereas type II c alcite s hows CL
microzoning (white arrow). Note the high content of
K -feldspars and plagioclase. Q uartz grains do not show
l uminescence. AB-1 well; 324 4 . 4 5 m; oil zone.
The Namorado Sandstone concretions include
intergranular calcite cements with textures varying
from microcrystalline to poikilotopic coarse calcite
mosaic. In addition, blocky calcite mosaic is replac­
ing bioclasts. Calcite cement is divided into three
types, based on textural features, cathodolumines­
cence appearance and isotope signatures. Type I fills
mouldic pores generated by bioclast dissolution, or
forms rims around the inner surface of micritic
envelopes of mouldic pores. Under cathodolumi­
nescence (CL) this calcite appears as large crystals
showing a well-defined concentric oscillatory zona­
tion pattern, which varies in colour from yellow to
dark orange (Fig. 7A,B). Because of difficulties in
 Carbonate cementation in the Namorado Sandstone
separating this type of calcite, no isotope analysis
was performed.
Type II calcite occurs as microcrystalline ( 1-
60 J.lm) aggregates that exhibit complex, thin CL
zonation (Fig. 7A,B). It is more common in the oil
zone, yet its occurrence is mainly related to the
unevenly distributed bioclasts and bioclastic
moulds in the cored interval. Below the water-oil
interface calcite II is scarce and occurs always in
association with type III (Fig. 8). Type II calcite has
oxygen isotope compositions ranging between -5o/oo
and -1o/oo.
Type III calcite occurs in the water zone as
coarsely mosaic (100-500 J.lm) and locally poikilo­
topic (> 500 J.lm). It exhibits a homogeneous yellow
cathodoluminescence (Fig. 9). In thin sections, type
II calcite shows neither dissolution features nor
evidence of replacement by type III calcite. The
contact between both is often sharp. This relation­
ship is most obvious in the water zone, where types
II and III occur together. Although zones cemented
by type III calcite show normal grain packing (43%),
the packing proximity index (Kahn, 1956) is higher
than the loose grain packing (34%) measured in
zones cemented by type II calcite. Type III calcite
has overall lower 8180 (-9o/oo to -5o/oo) than the
other types of calcite cement.
In the Namorado Sandstone the intergranular
calcite cement (types II and III) have o13Cp08
values ranging from I o/oo to 3o/oo (Table I; Fig. I 0).
Although the range of oxygen isotope values for
the Namorado concretions is large (-9o/oo to -2o/oo),
there is a significant difference between the average
oxygen isotope composition of the concretions
above and below the oil-water interface: 4. 5o/oo and
-7o/oo, respectively (Figs 8 and II).
In thin sections from the oil zone, we observe that
calcite II is associated with oil staining. On the
other hand, in the water zone type III calcite shows
no association with oil. Below the water-oil contact,
in zones where there is no or only scarce calcite
cement development, mechanical and chemical
compaction was the main diagenetic process re­
sponsible for porosity loss.
Chemical composition of calcite cements
Types II and III calcite cements contain a trace
element composition (Table 2) that differs from the
calcite that replaces bioclasts (type 1). Calcite types
II and III have greater values of (MgO + SrO)/CaO
than (MnO + FeO)/CaO, whereas type I calcite dis-
315
plays lower values of (MgO + SrO)/CaO than
(MnO + FeO)/CaO (Fig. 12). Type I calcite revealed
a (MgO + SrO)/CaO ratio of 0. 77, whereas types II
and III calcites have values of 1.0I and 1.08,
respectively. In contrast, calcite type I has a
(MnO + FeO)/CaO ratio of 0.87, whereas types II
and III calcite have ratios of 0.67 and 0.81, respec­
tively.
Burial history
The burial history of the Namorado Sandstone
(Fig. 13) is relatively simple. The unit was depos­
ited during the Albian-Cenomanian and was grad­
ually buried, remaining at shallow depths (less than
1000 m) for the initial 75 Myr. From the Oligocene
to the Recent, the Namorado Sandstone underwent
accelerated subsidence until it attained its present
depth (3200 m). This relatively recent high subsid­
ence rate is due to the effect of sedimentary loading
of the lithosphere in response to a high clastic
supply (Chang et at., 1992).
DISCUSSION
Depth and timing of calcite precipitation
The distribution patterns and diagenetic evolution
of calcite cement in sandstone sequences are often
complex and controlled by several factors, includ­
ing:
1 flow paths with different permeabilities due to
primary textural attributes (James, 1985; Pirrie,
1987);
2 distribution of shell-rich layers, which function as
a source of carbonate and also serve as nuclei for
cementation (Davies, 1967; Bj0rkum & Walder­
haug, 1990);
3 distribution of previously formed carbonate
crusts, which serve as nuclei for later cements
(Kantorowicz et a!., 1987);
4 the relative influence of diffusion versus advec­
tion during cementation process (Bj0rkum &
Walderhaug, 1990);
5 association of calcite-cemented bodies with folds
and faults (Johansen, 1993).
Two hypotheses have been put forward to explain
the origin of calcite concretions in the Namorado
Sandstone: (i) growth was essentially controlled by
local sources of carbonate and early bacterial pro­
cesses, with the initial calcite precipitating either at
316 R.S. de Souz a and C.M. de Assis Silva

Fig. 8. Continuously cored interval

with high recovery (I 00%) of t he
Namorado Sandstone composed of
numerous indi" v idual turbidite
layers. O bsei'Ve t hat density,
neutron and soni c logs clearly
show decreasing porosity values
below t he oil-water interface.
Arrows i ndicate t he strongly
cemented sandstones. Note also
t hat t ype II cal cite commonly
occurs in t he oil zone, whereas
t ype III dominantly appears below
t he oil-water con tact.
 Carbonate cementation in the Namorado Sandstone 317

� 13(%.
0
-2 -1 2 3 4 5.
-1

-2 •
•
-3-

-4.

-5
.I •
•
•
-6 •
•
•
7
•
- -
..
•
-8 e CALCITE II
•
o"O%o • +CALCITE Ill
-9 • CALCITES II + Ill

Fig. 10. o13Cp08 versus o180p08 compo&ition 9f the

intergranular carbonate cement s in t he Namor;tdo
Sandstone of t he Albacora Field.

Table 1. I sotope c;:omposition of t he cal cite eement

(t ype s II and III) of Namorado Sandstone in a sin gle
well (%o PDB)

Depth (m) Calcite t ype

3216.85 II 2,28 -3.91

Fig. 9. (A) Coarse mosai c t ype III cal cite, l ocally
3220. 35 III 2,65 -7.29
showing a poikilotopi c text ure. This t ype of cal cite i s
3231. 15 III 1.77 - 5 . 87
very common below t he oil-water interface (see Fi g. 8).
3233. 5 8 II 1. 86 -2.15
Reservoir intervals which are totally cemented by t ype
3237. 00 II 2.15 - 4 . 36
III cal cite have hi gher packing proximity values
3239.05 II 2. 2 -4. 61
(a verage= 4 5 %) t han reservoirs which are cemented
3243.93 II 4 . 08 -4._56
by type II c;:al cite (uncrossed polars). (B) CL
3249. 4 5 II 2.28 -4. 1:J
photomi crography of t he same field of view as shown
325 6.35 II 3,2 -4. 84
abo ve. O bserve t he cathodoluminescence pattern with
3263. 00 II 2.29 - 4 . 25
homogeneous te xt ure. AB-2 well; 3281. 75 m ; water
3285. 00 III 2. 29 -8. 5 !
zone.
3286. 90 III 1. 81 -7. 32
3290. 80 III 1.92 -2. 78
3291. 35 III 1. 82 - 4 . 55
3291.70 III + II 1.83 -7.81
or near the sediment-water interface (Carvalho, 3295 . 60 III + II 0.77 -6. 21
1990; Carvalho et al., 1995), acquiring the Albian 3300. 70 III + II - 1. 05 �5 . 5 8
marine water isotope signature; (ii) growth was due 3308. 4 5 III 1.94 .... 7.38
to a flow of cation-rich fluids generated by compac­ 3317. 00 III 1. 5 -6.58
3323. 00 III 2.04 -7.23
tion of the enclosing calcilutites of the Macae 3326. 00 III 1.84 -8. 24
Formation, following initial burial of this formation
(Moraes, 1985).
The mean temperature of precipitation of the two
intergranular types of calcite cement (II and III) was calcilutites of the Macae Formation by Spadini et al
calculated using the equation of O'Neil & Clayton (1988). The precipitation temperature of type II
(1964), assuming a value of -2%o for the oxygen calcite, which is dominant in the oil zone, is c.
isotope composition of the original pore water, 25•C, whereas for type III calcite, dominant in the
which is the average value that was determined for water zone, it is c. 40•c.
318 R.S. de Souza and C.M. de Assis Silva

li"C %o(PDB) li"O%o(PDB)

-4 -3 ·2 -1 0 1 2 3 4 5 6 ·9 -8 -7 ·6 -5 -4 ·3 ·2 ·1 0
�-r+.r4-r��3200
3200 1.2

• •
3220 • • 3220
• •
• i •
I !�
0.8
3240 3240
•
•
�·
•
0.6
•
3260 •
:·
•
3260.
0.4

3280 Oil 3280 0.2

--�----------.----y-------
• WATER • !
• i. • •
33fl • •
•! 3300
0
Ill
• •
Calcite type
• •
3320 • •
3320
• • Fig. 12. Comparative semiquantitative chemical
composition of t he t hree different textural types of
3340 13340 calcite identi fied in t he Namorado S andstone. The
vertical axis represents t he proportions of
depth (m) CALCITE It CALCITES 11+111 depth (m)
(MgO + SrO)/CaO and (FeO + MnO)/CaO.
• •

+CALCITE Ill

Fig. ll. Variation in the stable isotope composition of

the different types of intergran ul ar calcite, relative to The 813C values of calcite cement are similar to
depth and t he modern-day position of t he oil-water
contact.
those of the calcilutites ( +3%o to +4o/oo) of the Macae
Formation (Fig. 14) (Spadini et a!. 1988). These
authors considered these values to reflect the isotope
The depths of calcite precipitation were esti­ composition of a hypersaline Albian sea, which pre­
mated using the calculated temperatures of the two vailed during deposition of the Macae Formation.
intergranular calcite types (II and III) and the burial The presence of evaporites just below the Macae
subsidence and temperature history of the sedi­ sequence (Fig. 2) was believed to support this.
ments (Fig. 13). Type II calcite is inferred to have Although the carbon isotopic signature of calcite
formed between about 500 and I 000 m, and type cement in the Namorado Sandstone could indicate
III between approximately I 000 and 1500 m. This a marine origin, sea-water diffusion downwards to
depth range conflicts with the hypothesis of calcite the turbiditic sandstone is unlikely because the
II formation in a marine environment after initial 8180 values (Table 1) suggest precipitation at a
burial. Therefore, it is more likely that the Macae greater temperature and depth than the surface, and
calcilutite/calcarenite supplied ions for the forma­ the impermeable enclosing calcilutite should have
tion of the concretion. blocked the sea-water percolation to the turbiditic
A simple mass-balance calculation (Table 3) was sandstone. Moreover, the intergranular volume­
done to check whether if Macae Formation carbon­ cement diagram (Fig. 15) indicates that about 15%
ates could supply the amount of calcite necessary to of the average original porosity of the Namorado
cement Namorado Sandstone in the Albacora Field. Sandstone has been destroyed by compactional
The calculation shows that Macae Formation orig­ processes prior to the first event of intergranular
inal water could supply a mass of 4.19 x I 010 kg of calcite cementation (type II calcite). This porosity
dissolved CaC03 at 4o·c, which is greater than the loss indicates that calcite precipitation was after
calculated mass of 1.32 x I 09 kg of calcite cement of some sediment burial.
the Namorado Sandstone. Therefore, the Macae The proximity of the turbiditic deposits with the
carbonate rocks are a possible source for the calcite underlying calcilutites of the Macae Formation and
cement of the Namorado Sandstone. a suitable hydrologic gradient, favoured by faults,
Table 2. Calcite cement composition based on EDS analysis

CaO (wt%) M gO (wt %) FeO (wt %) MnO (wt%) SrO (wt%)

Dept h (m) n Calcite type Av. Ran ge Av. Ran ge Av. R an ge Av. Ran ge Av . R an ge

324 4 . 4 5 7 I 98. 82 98. 35-99.78 0. 38 0-0. 67 0. 4 1 0-0.66 0.24 0-0. 4 5 0. 20 0-0.71 Q

3243. 5 0 I I 99. 79 0. 19 0. 02 0.00 0. 00 Cj.
0
324 5 . 80 2 I 98. 75 98. 22-99. 27 0. 49 0. 21-0. 76 0. 26 0. 21-0. 3 0. 4 1 0- 0. 82 0. 11 0-0.22 ;::,:
3291. 35 I I 98. 69 0. 05 0. 34 0. 5 4 0. 38 !::>
;:;;
324 4 . 4 5 6 I 98.00 97. 06- 99. 60 0. 5 3 0-0. 93 0. 89 0- 1. 7 0. 14 0-0.23 0. 4 4 0-1. 24 r,
324 3. 5 I I 98.18 0.27 0. 74 0. 39 0. 4 1 ""
324 5 . 80 2 I 97. 73 97. 39-97. 62 0. 5 2 0. 45-0. 84 0. 72 0. 03-1. 4 0. 5 3 0. 47-0. 74 0. 4 9 0.05-1
�
""
3243. 5 0 2 I 98.73 98. 01- 99. 4 5 0. 4 7 0.15-0.79 0. 20 0- 0. 4 0. 11 0.09-0.13 ;::,:
0.36 0- 0.71
�
324 4 . 4 5 13 II 98. 5 5 96. 77- 99. 4 9 0. 4 9 0-0.87 0. 45 0.07-1.11 0. 15 0-0. 39 0.36 0-1. 5 7
3243. 5 0 4 II 98.00 97. 01-99. 63 1:21 0. 29-1.83 0.33 0. 08-0. 4 6 0. 25 0-0. 74 0. 22 0-0.65 6·
;::,:
324 5 . 80 II II 98. 24 97. 22-99. 4 8 0. 5 9 0. 02-1. 4 6 0.63 0- 1. 2 0.13 0-0. 5 3 0. 41 0- 1. 31 s·
3220. 35 I III 97. 31 0.86 0. 4 0 0. 4 2 1.01
I ;;.
3224. 35 III 98. 03 0. 4 0 0.61 0. 06 0.90 ""
3225 . 35 I III 97. 73 0. 66 0. 5 4 0.00 1. 08
3226. 35 I III 99.37 0.00 0. 03 0. 48 0. 12 �
3229. 35 I III 97. 91 0. 77 0. 19 0.27 0. 85
�
0
3231. 35 I III 98. 24 0. 63 0. 4 5 0.14 0. 5 5 �
3281. 75 7 III 98. 4 6 96. 5 7- 99. 5 7 0.32 0- 0. 90 0. 4 1 0-0.97 0.35 0.01-0. 79 0. 4 7 0- 1. 42 �
3285 . 00
3291. 35
4
I
III
III
98.28
98.39
97. 93-98.79 0. 4 2
0. 5 0
0. 06-0.75 0. 5 8
0. 09
0. 23-0.88 0. 36
0. 4 4
0. 15 -0. 71 0.37
0. 5 7
0.12-0.68
�
;::,:
3221. 35 I III 96.85 0. 88 1. 4 0 0.34 0. 5 3 �
3222. 35 I III 97. 01 0. 76 1. 37 0.65 0. 21 0
;::,:
3223. 35 I III 96. 5 7 0. 65 1. 96 0. 25 0. 5 6 ""
3227. 35 I III 96. 85 0.32 1. 90 0. 23 0.69
3228. 35 I III 96. 5 6 0.84 1. 73 0. 35 0. 5 2
3291. 35 2 III 98. 01 97. 83-98. 19 0. 5 8 0. 24 -0. 92 0. 5 3 0. 17-0.89 0. 31 0. 26-0.36 0. 5 7 0.32-0.82

n, number of analyses in each sample.

\0
320
100
(A)
TIME(M.a.)
� Calcite I precipitation
• Calcite II precipibltion
r:Z2 Calcite Ill precipit.atjon
CJ Reservoir
' (B)
provided the ideal conditions for upward fluid
movement from the calcilutite to the sandstone
(Fig. 4). After reaching the reservoir, the fluid flow
was focused along the direction of bedding plane
following the most permeable pathways. Thus, this
flow should have favoured the greatest concretion­
ary growth along the direction of bedding plane,
which was responsible for the formation of appar­
ently tabular (laterally extensive) concretions (Som­
bra et a/., 1995) within the Namorado Sandstone,
similar to those observed in outcrops of Tertiary
sandstones in the northern Apennines (McBride et
at., 1995).
Further, the availability and proximity of a large
source of carbonate material, namely the bioclasts
and carbonate intraclasts within the Namorado
Sandstone, probably provided supersaturation con­
ditions and sites of nucleation for intense calcite
cementation. Sombra et at. (199 5) attributed the
tabular geometry of calcite-cemented zones to the
lateral coalescence of concretions in carbonate clast­
rich units in the Namorado Sandstone.
R.S. de Souza and C.M. de Assis Silva
t
1.0 Ci
"'
"' =
�z �
0 �
.s
�
�
TIME
o SCALE
Fig. 13. Burial and t hermal history
diagram of t he Namorado
t urbidites in t he Albacora Field
RfSEKVOII.
LOWEitllMIT ( B) . Note in t he relati ve sea-level
curve of Vail et a/. ( 1977) (A) t hat
at t he time of t he t urbidite
depositi on t here was a signi ficant
sea-level fall (Cenomanian). The
low sea-le vel stand favoured
meteoric water invasi on at this
time.
The geochemical composition of the different
calcite types (Table 2) suggests that type I may have
been precipitated under a meteoric influence be­
cause the Sr/Ca and Mg/Ca ratios of marine water
are higher, and the Fe/Ca and Mn/Ca ratios lower
than those of meteoric water (Veizer, 1983). More­
over, the oscillatory cathodoluminescence patterns
identified in the type II calcites have been ascribed
to oscillations of Eh in meteQric environments
(Moore, 1989).
Even though this sediment represents marine
offshore deep-water turbidites, meteoric flow could
have reached the reservoir just after turbidite dep­
osition, during a sea-level lowstand (Fig. 13), when
both the platform and part of the slope were
partially exposed. The relatively low burial rate
during the early history of the Namorado Sandstone
favours this hypothesis. Meteoric waters resulted in
the formation of blocky calcite that replaces bio­
clasts or occurs as the inner rims of mouldic pores
(see Carvalho et at., 1995).
 Carbonate cementation in the Namorado Sandstone 321

CALCITE CEMENT(%)

0 5 10 15 20 25 30 35 40 45
DEPTH GAMMA
(m) RAY
I SONIC
45

40

� 35
w
::;;
:J 30
....1
0
>
a: 25
::i
:J 20
z
<( intergranular
2550- a:
CJ 15 porosity(%)
a:
w
1-
<
'W
� 10
u
<
::E e oil zone
.. 2600- 5
� 6"0
D waterzone
9 0
0 50 100

Original porosity destroyed by

2650-
cementation {%)

Fig. 15. Intergranular volume versus calcite cement

amounts in the Namorado Sandstone, based on the
diagram of Houseknecht ( 1987).
- 2 - 1 0 1 2 3 4 5 013C PDB
-1 -2 -3 -4 -5 -6 -7 o'"O,oa

LEGEND

c::=J <;ALCARENITES - SHALES

Table 3. Mass-balance calculations
c::=J CALCISSILTITES - MARLS

[3 C�LCILUTITES .
·
Data considered:
Albacora Field area= 40 km2

Fig. 14. General trend for composition of carbon and
oxygen isotopes in t he calcilutites of t he Macae
Formation. Modified from Spadini et a/. ( 1988).
Oil emplacement: influence on carbonate
cement distribution.
Several authors have point.ed out that relatively
early hydrocarbon saturation can retard diagenetic
processes (Nagtegaal, 1980; Dixon et al., 1989; De
Ros, 1990; Saigal et al., !992; Gluyas et al., 1993).
Other authors, however, have concluded that hy­
drocarbon saturation can reduce but probably not
completely halt diagenetic reactions, owing to the
small water film that borders most grains, providing
a medium for ionic transfer (Walderhaug, 1994).
Experimental analysis performed by De Boer
(1974), for instance, demonstrated that fluids
within a porous space are important in porosity loss
Macai Formation
Thickness= 600 m �rock volume = 2.4 X I 010 m 3
Macae calcilutite volume ( 25 %) = 6. 0 x I 09m 3
Macae calcarenite volume ( 75%) = 1.8 x 1010 m3
Calcilutite depositional porosity= 60% (Ginsburg, 1956;
apud Scho lle & Halley, 1985)
Calcarenite depositional porosity= 40% (Enos &
Sawatsky, 1979, 1981; Halley & Harris, 1979; apud
Scholle & Halley, 1985)
Primary volume of Macae Fm water= ( 6. 0 x I 09x 0.6) +
( 1. 8 X 1010 X 0.4) = 1.08 X 109m 3 ,
Calcite solubility in a 1. 0 molar NaC l solution at
40"C = 3.88 gil (Atkinson & Raju, 1991)
Primary mass of dissolved CaC03 = 4.19 x 1010 kg
Namorado Sandstone
Thickness= 70 m �rock volume = 2.8 x 109m3
Porous sandstone volume ( 5 5 %) = 1. 5 4 x I 09m3
Cemented sandstone volume ( 45%) = 1.26 x I 09m3
Average calcite content of porous sandstone= 7%
Average calcite content of cemented sandstone= 30%
Total volume of calcite cement = ( 1.54 x I 09x 0. 07) +
( 1.26 X 109X 0.30) 4. 86 X 108 m 3
=

by mechanical compaction. The author pointed out Calcite density= 2. 71

that porosity loss due to compaction is reduced
Total mass of calcite cement= 1.32 x 1()9 kg
when the fluid is oil rather than water.
322 R.S. de Souza and C.M. de Assis Silva

Type II calcite started to form before oil emplace­
ment (Fig. 13), partially controlled by bioclast oc­
currence in the sediment. The oil migration to the
reservoir occurred during or after the Early Tertiary
(Rizzo et a!., 1990; Soldan et a!., 1990). In the water
zone, cementation of type III calcite went on oc­
cluding the reservoir porosity, as indicated by the
calculated crystallization temperature of 40'C. The
paragenetic relationship between type II calcite and
oil staining observed in thin section, and the lack
of this relationship concerning type III calcite,
matches this interpretation.
Density, neutron and sonic logs show a distinct
shift below the oil-water interface identifying re­
duced porosity because the frequency of calcite­
cemented beds is greater below than above the
interface. We also note that type II calcite occurs
mostly above the oil-water contact, whereas type
III calcite appears below this limit (Fig. 8). Abreu et
a!. (1992) suggested that the interval below the
water-oil interface was continuously cemented by
calcite after oil had filled the crest structure.
A plot of porosity and permeability data against
depth (Fig. 16) clearly shows that diagenetic pro­
cesses (cementation plus compaction) have prima­
rily affected permeability below the oil-water
contact. This is expected, as the coarse type III
calcite and compaction have produced a poorly
connected pore system. Although permeability
values are very low, the remaining porosity indi­
cates that the lower portion of the Namorado
Sandstone (below the oil-water contact) is not a
seal.
When plotting calcite texture and isotope compo­
sition against depth (Fig. 11), the data suggest that
there was a strong influence of oil charging on the
diagenetic evolution of the Namorado Sandstone.
The plots show that type II calcite with higher 8180
values prevails in the oil zone, whereas type III with
lower 8180 values prevails in the water zone. In
fact, the diagenetic evolution cannot be entirely
understood if the oil emplacement history is not
considered. It is suggested here that within the oil
zone, partial cementation by type II calcite and oil
emplacement prevented further mechanical and
chemical compaction and maintained good reser-

Porosity (�•) Permeability (mD)

0 10 20 30 40 0 500 1000 1500 2000

3200 3200
average permeability=

3220 • • •• • • 3220 •• • 137.1mD

n=106

· •
• .#.
• • 3240
3240 · , .. . � �. •
�· • • • •. •. • • •• • •
� .
3260
.. .. �.
� 3260
�

.§.
•• �

.§.
• �>-•• ·
• ••
.c:
.c:
'ii. 'ii.
" •
"
3280 �� 3280
•• • t,rc
� c
c

•�• �··..· ..
__..

• •• r... 3300
average permeability=
24.5mD
3300 • $ # ••• n=90

•. .. .
•
• .�
• . 3320
3320 ,
r�. • •
•• • •
3340 3340

Fig. 16. Helium permeability and porosity versus depth, showing generally higher val ues of permeability and porosity
above the oil-water interface due to contin uation of cementation and compaction in the water zone.
 Carbonate cementation in the Namorado Sandstone 323

voir properties. Furthermore, precipitation of cal­ between calcite and packing proximity index in
cite at different successive periods of the burial both zones, suggesting that calcite cementation was
history probably explains the different 8180 values relatively early.
of calcite cement below (-9o/oo to -So/oo) and above -

(-So/oo to -1 o/oo) the oil-water interface in the Namo­

rado Sandstone. CONCLUSIONS
The relative importance of compaction and ce­
mentation processes in porosity reduction was eval­ Three types of calcite cement were detected in the
uated using the graphical method of Houseknecht Namorado Sandstone: (i) blocky calcite with an
(1987). Calcite cementation was the dominant oscillatory zoning pattern, formed by bioclast re­
diagenetic process affecting porosity reduction. placement and in response to meteoric water inva­
However, the relative importance of compaction sion during eodiagenesis; (ii) a very fine calcite
processes increases below the oil-water contact mosaic showing microzoning, formed by nucleation
(Fig. 15). This interpretation is in agreement with on bioclasts and cation supply from underlying
burial and thermal histories, and with the diage­ carbonate sediments; (iii) a coarse mosaic with
netic evolution of the calcite cement and the rela­ poikilotopic texture formed by cation supply from
tive timing of oil emplacement. the carbonate sediments, later in diagenesis.
In order to better evaluate the effects of compac­ The type II calcite is the most common cement in
tion on porosity loss, the packing proximity index the Namorado Sandstone. It was precipitated rela­
(Kahn, 19 56) was measured in 44 samples derived tively early, at an estimated temperature of 25 oc,
from both oil and water zones in the Albacora Field. prior to oil emplacement, whereas type III calcite
Plotting these data against calcite cement volumes was precipitated after oil charging in the water zone,
(Fig. 1 7) indicates that for the same calcite cement at an estimated precipitation temperature of 40"C.
content, the packing proximity index is lower in oil Below the oil-water interface mechanical and
zone samples than in samples derived from the chemical compaction, together with type III calcite
water zone. This means that compaction was less cementation, played an important additional effect
intensive in the oil zone than in the water zone. in occluding porosity. The data suggest that some
Furthermore, there is a relatively good correlation porosity was lost by mechanical compaction prior
to initial calcite cementation (type II).
Isotopic data and mass-balance calculations indi­
cate that the underlying Albian-Turronian carbon­
60 ate rocks were the principal carbonate source for
calcite precipitation within the reservoir.
The average porosity and permeability values of
50
the oil zone are higher than those of the water zone
><
Cll because of the inhibition of late calcite cementation
'C
.5 40 (calcite type III) and mechanical compaction, owing
� to oil emplacement within the reservoir.
.E
)(
'
30
0
...
a.. ACKNOWLEDGEMENTS
01
c
:;: 20
u The authors thank PETROBRAS for support, ac­
as •
a.. cess to data, samples and permission to publish this
10
• oil zone work. The paper was measurably improved by the
o water zone
----water zone trendline suggestions of Dr Marco A.S. Moraes, Cristiano L.
-- oil zone trendline Sombra and Alberto S. Barroso. Revisions and
0
20 30 40
suggestions by Drs Sadoon Morad, Ian Hutcheon
0 10
and Jorg P. Schulz-Rojahn helped to substantially
Calcite(%) improve the manuscript. We are grateful to Dr
Fig. 17. Packing proximity versus c alcite cement content Rene Rodrigues, Dr Sylvia M.C. Anjos, Maria C.M.
for samples derived from both t he oil and water zones. Bezerra, Francisca F. Rosario and Carlos J. Abreu
324 R.S. de Souza and C.M. de Assis Silva
for useful discussions. We thank Ailton L.S. Souza
for aid with the SEM/EDS analysis, Tikae Takaki
for helping with isotope analysis, Ricardo Bedregal
for assistance with the burial history reconstruction,
and Hilda V. Barbosa for drafting the figures.
REFERENCES
ABREU, C.J., ARIENTI, L.M., BACKHEUSER, Y., CANDIDO, A.
& ADAMS, T. ( 1992) Qualidade dos Reservat6rios Tur­
bidaicos do Arenito Namorado no Campo de Albacora­
Bacia de Campos. PETROBRAS/CENPES/DEPEX
Internal Report, 182 pp.
ATKINSON, G. & RAJU, K.U.G. ( 1991) OKSCALE Program
Manual.
BARROS, M.C., GUEIROS, E., APPI, C., DELLA FA VERA, J.C. &
FREITAS, L.S.C. ( 1982) Ditribuicao e Modelo Deposicio­
nal das Areias Albo-Santonianas e Campo­
Maestrichtianas da Bacia de Campos, Vols I and II.
PETROBRAS/DEPEX/CENPES/DEPRO Internal Re­
port.
BECKER, M.R., ARIENTI, L.M. , BONET, L. et a/. (198 9)
Caracterizacao Geologica e Petroffsica do Reservat6rio
Principal (Membra Carapebus Cretdcico) do Campo de
Pargo. Bacia de Campos, Rio de Janeiro. PETRO BRAS/
CENPES/DEPEX Internal Report, 12 1 pp.
BJ0RKUM, P.A. & WALDERHAUG, 0. (1990) Geometrical
arrangement of calcite cementation within shallow ma­
rine sandstones. Earth-Sci. Rev., 29, 145- 161.
BJ0RKUM, P.A. & WALDERHAUG, 0. (1993) Isotopic com­
position of a calcite-cemented layer in t he Lower Juras­
sic Bridport Sands, sout hern England: implications for
formation of laterally extensive calcite-cemented layers.
J. sediment. Petrol. , 63, 678-682.
BRUHN, C.H.L. & MORAES, M.A.S. (198 8 ) Turbiditos
Brasileiros: caracterizacao geometrica e faciol6gica. In:
Proceedings of the XXV Cong. Bras. Geol. , Belem-PA, 2,
22 4-238.
CANDIDO, A. & CoRA, C.A.G. (1991) The Marlim and
Albacora giant fields, Campos Basin, offshore Brazil. In:
Giant Oil and Gas Fields ofthe Decade 1 978-1 988 (Ed.
Halbouty, M.T.). Mem. Am. Ass. Petrol. Geol., Tulsa,
54, 123-135.
CARVALHO M.V., DE Ros, L.F. & GoMES, N.S. (1995)
Carbonate cementation patterns and diagenetic reser­
voir facies in the Campos Basin Cretaceous turbidites,
offs hore eastern Brazil. Mar. Petrol. Geol. , 1 2 , 741-758.
CARVALHO, M.V.F. (1990) Diagenese dos arenitos turbid{t­
icos cretdcicos no compartimento Norte-Nordeste da
Bacia de Campos, Brasil. Unpubl. MSc t hesis, Univer­
sidade Federal de Ouro Preto, 182 pp.
CHANG, H.K., BENDER, A.A., MELLO, U.T. & KOWSMANN,
R.O. (1991) Versao 2. 0 do Man ual do Sistema de
Simulacao de Bacias. PETROBRAS/CENPES/DIVEX
Internal report, 412 pp.
CHANG, H.K., KOWSMANN, R.O., FIGUEIREDO, A.M.F. &
BENDER, A.A. ( 1992) Tectonics and strati graphy of t he
East Brazil Rift system: an overview. Tectonophysics,
2 13, 97-138.
CRAIG, H. (1957) Isotopic standards for carbon and
oxygen correction factors for mass spectrometer analy­
sis of carbon dioxide. Geochim. Cosmochim. Acta, 12,
133- 149.
DAVIES, O.K. ( 1967) Shelf sedimentation: an example
from t he Jurassic of Britain. J. sediment. Petrol. , 12,
1334-1370.
DE BoER, R.B. (1974) Thermodynamical and experimen­
tal aspects of pressure solution. In: Proceeding of Inter­
national Symposium on Water-Rock Interactions. (Eds
Cadek, J. & Paces, T.), pp. 38 1-38 7. Geological Survey,
Prague.
DE Ros, L.F. ( 1990) Preservacao e geracao de porosidade
em reservat6rios clasticos profundos: uma revisao. Bol.
Geoc. PETROBRAS, 4, 38 7-404.
DIAS, J.L., VIEIRA, J.C., CATTO, A.J. et a/. (198 7) Estudo
Regional da Formacao Lagoa Feia. PETRO BRAS Inter-
nal Report, 143 pp. .
DIXON, S.A., SUMMERS, D.M. & SURDAM R.C. (198 9)
Diagenesis and preservation of porosity in Norphlet
Formation (Upper Jurassic), sout hern Alabama. Bull.
Am. Ass. Petrol. Geol. , 73, 707-728.
ENos, P. & SAWATSKY, L.H. ( 1979) Pore space in Holocene
carbonate sediments (abstract) Bull. Am. Ass. Petrol.
Geol. , 63, 445.
ENos, P. & SAWATSKY, L.H. (198 1) Pore networks in
Holocene carbonate sediments J. sediment. Petrol. , 5 1 ,
961-985.
FIGUEIREDO, A.M.F. (198 5) Geologia das Bacias Brasilei­
ras. In: Avaliacao de Formacoes no Brasil (Ed. Viro,
E.J.), pp. I -1-I-38. Sc hlumberger Spec. Pub!.
FREITAS, L.C.S. (198 7) Estudo de Reservat6rio do Membro
Carapebus (Cretdceo), Campo de Carapeba, Bacia de
Campos, Estado do Rio de Janeiro, Brasil. Unpublis hed
MSc t hesis, Universidade Federal de Ouro Preto,
126 pp.
GINSBURG, R.N. ( 1956) Environmental relationships of
grain size and constituent particles in some south
Florida carbonate sediments. Bull. Am. Ass. Petrol.
Geol. , 40, 2 38 4-2 42 7.
GLUYAS, J.G., ROBINSON, A.G., EMERY, D., GRANT, S.M. &
OxTOBY (1993) The link between petroleum emplace­
ment and sandstone cementation. In: Petroleum Geol­
ogy of Northwest Europe: Proceedings of the 4th
Conference (Ed. Parker, J.R.), pp. 367-375.
HALLEY, R.B. & HARRIS, P.M. ( 1979) Freshwater cementa­
tion of a 1000-year-old _ oolite. J. sediment. Petrol. , 49,
969-988.
HousEKNECHT, D.W. (198 7) Assessin'g t he relative impor­
tance of compaction processes and cementation to
reduction of porosity in sandstones. Bull. Am. Ass.
Petrol. Geol. , 7 1 , 633-642 .
JAMES, W.C. (198 5) Early diagenesis, At herton Formation
(Quaternary): a guide for understanding early cement
distribution and grain modifications in non-marine
deposits. J. sediment. Petrol. , 55, 135- 146.
JOHANSEN, S.J. (1993) Depositional and structural controls
on t he diagenesis of · Lockhart Crossing reservoir
(Wilcox), Gulf Coast of Louisiana (U.S.A.). In; Marine
Clastic Reservoirs: Examples and Analogs (Eds Moslow,
T.F. & Rhodes, E.G.), pp. 117- 134. Springer-Verlag,
New York.
KAHN, J.S. (1956) The analysis and distribution of the
'
properties of packin g in sand-size sediments: I . On the
 Carbonate cementation in the Namorado Sandstone
measurement of packing in sandstones. J. Geol. , 64,
385-395.
l<ANTOROWICZ, J.D., BRYANT, J.D. & DAWANS, J.M. ( 198 7)
Controls on the geometry and distribution of carbonate
cements in Jurassic sandstones: Bridport sands, south­
ern England and Viking Group, Troll Field, Norway. In:
Diagenesis of Sedim'entary Sequences (Ed. Marshall,
J.D.). Spec. Pub!. Geol. Soc. London, 36, 103-118.
McBRIDE, E.F. ( 1963) A classification of common sand­
stones. J. sediment. Petrol. , 33, 664-669.
McBRIDE, E.F., MILLIKEN, K.L., CAVAZZA, W. et al. ( 1995)
Heterogeneous distribution of calcite cement at the
outcrop scale in Tertiary sandstones, Northern Apenni­
nes, Italy. Bull. Am. Ass. Petrol. Geol. , 79, 1044-1063.
MooRE, C.H. ( 198 9) Carbonate Diagenesis and Porosity.
Elsevier, New York, 338 pp.
MORAES, M.A.S. ( 198 5 ) Petrologia dos arenitos turbidfticos
cretdcicos e tercidrios da Bacia de Campos. MSc thesis,
Universidade Federal de Ouro Preto, 112 pp.
MORAES, M.A.S. ( 198 9) Diagenetic evolution of
Cretaceous-Tertiary turbidite reservoirs, Campos Ba­
sin, Brazil. Bull. Am. Ass. Petrol. Geol. , 73, 5 98-612.
NAGTEGAAL, P.J.C. ( 1980) Diagenetic models for predict­
ing clastic reservoir quality. Rev. Inst. Invest. Geol. , 34,
5-19.
O'NEIL, J.R. & CLAYTON, R.N. ( 1964) Oxygen isotope
geothermometry. In: Isotope and Cosmic Chemistry
(Eds Craig, H., Miller, S.L. & Wasserburg, G.J.), pp.
15 7-168. North Holland, Amsterdam.
OJEDA Y OJEDA, H. A. ( 198 2) Structural framework, strati­
graphy and evolution of Brasilian marginal basins. Bull.
Am. Ass. Petrol. Geol. , 77, 732-749.
PIRRIE, D. ( 198 7) Oriented calcareous concretions from
James Ross Island, Antarctica. Br. A n tarct. Surv. Bull.,
75, 4 1- 5 0.
PONTE, F. C. & ASMUS, H.E. ( 1976) The Brazilian marginal
basins: current state of knowledge in continental mar­
gins of Atlantic type. A nais Acad. Bras. Cienc., 48,
121- 147.
RANGEL, H.D., MARTINS, F.A.L., ESTEVES, F.R. & FEIJO,
F.J. ( 1994) Bacia de Campos. Bull. Geoc. PETROBRAS,
8, 203- 217.
325
RIZZO, J.G., PANTOJA, J.L. & PESSOA, J. (1990) Aspectos,
tempo :de geracao, migracao secundaria, trapeamento e
preservacao de hidrocarbonetos. In: 4th Cong. Bras. de
Petr6leo, Proc. , pp. 120- 128. Inst. Bras do Petroleo, Rio
de Janeiro.
SAIGAL, G.C., BJ0RLYKKE, K. & LARTER, S. (1992) The
effect of oil emplacement on diagenetic process: exam­
ples from the Fulmar reservoir sandstones, Central
North Sea. Bull. Am. Ass. Petrol. Geol. , 76, 1024-1033.
SCHOLLE, P.A. & HALLEY, R.B. ( 198 5) Burial diagenesis:
out of sight, out of mind. In: Carbonate Cements (Eds
Schneidermann, N. & Harris, P.M.). Spec. Pub!. Soc.
Econ. Paleont. Miner., Tulsa, 36, 309-334.
SOLDAN, A.L., CERQUEIRA, J.R., FERREIRA, J.C., SCARTON,
J.C. & CoRA, C.A.G. ( 1990) Aspectos Relativos ao
Habitat do Oleo dos Campos de Marlim e Albacora,
Bacia de Campos. PETROBRAS/CENPES Internal
Report.
SOMBRA, C.L., SOUZA, R.S., SILVA, C.M.A. & PEREIRA,
C.A.F. ( 1995) Camadas Cimentadas por Calcita no
Arenito Namorado de Albacora. PETROBRAS/
CENPES Internal Report.
SPADINI, A.R., ESTEVES, F.R., DIAS-BRITO, D., AZEVEDO,
R.L.M. & RODRIGUES, R. ( 198 8) The Macae Formation,
Campos Basin, Brazil: its evolution in the context of the
initial history of the South
. Atlantic. Rev. Bras. Geoc. ,
18, 261- 272.
VAIL, P.R., MITCHUM, R.M. & THOMPSON, S. ( 1977) Seis­
mic stratigraphy and global changes of sea level, part 4:
Relative changes of sea level from coastal onlap. In:
Seismic Stratigraphy-Applications to Hydrocarbon Ex­
ploration (Ed. Payton, C.E.). Mem. Am. Ass. Petrol.
Geol., Tulsa, 26, 8 3-97.
VEIZER, J. ( 1983) Chemical diagenesis of carbonates:
theory and application of trace element technique. In:
Stable Isotopes in Sedimentary Geology (Eds Arthur,
M.A., Anderson, T.F., Kaplan, I.R., Veizer, J. & Land,
L.S.). Soc. Econ. Paleont. Miner. Short Course 10.
WALDERHAUG, 0. (1994) Temperatures of quartz cemen­
tation in Jurassic sandstones from the Norwegian con­
tinental shelf-evidence from fluid inclusion. J.
sediment. Res. , A64, 311-323.
Spec. Pubis int. Ass. Sediment. (1998) 26, 327-362

Structural controls on seismic-scale carbonate cementation

in hydrocarbon-bearing Jurassic fluvial and marine sandstones
from Australia: a comparison

J . S C H ULZ-ROJAH N*1, S . RYAN - G R I G O R* 2 a n d A. A N D E R S O Nt

*Australian Petroleum Cooperative Research Centre (APCRC),
National Centre for Petroleum Geology and Geophysics (NCPGG),
Thebarton Campus, University of Adelaide, SA 5005, Australia; and
tBHP Petroleum (Americas) Inc., 1360 Post Oak Boulevard, Suite 500,
Houston, TX 77056, USA, e-mail anderson.alan.ad@bhp.com.au

ABSTRACT

Wireline log responses, 2D and 3D seismic data, petrographic and isotope results were used to compare
major carbonate-cemented zones in Jurassic marine (Angel Field, Carnarvon basin) and fluvial
sandstones (Gidgealpa Field, Eromanga basin), Australia. In both fields the carbonate-cemented zones
concentrate near the crest of major structures, above areas where regional seals in older formations are
breached. In the Angel Field, poikilotopic dolomite-cemented zones with a cumulative thickness of up
to 1 6 5 m occur in the Upper Jurassic Angel Formation of submarine fan origin, both above and below
the present-day gas-water contact. In this field the dolomite cement volume is of the order of 0.6 km3
to 1 .4 km3, distributed over an area of 300 km 2 . At Gidgealpa, poikilotopic calcite-cemented zones
with a cumulative thickness of up to 65 m concentrate in the lower portion of the fluvial Namur
Sandstone, about 1 00 m below the present-day oil-water contact. In this field, the calcite-cemented
zones extend over an area 7.5 km wide and 20 km long, with the total volume of calcite cement being
0.22-0.37 km3. Both geological areas are characterized by: (i) rapid initial burial; (ii) continuous
subsidence; (iii) late-stage Tertiary compression, which triggered structural growth and closure
development; (iv) coincidence of timing of peak hydrocarbon generation and migration with the
Tertiary compression; and (v) the availability of an effective vertical plumbing system (locally breached
regional seals in sequences underlying the carbonate-cemented reservoirs). These observations point
towards a migration-related control on carbonate cementation broadly synchronous with hydrocarbon
charging into structures. In the Eromanga basin, a statistical correlation exists between major calcite
cement occurrence and oil pools in Jurassic reservoirs. The data suggest that seismic (predrill)
identification of high-amplitude events related to major carbonate cementation can be useful for
highgrading prospects and leads for drilling in clastic petroleum provinces that are character�zed by a
relatively late-stage compressive tectonic regime.

INTRODUCTION

This chapter describes the occurrence of large,
migration-related carbonate cement bodies near the
1 Present address: Shell Development (Australia) Pty Ltd,
Shell House, I Spring Street, Melbourne, Victoria 300 I,
Australia, e-mail jorg.p.schulzrojahn@shell.com.au.
2Present address: Schlumberger Cambridge Research,
High Cross, Madingley Road, Cambridge CB3 OEL, UK,
e-mail ryan@cambridge.scr.slb.com.
crests o f two Australian petroleum fields in different
geological settings. The study shows that the seismic
delineation of the geometry and spatial extent of
these carbonate cement bodies can assist in the
reconstruction of closure development over geolog­
ical time, and the ranking of prospects and leads for
drilling.
The origin of carbonate cements in clastic se-

Carbonate Cementation in Sandstones Edited by Sadoon Morad 327

© 1998 The International Association of Sedimentologists
ISBN: 978-0-632-04777-2
328 J. Schulz-Rojahn, S. Ryan-Grigor and A. Anderson
quences is generally discussed on the microscopic
scale. Petrographic, isotope and electron microprobe
data have provided valuable insight into the deriva­
tion and diagenetic evolution of carbonate cements
(e.g. Kantorowicz et a!., 1 9 87; Machel, 1987; Saigal
& Bj0rlykke, 1987; Hutcheon et a!., 1989; Taylor,
1990; Morad et a!., 1992; Souza et a!., 1 995). How­
ever, few published data are available concerning
the subsurface distribution of carbonate-cemented
zones in sandstones away from the well or outcrop
locations. Most of the data are based on two­
dimensional wireline log correlations, showing that
individual carbonate-cemented intervals with thick­
nesses up to 2 m can be correlated over distances of
several kilometres in some regions (e.g. Bryant et a!.
1 988; Walderhaug et al., 1989; Prosser et a!., 1993;
Williams et a!., 1994). Rarely are seismic data inte­
grated with petrographic and geochemical tech­
niques. Woods (199 1 ) and O'Brien & Woods (1995)
identified anomalously high velocities associated
with carbonate- cemented sands in the Vulcan sub­
basin, Timor Sea. Hanneman et a!. (1994) showed
that calcic palaeosols produced bright seismic reflec­
tions in southwestern Montana.
These authors demonstrate that both wireline log
and seismic investigations can provide important
tools for delineating the distribution of carbonate
W�NCOOCHA.
- -- ""-BI'S\1'4
,--··--aoPER
/ UM\i Of C
'
1 STRZELECKI
2MARANA
0 10
50
3KIOMAN
�DELLA
cements in sandstones where the thickness of these
cemented zones is above seismic resolution. How­
ever, in the case studies the carbonate cements are
interpreted to have formed either during relatively
early diagenesis (Bryant et a!., 1988; Prosser et a!.,
1993; Hanneman et a!., 1994; Morad & De Ros,
1994; Williams et a!., 1 994), including possibly via
later-stage diagenetic alteration of biogenic carbon­
ate accumulations (Walderhaug et a!., 1 989), or via
the microbial oxidation of migrating hydrocarbons
in the relatively shallow subsurface (O'Brien &
Woods, 1995). There is a lack of data in the
literature concerning the application of geophysical
methods to delineate major carbonate-cemented
zones that formed under conditions unrelated to
depositional or shallow diagenetic controls, but
mainly under relatively deep burial conditions.
Attention is focused on two Australian petroleum
fields, located in different geological settings more
than 2500 km apart: the Gidgealpa Field in the
Jurassic-Cretaceous Eromanga basin of Central
Australia (Fig. 1 A) and the Angel Field, located in
the Dampier sub-basin (Carnarvon basin) of Aus­
tralia's North West Shelf (Fig. 1 B). The case studies
exemplify the occurrence of major poikilotopic
carbonate-cemented zones in sandstones buried to
depths between 1 500 and 3000 m, namely the
V TAS
Fig. I. Location maps of (A) the
Gidgealpa Field in the Eromanga
basin, and (B) the Angel Field in
the Dampier sub-basin, Carnarvon
basin. Both fields occur along
broadly NE-SW-trending
anticlinal trends. The small black
D ����D squares in (A) represent the
20 30
km
II �i€Lo location of the wells listed in
Table I.
 Structural controls on seismic-scale carbonate cementation 329

fluvial Namur Sandstone of Upper Jurassic age STRATIGRAPHIC NOMENCLATURE. SOUTH AUSTRALIA
L.AT -
(Gidgealpa Field) and the marine Angel Formation - WINTON FORMATION - - · -

of the same age (Angel Field). In both fields a major � o.. MACKUNDA FORMATION ·

� ����������-=-=�� ���
structural control on carbonate cement occurrence
� � g; :�LL�U- Ml!_� S_IOr::!:_:
: : 1:: :
is demonstrated. The results emphasize that seismic 0 � � _. OODNADATTA. _t- TOOLEBUC FORMATION---
data can be useful in constraining the geological � iS! � FORMATION
- - - - - -
timing of carbonate cement development, and show
�i:t :J
w Ul Ul
- TO,.C: N=-
:- : - WALLUMBILLA:-
- =-
E .-+ :: - FORMATION
:-
that large carbonate cement volumes can precipi­ 5 � tlJ I-. ':-, CO,.: :=R::':IK7-:-IA.C.,NA:-S-=':AccN':::DS
:::::=: - _- _
�
�
� �
z ::;
tate in a relatively short time in some geological z
<! ·c7 SEISMIC HORIZON -:-:-:-:-:-:-:-:
w - CA DNA.QWIE· . MUUN rANNA�'w� ou"c McM�cK
settings. �
<!
FORMATION ·
� - : -- : -- : - : -MURTA·MEMBER-:-:-:--
- - - - - - -
<! -
::; "McKINLAY MEMBER"
0
�
TECTONIC SETTINGS w NAMUR SANDSTONE MEMBER
AND REGIONAL GEOLOGY WESTBOURNE FORMATION

The Gidgealpa Field is located along the

.I· ADORI SANDSTONE.
BIRKHEAD FORMATION -------
Gidgealpa-Merrimelia-Innamincka (GMI) anticli­ -
nal trend in the southern part of the Cooper­ HUTTON SANDSTONE
Eromanga basin system, South Australia (Fig. 1 A).
The Permo-Triassic Cooper basin is Australia's . . . . . . . .
largest onshore hydrocarbon province, containing -- - - POOLOWANNA FORMATION ... -
a:
6 TCF of recoverable gas and 300 MMSTB of oil ()

and gas liquids in over I 00 fields. The Cooper basin

1-- �
!>
U5
U) ffi �
��
:-_-:(Regional-:.:.�
::: : :
: : :. se�l)::::
� ·-:-:-----:- --:-_-_
�:-:-::-��':_9��:�;:::::
is unconformably overlain by up to 2 km of Jurassic ... z :- 'P' S EISM IC HORIZON· · -- - -
- - - - - -- -
and Cretaceous sediments of the oil-bearing Ero­ TOOLACHEE FORMATION
DARALINGIE FORMATION
manga basin, which forms part of the I . 7 million -·ROSE NEATH SHALE--
km2 area of the Great Artesian basin (Fig. 1 A). EPSILON FORMATION
- - -- MURTEREE SHALE -
Cooper basin sediments consist of glaciofluvial, -· .. PATCHAWARRA FORMATION .. - .
fluvio-lacustrine and deltaic clastics (Fig. 2) (Bat­ TIRRAWARRA SANDSTONE
·

tersby, 1976; Thornton, 1979). The Jurassic section
of the Eromanga basin sequence was deposited in
an intercratonic basin sag, and is made up of
non-marine clastics deposited under fluvial, deltaic
and lacustrine conditions. The Early Cretaceous
Murta Member to Cadna-Owie Formation (Fig. 2)
show deposition changing from lacustrine to ma­
rine, with marine conditions prevailing from the
Aptian until the Upper Albian, when a return to
paralic and fluvio-lacustrine conditions is indicated
(Senior et a!., 1978; Armstrong & Barr, 1 9 86).
The Angel Field is located at the northern end of
the NE-SW-oriented Madeleine Trend in the
Dampier sub-basin, which forms part of the Sil­
urian to Holocene Carnarvon basin, offshore West­
em Australia (Fig 1 B). The Madeleine Trend is a
major fault-controlled anticlinal feature that follows
the main depositional axis of the Dampier sub­
basin and along which several other oilfields are
located, including the nearby Cossack and Wanaea
Fields (Fig. IB). The stratigraphy of the Dampier
sub-basin consists of marine and deltaic clastics and
Tertiary carbonates (Fig. 3).
MERRIMELIA FORMATION ¢. O·
Fig. 2. Stratigraphical column for the Cooper/Eromanga
basin province. The seal integrity of the regional
Nappamerri Group (Triassic) is a major controlling
factor in the distribution of Permian-sourced
hydrocarbons in Jurassic sandstones. Much of the
Jurassic Eromanga basin sequence forms part of the
regional J-aquifer system of the Great Artesian Basin
(Fig. lA), including the Namur Sandstone (Habermehl,
I 980). Major seismic markers are shown: P, top
Permian; C, top Cadna-Owie Formation.
Both the Dampier sub-basin and the Cooper­
Eromanga basin system have in common the rapid
initial burial of Jurassic and younger sediments;
late-stage Tertiary compression broadly coincident
with hydrocarbon migration; the availability of an
effective vertical plumbing system that has allowed
hydrocarbon migration via locally breached re­
gional seals into Angel and Gidgealpa Field struc­
tures.
330 J. Schulz-Rojahn, S. Ryan-Grigor and A. Anderson

Barrow Dampier Burial subsidence and temperature histories

Epoch Stage S Sub-basin Sub-basin N

' BASINMOD modelling shows that Upper Jurassic

MI ,MARL=
sediments (Angel Formation) attained a depth of
C/)
::1
SER
WITHNELL ;;-r,0;!:.o.c
.
approximately 1 km by the Late Upper Jurassic to
0
w
()
. :.;:.;:.;:.; �GEARLE
· • • • • · • • ·SILTSTONE .
WINDALIA- ----
�f 'I'��
MAR �
HAYCOCK
Lowermost Cretaceous in the Angel Field (Fig. 4) .
In the Gidgealpa Field, sediments of the same age
15
ALB .
RADI OL. � :.;:.;· r-o::;;;_;; KA (Namur Sandstone) reached a depth approaching
w
a:
APT � MARDIE-MUDERONG
-.....:..
�
BIRDRONG -
2 km by Mid-Cretaceous times (Fig. 5). In both
() HAU
VAL �-:::·:.. �A�RQ\'[f'D - __!L geological provinces subsidence continued through­
BER out the Cretaceous, Tertiary and Quaternary. Angel
TITH
KIM
OXF
�UP��?>
�DINGOFM
.
� �
. .FM.
�
.•
Formation sandstones occur at depths between
2700 and 3000 m, whereas the Namur Sandstone is
() CALL � found between 14 70 and 1900 m.
en
�
�
C/)
BATH
�LEARMONTH....:.:: -� �.� . · In the Angel Field, Angel Formation tempera­
BAJ � ,....';::
.. -, .-- .-�. -
tures exceeded 8o·c from the Pliocene onwards,
<(
a:
::1
AAL � ��GNR ATHOT
EMBER
� •
·.;;;,.� .� .�
� and first attained 10o · c during the Oligocene
., TOA
•
E E D E
.
• FM.... - (Fig. 4). During the Miocene, temperatures were
PLIE
close to the present-day maximum burial tempera­

��-"'�'�: :;.;:
SIN � �
HET
MURAT --c:z:: · - ture ooo · c- 1 o4·q.
MEMBER
RHA
In the Gidgealpa Field, temperatures were be­
() NOR --. •
...... -._ ......

-.-.-.
• • •
..

--.--.
-.
0
..
..
0
.......................

-.
0
--. • ._
-.-. --.•
--. tween 60 and 70"C from Mid-Cretaceous to Late
en � ·�� ·� ·� M!JN!JA.RO.O fM.. ·� ·� ·� Tertiary times in the Namur Sandstone (Fig. 5). In
C/) CAR
<( LAD � � � � � � � � :� � � � � �� the last 1 0 Ma, and possibly as recently as the last
a: ANI
1-2 Ma, a basinwide increase in thermal gradients
1- SCY
LOCKER
SHALE
occurred, probably in response to deep-seated igne­
ous activity, as indicated by apatite fission track
Fig. 3. Generalized stratigraphy for the Dampier
data (Gieadow et a/., 1 988). In the Namur Sand­
sub-basin. Major seismic markers are indicated: MUC,
Callovian-Oxfordian 'main rift' or 'breakup' stone, present-day temperatures are between at least
unconformity; K, near top Angel Formation; KA, near 80 and 105·c.
top Aptian unconformity. Modified from Kopsen &
McGann ( 1985).

TREALLA LST.

CAPE RANGE

CARDABIA GP.

TOOLONGA CAL.
HAYCOCK MARL
MUDERONG SH.
Fig. 4. Geohistory plot for
EARLY MATURE (OIL) sediments in the Dampier
o.s - o. 7 (%Ro) sub-basin (Angel-2). Note that
II MID MATURE (OIL)
0.7- 1.0 (%Ro)
Upper Jurassic Angel Formation
sediments underwent rapid initial
burial, and that temperatures in
this formation exceeded so·c
4-r---,----,---,---� from the beginning of the Tertiary,
150 100 50 0 and from the Oligocene onwards
TIME (Ma) have been 10o·c or greater.
 Structural controls on seismic-scale carbonate cementation
'E
::.
J:
I­
C.
w OODNADATTA
c
<C
w
en
III
:::>
en
Fig. 5. Geohistory plot for
Eromanga basin sediments
(Gidgealpa-7). Note that the
Upper Jurassic Namur Sandstone 2 EARLY MATURE (OIL)
0.5-0.7
and younger Mesozoic sediments
underwent rapid initial burial, the
same as in the Angel Field area 200
(Fig. 4).
Tectonic histories
In both geological provinces the tectonic style is the
result of reactivation of older structures and faults
during the Tertiary.
In the Dampier sub-basin, the main structural
elements such as the Rankin Trend, the Lewis
Trough and the Madeleine Trend and associated
faults (Fig. l B) formed during a period of NW-SE­
oriented Early Jurassic extension associated with
the break-up of Gondwana (Bradshaw et a!., 1988;
Veevers, 1988). In the Callovian-Oxfordian, conti­
nental breakup to the northwest resulted in relative
uplift, producing the 'main' or 'breakup' unconfor­
mity that takes the form of a buried escarpment
(seismic reflector 'MUC') (Fig. 3) onto which dom­
inantly marine Upper Jurassic, Cretaceous and
Tertiary sequences were deposited, including Angel
Formation sediments (Woodside, 1 988). During the
Middle to Late Miocene, the pre-existing structures
were reactivated during a major compressional
event caused by the northward movement of the
Australian plate and its subduction beneath Indo­
nesia (Denham & Windsor, 1991; Etheridge et a!.,
1 991; Etheridge & O'Brien, 1994). The tectonic
event, which is continuing, has produced an ap­
proximately E-W maximum compressional stress
vector which formed many of the hydrocarbon-
331
COORIKIANA
POOLOWANNA
(%Ro)
150 100 50 0
TIME (Ma)
bearing structures in the North West Shelf ( Kopsen
& McGann, 1985; Parry & Smith, 1988; Etheridge
et a!., 199 1 ; Etheridge & O'Brien, 1994). This
tectonic event also caused a widespread down­
warping or tilting of structures towards the north­
west, including the Madeleine Trend (Apthorpe,
1988). The orientations of the structures on the
Madeleine Trend (e.g. Angel, Wanaea, Cossack,
Dampier, Withnell) (Fig. IB) are consistent with
right-lateral movement that was initiated during the
Miocene (Woodside, 198 8).
In the Cooper basin, rejuvenation of pre-Permian
faults along the flanks of many structures occurred
contemporaneously with Permo-Triassic deposition
(Battersby, 1976; Stuart, 1976; Apa)< et a!., 1993).
Cooper basin sediment deposition terminated at
the end of the Early to Mid-Triassic, when wide­
spread compressional folding, regional uplift and
erosion occurred (Battersby, 1 976). The Eromanga
basin tectonic style is the result of reactivation of
Late Triassic structures (Battersby, 1976; Frears,
1 995) during the Tertiary (Krieg, 1986; Shaw,
1 991 ). Based on outcrop studies of the Dalhousie
Anticline in the southwest Eromanga basin, Krieg
(1986) concluded that anticlinal growth was in
progress during the Oligocene and Early Miocene,
but that major uplift did not occur until the Late
Tertiary and Quaternary. Krieg (1986) considers
332 J. Schulz-Rojahn, S. Ryan-Grigor and A . Anderson

that anticlinal growth is still active today in the
Eromanga basin.
Petroleum systems
Dampier sub-basin
The major source of gas in the Angel Field and
other petroleum fields in the region is the Triassic
Locker Shale (Fig. 3), consisting of carbonaceous
claystones and coals (Woodside, 1 988; Di Toro,
1 994). The Locker Shale first entered the oil win­
dow in the Jurassic, but major gas generation did
not occur until the Miocene, synchronous with early
oil generation from the Jurassic Dingo Claystone
(Fig. 6a) (Brikke, 1 982). The Dingo Claystone on­
laps onto the Angel Field structural high, which
allowed migration of hydrocarbon gases from the
Locker Shale into Upper Jurassic Angel Formation
sandstones at this location (Di Toro, 1994). In
contrast, gas is absent in Angel Formation reser­
voirs at the nearby Wanaea Oilfield (Fig. !B), owing
to the presence of an unbreached Dingo Claystone
(Di Toro, 1994). In the Angel Field the timing of
hydrocarbon migration into Angel Formation reser­
voirs broadly coincides with the Miocene compres­
sive event that triggered widespread structure
development in the North West Shelf (Fig. 6a).
Cooper-Eromanga basins
The Gidgealpa Field currently produces gas from
the Early Permian Patchawarra Formation and
Tirrawarra Sandstone, the Late Permian Toolachee
Formation, and oil from the Jurassic Namur, Birk­
head, Hutton and Poolowanna Formations (Fig. 2).
The abundant dispersed organic matter in the
intraformational shales and siltstones, and possibly

(a)
Age of
Main Source
l
Locker
Shale
Legendre & Athol Fms.
•••• � u/C
� ��
0.

( )
go Claystone

l
Rocks i and gas) (oil or gas?)

Deposition Angel
of Reservoir Formation
Sandstone (Kim.·lith.)

Seal
Deposition
II Barrow Group Shales

Early MID·LATE MIOCENE:

Tectonic Jurassic Major compression -
History extension collision of Australian
and Indonesian Plates

Timing of Main
Hydrocarbon
Charge

Gidgealpa Field area,

Cooper-Eromanga Basin Systems

(b) Jurassic Cretaceous

Age of _______ !��sib/e)

Main Source
Rocks (minor)

Deposition
of Reservoir
Sandstone
Namur Sandstone
of Mooga Fonnation
Ill Fig. 6. Petroleum systems of (a)
the Dampier sub-basin (Carnarvon
basin) and (b) the

Seal
Deposition
I Murta
Member
Cooper-Eromanga basin province.
Note that both geological
provinces are characterized by
Reactivation of Permian and Late Triassic structures _
late-stage (Tertiary) compression
Tectonic during the Tertiary and Quaternary which produces ? _ that triggered trap development in
History major uplift and erosion •
Mesozoic sequences, broadly
Timing of Main synchronous with peak
Hydrocarbon hydrocarbon generation and
Migration
migration from older sequenq:s,
 Structural controls on seismic-scale carbonate cementation 333

the coals, represents the source of the gaseous

A A'
hydrocarbons in the Permian sandstones (Brooks et
DEPTH 27 30 20 17
al., 1971; Battersby, 1976). The Permian sediments SUBSEA

were also the source of the bulk of the oil trapped in

(METRES) *-<>- • *
1 0
the Jurassic sandstones of the Eromanga basin 4 0 1--�+- r - � � - - - - 'TaoPPi:c!M,""'=-o"'wo""'•'"'"�
M

(Heath et al., 1989; Jenkins, 1989). Based on 1500 -=--�=-=--=�-;':_::.::=�====:-::_;;�;;:;;a��

distributional patterns of age-dependent biomark­
1600
ers, at least 60% by pool and over 75% by volume of
the Eromanga basin crude oils were derived from
Permian source rocks (Jenkins, 1989).
Cooper basin hydrocarbons and carbon dioxide
were expelled from terrestrial organic matter over a 1900

wide maturity range, and therefore the timing of

expulsion varied across the basin (Kantsler et al.,
2ooo-v.---
l""""'
�--..... __�� --�r---
GROUP�
-;;Pooo:;;;;,;o""w•"""""='"�-
NAPPAMERAI
��""' ·rOP TOC:X......CHEE FM.
1983; Tupper & Burckhardt, 1990). Oil expulsion 21 QQ

from the Permian sequence locally may have begun -Oil km

as early as in the Triassic in deep sequences of the

� �0��� calcite·cemented
2 3 4

Nappamerri Trough (Fig. I A) (Tupper & Burck­

hardt, 1990), but the onset of peak oil generation
from the Permian source rocks occurred during the
Late Cretaceous and the Early Tertiary in most areas
(Kantsler et al., 1983; Heath et al., 1989; Tupper &
Burckhardt, 1990). Much of the Permian sequence
in the Patchawarra Trough (Fig. lA) remains within
the oil window, and in other regions the Permian
sequence continues to be mature for wet or dry gas
generation at the present time (Kantsler et al., 1983;
Tupper & Burckhardt, 1990). Hydrocarbon migra­
tion and entrapment within the Eromanga basin
sequence probably occurred relatively recently, and
is perhaps still occurring (Heath et al., 1989). The
timing of major hydrocarbon generation broadly
coincided with the reactivation of Late Triassic
structures during the Tertiary (Fig. 6b).
Hydrocarbon migration was influenced by multi­
ple seals within the Cooper basin sequence, includ­
ing intraformational shales and siltstones, the Mur­
teree Shale, the Roseneath Shale and the Triassic
Nappamerri Group (Fig. 2) (Heath, 1989; Powis,
1989). In particular, the seal integrity of the regional
Nappamerri Group controls the distribution of
Permian-sourced hydrocarbons in Jurassic sand­
stones (Gilby & Mortimore, 1989; Powis, 1989).
Erosional truncation of Permian carrier beds be­
neath Jurassic sequences, coupled with a breached
Nappamerri Group seal, accounts for Jurassic oil
pools in many Eromanga basin fields (see Gilby &
Mortimore, 1989; Heath et al., 1989), including the
multiple stacked oil pools in the Gidgealpa Field
(Fig. 7) (Mcintyre et al., 1989). The regional Nappa­
merri Group is locally breached due to non-deposi­
tion or erosion triggered by structural rejuvenation
of deep-seated basement faults along structural
Fig. 7. Structural cross-section along line A-A' (see
Fig. 2 1 ) showing the location of the major
calcite-cemented zones in relation to the oil pools in the
Upper Jurassic Namur Sandstone, and the erosive edge
of the regional Nappamerri Group seal (Triassic).
Modified from Mcintyre et a!. ( 1 9 89).
highs. Where wells encounter thick Nappamerri
Group seal development, Jurassic reservoirs are
typically water saturated (Heath et al., 1989).
The Permian-derived wet gas-high GOR oil
migrates vertically through up to 300 m of the
Eromanga basin aquifer system, where the hydro­
carbons are stripped of their gaseous components
by water washing, leading to changes in their alkane
distribution (Heath et al., 1989). The dissolved gas
is dissipated by flushing (Heath et al., 1989) and/or
diffusion (Dunlop et al., 1992) away from the
Eromanga basin closures towards the basin margin,
escaping into the atmosphere as high methane-cut
artesian water flows (Habermehl, 1980, 1986). The
resultant residual Permian-sourced oil is therefore
almost devoid of associated gas and continues to
migrate vertically until a competent Eromanga
basin seal is reached (Heath et al., 1989).
FIELD DESCRIPTIONS
Traps
Both the Angel Field and Gidgealpa Field represent
structural four-way dip closures. The Angel Field is
mapped at reservoir (Angel Formation) level as a
broad, low-relief drape closure with three separate
334 J. Schulz-Rojahn, S. Ryan-Grigor and A . Anderson

culminations over the regional 'breakup' uncon­ Table 1. Examples of major carbonate-cemented zones
formity (Woodside, 1988). The Gidgealpa Field as identified from sonic and bulk density logs and
lithological descriptions of cuttings for a variety of
represents an elongated NE-SW trending anticlinal structures in different parts of the Eromanga basin (Fig.
feature that comprises an independent north and lA). Note that the carbonate-cemented zones in Jurassic
south dome separated by a relatively shallow sandstones vary in cumulative thickness, from a few
saddle. metres (e.g. Strzelecki- 1 0) to about 1 1 0 m (Spencer
West-!), including over short distances (e.g. Strzelecki).
All wells occur near the Cooper basin margin, or along
Reservoirs major fault-bounded structural trends where the regional
Nappamerri Group seal (Triassic) underlying the
In both fields, reservoirs of Upper Jurassic age Jurassic sandstones is incompetent or missing
consist of fairly massive, porous, stacked sandstone
Cumulative
sequences with minor shale and siltstone interbeds, Stratigraphic thickness Depth
albeit deposited in different environments. In the Well name interval (m) (mkb)
Angel Field, the Angel Formation sandstones
(Fig. 3) contain between 5 and 10% of glaucony and Boukabourdie-2 Adori 1 4.6 2033-2057
Della-7 Namur 83.5 1 450- 1 666
minor pyrite, and were deposited under deeper­ Kidman- ! Namur 64.6 1 5 5 1 - 1 688
water marine conditions, by a mass flow and/or Marana- ! Namur 69.5 1 5 52- 1 658
grain flow-dominated submarine fan system (Di­ Spencer-3 Namur 5 1 .8 1 490- 1 592
bona & Scott, 1990; Hocking, 1992). In the Spencer-3 Hutton 36.6 1 734- 1 8 1 4
Spencer West-! Namur 1 09.7 1 478- 1 594
Gidgealpa Field, the Namur Sandstone Member of
Strzelecki-3 Namur 39.6 1 563- 1 64 1
the Mooga Formation (Fig. 2) was deposited in Strzelecki-5 Namur 1 5.2 1 600- 1 6 1 8
high-energy braided to meandering stream environ­ Strzelecki-! 0 Namur 67 1 59 1 - 1 6 8 1
ments (Nugent, 1969; Bowering, 1982; Ambrose et Strzelecki-! 0 Hutton 4.3 1 789- 1 795
a!., 1986). Whereas Angel Formation sandstones Tantanna-1 Adori 29.9 1 474- 1 530
Tantanna-4 Adori 1 2.2 1 600- 1 6 1 3
are dominantly medium grained and moderately Tantanna-5 Namur 6 1 .6 1 478- 1 5 8 1
sorted (Buswell, 1989; Miller, 199 5) the Namur Wancoocha-2 Namur 20. 1 1 499- 1 543
Sandstone is fine to coarse grained and often poorly Wancoocha-2 Poolowanna 6.7 1 65 5- 1 662
sorted (Armstrong & Barr, 1986). In both fields, the Wancoocha-2 Hutton 39 1 600- 1 652
Upper Jurassic reservoirs are characterized by lo­
cally pervasive carbonate cementation.
In the Namur Sandstone the major carbonate­
cemented zones have long been known to cause seis­ voirs have oil pools in the Namur Sandstone or
mic velocity 'pull-ups', of the order of10 ms or more, other Jurassic sandstone intervals (B. Jensen­
producing problems for accurate reservoir delinea­ Schmidt, MESA, personal communication).
tion and depth prognosis of underlying reservoirs In the Dampier sub-basin no such statistical
(Anderson, 1985; Singh, 1990). The Gidgealpa Field relationship between hydrocarbon pools and major
is one of many Eromanga basin fields in which major carbonate-cemented zones is known in Angel For­
carbonate-cemented zones occur in the Namur mation reservoirs.
Sandstone (Table 1 ). In other Eromanga basin fields,
major carbonate-cemented zones are also found at
Seals
different stratigraphical levels, including the Adori
Sandstone, the Hutton Sandstone and the Poolo­ In the Gidgealpa Field the Namur Sandstone is
wanna Formation of Jurassic age (Fig. 2; Table 1) sealed by the Early Cretaceous shales of the Murta
(Anderson, 198 5 ; Staughton, 1985; Wall, 1987; Member of the Mooga Formation (Figs 2 and 7). In
Stienstra, 1992; Schulz-Rojahn, 1993; Townshend, the Angel Field, Angel Formation reservoirs are
1 993). sealed by the conformable Neocomian shales of the
Based on the petrophysical review of 475 wells on Barrow Group (Fig. 3).
169 structures, 50% of all Eromanga basin struc­
tures with major carbonate-cemented zones in Ju­
Hydrocarbon types and reserves
rassic reservoirs have oil pools in reservoirs of the
same age. In contrast, only 1 3% of structures with­ The Angel Field contains 30 billion m3 of gas and
out identifiable carbonate cement in Jurassic reser- 50 million barrels of condensate (Playford, 1 97 5).
 Structural controls on seismic-scale carbonate cementation
The discovery well intersected an 8 2 m net gas
column in Upper Angel Formation sandstones.
Three appraisal wells established major hydrocar­
bon reserves at the same stratigraphical horizon,
with a maximum gas column of 140 m (Vincent &
Tilbury, 1988). A thin oil leg was discovered be­
neath the gas cap in all four wells, but is thickest at
Angel-3 (20 m), the only well drilled south of a
major ENE-WSW -oriented strike-slip fault zone
that traverses the field (Ryan-Grigor & Schulz­
Rojahn, 1995). The gas composition in Angel For­
mation reservoirs consists of dominantly methane
(80% by molar volume), ethane (6%), propane (3%)
and heavier hydrocarbon gases (up to c7+ ), includ­
ing minor carbon dioxide (less than 3%) (Woodside,
1971, 1972, 1990).
In the Gidgealpa Field, original gas-in-place in
the proven and probable categories is 348 BCF in
the Permian reservoirs, whereas original oil-in­
place is 12 MMSTB in the Jurassic reservoirs
(Robertson Research, 198 8). The bulk of the oil
reserves are contained within the Jurassic sand­
stones of the South Dome, whereas the majority of
the gas is in the North Dome Toolachee Formation
(Mcintyre et a!., 1989). The majority of the current
oil production is from the Hutton Sandstone (Mcin­
tyre et a!., 1989). The structural closure at the top
Hutton Sandstone horizon is at least 5 3 m in the
South Dome (Singh, 1990). Crude oils in Jurassic
reservoirs have a high gravity of 47° API, a low
gas/oil ratio and a high wax content indicative of
a land-plant origin (McKirdy, 1982; Kantsler et
a!., 1983; Heath et a!., 1989; Armstrong & Barr,
1986).
PREVIOUS INVESTIGATIONS
Both fields contain seismically identifiable carbon­
ate cement bodies at reservoir level. In the
Gidgealpa Field, Singh (1990) seismically mapped
Namur Sandstone carbonate cement thickness over
the South Dome and noted that these cements
concentrate in a crestal position, confirming earlier,
fieldwide results by Anderson (1985). In this field,
based on results obtained from cuttings, the main
pore-filling carbonate cement is calcite, which post­
dates quartz cementation and minor ankerite and
siderite cement (Stienstra, 1992). Bulk-rock isotope
compositions of eight calcite-cemented samples de­
rived from cuttings in the Lower Namur Sandstone
yielded o 1 3C values between -12.9 and -2.5o/oo PDB
335
(mean -6.25 ± 3. 7%o PDB) (Fig. 8a) and o180 values
from + 7 . 3%o to +13. 8%o SMOW, with an average
composition of +11.1 ± 1 .8o/oo SMOW (Stienstra,
1992).
In the Angel Field, Ryan-Grigor & Schulz-Rojahn
(199 5) identified a high-amplitude zone caused by
major carbonate cementation in the northern sector
of the Angel Field, along the structural axis of the
Madeleine Trend. Bulk-rock XRD analysis of 1 5
core samples shows that dolomite i s a major rock
constituent at Angel-2, but that the same mineral is
only present in trace to minor quantities at Angel- l
(Woodside, 1971, 1972). Other rock constituents
include quartz, feldspar, kaolin and illite at Angel- l
and Angel-2, with minor anhydrite occurring in
some Angel Formation samples derived from core 3
at Angel-2 (Fig. 9) (Woodside, 1972). Vincent &
Tilbury (1988) noted that reservoir quality is gener­
ally poorer in the vicinity of Angel-2 owing to
diagenetic effects, but did not comment on the
nature of the diagenetic modifications. At Angel-3,
Buswell (1989) petrographically observed minor to
trace amounts of rhombic and poikilotopic carbon­
ate cements that lack dissolution textures in Angel
Formation reservoirs. At Angel-4, and near the base
of core I at Angel-2 (Fig. 9), the dolomite cement
completely or partly fills subvertical fractures that
are 5-25 mm wide (Woodside, 1990; Miller, 1995)
and oriented NW-SE (L. Tilbury, Woodside, per­
sonal communication).
MET HODS
Core samples and cuttings were collected from the
Angel and Gidgealpa Fields following on-site corre­
lation of the lithologies against gamma-ray and
sonic log responses (Table 2).
Thin sections were prepared for 28 samples
following impregnation with blue-dye epoxy resin,
and stained for calcite and K-feldspar using alizarin
red S and potassium ferrocyanide, respectively. The
thin sections were systematically scanned by the
senior author to determine rock composition, po­
rosity and textural relationships.
Semi-quantitative bulk-rock XRD analyses were
carried out on 23 samples (Table 3), which were
ground in a Siebtechnick mill and prepared as
pressed powder mounts. Continuous scans were run
of these powder pressings from 3 ° to 7 5° 2 e, at
2 /min, and the Co X-ray tube was operated at
o
50 kV and 30 rnA, on a Philips PW1050 diffracto-
336 J. Schulz-Rojahn, S. Ryan-Grigor and A . Anderson

meter. Peak identification was based on comparison I OO " C overnight. The carbon dioxide was purified
with JCPDS files stored in the CSIRO software according to conventional techniques (McCrea,
XPLOT. 1 9 50) and analysed on a 6-inch dual-collector VG
Four core samples were selected for carbon and Micromass 602E mass spectrometer. The acid cor­
oxygen isotope analysis following identification of rection factors of Rosenbaum & Sheppard ( 1 986)
the mineralogy under bulk-rock XRD. Only sam­ were used to compensate for the oxygen isotope
ples dominated by dolomite cement were chosen for fractionation. Stable isotope values are reported in
isotope analysis, as identified by XRD (Table 3). the 8 notation in parts per thousand (%o). All oxygen
The samples were crushed to a fine dry powder. isotope ratios are expressed relative to SMOW (Craig,
Carbon dioxide was extracted from the powdered 1 96 1 ) and all carbon values relative to PDB (Craig,
samples by reaction with 1 00% phosphoric acid at 1 95 7).

(a) Gidgealpa Field (b) Other Eroman lJa Basin Fields 1

Lower Namur Sandstone - calcite cement Namur & Adon Sandstones - calcite cement
Cuttings Mostly cuttings
3 15

• Stlenstra(19�2)
D
Wall(1987)
Namur/Adorl Sandstone

ci
(cuttings)

0
10
• (cuttings)
Townshend(1993)
Upper Namur Sandstone

z z (core)

·14 ·16 ·18 ·20 · 22 ·24

2
(c) Various Eromanga Basin Fields (d) Cooper Basin
Jurassic sediments - siderite cement Carbon dioxide gases
Mostly cuttings
3 10

0 Blrkhud
Waii(HI87)
Formation
• R�by&Sm'h
(1081)

• (cuttings)
Townshend(1!l93)

ci
Upper Namur SandstoM
(core)
ci
z D Vincent et al.
(1985)

z 0 Schulz·Aojahn (unpublish.ci)
U�r Namur Sandstone
4
(core)

0 ·2 -4 .e ·8 -10 ·12 -14 ··16 -18 -20 ·22 ·24

a13c

(e) Eroman g a Basin

Carbon dioxide gases

• Rlot>y&Smlth
(1981)

0 Vlncent etal.
(1i85)

ci
z

o-1-----�-
4 0 ·2 ·4 .e ·8 ·10 ·12 -14 ·16 ·18 ·20 22
· 24
·

�3c
Fig. 8. Histogram of o 1 3C frequency distribution for (a) calcite cement in the Lower Namur Sandstone of the Gidgealpa
Field; (b) calcite cement in the Namur Sandstone and the Adori Sandstone, its lateral equivalent, for different
petroleum fields, including the Big Lake, Kerna, Marana, Moomba, Spencer, Strzelecki, Tantanna and Warana Fields
(see Fig. l A); (c) siderite cement in Jurassic clastics; (d) Cooper basin C0 2 gases; (e) Eromanga basin C0 2 gases.
Compiled from various sources (as shown). Note that the o 1 3C character of Eromanga basin calcite cements is similar
to that of Cooper basin carbon dioxide gases. See text for explanation.
 Structural controls on seismic-scale carbonate cementation 337

SE 5.2 km NW 3km 6.2 km -------1 NE

Angel-3 Angel-1 Angel-4 Angel-2

v
DEPTH
vDEPTH
vDEPTH
v
DEPTH
GR mkb DT GR mkb DT GR mkb DT

MAJOR
[77) DOLOMITE­
l:::::LJ CEMENTED
ZONES

CORE
c1
I LOCATION
AND
NUMBER

Fig. 9. Stratigraphical cross-section, Angel Formation, Angel Field. The Upper Angel Formation is defined by the
relatively clean, massive gamma-ray response, representing stacked mass-flow sandstones of marine origin that extend
down to the Mid D. jurassicum boundary. All four wells encountered gas and condensate reserves within this
stratigraphical interval (see Fig. 20 for a location map). Only at Angel-2 were major dolomite-cemented zones
intersected (shaded), which cannot be explained by facies variations between the well locations, based on GR log
motives and core descriptions of Upper Angel Formation sandstones.

Quantitative determination of elemental dolo­
mite composition was carried out on five polished
thin sections covered with a thin layer of carbon
and using a CAMECA SX 5 1 electron microprobe
at 1 5 kV, a 20 nA beam current and a 0.2 J.Lm beam
diameter. The BSE imaging system linked to the
electron microprobe was used to detect zonation in
the dolomite cement, and compositional analyses
were carried out for each zone. Results were nor­
malized to 1 00 mol% Fe, Mg and Ca. The precision
of the analyses was 1 00% ± 2.
DIAGENETIC OBSERVATIONS
(MICROSCOPIC SCALE)
Angel Field
Semi-quantitative XRD traces show that Angel
Formation sandstones are dominated by quartz and
dolomite, and typically contain trace to minor
amounts of pyrite, glaucony, feldspar and kaolinite
(Table 3). Two samples in core 3 at Angel-2 (Fig. 9)
contain minor to subdominant amounts of anhy­
drite. There also is an indication of calcite in three
samples.
Under the petrographic microscope, porous An­
gel Formation sandstones are generally character­
ized by a loose grain packing (Fig. 1 OA). There is a
predominance of tangential grain contacts, with
some long contacts and rare sutured contacts be­
tween detrital quartz grains. The rocks contain
abundant macroporosity (Fig. 1 OA), chiefly rem­
nants of primary porosity, with some secondary
porosity caused by partial dissolution of infrequent
detrital feldspars. Where poikilotopic dolomite is
abundant, porosity is occluded (Fig. 1 OB).
The poikilotopic dolomite cement occurs in irreg­
ular patches (Fig. 1 OA) and locally completely ce­
ments the rock (Fig. 1 OB). The cement is common
338 J. Schulz-Rojahn, S. Ryan-Grigor and A. Anderson

Table 2. Derivation, lithological and textural characteristics of core samples and cuttings derived from the Upper
Angel Formation in the Angel Field and the Namur Sandstone in the Gidgealpa Field

Sample Core Depth Mean Average

Well no. Formation Type no. (mkb) Lithology grain size sorting

Dampier sub-basin
Angel- l 1 1 60 Angel Fm. Core I 2666. 1 Sandstone Fine-medium Moderate
Angel-2 1 1 53 Angel Fm. Core I 2704.2 Sandstone Medium Well-moderate
Angel-2 1 1 57 Angel Fm. Core I 2704.5 Sandstone Fine Well
Angel-2 1 1 54 Angel Fm. Core I 2705.5 Sandstone Fine Well-moderate
Angel-2 1 161 Angel Fm. Core 3 2749 Sandstone Medium Well-moderate
Angel-2 1 1 56 Angel Fm. Core 3 2754.2 Sandstone Medium Moderate-poor
Angel-3 1 1 58 Angel Fm. Core 2 2748.3 Sandstone Medium Moderate
Angel-3 1 1 62 Angel Fm. Core 2 2752.3 Sandstone Medium Moderate-poor
Angel-4 1 1 63 Angel Fm. Core I 2706.9 Sandstone Medium Well-moderate
Angel-4 1 1 64 Angel Fm. Core 2 2730.8 Sandstone Medium Moderate
Angel-4 1 1 59 Angel Fm. Core 3 2757.6 Sandstone Medium Moderate
Angel-4 1 1 65 Angel Fm. Core 4 2790.4 Sandstone Fine-medium Moderate

Eromanga basin
Gidgealpa- 1 9 1 1 66 Upper Namur Core I 1 549.7 Sandstone Fine-medium Well-moderate
Gidgealpa- 1 9 1 1 67 Upper Namur Core 2 1 5 5 1 .7 Sandstone Fine-medium Well-moderate
Gidgealpa- 1 9 1 1 68 Upper Namur Core 2 1 5 85.2 Sandstone Fine Well
Gidgealpa- 1 9 1 1 69 Upper Namur Core 2 1 588 Sandstone Coarse Moderate-well
Gidgealpa- 1 9 1 1 70 Lower Namur Cuttings - 1 696.2 Sandstone V. fine-medium -
Gidgealpa- 1 9 1 171 Lower Namur Cuttings - 1 760.2 Sandstone Fine-coarse
Gidgealpa-23 1 1 72 Lower Namur Cuttings - 1 684 Sandstone V fine-coarse
Gidgealpa-23 1 1 73 Lower Namur Cuttings - 1 760.2 Sandstone Fine-medium
Gidgealpa-24 1 1 74 Upper Namur Core I 1 5 57.5 Siltstone Silt Poor
Gidgealpa-24 1 1 75 Upper Namur Core 2 1 5 83.4 Sandstone Fine-coarse Moderate
Gidgealpa-24 1 1 76 Upper Namur Core 2 1 5 85.5 Sandstone Med.-coarse Very poor
Gidgealpa-24 1 1 77 Upper Namur Core 2 1 5 87.4 Sandstone Fine Well-moderate
Gidgealpa-24 1 1 78 Lower Namur Cutting 1 7 1 4.5 Sandstone V. fine-coarse
Gidgealpa-26 1 1 79 Birkhead Fm. Core 1 8 1 7.8 Sandstone V. fine-fine Poor
Gidgealpa-32 1 1 80 Upper Namur Core 1 700 Sandstone V. fine-fine Moderate
Gidgealpa-32 1 181 Upper Namur Core 1 822.5 Sandstone V. fine-fine Moderate

in samples derived from core 1 at Angel-2 and, to a
lesser extent, in one sample at Angel-4 (Table 3;
Fig. 9). In these samples no difference in grain
packing is evident between areas where the detrital
grains are engulfed by the dolomite, and unce­
mented areas which contain abundant macroporos­
ity in the same thin section (Fig. 1 OA). The
dolomite cement displays euhedral rhombic termi­
nations (Fig. 1 0D).
The dolomite cement locally engulfs glaucony
pellets, small pyrite cubes, detrital quartz grains
with minor, rare syntaxial overgrowths, minor to
rare K-feldspars displaying various stages of disso­
lution, and rare kaolin booklets. In two samples
derived from core 3 at Angel-2 (Fig. 9), pore-filling
anhydrite cement locally surrounds the poikilotopic
dolomite cement patches, which are characterized
by straight crystal edges.
In two samples containing abundant intergranu­
lar porosity, derived from Angel-2, poikilotopic
dolomite cement fills subvertical microfractures
that are between 20 and 50 Jlm wide and extend
across the length of the thin sections (Fig. 1 OC). In
both samples the dolomite cement engulfs a high
proportion of angular to subangular quartz grains
that are smaller than the more rounded quartz
grains in adjacent porous areas. The fracture-filling
dolomite cement displays euhedral to subhedral
terminations, and locally is intergrown with patchy
(pore-filling) poikilotopic dolomite cement that also
exhibits euhedral to subhedral terminations. The
centres of the fractures healed by dolomite cement
are filled with oil (Fig. I OC). Adjacent pores also
contain some oil.
These petrographic data indicate that the over­
all paragenetic sequence is: (i) syndepositional
 Structural controls on seismic-scale carbonate cementation 339

Table 3. Semi-quantitative bulk-rock XRD results for Angel Formation reservoirs in the Angel Field, and Namur
Sandstone samples from the Gidgealpa Field. The dominant carbonate cement is dolomite in the Angel Field, and
calcite in the Gidgealpa Field

Well Sample Qtz Dol Cal Sid Fsp Kao 1/M Pyr Glauc Anh

Dampier sub-basin
Angel- l 1 1 60 D Tr Tr-M Tr Tr
Angel-2 1 1 53 D SD Tr M Tr M M
Angel-2 1 1 57 D CD ? Tr-M M
Angel-2 1 1 54 D SD Tr M Tr Tr Tr
Angel-2 1 1 61 D M Tr M M M
Angel-2 1 1 56 D ? Tr Tr Tr M SD
Angel-3 1 1 58 D Tr M Tr-M
Angel-3 1 1 62 D Tr-M Tr Tr-M
Angel-4 1 1 63 D SD M Tr-M Tr
Ange1-4 1 1 64 D Tr M Tr Tr
Angel-4 1 1 59 D ?Tr M-SD Tr Tr-M
Ange1-4 1 1 65 D Tr Tr-M Tr Tr

Eromanga basin
Gidgealpa-1 9 1 1 66 D Tr Tr M CD SD M
Gidgealpa-1 9 1 1 68 D M-Tr M Tr SD M Tr-M
Gidgealpa- 1 9 1 1 69 D Tr-M Tr-M M Tr-M Tr
Gidgealpa- 1 9 1 1 70 D M CD M M-SD M Tr
Gidgealpa- 1 9 1 171 D Tr-M SD Tr Tr M Tr?
Gidgealpa-23 1 1 72 D SD CD M M-SD M M
Gidgealpa-23 1 1 73 D M-SD SD M Tr M
Gidgealpa-24 1 1 74 SD Tr? D M-SD M Tr
Gidgealpa-24 1 1 76 D Tr Tr M-SD M-SD
Gidgealpa-24 1 1 77 D Tr Tr Tr SD SD Tr-M
Gidgealpa-24 1 1 78 D SD Tr? Tr-M Tr M M

Qtz, quartz; Dol, dolomite; Cal, calcite; Sid, siderite; Fsp, K-feldspar; Kao, kaolin; liM, illite/muscovite; Pyr, pyrite;
Glauc, glaucony; Anh, anhydrite; D, dominant; CD, codominant; SD, subdominant; M, minor; Tr, trace.

glaucony pellets; (ii) early pyrite cement; (iii)

quartz overgrowths, partial dissolution of minor
K-feldspars, and trace to minor kaolin precipita­
tion; (iv) major pore-filling dolomite cement,
broadly synchronous with microfracturing and oil
migration; and (v) local late-stage anhydrite cement
(Fig. 1 1 ).
BSE data reveal that the dolomite cement is
relatively Ca rich (Fig. 1 2) and consists of multiple
generations. An early homogeneous Fe-poor dolo­
mite surrounds the surfaces on some detrital grains,
and is followed by multiple zonations of a relatively
Fe-rich dolomite (the main pore-filling event) and
then a homogeneous Fe-poor dolomite cement
(Fig. 1 3).
Bulk-rock 8 1 3C signatures of the dolomite ce­
ments vary between - 1 4. 1 and - 1 5. 5o/oo, with a
mean of - 1 4. 5 ± 0. 7o/oo. Bulk-rock 8 1 80 values range
from +20.08 to +20.72o/oo, with an average 8 1 80
composition of +20. 39 ± 0.27o/oo (Table 4).
Gidgealpa Field
Semi-quantitative bulk-rock XRD analyses show
that the Namur Sandstone is generally dominated
by quartz, calcite and feldspars, and contains minor
to subdominant amounts of kaolinite, trace to
minor amounts of illite/muscovite, and minor to
subdominant amounts of dolomite, siderite and
ankerite (Table 3).
In thin section, the Lower Namur Sandstone is
characterized by poikilotopic calcite cement that
engulfs quartz grains with local, minor syntaxial
overgrowth development but which more typically
display serrated edges (Fig. 1 4A). The quartz grains
generally have tangential grain contacts, rarely long
contacts, and not uncommonly 'float' in the carbon­
ate cement matrix. The poikilotopic calcite cement
also engulfs remnants of K-feldspars that have
undergone partial dissolution, associated authigenic
kaolin patches, minor siderite micrite, microspar
340 J. Schulz-Rojahn, S. Ryan-Grigor and A. Anderson
 Structural controls on seismic-scale carbonate cementation 341

....
CaO
EARLY LATE

-
Glauconite
Pellets <

Mechanical
Compaction 111111111 1111111 .I o light-coloured

+ intenmediate
-
Pyrite
Cement

Quartz • dark-coloured
Overgrowths ? 11 1 -1 1 11 I ?
0

? I I-II? <9' 0
Feldspar
Dissolution

? I I-III?
Authigenic
Kaolin

Dolomite
Cement ......

Microfracturing
•
•
Anhydrite
Cement

l �o�IGR��ON? I MgO FeO

Fig. 1 1 . Paragenetic sequence for the Upper Angel Fig. 12. Ternary diagram showing the elemental
Formation, Angel Field. Dolomite cementation was a composition of CaO, MgO and FeO (mol%) for the
relatively late, pore-filling diagenetic event that occurred, dolomite cement in Angel Formation reservoirs, Angel
at least in part, synchronously with microfracturing and Field.
hydrocarbon migration (see Fig. l OC). Later-stage
anhydrite cement precipitated in some Angel Formation
sandstones at Angel-2 and, to a lesser extent, at Angel-4.
and spar (Fig. 1 48), and rare euhedral dolomite
Table 4. Bulk-rock carbon and oxygen isotope results for rhombs (Fig. 1 4C). Locally, fractured feldspar
dolomite cement in the Angel Formation, Angel Field.
grains are healed by the calcite. The micritic siderite
Precipitation temperatures for dolomite cement were
calculated using the fractionation factor of Northrop & blotches are typically surrounded by siderite mi­
Clayton ( 1 966) and assuming a marine composition for crospar, and euhedral to subhedral siderite spar,
the original 8 1 80 pore water (81 80 O%o). When
=
which in turn is engulfed by the calcite cement
integrating the calculated dolomite precipitation (Fig. 1 48). The siderite spar developed locally on
temperatures (93-97 'C) with the geohistory plot for
Angel-2 (Fig. 4) an Eocene to Late Miocene age for the the edges of the detrital quartz, and can be seen to
dolomite cement is suggested, closely matching the be partly surrounded by quartz overgrowths which
seismic evidence (Fig. 22c) in turn were engulfed by a second generation of
siderite spar. The remaining pore space was filled by
Yield 8 1 3C 8 1 80 T ( ' C)
the poikilotopic calcite cement. The broad parage­
Well No. (%) (PDB) (SMOW) (0%o)
netic sequence is: (i) early micritic siderite; (ii)
Angel-2 1 1 53 8 - 1 4. 1 5 - 1 0.5 20.08 98.4 feldspar decomposition and/or kaolinization, prob­
Angel-2 1 1 54 26 - 1 4. 1 5 -9.88 20.72 93.4 ably accompanied by minor quartz cementation;
Angel-2 1 1 5 7 22 - 1 4. 1 7 - 1 0 . 1 20.49 95. 1
(iii) small dolomite rhombs, minor siderite mi­
Angel-4 1 1 63 7 - 1 5.54 -1 0.32 20.27 96.9
crospar and spar, and rare ankerite spar; and (iv)

Fig. 10. (Opposite) (A) An example of a porous Upper Angel Formation sandstone with patchy dolomite cementation.
Observe the loose grain packing, which is about the same for both dolomite-cemented and porous areas in the field of
view. Despite its depth, the rock retains a high proportion of intergranular porosity. Plane-polarized light (PPL).
Sample 1 1 5 3, Angel-2, 2704.2 m. (B) Crossed polar view showing the poikilotopic nature of the dolomite cement and
the relatively loose grain packing, which is consistent with a relatively recent origin of the dolomite cement in fairly
unconsolidated sands. Sample 1 1 5 7, Angel-2, Upper Angel Formation, 2704.5 m. (C) Example of a fine-grained,
porous Upper Angel Formation sandstone, locally cemented by dolomite. Note the oil staining (black) that fills the
middle of the microfracture, which is healed by dolomite cement and extends along the length of the field of view
(PPL). In core, this fracture is oriented NW-SE, which is consistent with a Miocene origin. Sample 1 1 54, Angel-2,
2705.5 m. (D) Close-up view (PPL) of pore-filling dolomite cement with euhedral terminations, which is typical for the
Angel Formation samples. Sample 1 1 5 3, Angel-2, Upper Angel Formation, 2704.2 m.
342 J. Schulz-Rojahn, S. Ryan-Grigor and A. Anderson

In the Upper Namur Sandstone the poikilotopic

calcite cement is absent in moderately to poorly
sorted quartz arenites and feldspathic quartz aren­
ites that contain abundant primary porosity and
some secondary porosity. Only minor patches of
siderite micrite and microspar are observed in these
clastics, where euhedral quartz overgrowths are well
developed, albeit not volumetrically significant
(< 5- 1 0%) (Fig. 1 40). The detrital grains are dom­
inated by tangential and long contacts, with rare
sutured contacts.

CARBONATE CEMENT DISTRIBUTION

(MACROSCOPIC SCALE)

Wireline log responses

In both fields, integration of core data with wireline

Fig. 13. BSE image of Angel Formation dolomite
cement, Angel Field. High-Fe dolomite cement is light
grey whereas low-Fe dolomite cement is dark grey. Note
the euhedral crystal terminations and the broad multiple
zoning. Low-Fe dolomite cementation was the final
pore-filling event. There is no evidence for dolomite
dissolution or recrystallization. Slight irregularities in the
dolomite zoning are attributed to irregular crystal
growth. Sample 1 1 63, Angel-4, Angel Formation,
2706.9 m. Scale bar represents 1 00 11m.
major pore-filling poikilotopic calcite cement
(Fig. 1 5). Where carbonate crystals are in contact
with the blue-dye epoxy resin, generally straight
crystal edges are observed in the thin sections.
logs shows that carbonate-cemented zones produce
marked increases in bulk density and resistivity
readings, and are characterized by much lower
neutron porosity values and sonic travel times than
vertically adjacent reservoir zones that lack signifi­
cant carbonate cement.
In the Gidgealpa Field, Namur Sandstone inter­
vals that lack major calcite cementation have sonic
velocities that range from 2978 m/s to 3 8 3 8 m/s
(Fig. 1 6) and bulk log densities of 2 . 3 1 -2 . 3 8 g/cm3 .
In contrast, i n sandstones with significant calcite
cement sonic velocities range from 4 7 1 2 m/s to
5 346 m/s (Fig. 1 6), and bulk densities vary between
2.48 and 2 . 6 5 g/cm3 .
I n the Angel Field (Angel-2), bulk density can be
as high as 2.6 g/cm3 in dolomite-cemented sand­
stone intervals, compared with 2 . 3 5 g/cm3 in sand­
stones without significant dolomite cement
(Fig. 1 7). Examination of resistivity character
shows that the MSFL, LLS anq LLD curves each

Fig. 14. (Opposite) (A) Crossed polar (XP) view of poikilotopic calcite cement in the Lower Namur Sandstone. Sample
1 1 70, Gidgealpa- 1 9, 1 696.2 m (cuttings). (B) Different field of view (XP) of the same sample as shown in (A). Note the
siderite cement (S), which is engulfed by the pore-filling calcite cement. The siderite consists of a micritic variety (dark
core), followed by a microspar and outer sparry siderite cement (arrow). Bulk-rock XRD results also suggest the
presence of minor dolomite in this sample (Table 2). (C) Close-up view (PPL) showing small dolomite rhombs (D) that
precipitated on detrital quartz surfaces (Q) and were engulfed by minor poikilotopic calcite (C) in the Upper Namur
Sandstone. The same diagenetic relationship is consistently evident in cuttings from the Lower Namur Sandstone,
where the calcite cement is abundant. Sample 1 1 8 1 , Gidgealpa-32, 1 822.5 m (core). (D) Example of a well to
moderately sorted sandstone in the Upper Namur Sandstone, showing interconnected primary porosity between
well-developed quartz overgrowths (arrows) on detrital quartz grains. The Upper Namur Sandstone lacks major
calcite-cemented zones, and here quartz cementation was the most recent pore-filling event. Note the weathered
feldspar (F) in the upper right of the field of view (PPL). Sample I I 77, Gidgealpa-24, 1 5 87.4 m (core).
Structural controls on seismic-scale carbonate cementation 343
344 J. Schulz-Rojahn, S. Ryan-Grigor and A. Anderson
EARLY LATE
Compaction 11111111111 111111111111
Quartz
Overgrowths ? 111 1111111111111111
Feldspar
Decomposition ? I l-l?
? Il - l ?
Authigenic
Kaolin
Siderite
Cement ?II -I?
Ankerite
Cement ? Il l ?
Dolomite
Rhombs ? II
Poikilotopic
Calcite
Fig. 15. Paragenetic sequence for the Namur Sandstone,
Gidgealpa Field. Calcite cementation is the major, most
recent pore-filling event in the Lower Namur Sandstone,
where syntaxial quartz overgrowths are minimal. In the
Upper Namur Sandstone, where major calcite-cemented
zones are absent, quartz cementation could have
continued until relatively recent times.
can reach values of 40 Q/m in dolomite-cemented
sandstone intervals of the Angel Formation. These
values are significantly higher than in porous reser­
voir zones of the study area. The sonic log reads
about 3 8 50 m/s in these relatively uncemented
zones, and in dolomite-cemented zones sonic veloc­
ity increases to 5000- 5 5 5 0 m/s (Fig. 1 7).
Correlation studies
In both fields, wireline log correlation shows that
the major carbonate-cemented zones share the fol­
lowing characteristics: (i) they are found in clean
stacked sandstone sequences; (ii) they occur in
discrete layers, separated by porous sandstone inter­
vals; (iii) they are intimately associated with hydro­
carbon pools; (iv) the thickness and relative spatial
configuration of the carbonate-cemented zones vary
over relatively short distances and are unrelated to
facies controls (Figs 9 and 1 6). The major difference
between the carbonate-cemented zones in each field
is their stratigraphical position in relation to the
hydrocarbon pools. In the Gidgealpa Field, the
calcite-cemented zones occur below the oil-water
contact in the lower and middle portions of the
Namur Sandstone, in a stacked sandstone sequence
(Fig. 1 6). In the Angel Field, the dolomite-cemented
zones occur in the upper part of a massive sand­
stone sequence (Fig. 9), both below and above the
present-day gas-water contact in the Angel Forma­
tion (Fig. l 7).
Angel Field
Gamma-ray log correlations show that the Upper
Angel Formation (Mid D. jurassicum to P. iehiense
zones) comprises a fairly massive sandstone se­
quence up to 300 m thick, with a net-sand-to-gross
ratio of more than 90% (Fig. 9). Log correlations
show that major dolomite-cemented zones are ab­
sent at this stratigraphical level in Angel- l , Angel-3
and Angel-4 (Fig. 9). Only at Angel-2 are major
carbonate-cemented intervals identified in the Up­
per Angel Formation, which is consistent with the
available petrographic and XRD data (Table 3). In
Angel-2, individual dolomite-cemented zones are
from a few metres up to 40 m thick, separated by up
to 20 m of relatively clean, porous sandstone (Fig.
1 7). The cumulative thickness of these carbonate­
cemented intervals that occur both above and below
the gas-water contact (Fig. l 7) is 1 64 m, or 67% of
the P. iehiense and D. jurassicum zones at this loca­
tion (Fig. 9).
Gidgealpa Field
The two major calcite-cemented intervals in the
lower and middle portions of the Namur Sandstone
are confined to the sandstones with the lowest API
gamma-ray readings, indicative of a relatively low
clay content (Fig. 1 6). The intervals are separated
by up to 45 m of porous sandstone, characterized by
a relatively high argillaceous content. Individual
carbonate-cemented zones within the two intervals
vary in thickness from 30 em to l l m, and are also
separated by relatively thin, argillaceous sandstone
intervals and minor clayey stringers. The two
carbonate-cemented intervals attain a maximum
'
cumulative thickness of 66 m at Gidgealpa-23
(Fig. 1 6). The thickest carbonate-cemented zone
occurs in the first clean stacked Namur Sandstone
sequence directly overlying the fluvio-lacustrine
Birkhead Formation (Figs 2 and 1 6). In the
Gidgealpa Field, the Birkhead Formation acts as a
reservoir in its lower portion (quartz arenites) and
as a partly ineffective local seal in its upper portion
(volcanic arc-derived sediments containing abun­
dant Al-silicates). The reservoir in the Lower Birk­
head Formation (Fig. 7) is filled to capillary seal
capacity, and about 4 million barrels of dominantly
Permian-sourced oil have leaked through the Upper
 Structural controls on seismic-scale carbonate cementation 345

a Gidgealpa-23 Gidgealpa-20 Gidgealpa-17 a·

GR
(API)
-o- DT
(mslft)
GR
(API)
• DT
(mslft)
GR
(API)
• DT
(mslft)
0 200 1 40 40 0 200 1 40 40 0 200 1 40 40
L._____J L._____J L._____J L______j L._____J L______j

Cadna-Owie
Formation

Murta
Member
DATUM:
To Nll mur Sat. (Seal)

-1600-

-1 800-

Fig. 16. Stratigraphic cross-section along line B-B' (see Fig. 2 1 ) for the Namur Sandstone, Gidgealpa Field. The major
calcite-cemented zones in the Lower Namur Sandstone are easily identifiable on the basis of their relatively high sonic
velocities (shaded), and occur in relatively clean massive sandstone sequences. Observe that the calcite-cemented zones
vary in thickness and spatial configuration between well locations which are only a few kilometres apart. The zones are
not fully cemented but retain some residual porosity, and cannot be correlated.

Birkhead Formation up into the Namur Sandstone
(Boult et a!. , 1 997).
The variability in thickness and lateral continuity
ofboth the individual and the gross calcite-cemented
zones in the Namur Sandstone is demonstrated in
Fig. 1 6. Whereas the major carbonate-cemented
zones are well developed in Gidgealpa-20, carbonate
cementation is restricted to several thin zones in
the basal Namur Sandstone at Gidgealpa- 1 7, lo­
cated less than 1 km to the northeast (Fig. 1 6). At
Gidgealpa-20, the calcite cements increase in abun­
dance toward the top of the first basal stacked
sandstone sequence directly overlying the Birkhead
Formation, whereas at Gidgealpa-2 3, about 1 km to
the south, no such trend is apparent in the same
lithological unit. The data demonstrate that the
calcite-cemented zones are not stratabound (see
Schulz-Rojahn, 1 993, his Fig. 3; Townshend,
1 993), even though a broad stratigraphical control
is indicated for carbonate cement occurrence in the
Gidgealpa Field.
Seismic response
In both the Angel Field and Gidgealpa Field, the
acoustic impedance produced by the major
346 J Schulz-Rojahn, S. Ryan-Grigor and A. Anderson

GR RHOB RESISTIVITY DT SMOOTHED 3D SEISMIC RESPONSE

1 .95 glee 2.95 ACOUSTIC
IMPEDANCE
NPHI SYNTHETIC
0 200 .45 -. 1 5 0.2 2000 500 1 00 5000 1 5000 SEISMIC
API -·-------- F.r�_c;, __ . _ _ _ _ _ _ _ ohm-m �s/m gm/e and GR

2600

2700 ��=-----��
Ill
�
E
i!= 2800
c.
w
Q

Fig. 17. Wireline log characteristics, smoothed acoustic impedance curve and 3D seismic response over the Upper
Angel Formation at Angel-2. Note that the major dolomite-cemented zones (black bars) are identifiable on the basis of
neutron, density, resistivity and sonic log profiles. The zones appear as discrete layers at this location, with a
cumulative thickness of 1 64 m, and are not fully cemented but contain some residual porosity. The dolomite-cemented
zones occur both above and below the gas-water contact (GWC). The smoothed acoustic impedance curve shows that
the zones produce a visible seismic response which is mappable. For an example of a line through the 3D seismic
volume see Ryan-Grigor & Schulz-Rojahn ( 1 995; their Fig. ! Oa,b).

carbonate-cemented zones is considerably higher
than in vertically adjacent lithologies (Figs 1 7 and
1 8).
In the Angel Field, a synthetic seismogram was
generated using a 3 5 Hz zero-phase Ricker wavelet
of reverse polarity, providing a good match with 3D
seismic data (Fig. I 7). The filtered acoustic imped­
ance correlates well with individual seismic reflec­
tions observed below the top Angel Formation
seismic marker at Angel-2, corresponding to the
transition from Tithonian sandstones to the con­
formably overlying Cretaceous claystones (Fig. 3)
(Ryan-Grigor & Schulz-Rojahn, 1 995). Each major
carbonate-cemented zone has produced a separate,
seismically visible change in acoustic impedance
(Fig. 1 7).
In the Gidgealpa Field, reflection coefficients
calculated from the sonic and bulk density logs for
carbonate-cemented sandstone interfaces with clas­
tics that lack major carbonate cement are in the
order of 0.22-0.26, demonstrating a relatively
strong 2D seismic response. The seismic response is
the relatively strong, generally continuous two-peak
event that occurs immediately above the top of the
Birkhead Formation reflector (Fig. 1 9). The ampli­
tudes of the separate peak events vary considerably
between the well locations because of the variable
carbonate cement development. At Gidgealpa-20,
the two major carbonate-cemented zones that occur
within the basal Namur Sandstone (Fig. 1 6) corre­
spond to separate high-amplitude events on the
seismic data (Fig. 1 8). In contrast, at Gidgealpa- 1 7,
where wireline log data indicate the presence of
several thin carbonate-cemented zones at the same
stratigraphical horizon, the low,er peak event is
absent. At Gidgealpa-23, which contains the great­
est cumulative thickness of carbonate-cemented
Namur Sandstone in the Gidgealpa Field (66 m),
the two peaks are separated by a broad trough
(Figs 1 6 and 1 8).
In both study areas the data show that the seismic
response closely matches the observed occurrence
of major carbonate-cemented zones at the well
locations. However, thin bed tuning and limitations
in seismic resolution prohibit delineation of the
thickness and spatial separation of individual
carbonate-cemented zones away from the well loca­
tions. Only the gross carbonate-cemented intervals
 Structural controls on seismic-scale carbonate cementation 347

B Gidgealpa-23 Gidgealpa-20 Gidgealpa-1 7

Tlme GR Velocity Depth Tlme GR Velocity Depth Tlme GR Velocity Depth
(sec) (API) (ft/s) (m) (sec) (API) (ft/s) (m) (sec) (API) (It/a) (m)

§ � �0 � � � �

1.

1753

1 829

Fig. 18. The same line of traverse as shown in Fig. 1 6, illustrating the variable synthetic seismic response over the
major calcite-cemented zones (arrows) in the Lower Namur Sandstone. Note that where the calcite-cemented zones are
thickest (Gidgea1pa-23) two peaks are produced which are separated by a broad trough. In contrast, where only a
relatively thin carbonate-cemented zone is present (Gidgealpa- 1 7) only a small peak is produced. The gross
carbonate-cemented interval is mappable on seismic sections below the 'C' horizon (top Cadna-Owie Formation).

within the reservoir sandstones are mappable on the
basis of their high-amplitude seismic response m
the Angel and Gidgealpa Fields.
Seismic mapping
In both fields carbonate cement distribution follows
the regional structural trends. The carbonate ce­
ments concentrate near the crest of the fields, with
amplitudes decreasing downslope (Figs 20 and 2 1 ).
Both fields have in common that the carbonate
cements concentrate in sandstones in areas where
The fault zone traverses the width of the Angel
Field. Other secondary faults include NW-SE­
trending normal faults, NNW-SSE-trending strike­
slip faults, and NNE-SSW-trending reverse faults.
The southern boundary of the high-amplitude
zone thought to reflect major dolomite cementation
closely follows the northern margin of the major
east-west-trending fault system for about lO km
(Fig. 20). The high-amplitude zone extends in a
wedge-like shape for 27 km towards the northeast,
along the structural axis of the Angel Field/
Madeleine Trend (Fig. I B) covering an area of
2 . •

underlying regional seals are breached. The fields
differ on the basis of the greater structural complex­
ity of the Angel Field relative to the Gidgealpa
Field.
Angel Field area
The 3D seismic data show that four different sets of
faults occur at reservoir level in the Angel Field area
(Fig. 20). The most prominent are ENE-WSW­
oriented strike-slip faults that concentrate in a zone
l km wide south of Angel- l , Angel-2 and Angel-4.
300 km .
The western and eastern boundaries of the high­
amplitude zone are irregular and less well defined,
and do not appear to be associated with any faults.
The high-amplitude zone is restricted to the north­
ern and northeastern sectors of the Angel Field,
with the major proportion of the high-amplitude
zone extending well beyond the present-day field
outline, towards the northeast.
Amplitudes are strongest near the crest of the
Angel Field structure/Madeleine Trend, and de­
crease downslope until they phase out completely in
more basinal areas of the Angel Field, well below
3 48 J. Schulz-Rojahn, S. Ryan-Grigor and A. Anderson

B B'
G I D G EALPA - 2 3 GIDGEALPA - 20 GID GEALPA - 1 7

T
w
0

w
A
y

T
I
M
E

I
N

s
E
c
0 ·
N
0
s

Fig. 19. Seismic traverse along line B-B' (see Fig. 2 1 ) illustrating the major seismic horizons and the high-amplitude
zone caused by major calcite cementation in the Lower Namur Sandstone, Gidgealpa Field. Note the lateral variation
in seismic character of the high-amplitude zone caused by variable carbonate cementation. Horizons: C, top
Cadna-Owie Formation; P, top Permian (Toolachee Formation).

the present-day gas-water contact. The downflank
decrease in amplitudes occurs at roughly the same
rate along both the eastern and western margins of
the northern Angel Field structure. The geographi­
cal centre of the highest amplitude zone is offset
relative to the centre of the Angel Field (as defined
by the gas-water contact) by 1 2 km towards the
northeast. The highest amplitude zone occurs over
an area of about the same size as the present-day
field area (Fig. 20).
Isochron maps produced for the interval between
the top Angel Formation horizon and several shal­
lower reflectors, including reflectors at Mid-Eocene,
Mid-Miocene and Late Miocene level, reveal the
structural evolution of the Angel Field area
(Fig. 22). The reflectors used here are present over
the entire northwestern margin of Australia, and
lack evidence of major erosion (Apthorpe, 1 9 88).
The Mid-Eocene reflector marks the regional
boundary between predominantly clastic to pre-
dominantly carbonate sedimentation in the North
West Shelf, and corresponds to the boundary be­
tween cycles 1 and 2 as described by Apthorpe
( 1 98 8). Cycle 2 is Mid-Miocene to Early Eocene in
age, and consists of the Walcott Formation, a cherty
calcilutite which was deposited in a middle to outer
shelf environment (Apthorpe, 1 98 8). During the
'
deposition of this cycle there is evidence of mild
tectonism and downwarping, which is expressed as
tilting towards the west in the Angel Field area.
Importantly, the seismic reflectors are conformable
both above and below this Mid-Eocene disconfor­
mity, indicating that the sea-floor topography was
essentially flat during the deposition of the Walcott
Formation.
The Late Miocene reflector represents a regional
discontinuity surface produced by a relative sea
level drop, and corresponds to the boundary
between Apthorpe's ( 1 988) cycle 3BLower and
3Bupper· Cycle 3BLower is the Early Miocene to Late
 Structural controls on seismic-scale carbonate cementation 3 49

• COWRALLI·1
•COWAALLI-3
• COWRALLI·2

Amplitude Anomalies

Strong
PRESENT-DAY
STRUCTURAL
CLOSURE D Moderate

D Weak

�.�t:::r=
ANGEL-
• N
-2om- Isopach

�
ANGEL-:1
....
. , -...
"/. eANGEL-3
� •• c
G�y,
_, ••
, '

; 0 5 10

1 \ ;���;�:
A TE
I , t I
km
I , I

STRUCTURAL TREND

N
HIG HEST DECREASING
D AMPLITUDES D AMPLITUDES

Fig. 20. Distribution of the high-amplitude zone

representing major dolomite cementation in the Angel tO km

Formation, Angel Field area. The area depicted is that of

the 3D seismic survey. Major faults are shown in bold.
The high-amplitude zone follows the main structural axis
of the Madeleine Trend, which plunges gently towards the
NNE (dashed arrow). Note that the high-amplitude zone
is restricted to the northern part of the field, extending
beyond the gas-water contact (GWC) in this region. The
highest amplitudes occur near the crest of the Madeleine
Trend, with amplitudes decreasing off-structure until they
can no longer be differentiated on the seismic data.
Observe that the centre of the high-amplitude zone is
offset relative to the centre of the modern-day structural
closure (as defined by the GWC) by I I km in a NE
direction, along the axis of the Madeleine Trend.
Modified from Ryan-Grigor & Schulz-Rojahn ( 1 995).
Miocene Trealla Limestone, an inner-shelf marly
limestone/calcarenite. Cycle 3Bupper represents the
Late Miocene Bare Formation, an inner shelf/
coastal sediment consisting of quartz sandstone and
dolomites, prograding towards the west (Apthorpe,
1 98 8). The seismic character shows gentle pro­
grades of cycle 3BLower below, and gentle downlap
of cycle 3Bu pper above the Late Miocene horizon.
Clearly, the Late Miocene reflector also lacked
major regional topographical relief at the time of
sediment deposition.
Isochron maps between these reflectors and the
top Angel Formation horizon show that no time
closure was available at reservoir level in the Angel
Field area until the Late Miocene (Fig. 22c). A
Fig. 2 1 . Areal extent of the high-amplitude zone
produced by major calcite cementation in the Lower
Namur Sandstone, Gidgealpa Field. The map includes
carbonate cement isopachs produced from the available
well log data, showing a good correlation with the
seismic data. The high-amplitude zone follows the GMI
anticlinal trend (Fig. l A). The highest amplitudes occur
near the crest (darker shade) and amplitudes decrease
off-structure until they can no longer be differentiated on
the seismic data. The carbonate cement volume is largest
over the Gidgealpa South Dome, where the underlying
regional Nappamerri Group seal (Triassic) is breached,
which has allowed migration of Cooper basin-derived oil
into the Jurassic sandstones (see Figs 7 and 23). The
areal extent of the high-amplitude zone decreases
towards the northeast, where the Nappamerri Group seal
gains competence and major gas reserves are trapped in
Permian sediments in the North Dome (see Mcintyre et
a/., 1 989).
NNE-SSW-trending ridge first developed at reser­
voir level during the Mid-Eocene in response to
differential compaction over the Madeleine Trend
(Fig. 22a). From the Mid-Miocene onwards, the
incipient drape structure began to gently tilt to­
wards the northeast and relief increased (Fig. 22b).
By the Late Miocene, the northerly tilt and contin­
ued differential compaction had produced a time
closure at reservoir level, coinciding with the
present-day occurrence of the high-amplitude zone
350 J. Schulz-Rojahn, S. Ryan-Grigor and A . Anderson
thought to reflect major dolomite cementation
(Fig. 22c). From the Late Miocene onward, the con­
tinued northward tilt of the structural trend shifted
the Angel Field closure towards the south, until its
present-day position was attained (Fig. 22d).
Gidgealpa Field
Seismic mapping shows that carbonate cement dis­
tribution broadly follows the structural trend of the
GMI anticline, in a SSW-NNE direction (Figs l A
and 2 1 ) . The carbonate cements extend over an area
up to 7 . 5 km wide and at least 20 km long, covering
2
an area of 1 50 km • Comparison of the carbonate
cement isopach map with a depth structure contour
map for the top Toolachee Formation (Mcintyre et
al., 1 989) demonstrates that the cements concen­
trate near the crest of the Gidgealpa Field in the
Lower Namur Sandstone. The high amplitudes
produced by the carbonate cementation decrease
downflank, and phase out completely in more off­
structure regions of the field.
The 'P' (top Permian) to 'C' (top Cadna-Owie
Formation) isochron (Fig. 1 9), which includes the
Namur Sandstone interval, is fairly constant. In
contrast, the surface datum to 'C' horizon isochron
thins appreciably over the Gidgealpa Field struc­
ture, with major isochron variations of about 1 00
and 700 ms, consistent with major Tertiary uplift
and erosion. Shallow faulting in the Tertiary and
Fig. 22. Schematic diagram
illustrating the evolution of closure
development in the Angel Field,
relative to the present-day
occurrence of the high-amplitude
zone representing major dolomite
cementation in Angel Formation
sandstones. The figure is based on
isochron mapping of regional
reflectors at (a) Mid-Eocene, (b)
Mid-Miocene and (c) Late
Miocene level (see text). The data
show that no structural closure
existed at reservoir level prior to
the Late Miocene (c), and that
during this time the incipient
structure was at the same location
as the modern-day high-amplitude
zone. Since this time, the
continued northward tilt of the
Madeleine Trend has shifted the
Angel Field closure 1 2 km SSW of
the high-amplitude zone (d).
Quaternary sediments is evident over the Gidgealpa
Field structure.
The carbonate cements concentrate in three main
areas at Gidgealpa (Fig. 2 1 ). The largest and thick­
est area of carbonate cementation occurs over the
South Dome, where the bulk of the Jurassic oil
reserves are located in the Gidgealpa Field. In this
area, the Nappamerri Group seal thins on to the
Gidgealpa structure, primarily due to erosion, and
in some places may be absent (Mcintyre et al.,
1 989). Over the Gidgealpa North Dome, where the
Nappamerri Group seal is significantly thicker than
over the South Dome, Eromanga basin oil pools are
absent (Mcintyre et al., 1 989) and carbonate­
cemented zones in the Namur Sandstone are much
reduced in overall size and thickness relative to the
South Dome (Fig. 2 1 ). In the no,rthern areas, where
the smallest carbonate cement occurrence is identi­
fiable on seismic data, Cooper basin sediments
contain major gas reserves, below competent Nap­
pamerri Group and older seals.
Volumetric calculations
Calibration of seismic results with wireline log data
shows that the carbonate cement volume is several
times larger in the Angel Field than in the
Gidgealpa Field area.
In the Gidgealpa Field, the lowest closing contour
of the gross calcite-cemented zone occupies an area
 Structural controls on seismic-scale carbonate cementation
of 6 6 km 2 , and the highest closing contour an area
of 1 7 1 km 2 (Fig. 2 1 ). The bulk-rock volume of the
calcite-cemented interval (excluding sandstones
that lack major carbonate cementation) is estimated
to be 1 .22 km3. If we assume that the carbonate
cement constitutes between 20 and 30% of the
bulk-rock volume across the field (as indicated by
the available petrographic data), then the calcite
cement that concentrates in a crestal position could
form a body that is between 0.22 and 0 . 3 7 km3
large in the Lower Namur Sandstone.
In the Angel Field, the bulk-rock volume of the
highest amplitude zone is 1 . 6 km3 in the Upper
Angel Formation (Fig. 20). If a dolomite cement
volume of 30% is used, then the volume of carbon­
ate cement is 0.52 km3 in this highest-amplitude
area, to which we must add the bulk-rock volume of
the surrounding, lower-amplitude region (Fig. 20),
which we estimate at 4.5 km3. Using a dolomite
cement percentage of 1 0%, the volume of carbonate
cement is 0.45 km3 in this lower-amplitude region.
Thus, the cumulative volume of dolomite cement in
these. zones is almost I km3 in the Angel Field area,
with possible minimum and maximum values for a
range of different carbonate cement contents be­
tween 0.6 and 1 .4 km3.
SUMMARY OF
RESERVOIR CHARACTERISTICS
The fields have many of the same characteristics but
also differ substantially. Table 5 summarizes the
various characteristics of the hydrocarbon-bearing
formations in the Angel and Gidgealpa Fields in
which the major carbonate-cemented zones occur.
In both fields, the carbonate-cemented zones form
large bodies in Upper Jurassic sandstones and occur
in relatively massive, clean sandstones. The
carbonate-cemented zones are located at depths of
1 . 5-3 km at temperatures close to or in excess of
1 0o ·c. In the Gidgealpa Field, the dominant car­
bonate cement is calcite, occurring in the fluvial
Namur Sandstone. In the Angel Field, dolomite is
dominant in Angel Formation reservoirs of a ma­
rine mass-flow origin.
CONTROLS ON
CARBONATE CEMENT DISTRIBUTION
In both fields there are strong arguments for the
351
coincidence of timing between carbonate cementa­
tion, structural development and hydrocarbon mi­
gration. Evidence from seismic data shows that the
major carbonate-cemented zones concentrate along
major structural trends, near the crest of structures
that were not available at the time of deposition of
the Jurassic reservoir sandstones. Seismic evidence
and regional geological considerations show that the
structures formed during the Tertiary, in response
to different tectonic compressive events (Fig. 6a,b).
In the Angel Field area no structural closure existed
at reservoir level before the Late Miocene
(Fig. 22c). In both geological provinces, geochemi­
cal data show that major hydrocarbon generation
and migration also occurred during the Tertiary
(Fig. 6a,b). Therefore, the close association of the
major carbonate-cemented zones with hydro­
carbon-bearing structures strongly points towards a
Tertiary control on carbonate cementation in the
study areas.
In either field, the observed carbonate cement
distribution cannot be explained by preferential
cement dissolution in downdip position because of
(i) the lack of petrographic evidence, and (ii) the
lack of a C02 gas cap in the reservoirs. If it is
hypothesized that cement dissolution was initiated
prior to Tertiary structure development, it needs to
be explained why the location of the remnant
carbonate bodies should coincide with the location
of the modern-day structural trend in either geolog­
ical province. The dissolution model would need to
shed light on the areal extent of the carbonate
cements prior to major dissolution, provide an
estimate of the amount of cement that was re­
moved, and outline where the dissolved mineral
components were transported to. Certainly in the
Eromanga basin, major carbonate-cemented zones
with a cumulative thickness of up to I I 0 m are
common in Jurassic reservoirs of hydrocarbon­
bearing structures throughout the basin (Table 1 ;
Fig. I A). This fact makes it highly improbable that
regionally extensive dissolution could have oc­
curred in downflank areas, because of the diffi­
culties in envisaging where the required large
volumes of carbonate cement could have originated
in the first place (bearing in mind that the carbonate
cements are not stratabound). The conclusion is
supported by low modern-day C02 gas concentra­
tions in Jurassic reservoirs (2-3% by molar volume)
(Vincent et al. 1 98 5 ; Armstrong & Barr, 1 986) and
the fact the C0 2 has generally much lower 8 1 3C
values (mean - 1 1 .4o/oo) than the calcite cement that
352 J. Schulz-Rojahn, S. Ryan-Grigor and A . Anderson

Table 5. Comparison of the characteristics of the major carbonate-cemented zones and their Jurassic host sandstones
in the Angel and Gidgealpa Fields, including present-day temperatures and organic maturity for adjacent source rocks.

Angel Field (Dampier sub-basin) Gidgealpa Field (Eromanga basin)

Reservoir age Late Jurassic (Angel Formation) Late Jurassic (Namur Sandstone)
Reservoir thickness 220-250 m 1 77-284 m
Environment of deposition Marine: mass-flow or grain-flow Fluvial: high-energy braided to
dominated submarine fan system meandering stream environment
Reservoir depth (mkb) 2700-2950 m 1 470- 1 90 1 m
Reservoir temperature ( C)
• ;;. I 00- 1 04 ;;.80- 1 0 5
Rv max. (%) in adjacent source rocks 0. 5-0. 7 0.45-0.6
Cumulative thickness of cemented up to 1 64 m (Angel-2) up to 66 m (Gidgealpa-23)
zones
Approx. volume of carbonate cement I km3 (0.6- 1 .4 km3) 0.2-0.4 km3
Dominant cement Dolomite Calcite
Mean 0 1 3CPDB - 1 4.5 ± 0.7 -6.25 ± 3 . 7*
Mean o 1 80sMOW +20.39 ± 0.27 +1 1 .8 ± 1 .8*

*Stienstra ( 1 992).

occurs in these reservoirs (mean -6.25%o) (Fig.
8a,b,e). These considerations emphasize that
modem carbonate cement occurrence near the crest
of the Angel Field and Gidgealpa Field must be the
result of relatively recent structure-controlled pre­
cipitation, not dissolution.
In both fields, hydrocarbon pools and the major
carbonate-cemented zones occur in Jurassic sand­
stones above areas where regional seals are breached
by non-deposition or erosion (Figs 7 and 23). In the
Angel Field, gas and condensates that were sourced
in the deeper Triassic Locker Shale migrated upward
along the Jurassic Dingo Claystone seal (Fig. 3) that
onlaps the Angel Field high, and then spilled into the
Angel Formation reservoir (Fig. 23) (Di Toro, 1 994).
In the Gidgealpa Field, breaching of the Lower Nap­
pamerri Group seal (Fig. 2) made possible upward
migration of Permian-sourced oil into Jurassic res­
ervoirs in the South Dome, where the bulk of the
calcite cement is located (Fig. 23). In contrast, the
lower Nappamerri Group seal is present at the North
Dome, where oil pools in Jurassic reservoirs are ab­
sent (Mcintyre et a/. , 1 989) and calcite-cemented
zones are relatively minor in the Lower Namur
Sandstone (Fig. 2 1 ). These data strongly suggest that
a migration-related control from deeper sequences
was responsible for major carbonate cementation in
the Jurassic sandstones, possibly along fault planes,
broadly synchronous with hydrocarbon charging in
both areas.
Supporting evidence for a late-stage ongm of
carbonate cements along petroleum migration con­
duits includes the strong statistical correlation be­
tween oil pools and major calcite-cemented zones in
Jurassic sandstones of the Eromanga basin. In both
the Angel Field and Gidgealpa Field, petrographic
data show that the major pore-filling carbonate
cement is a relatively late diagenetic event (Figs 1 1
and 1 5). In the Angel Field, the association of
hydrocarbons with near-vertical fractures that are
oriented NW-SE and healed by dolomite cement
(Fig. 1 OC) is a good indication that Miocene tec­
tonic activity broadly coincided with hydrocarbon
migration and dolomite precipitation. In this field,
the fact that the dolomite cements occur both below
and above the gas-water contact is consistent with
the view that hydrocarbon empl,acement took place
synchronous with, or after, dolomite cementation
in the area. In both fields, the lack of facies control
on carbonate cement occurrence (vertical and spa­
tial variability) demonstrates that precipitation con­
trols must have been related to postdepositional,
relatively late geological events, rather than shallow
burial diagenesis. In Angel Formation reservoirs,
the fact that the same loose grain packing is evident
in both uncemented and carbonate-cemented zones
(Fig. 1 OA) provides evidence that dolomite cemen­
tation could have been a relatively recent phenom­
enon in relatively unconsolidated sands currently
buried to a depth of between 2 700 and 3000 m.
 Structural controls on seismic-scale carbonate cementation 353

Gidgealpa Field Towards Basin Margin

sw Structure

Gidgealpa Field Angel Field

(Tertiary-Recent) (Miocene)
D····· '
M aJor seal s
�
�
Major carbonate-
cemented zones
11 Major migration fairways of
__./ · hydrocarbons and C02-charged fluids
,.___ Direction of modern
groundwater flow

Fig. 23. Schematic illustration of the role of closure availability and seal integrity on major carbonate cement
occurrence (shaded) in the Gidgealpa Field and Angel Field. In both fields, the carbonate-cemented zones concentrate
in Upper Jurassic sandstones above areas where regional seals are breached in older sequences owing to
non-deposition or erosion. The availability of plumbing has allowed the partial filling of the' Jurassic reservoirs in the
Tertiary structures with hydrocarbons sourced from deeper sequences. The observations point towards a major
migration-related control on carbonate cementation in both fields.

Last but not least, Stienstra ( 1 992) cites fluid the original pore water had a marine () 1 80 compo­
inclusion microtherrnometry data by Bone ( 1 989) sition (o 1 80 Oo/oo). Using this value, and the frac­
=

showing that poikilotopic calcite cements precipi­
tated at temperatures between 85 and 1 20 " C in the
Jurassic Adori Sandstone (Fig. 2) of the Tantanna,
Marana and Strzelecki Fields (Fig. 1 A) (Bone,
1 989). In these fields, the calcite cement has the
same petrographic and isotope characteristics
(Fig. 8b) as the calcite in the Namur Sandstone of
the Gidgealpa Field.
The relatively low o 1 80 values of the Angel Field
and Gidgealpa Field carbonate cements can be
explained in terms of either a meteoric-water incur­
sion and/or elevated precipitation temperatures.
Low () 1 80 values in minerals can also be achieved
through the interaction between sea water and
volcaniclastic sediments under low-temperature
conditions (Morad & De Ros, 1 994). However,
because of the late-stage origin of the carbonate
cements and the fluvial nature of the Namur Sand­
stone, this possibility need not be considered in the
present context.
In the Angel Formation, given the marine nature
of the sediments, it seems reasonable to assume that
tionation equation for dolomite of Northrop &
Clayton ( 1 966), the Angel Formation dolomite
cements must have formed at temperatures be­
tween 93 and 9 7 " C (Table 4). When integrating
these results with the geohistory plot at Angel-2
(Fig. 4), a broadly Eocene to Late Miocene age for
the dolomite cements is suggested, which is in good
agreement with the seismic evidence (Miocene)
(Fig. 22c).
Although the marine o 1 80 composition of the
original pore water could have been modified dur­
ing sediment subsidence, such as in response to
water-rock interactions or meteoric incursion (e.g.
Longstaffe & Ayalon, 1 987; Longstaffe et a!., 1 992),
which would lead to different estimates of dolomite
precipitation temperatures, we see no compelling
reason to make this assumption. The minor kaolin­
ite which is present in the Angel Formation sand­
stones (Table 3) does not necessarily require
meteoric invasion, but can also be produced by
generation of C0 2 or organic acids generated from
the thermal decarboxylation of organic matter
354 J. Schulz-Rojahn, S. Ryan-Grigor and A . Anderson
(Souza et al., 1 995). The good general agreement
between the isotopic and seismic data provides
strong evidence that the Angel Formation dolomite
cements precipitated from relatively hot fluids dur­
ing Miocene times.
In the Gidgealpa Field, 8 1 80 compositions of
pore water from which the Namur Sandstone cal­
cite cements precipitated are uncertain. Present-day
formation waters in the Jurassic aquifers have 8 1 80
compositions close to -6.6%o over large areas (Airey
et al., 1 979). As Australia is at its northernmost
latitude since at least the Late Carboniferous
(Veevers, 1 984) the isotopic composition of Austra­
lian meteoric water is unlikely to have been higher
than -6.6%o during the last 280 million years (Late
Carboniferous). Therefore, using this 8 1 80 value
and the fractionation equation of Friedman &
O'Neil ( 1 97 7), the maximum temperature at which
the calcite cements could have precipitated is be­
tween 66 and 1 34 · c (mean 9 5 S C), which partly
exceeds current reservoir temperatures (Table 5).
More probably, the 8 1 80 composition of the origi­
nal meteoric water was lower than -6.6%o. Because
of Australia's northward drift, the 8 1 80 of meteoric
waters is thought to have become progressively 1 80
enriched since the Mid-Cretaceous on this conti­
nent (Bird & Chivas, 1 988). Accordingly, if we
assume a 8 1 80 value of -9%o for the original
meteoric water (i.e. less 1 80-enriched in the past
than in the present), the calcite cements must have
formed at temperatures between 50 and l O ? " C
(mean 7 5 . 4 • C). Although the 8 1 80 o f the original
meteoric water cannot be precisely determined, it is
clear that the calcite cement formed from relatively
hot fluids, consistent with relatively deep burial
(> 1 km) (Fig. 5).
The carbon isotopic values of Angel Formation
dolomite cements (mean - 1 4. 5 ± 0. 7%o) are very
consistent between samples (Table 4) and indicate a
major organic derivation of carbon, either from
kerogen maturation or from a hydrocarbon source.
In contrast, the 8 1 3C values of Namur Sandstone
calcite cements (-2.5%o to - 1 2. 9%o) (Fig. 8a,b) can
be explained by mixing of a wide range of carbon
sources, including marine carbon (Hudson, 1 97 7),
atmospheric carbon (Platt, 1 992; De Ros et al.,
1 994), carbon produced from bacterial fermenta­
tion reactions (!twin et al., 1 97 7), and carbon
derived from organic matter (Hudson, 1 97 7 ; !twin
et al. 1 977), including from hydrocarbons (Dono­
van, 1 974; Donovan et al., 1 974; Hovland et al.,
1 98 7). Both carbon derived from kerogen and
hydrocarbons could partly account for i) 1 3C values
as negative as - 1 2. 9%o for some Namur Sandstone
calcite cements (Fig 8a).
In view of the intimate association of the major
carbonate-cemented zones with hydrocarbon­
bearing reservoirs, and the fact that carbonate
cementation appears to have occurred broadly syn­
chronously with hydrocarbon generation, we must
consider the possibility of a hydrocarbon-related
origin of the c2rbonate cements in both study areas.
Carbonate cement 'haloes' associated with hydro­
carbon pools are well documented, and commonly
attributed to the microbial oxidation of crude oil or
methane in different geological settings (Gould &
Smith, 1 97 8 ; Smith, 1 978; Faber & Stahl, 1 984;
Oehler & Sternberg, 1 984; Hovland et al., 1 987;
O'Brien & Woods, 1 99 5). However, a number of
observations point towards this type of precipita­
tion mechanism not being appropriate in the con­
text of the Angel Field and Gidgealpa Field areas:
I Based on gas chromatography results, conden­
sates show no evidence of biodegradation in the
Angel Formation at Angel- l and Angel-2 (Brikke,
1 982), which is consistent with regional geochemi­
cal data from nearby wells (Gould & Smith, 1 97 8 ;
Philp e t al., 1 98 1 , 1 982; Volkman e t al., 1 98 3 ;
Kopsen & McGann, 1 985). Cooper and Eromanga
basin hydrocarbons are also devoid of any signs of
microbial alteration, based on a large geochemical
database incorporating data from hundreds of ap­
praisal and production wells. Only oil from a single
Eromanga basin well (Bodalla South-2) shows evi­
dence of slight biodegradation in its gasoline frac­
tion in the Namur Sandstone (D. McKirdy,
University of Adelaide, personal communication).
2 Bacterial alteration of hydrocarbons is generally
restricted to reservoirs cooler than 82 • C because
biological activity ceases at higher temperatures
(Hunt, 1 979). Angel FormatiQn temperatures ex­
ceeded 800C from the Pliocene onwards, and were
close to the present-day maximum burial tempera­
ture ( 1 00- 1 04 • C) during the Miocene (Fig. 4).
3 In the Gidgealpa Field, the major calcite­
cemented zones occur up to l 00 m below the
present-day oil-water contact (Fig. 7), with the
thickest interval in the basal portion of the Namur
Sandstone, furthest away from the existing oil pool
(Fig. 1 6). In the Angel Field, major dolomite­
cemented zones are absent within hydrocarbon­
bearing sandstones on the south side of the field
(Fig. 20).
4 In the Gidgealpa Field, multiple stacked oil pools
 Structural controls on seismic-scale carbonate cementation
occur i n the Hutton Sandstone underlying the Na­
mur Sandstone, but at this stratigraphical interval
major calcite-cemented zones are absent (Fig. 7).
5 In the Cooper-Eromanga basin province, liquid
and gaseous hydrocarbons have much lower o 1 3C
values (-42%o to -20%o) than the calcite cement
(- 1 2.9 to -2.5%o) (see Rigby & Smith, 1 98 1 ;
McKirdy, 1 982; Vincent et al., 1 985). Although
Dimitrakopoulos & Muehlenbachs ( 1 987) suggest
that biodegradation of petroleum can provide a
source of 1 3C-enriched C02 in the formation of
carbonate cement (o 1 3C - l .2%o to + 1 4. 3%o), they
= advanced diagenesis, C02 gas is effectively retained,
consider this diagenetic process to be a near-surface
phenomenon, which is clearly inappropriate in the
present context. No isotopic values of hydrocarbons
are available in the Dampier sub-basin.
We conclude that the major source of carbon for
dolomite cementation in Angel Formation reser­
voirs was probably C0 2 derived from kerogen mat­
uration, rather than microbial oxidation of hydro­
carbons. In the Gidgealpa Field, carbon for calcite
cementation could have originated from a variety of
sources, including organic matter, sporadic bioclast
material and early (methanogenic) siderite cement,
which has o 1 3C values between -2%o and +4%o in
Jurassic Eromanga basin sediments (Fig. 8c). How­
ever, these carbon sources do not readily explain the
late-stage, structurally controlled mode of Namur
Sandstone calcite cementation above the locally
breached Nappamerri Group seal.
In the simplest case, the availability of plumbing
for hydrocarbons probably also means the availabil­
ity of plumbing for brines. Because the microbial
degradation of hydrocarbons is not a probable
source of carbon for carbonate cementation, up­
welling of C0 2 -charged brines along petroleum mi­
gration conduits must be considered to explain the
observed geological phenomena in both study areas.
In the Gidgealpa Field the o 1 3C signatures of the
Namur Sandstone calcite cement (- 1 2. 9 to -2.5%o)
broadly overlap with the o 1 3C character of C02 gases
in Permian reservoirs (-1 l . 7%o to +0. 3%o)
(Fig. 8a,b,d), which supports the theory of brine up­
welling. Because C02 solubility is pressure depen­
dent (Bray & Foster, 1 980) and the C02 gases are
derived from Permian sediments located at depths
between 2 and 3 km (Rigby & Smith, 1 9 8 1 ), the C02
must have been mostly in solution at Permian reser­
voir level. Even when taking into account the effects
of calcite-C02 fractionation, it is clear that a major
proportion of the carbon incorporated into the Na­
mur Sandstone calcite cement could have been
355
sourced from the Cooper basin, like the hydrocar­
bons trapped in this sandstone.
Regional patterns of C02 concentration in Cooper
basin reservoirs (Fig. 24a) are consistent with C0 2
migration into Eromanga basin reservoirs near the
Cooper basin margins and along structural highs,
primarily because of breached Nappamerri Group
seal integrity in these areas (Fig. 24b) (Schulz­
Rojahn, 1 993). In contrast, where Cooper basin sed­
iments are characterized by competent seal develop­
ment owing to downbasin facies changes and
leading to concentrations in excess of 40% by molar
volume in some basinal areas (Fig. 24b) (Hunt et al.,
1 989).
Brine upwelling can be accomplished by a variety
of mechanisms, including compactional dewatering
(Magara, 1 976, 1 98 1 ), thermal convection currents
(Wood & Hewlett, 1 982; Bj0rlykke & Egeberg,
1 993), temperature-dependent clay dehydration
(Magara, 1 97 5), seismic pumping (Sibson et al.,
1 97 5 ; Sibson, 1 987) and/or tectonic compression
(Oliver, 1 986).
In the North West Shelf, irrespective of the
mechanism of upwelling fluid movement, various
lines of evidence indicate that such upwelling has
occurred and may be continuing. Upwelling of hot
basinal brines is evidenced by high heat-flow re­
gions in the North West Shelf (Swift et al., 1 988),
including fluid inclusion studies (Eadington &
Hamilton, 1 990). Apatite fission track data provide
evidence for a well-developed Late Tertiary heating
event in the Timor Sea, which O'Brien ( 1 996)
attributes solely to the upward migration of hot
saline brines from deeper sequences. The formation
of anhydrite cement after carbonate cementation,
as observed in Angel Formation reservoirs, is com­
monly attributed to the upwelling of saline basinal
brines in different clastic settings. (e.g. Fisher &
Kreitler, 1 9 87; Lee & Bethke, 1 994; Morad et al.,
1 994; McNeil et al., 1 995).
In the Angel Field, Angel Formation waters are
characterized by salinities between 35 000 and
37 000 ppm (Woodside, 1 9 72), consistent with this
interpretation. Brine-upwelling probably occurred
episodically, as indicated by the multiple zoning in
the dolomite cement. In this field, whatever the
trigger for the major dolomite cementation, the
bulk of the cements must have formed in less than
1 0 million years because the incipient Angel Field
structure developed between the Mid-Eocene and
the Late Miocene (Fig. 22a-c), and since the Late
356 J. Schulz-Rojahn, S. Ryan-Grigor and A. Anderson

(a) (b) NE

• TERTIARY & CRETACEOUS • TRIASSIC CAP ROCK

(Nappamerri Group) 0 PRE·PERMIAN

0 JURASSIC (J·AOUIFERS) o. PERMIAN SEDIMENTS \ FAULT

incl. Namur Sandstone

Fig. 24. (a) C0 2 concentrations (% molar volume) within Toolachee Formation (top Permian) gas reservoirs, southern
Cooper basin. The same trend of increasing C02 concentrations towards the basin depocentres is evident in older
Permian formations in this basin (modified from Hunt et a/., 1 989). (b) Schematic cross-section from the basin margin
to the central Nappamerri Trough showing the possible interrelationship between porosity, seal development and C0 2
concentrations in the Permian gas reservoirs. Towards the Cooper basin depocentres, the predominance of
microporosity, coupled with overall decreasing sand percentages and more competent seal development, could have
led to the concentration of C02 over geological time. Upwelling of COr rich brines from the Permian sediments into
Jurassic sandstones may have been facilitated where the Triassic regional Nappamerri Group seal is breached owing to
erosion or non-deposition, near the Cooper basin margin or along fault-bounded structures. Modified from
Schulz-Rojahn ( 1 993).

Miocene the closure has moved to the south be­
cause of structural tilting. Wood & Boles ( 1 9 9 1 )
consider that large-scale tectonic processes can trig­
ger major carbonate cementation within years or
even months, and it is possible that the Angel
Formation dolomite cements formed on the same
short timescale.
In both the Angel and Gidgealpa Field areas, ma­
jor carbonate cementation could have been triggered
as the result of the decrease in fluid temperature,
pressure and C02 solubility in the ascending fluids.
The precipitation mechanism was proposed by var­
ious workers on the basis of theoretical (Capuano,
1 990) and geological observations in different geo­
logical settings (e.g. Lundegard & Land, 1 986; Leach
et a!., 1 9 9 1 ; Wood & Boles, 1 9 9 1 ). Alternatively, in
the presence of pH-buffering agents the C0 2 in the
ascending brines could have been sequestered during
massive carbonate cementation in the reservoir
sandstones of the study areas. The role of carboxylic
acid anions in pH-buffering reactions has been ex­
tensively discussed in recent years (e.g. Hanor et a!.,
1 993; Lundegard & Land, 1 993; MacGowan & Sur-
dam, 1 993). Other workers have emphasized the im­
portance of Al-silicates for the pH levels and buffer
capacity of formation waters (Hutcheon, 1 989;
Smith & Ehrenberg, 1 989; Barth & Bj0rlykke, 1 993;
Hutcheon e t a!., 1 993). Although insufficient data
are available to differentiate between these different
models, clearly several potential mechanisms for
transport and precipitation of C02-rich brines exist
in the study areas. The important role of variable
pore fluid chemistry and reservoir interconnection
in sandstones probably accounts for the fact that
carbonate cementation occurs in discrete layers at
certain stratigraphical levels in the study areas. The
case for late-stage carbonate cementation controls is
not limited by having no identifiable transport or
precipitation mechanisms.
I MPLICATIONS FOR
PETROLEUM EXPLORATION
The conclusion that major carbonate-cemented
zones formed near the crest of hydrocarbon-bearing
 Structural controls on seismic-scale carbonate cementation
structures i n response t o a migration-related control
from deeper sequences has important implications
for petroleum exploration in the Eromanga basin
and the North West Shelf (Schulz-Rojahn, 1 99 3 ;
Ryan-Grigor & Schulz-Rojahn, 1 995). The two
case studies show that, owing to the large contrasts
in acoustic impedance associated with major
carbonate-cemented zones, seismic delineation
of this phenomenon can guide explorers to
petroleum-bearing structures, including probably
subtle traps.
In the Eromanga basin this conclusion is sup­
ported by the strong statistical correlation between
the occurrence of major calcite-cemented zones and
oil pools in Jurassic sandstones. This concept is
particularly valuable if a correlation can be proved
between the size of the hydrocarbon pools and the
size of the carbonate cement volume (intensity and
areal extent of high-amplitude events).
In cases where Eromanga basin structures contain
major carbonate-cemented zones but are without
oil discoveries in Jurassic sandstones, it is possible
that either the hydrocarbons have leaked into
younger sequences via partially breached seals, or
that drilling has been off the crest of these structures
and that updip oil pools remain to be discovered (B.
Jensen-Schmidt, MESA, personal communication).
However, as the relationship between the oil
pools and the major calcite-cemented zones is only
indirect, not all calcite-cemented zones will be
associated with oil pools.
In the North West Shelf insufficient data are
available to establish whether major carbonate­
cemented zones are common in petroleum-bearing
reservoirs in many different fields; however, at least
in some deeply buried reservoirs (> 2 km) signifi­
cant dolomite cementation is known. In the Talis­
man Field, located 1 2 km east of the Angel Field,
ferroan dolomite is the dominant cement within the
oil-bearing Talisman Sandstone that overlies the
Angel Formation (Ellis, 1 988). Ellis ( 1 988) inter­
preted the dolomite cement to have developed
synchronously with major oil entrapment during
the Late Cretaceous to Early Tertiary. In the Wa­
naea and Cossack Fields (Fig. I B) locally very
significant dolomite cementation in Angel Forma­
tion reservoirs was noted by Di Toro ( 1 994). Di
Toro ( 1 994) considered the dolomite cement to be
early and pre-dating quartz cementation, even
though he observed that the dolomite is more
common towards the base of the oil column.
The data encourage explorers to investigate the
357
areal extent of major carbonate cementation in the
North West Shelf. The possibility exists that major
carbonate cement occurrence can provide impor­
tant clues towards understanding petroleum migra­
tion pathways, the evolution of structural closure
and the location of subtle traps in this region.
Minimum criteria for identification of other clas­
tic provinces where this exploration concept may be
applicable include (i) geological settings character­
ized by a late-stage compressional regime; (ii) coin­
cidence of timing of late-stage hydrocarbon
migration with the late-stage compressive event;
and (iii) the availability of plumbing for hydrocar­
bon and brine migration from deep sequences into
structures (locally breached regional seals acting as
focused migration conduits).
CONCLUSIONS
The use of 2D and 3D seismic data and wireline
logs can greatly assist in the understanding of the
subsurface extent, geometry and structural timing
of carbonate cementation in some clastic provinces.
When integrated with routine diagenetic techniques
such as petrography, isotope analysis, SEM and
BSE imaging, these geophysical methods can pro­
vide important clues towards explaining carbonate
cementation processes in some clastic reservoirs.
Poikilotopic carbonate cements can reduce reser­
voir porosity in relatively clean, massive sandstones
2
over large areas, of the order of at least 300 km
(Angel Field). Based on log characteristics, major
carbonate-cemented zones can attain a cumulative
thickness of at least 1 65 m in marine sandstones
(Dampier sub-basin, Carnarvon basin) and 1 1 0 m
in fluvial sandstones (Eromanga basin). The total
volume of carbonate cement in petroleum fields can
approach I km3, as exemplified by, the Angel Field
case study.
A major structural control on carbonate cemen­
tation is demonstrated in both the Angel and
Gidgealpa Field areas. In both fields, poikilotopic
carbonate cements concentrate in hydrocarbon­
bearing sandstones near the crest of the fields, along
the axis of the main structural trends.
The areal concentration of major carbonate­
cemented zones can reflect the location of incipient
structural closure during the time of carbonate
cementation, rather than the modern-day closure,
as exemplified by the Angel Field case study.
Major carbonate cementation in sandstones can
358 J. Schulz-Rojahn, S. Ryan-Grigor and A . Anderson
take place in a relatively short time, as shown by the
Angel Formation dolomite cements. These dolo­
mite cements formed after the development of
closure at reservoir level during the Miocene,
broadly synchronous with major hydrocarbon
charging into the Angel Field structure. In the
Gidgealpa Field, a Tertiary origin for major calcite
cementation is also suggested by the available struc­
tural evidence, again broadly synchronous with
petroleum generation and migration from deeper
sequences.
The integrity of underlying regional seals is a
major controlling factor on carbonate cement dis­
tribution in sandstones in both the Angel Field and
Gidgealpa Field. These data strongly suggest a
migration-related control from deeper sequences on
carbonate cementation.
Seismic delineation of major carbonate cement
bodies in sandstones may assist petroleum explorers
in identifying the location of subtle traps, including
stratigraphical traps, and the ranking of prospects
and leads for drilling in the Eromanga basin and the
North West Shelf. This exploration concept may be
applicable to other clastic petroleum provinces,
characterized by a late-stage compressive tectonic
regime which triggered structural growth and fluid
upwelling, coinciding with peak hydrocarbon gen­
eration and migration.
ACKNOWLEDGEMENTS
We thank the Cooper Basin Consortium Group of
Companies, Woodside Offshore Petroleum Pty Ltd,
BHP Petroleum (North West Shelf) Pty Ltd, BP
Developments Australia Pty Ltd, Chevron Asiatic
Ltd, Japan Australia LNG (MIMI) Pty Ltd, Shell
Development (Australia) Pty Ltd, and Mines and
Energy South Australia (MESA) for their support.
We are further indebted to Ms Eleanor Alexander,
Mr Peter Hough and Alan Sansome (MESA), Mr
Larry Tilbury (Woodside), Mr Peter Boult (Univer­
sity of South Australia), Mr Alex Kaiko (NCPGG),
Dr Keith Turnbull and Mr John Stanley (University
of Adelaide) for their kind assistance. The manu­
script was greatly improved by the constructive
criticisms of lAS reviewers Drs Lori L. Summa
(Exxon Production Research Company, Houston),
Thomas L. Dunn (University of Wyoming) and
Professor Sadoon Morad (University of Uppsala).
Alan Anderson thanks Mr Andrew Mitchell
(NCPGG), Professor David Boyd (University of
Adelaide) and Mr Doug Roberts (SAGASCO) for
their supervision of his Honours thesis in 1 985, on
which some of the data in this paper are based.
Opinions expressed are those of the authors. Sarah
Ryan-Grigor gratefully acknowledges financial sup­
port by the University of Adelaide, and Jorg Schulz­
Rojahn by the Australian Research Council (ARC)
and the NCPGG.
REFERENCES
AIREY, P.L., CALF, G.E., CAMPBELL, B.L. et a/. (1979)
Aspects of the hydrology of the Great Artesian Basin,
Australia. In: Isotope Hydrology Proc. Int. Symp. Iso­
tope Hydrology, International Atomic Energy Agency,
Vienna, 1, 205-2 19.
AMBROSE, G., SUTTILL, R. & LAYERING, I. (1986) The
geology and hydrocarbon potential of the Murta Mem­
ber (Mooga Formation) in the southern Eromanga
Basin. In: Contributions to the Geology and Hydrocar­
bon Potential of the Eromanga Basin (Eds Gravestock,
D.I., Moore, P.S. & Pitt, G.M.). Geol. Soc. Aust. Spec.
Pub!., 12, 7 1-84.
ANDERSON, A. ( 1 985) A geoseismic investigation ofcarbon­
ate cementation of the Namur Sandstone in the
Gidgea/pa Field, Eromanga Basin. BSc Honours thesis,
University of Adelaide, Department of Geology and
Geophysics.
APAK, S.N., STUART, W.J. & LEMON, N.L. (1993)
Structural-stratigraphic development of the Gidgealpa­
Merrimelia-lnnamincka Trend with implications for
petroleum trap styles, Cooper Basin, Australia. Aust.
Petrol. Exp/. Ass. J., 33, 94-104.
APTHORPE, M. ( 1988) Cainozoic depositional history of
the North West Shelf. In: The North West ShelfAustralia
(Eds Purcell, P.G. & Purcell, R.R.), pp. 56-84. Petro­
leum Exploration Society of Australia, Perth.
ARMSTRONG, J.D. & BARR, T.M. (1986) The Eromanga
Basin-an overview of exploration and potential. In:
Contributions to the Geology and Hydrocarbon Potential
of the Eromanga Basin (Eds Gravestock, D.I., Moore,
P.S. & Pitt, G.M.). Geol. Soc. Aust. Spec. Pub!., 12,
25-38.
BARTH, T. & BJORLYKKE, K. (1993) Organic acids from
source rock maturation: generation potentials, transport
mechanisms and relevance for mineral diagenesis. App/.
Geochem. , 8, 325-337.
BATTERSBY, D.G. ( 1976) Cooper Basin oil and gas fields.
In: Economic Geology ofAustralia and New Guinea, 3.
Petroleum (Eds Leslie, R.B., Evans, H.J. & Knight,
C.L.). Aust. Inst. Min. Metall. Monogr., 7, 321-370.
BIRD, M.I. & CHIVAS, A.R. (1988) Oxygen isotope dating of
the Australian regolith. Nature, 331, 5 1 3-516.
BJORLYKKE, K. & EGEBERG, P.K. (I 993) Quartz cementa­
tion in sedimentary basins. Bull. Am. Ass. Petrol. Geo/. ,
77, 1 5 38-1 548.
BoNE, Y. ( 1989) Paleotemperature analysis, Eromanga
Basin, South Australia-fluid inclusion microthermome­
try auto-fluorescence microscopy and cathodo/umines-
 Structural controls on seismic-scale carbonate cementation
cence studies of samples from Tantanna-1. Marana- I
and Strze/ecki-10. Report for Santos Ltd, Adelaide
(unpublished).
BOULT, P., THEOLOGOU, P. & FODEN, J. ( 1 997) Capillary
seals within the Eromanga Basin, Australia­
implications for exploration and production. In: Seals.
Traps and the Petroleum System (Ed. Surdam, R.C.).
Mem. Am. Ass. Petrol. Geol., Tulsa, 67, 1 43- 1 67.
BowERING, A.J.W. ( 1 982) Hydrodynamics and hydrocar­
bon migration-a model for the Eromanga Basin. Aust.
Petrol. Exp/or. Ass. J. , 22, 227-236.
BRADSHAW, M.T., YEATES, A.N., BEYNON, R.M. et a/.
( 1 988) Palaeogeographic evolution of the . North West
Shelf region. In: The North West She/f Australia (Eds
Purcell, P.G. & Purcell, R.R.) pp. 29-54. Petroleum
Exploration Society of Australia, Perth.
BRAY, E. E. & FOSTER, W.R. ( 1 980) A process for primary
migration of petroleum. Bull. Am. Ass. Petrol. Geo/. , 64,
1 07- 1 1 4.
BRIKKE, I. ( 1 982) Geochemical interpretation of some oils
and condensates from the Dampier Sub-basin of West­
em Australia. A ust. Petrol. Exp/or. Ass. J., 22, 1 79- 1 87.
BROOKS, J.K., HESP , W.R. & RIGBY, D . ( 1 9 7 1 ) The natural
conversion of oil to gas in sediments in the Cooper
Basin. Aust. Petrol. Explor. Ass. J. , 1 1 , 1 2 1 -5.
BRYANT, I. D., l<ANTOROWICZ, J.D. & LovE, C.F. ( 1 988) The
ongm and recognitiOn of laterally continuous
carbonate-cemented horizons in the Upper Lias Sands
of southern England. Mar. Petrol. Geo/. , 5, 1 08- 1 3 3.
BUSWELL, A.J.G. ( 1 989) The Angel Formation at Ange/-3,
Dampier Sub-basin, Western Australia. BApp!Sc Hon­
ours thesis, Curtin University of Technology, Depart­
ment of Geology and Geophysics, Perth.
CAPUANO, R.M. ( 1 990) Hydrochemical constraints on
fluid-mineral equilibria during compaction diagenesis
of kerogen-rich geopressured sediments. Geochim. Cos­
mochim. Acta, 54, 1 283-1 299.
CRAIG, H. ( 1 957) Isotope standards for carbon and oxygen
and correction factors for mass spectrometric analysis
of carbon dioxide. Geochim. Cosmochim. Acta, 12,
1 33 - 1 49.
CRAIG, H. ( 1 96 1 ) Standards for reporting concentrations
of deuterium and oxygen- 1 8 in natural waters. Science,
133, 1 833-1 834.
DENHAM, D. & WINDSOR, C.R. ( 1 99 1 ) The crustal stress
pattern on the Australian continent. Explor. Geophys.,
22, 1 0 1 - 1 06.
DE Ros, L.F., MORAD, S. & PAIM , P.S.G. ( 1 994) The role of
detrital composition and climate on the diagenetic
evolution of continental molasses: evidence from the
Cambro-Ordovician Guaritas Sequence, southern Bra­
zil. Sediment. Geo/. , 92, 1 9 7-228.
DIBONA, P. & ScoTT, J. ( 1 990) Heavy mineral study of the
Late Jurassic and Early Cretaceous section from fifteen
wells within the Barrow-Dampier Sub-basins, North West
She/f. Western Australia. Report for Centre for Petro­
leum Exploration, Curtin University (unpublished).
DIMITRAKOPOULOS, R. & MUEHLENBACHS, K. ( 1 987) Bio­
degradation of petroleum as a source of 1 3C-enriched
carbon dioxide in the formation of carbonate cement.
Chern. Geol. , 65, 283-29 1 .
DI ToRO, G.A.E. ( 1 994) The reservoir geology of the
Wanaea and Cossack Oil Fields. In: The Sedimentary
359
Basins of Western A ustralia (Eds Purcell, P.G. & Purcell,
R.R.), pp. 5 5 7-572. Petroleum Exploration Society of
Australia, Perth.
DoNOVAN, T.J. ( 1 974) Petroleum microseepage at Ce­
ment, Oklahoma: evidence and mechanism. Bull. Am.
Ass. Petrol. Geo/. , 58, 429-446.
DONOVAN, T.J., FRIEDMAN, I. & GLEASON, J.D. ( 1 974)
Recognition of petroleum-bearing traps by unusual
carbon isotopic compositions of carbonate-cemented
surface rocks. Geology, 2, 3 5 1 -354.
DuNLOP, E.C., BROWNE, M.V. & TADIAR, E.F. ( 1 992)
Depletion of gas reservoirs by molecular diffusion-a
case study. Petrol. Exp/or. Soc. A ust. J., 32, 369-390.
EADINGTON, P.J. & HAMILTON, P.J. ( 1 990) Fluid migration
and thermal history of Tithonian sandstones, Lower
Swan Formation. Octavius- ] borehole, Timor Sea. Re­
port for Commonwealth Scientific Industrial Research
Organisation, Australia (unpublished).
ELLIS, G.K. ( 1 988) Diagenesis of the Talisman-! reservoir
sequence. In: The North West She/f. Australia (Eds
Purcell, P.G. & Purcell, R.R.), pp. 237-249. Petroleum
Exploration Society of Australia, Perth.
ETHERIDGE, M.A. & O'BRIEN, G. ( 1 994) Structural and
tectonic evolution of the Western Australian margin
basin system. Petrol. Exp/or. Soc. Aust. J. , 34, 45-63.
ETHERIDGE, M., McQUEEN, H. & LAMBECK, K. ( 1 9 9 1 ) The
role of intraplate stress in Tertiary (and Mesozoic)
deformation of the Australian continent and its mar­
gins: a key factor in petroleum trap formation. Exp/or.
Geophys., 22, 1 23-1 28.
FABER, E. & STAHL, W. ( 1 984) Geochemical surface explo­
ration for hydrocarbons in North Sea. Bull. Am. Ass.
Petrol. Geo/. , 68, 363-386.
FISHER, R.S. & KREITLER, C.W. ( 1 987) Geochemistry and
hydrodynamics of deep-basin brines, Palo Duro Basin,
Texas, USA. App/. Geochem. , 2, 459-476.
FREARS, B. ( 1 995) Regional focus-South Australian ba­
sins overview. Petrol. Exp/or. Soc. Aust. News, April!
May, 38-42.
FRIEDMAN, I. & O N EIL, J.R. ( 1 977) Compilation of stable
'
isotope fractionation factors of geochemical interest. In:
Data of Geochemistry (Ed. Fleischer, M.). US Geol.
Surv. Prof. Paper, 440-KK.
GILBY, A.R. & MORTIMORE, I.R. ( 1 989) The prospects for
Eromanga oil accumulations in the northern Cooper
Basin region, Australia. In: The Cooper and Eromanga
Basins, A ustralia (Ed. O'Neil, B.J.), pp. 39 1 -403. Proc.
Petroleum Exploration Soc. Australia, Soc. Petroleum
Engineers, Australian Soc. Exploration Geophysicists
(S.A. Branches), Adelaide.
GLEADOW, A.J.W., DUDDY, I.R., GREEN, P.F. & LOVERING,
J.F. ( 1 988) Assessment ofhydrocarbon resource potential
in Australian sedimentary basins: development offission
track techniques. End-of-grant report for National Edu­
cation, Research Development and Demonstration
Council, Project No. 720, Australia (unpublished).
GouLD, K.W. & SMITH, J.W. ( 1 978) Isotopic evidence for
microbiologic role in genesis of crude oil from Barrow
Island, Western Australia. Bull. Am. Ass. Petrol. Geo/. ,
62 , 455-462.
HABERMEHL, M.A. ( 1 980) The Great Artesian Basin, Aus­
tralia. Bureau Miner. Res. J. Geo/. Geophys., 5, 9-38.
HABERMEHL, M.A. ( 1 986) Regional groundwater move-
360 J Schulz-Rojahn, S. Ryan-Grigor and A. Anderson
ment, hydrochemistry and hydrocarbon migration in
the Eromanga Basin. In: Contributions to the Geology
and Hydrocarbon Potential of the Eromanga Basin (Eds
Gravestock, D.I., Moore, P.S. & Pitt, G.M.). Geol. Soc.
Aust. Spec. Publ., 12, 3 5 3-376.
HANNEMAN, D.L., WIDEMAN, C.J. & HALVORSON, J.W.
( I 994) Calcic paleosols: their use in subsurface strati­
graphy. Bull. Am. Ass. Petrol. Geol. , 78, 1 360- 1 3 7 1 .
HANOR, J.S., LAND, L.S. & MACPHERSON, G.L. ( 1 993)
Carboxylic acid anions in formation waters, San
Joaquin Basin and Lousiana Gulf Coast, USA­
implications for clastic diagenesis. Critical comment.
Appl. Geochem., 8, 305-307.
HEATH, R. (I 989) Exploration in the Cooper Basin. Aust.
Petrol. Explor. Ass. 1., 29, 366-378.
HEATH, R., MCINTYRE, S. & GIBBINS, N. ( 1 989) A Permian
origin for Jurassic reservoired oil in the Eromanga
Basin. In: The Cooper and Eromanga Basins, Australia
(Ed. O'Neil, B.J.), pp. 405-4 1 6. Proc. Petroleum Explo­
ration Soc. Australia, Soc. Petroleum Engineers, Austra­
lian Soc. Exploration Geophysicists (S.A. Branches),
Adelaide.
HOCKING, R.M. ( 1 992) Jurassic deposition in the Southern
and Central North West Shelf, Western Australia. West­
em Australian Geological Survey Record 1 992/7 (un­
published).
HOVLAND, M., TALBOT, M.R., QVALE, H., 0LAUSSEN, S. &
AAsBERG, L. ( I 987) Methane-related carbonate cements
in pockmarks of the North Sea. J. sediment. Petrol. , 57,
8 8 1 -892.
HuDSON, J.D. ( 1 977) Stable isotopes and limestone lithifi­
cation. J. Geol. Soc. Land., 133, 637-660.
HuNT, J.W. ( 1 979) Petroleum Geochemistry and Geology.
W.H. Freeman, San Francisco.
HuNT, J.W., HEATH, R.S. & McKENZIE, P.F. ( 1 989) Ther­
mal maturity and other geological controls on the
distribution and composition of Cooper Basin hydro­
carbons. In: The Cooper and Eromanga Basins, Austra­
lia (Ed. O'Neil, B.J.), pp. 509-523. Proc. Petroleum
Exploration Soc. Australia, Soc. Petroleum Engineers,
Australian Soc. Exploration Geophysicists (S.A.
Branches), Adelaide.
HUTCHEON, I.E. ( 1 989) Application of chemical and iso­
tope analysis of fluids to problems in sandstone diagen­
esis. In: Short Course in Burial Diagenesis (Ed.
Hutcheon, I.E.), pp. 279-3 1 0. Mineralogical Associa­
tion of Canada, Calgary.
HUTCHEON, I., ABERCROMBIE, H.J., PUTNAM, P., GARDNER,
R. & KRousE, H.R. ( I 989) Diagenesis and sedimentol­
ogy of the Clearwater Formation at Tucker Lake. Bull.
Can. Petrol. Geol. , 37, 83-97.
HUTCHEON, I., SHEVALIER, M. & ABERCROMBIE, H.J. (I 993)
pH buffering by metastable mineral-fluid equilibria and
evolution of carbon dioxide fugacity during burial
diagenesis. Geochim. Cosmochim. Acta, 57, 1 0 1 7- 1 027.
IRWIN, H., CURTIS, C. D. & COLEMAN, M. ( 1 977) Isotopic
evidence for source of diagenetic carbonates formed
during burial of organic-rich sediments. Nature, 269,
209-2 1 3.
JENKINS, C. C. ( 1 989) Geochemical correlation of source
rocks and crude oils from the Cooper and Eromanga
Basins. In: The Cooper and Eromanga Basins, Australia
(Ed. O'Neil, B.J.), pp. 525-539. Proc. Petroleum Explo-
ration Soc. Australia, Soc. Petroleum Engineers, Austra­
lian Soc. Exploration Geophysicists (S.A. Branches),
Adelaide.
KANTOROWICZ, J.D., BRYANT, l.D. & DAWANS, J.M. ( 1 987)
Controls on the geometry and distribution of carbonate
cements in Jurassic sandstones: Bridport Sands, south­
em England and Viking Group, Troll Field, Norway. In:
Diagenesis of Sedimentary Sequences (Ed. Marshall,
J.D.). Spec. Publ. Geol. Soc. London, 36, 1 03- 1 1 8.
KANTSLER, A.J., PRUDENCE, T.J.C., COOK, A.C. & ZWIGU­
LIS, M. ( I 983) Hydrocarbon habitat of the Cooper/
Eromanga Basin, Australia. A ust. Petrol. Explor. Ass. 1. ,
23, 373-389.
KOPSEN, E. & McGANN , G. ( 1 985) A review of the
hydrocarbon habitat of the Eastern and Central
Barrow-Dampier Sub-basins, Western Australia. Aust.
Petrol. Explor. Ass. 1., 25, 1 54- 1 76.
KRIEG, G.W. (I 986) Stratigraphy and tectonics of the
Dalhousie Anticline, southwest Eromanga Basin. In:
Contributions to the Geology and Hydrocarbon Potential
of the Eromanga Basin (Eds Gravestock, D.I., Moore,
P.S. & Pitt, G.M.). Geol. Soc. Aust. Spec. Publ., 12,
7 1 -84.
LEACH, D.L., PLUMLEE, G.S., HOFSTRA, A. H. et al. ( I 99 1 )
Origin of late dolomite b y C0 2-saturated deep basin
brines: evidence from the Ozark region, central United
States. Geology, 19, 348-3 5 1 .
LEE, M.K. & BETHKE, C.M. ( I 994) Groundwater flow, late
cementation, and petroleum accumulation in the Per­
mian Lyon sandstone, Denver Basin. Bull. Am. Ass.
Petrol. Geol. , 78, 2 1 7-237.
LONGSTAFFE, F.J. & AVALON, A. ( 1 987) Oxygen-isotope
studies of clastic diagenesis in the Lower Cretaceous
Viking Formation, Alberta: implications for the role of
meteoric water. In: Diagenesis of Sedimentary Se­
quences (Ed. Marshall, J.D.). Spec. Publ. Geol. Soc.
London, 36, 277-296.
LONGSTAFFE, F.J., TILLEY, B.J., AYALON, A. & CONNOLLY,
C. A. ( 1 992) Controls on porewater evolution during
sandstone diagenesis, western Canada Sedimentary
Basin: an oxygen isotope perspective. In: Origin, Di­
agenesis, and Petrophysics of Clay Minerals in Sand­
stones (Eds Houseknecht, D.W. & Pittman, E.D.).
Spec. Publ. Soc. econ. Paleontol. Miner., Tulsa, 47,
1 2-32.
LUNDEGARD, P.D. & LAND, L.S. ( 1 986) Carbon dioxide
and organic acids: their role in porosity enhancement
and cementation, Palaeogene of the Texas Gulf Coast.
In: Roles of Organic Matter during Sediment Diagenesis
(Ed. Gautier, D.L.). Spec. Publ. Soc. econ. Palaeontol.
Miner., Tulsa, 38, 1 29- 1 46.
LUNDEGARD, P.D. & LAND, L.S. ( 1 993) Carboxylic acid
anions in formation waters, San Joaquin Basin and
Lousiana Gulf Coast, USA-implications for clastic
diagenesis (discussion). Appl. Geochem., 8, 297-300.
MACGOWAN, D.B. & SuRDAM, R.C. ( 1 993) Carboxylic acid
anions in formation waters, San Joaquin Basin and
Lousiana Gulf Coast, USA-implications for clastic
diagenesis (reply to discussion by P.D. Lundegard and
L.S. Land). Appl. Geochem. , 8, 30 1 -304.
MACHEL, H.-G. ( 1 987) Saddle dolomite as a by-product of
chemical compaction and thermochemical sulphate re­
duction. Geology, 15, 936-940.
 Structural controls on seismic-scale carbonate cementation
MAGARA, K. ( 1 97 5) Re-evaluation o f montmorillonite
dehydration as a cause of abnormal pressure and hydro­
carbon migration. Bull. Am. Ass. Petrol. Geol. , 59,
292-302.
MAGARA, K. ( 1 976) Water expulsion from clastic sedi­
ments during compaction, direction and volumes. Bull.
Am. Ass. Petrol. Geol. , 60, 543-55 3 .
MAGARA, K . ( 1 98 1 ) Fluid dynamics for cap-rock forma­
tion in the Gulf Coast. Bull. Am. Ass. Petrol. Geol. , 65 ,
1 334- 1 343.
McCREA, J .M. ( 19 50) On the isotopic chemistry of carbon­
ates and a palaeotemperature scale. J. Chern. Phys., 18,
849-857.
MciNTYRE, S.M., JAMAL, F.G., PIDOCK, C.L. & KABIR, M.I.
( 1 989) The Gidgealpa oil and gas field: a case history. In:
The Cooper and Eromanga Basins, Australia (Ed. O'Neil,
B.J.), pp. 9 1 - 1 0 1 . Proc. Petroleum Exploration Soc.
Australia, Soc. Petroleum Engineers, Australian Soc. Ex­
ploration Geophysicists (S.A. Branches), Adelaide.
McKIRDY, D.M. ( 1 982) Petroleum geochemistry and
source-rock potential ofthe Arrowie, Pedirka, Cooper and
Eromanga Basins, Central Australia. Report for Delhi
Petroleum Pty Ltd (unpublished).
McNEIL, B., SHAW, H.F. & RANKJN, A.H. ( 1 995) Diagene­
sis of the Rotliegend Sandstones in the V-Fields, south­
ern North Sea: a fluid inclusion study. In: The
Geochemistry ofReservoirs (Eds Cub itt, J.M. & England,
W.A.). Spec. Pub!. Geol. Soc. London, 86, 1 25 - 1 39.
MILLER, L.R. ( 1 995) Sedimentary architecture of rift­
related depositional systems within the Dampier Prov­
ince, North West Shelf, A ustralia. PhD thesis, National
Centre for Petroleum Geology and Geophysics, Univer­
sity of Adelaide.
MORAD, S. & DE Ros, L.F. ( 1 994) Geochemistry and
diagenesis of stratabound calcite cement layers within
the Rannoch Formation of the Brent Group, Murchison
Field, Northern Viking Graben (North Sea)-comment.
Sediment. Geol. , 93, 1 35- 1 4 1 .
MORAD, S., MARFIL, R., AL-AASM, l.S. & GOMEZ-GRAS, D.
( 1 992) The role of mixing-zone dolomitization in sand­
stone cementation: evidence from the Triassic Bunt­
sandstein, the Iberian Range, Spain. Sediment. Geol. ,
80, 53-65.
MORAD, S., BEN ISMAIL, H.N., DE Ros, L.F., AL-AASM, l.S.
& SERRHINI, N.E. ( 1 994) Diagenesis and formation
water chemistry of Triassic reservoir sandstones from
southern Tunisia. Sedimentology, 41, 1 2 5 3- 1 272.
NORTHROP, D.A. & CLAYTON, R.N. ( 1 966) Oxygen-isotope
fractionation in systems containing dolomite. J. Geol. ,
74, 1 74- 1 96.
NuGENT, 0. ( 1 969) Sedimentation and petroleum poten­
tial of the Jurassic sequence in the southwestern Great
Artesian Basin. A ust. Petrol. Explor. Ass. J., 9, 97-1 06.
O'BRIEN, G.W. ( 1 996) Late Tertiary fluid migration in the
Timor Sea: a key control on both thermal history and
seismic velocity. Petrol. Explor. Soc. Aust. News,
February/March, 1 3.
O'BRIEN, G.W. & Wooos, E.P. ( 1 995) Hydrocarbon­
related diagenetic zones (HPDZs) in the Vulcan Sub­
basin, Timor Sea: recognition and exploration
implications. A ust. Petrol. Explor. Ass. J. , 35, 220-25 3 .
OEHLER, D.Z. & STERNBERG, B.K. ( 1 984) Seepage-induced
anomalies, 'false' anomalies, and implications for elec-
361
trical prospecting. Bull. Am. Ass. Petrol. Geol. , 68,
1 1 2 1 - 1 1 45.
OLIVER, J. ( 1 986) Fluids expelled tectonically from oro­
genic belts: their role in hydrocarbon migration and
other geologic phenomena. Geology, 14, 99- 1 02.
PARRY, J.C. & SMITH, D.N. ( 1 988) The Barrow and
Exmouth Sub-basins. In: The North West Shelf, Austra"
lia (Eds Purcell, P.G. & Purcell, R.R.), pp. 1 29- 1 45 .
Petroleum Exploration Society o f Australia, Perth.
PHILP, R.P., GILBERT, T.D. & FRIEDRICH, J. ( 1 98 1 ) Bicyclic
sesquiterpenoids and diterpenoids in Australian crude
oils. Geochim. Cosmochim. Acta, 45, 1 1 73- 1 1 80.
PHILP, R.P., GILBERT, T.D. & FRIEDRICH, J. ( 1 982)
Geochemical correlation of Australian crude oils. A ust.
Petrol. Explor. Ass. J. , 22, 1 88- 1 99.
PLATT, N.H. ( 1 992) Fresh-water carbonates from the
Lower Freshwater Molasse (Oligocene, western Switzer­
land): sedimentology and stable isotopes. Sediment.
Geol. , 78, 6 6 1 -669.
PLAYFORD, P.E. ( 1 97 5) Petroleum in Western Australia: a
review of exploration, production, and future prospects.
Aust. Petrol. Explor. Ass. J. , 15, 72-79.
Powis, G.D. ( 1 989) Revision of Triassic stratigraphy at the
Cooper Basin to Eromanga Basin transition. In: The
Cooper and Eromanga Basins, Australia (Ed. O'Neil,
B.J.), pp. 265-277. Proc. Petroleum Exploration Soc.
Australia, Soc. Petroleum Engineers, Australian Soc.
Exploration Geophysicists (S.A. Branches), Adelaide.
PROSSER, D.J., DAWS, J.A., FALLICK, A.E. & WILLIAMS,
B.P.J. ( 1 993) Geochemistry and diagenesis of stra­
tabound calcite cement layers within the Rannoch
Formation of the Brent Group, Murchison Field, North
Viking Graben (northern North Sea). Sediment. Geol. ,
87 , 1 39- 1 64.
RIGBY, D. & SMITH, J.W. ( 1 98 1 ) An isotopic study of gases
and hydrocarbons in the Cooper Basin. Aust. Petrol.
Explor. Ass. J., 2 1 , 222-229.
ROBERTSON RESEARCH ( 1 988) Oi/ and gas.fields ofAustrala­
sia (A ustralia, New Zealand and Papua New Guinea).
Robertson Research (Australia) Pty Ltd.
ROSENBAUM, J. & SHE PPARD , S.M.F. ( 1 986) An isotopic
study of siderites, dolomites and ankerites at high
temperatures. Geochim. Cosmochim. Acta, 50 , 1 1 47-
1 1 50.
RYAN-GRIGOR, S. & SCHULZ-ROJAHN, J.P. ( 1 995) Seismic
delineation of structure-controlled carbonate cement
and potential economic implications, Angel Field, North
West Shelf. Aust. Petrol. Explor. Ass. 'J. , 35, 280-295.
SAIGAL, G.C. & BJ0RLYKKE, K. ( 1 987) Carbonate cements
in clastic reservoir rocks from offshore Norway­
relationships between isotopic composition, textural
development and burial depth. In: Diagenesis of Sedi­
mentary Sequences (Ed. Marshall, J.D.). Spec. Pub!.
Geol. Soc. London, 36 , 3 1 3-324.
ScHULZ-ROJAHN, J.P. ( 1 993) Calcite-cemented zones in the
Eromanga Basin: clues to petroleum migration and
entrapment? Aust. Petrol. Explor. Ass. J., 33, 63-76.
SENIOR, B.R., MONO, A. & HARRISON, P.L. ( 1 978) Geology
of the Eromanga Basin. Bureau Miner. Res. Geol.
Geophys. Bull. , 167 , 1 02 pp.
SHAW, D. ( 1 99 1 ) Tertiary structuring in southwest
Queensland: implications for petroleum exploration.
Explor. Geophys. , 22, 339-344.
362 J. Schulz-Rojahn, S. Ryan-Grigor and A. Anderson
SISSON, R.H. ( 1 987) Earthquake rupturing as a mineraliz­
ing agent in hydrothermal systems. Geology, 15, 70 1 -
704.
SmsoN, R.H., MOORE, J.M. & RANKIN, A.H. ( 1 975) Seismic
pumping-a hydrothermal fluid transport mechanism.
J. Geol. Soc. Lond., 1 3 1 , 653-659.
SINGH, R. ( 1 990) The seismic response of calcite-cemented
zones-Gidgealpa. MSc thesis, University of South An­
stralia.
SMITH, J.T. & EHRENBERG, S.N. ( 1 989) Correlation of
carbon dioxide abundance with temperature in clastic
hydrocarbon reservoirs: relationship to inorganic chem­
ical equilibrium. J. Mar. Petrol. Geol. , 6, 1 29- 1 36.
SMITH, J.W. ( 1 978) Carbonates: a guide to hydrocarbons.
J. Geochem. Explor. , 10 , 1 03 - 1 07.
SouzA, R.S., DE Ros, L.F. & MORAD, S. ( 1 995) Dolomite
diagenesis and porosity preservation in lithic reservoirs:
Carm6polis Member, Sergipe-Alagoas Basin, North­
eastern Brazil. Bull. Am. Ass. Petrol. Geol. , 79, 725-748.
STAUGHTON, D.B. ( 1 985) The diagenetic history and reser­
voir quality evaluation of the Strzelecki hydrocarbon
field, Cooper/Eromanga Basins, South Australia. BSc
(Hons) thesis, Department of Earth Sciences, Monash
University.
SnENSTRA, N. ( 1 992) Hydrocarbon-associated carbonate
accumulations within the Middle to Upper Jurassic
reservoir sandstones of the Cooper-Eromanga Basins,
Central Australia. BSc (Hons) thesis, University of
Melbourne, School of Earth Sciences.
STUART, W.J. ( 1 976) The genesis of Permian and Lower
Triassic reservoir sandstones during phases of southern
Cooper Basin development. A ust. Petrol. Explor. Ass. J. ,
16, 37-47.
SwiFT, M.G., STAGG, H.M.J. & FALVEY, D.A. ( 1 988) Heat
flow regime and implications for oil maturation and
migration in the offshore northern Carnarvon Basin. In:
The North West Shelf, A ustralia (Eds Purcell, P.G. &
Purcell, R.R.), pp. 5 3 9-5 5 1 . Petroleum Exploration
Society of Australia, Perth.
TAYLOR, T.R. ( 1 990) The influence of calcite dissolution
on reservoir porosity in Miocene sandstones, Picaroon
Field, offshore Texas Gulf Coast. J. sediment. Petrol. ,
60, 322-334.
THORNTON, R.C.N. ( 1 979) Regional stratigraphic analysis
of the Gidgealpa Group, southern Cooper Basin, Aus­
tralia. Geol. Surv. Aust. Bull. , 49, 1 - 1 40.
TOWNSHEND, G. ( 1 993) Distribution, chemistry, isotopic
composition, and origin ofdiagenetic carbonates: Namur
Sandstone, Cooper-Eromanga Basin, South Australia.
BSc (Hons) thesis, School of Earth Sciences, University
of Melbourne.
TUPPER, N.P. & BURCKHARDT, D.M. ( 1 990) Use of the
methylphenanthrene index to characterise expulsion of
Cooper and Eromanga Basin oils. Aust. Petrol. Explor.
Ass. J. , 30, 373-385.
VEEVERS, J.J. ( 1 984) Phanerozoic Earth History ofAustra-
lia. Oxford Geological Science Series, Oxford Univer­
sity Press, New York, 4 1 8 pp.
VEEVERS, J.J. ( 1 988) Morphotectonics of Australia's north­
western margin: a review. In: The North West Shelf,
Australia (Eds Purcell, P.G. & Purcell, R.R.), pp. 1 9-27.
Petroleum Exploration Society of Australia, Perth.
VINCENT, P. & TILBURY, L. ( 1 988) Gas and oil fields of the
Rankin Trend and northern Barrow-Dampier Sub­
basin. In: The North West Shelf, Australia (Eds Purcell,
P.G. & Purcell, R.R.), pp. 34 1 -369. Petroleum Explora­
tion Society of Australia, Perth.
VINCENT, P.W., MORTIMORE, I.R. & MCKIRDY, D.M.
( 1 985) Hydrocarbon generation, migration and entrap­
ment in the Jackson-Naccowlah area, ATP 259P, south­
western Queensland. Aust. Petrol. Explor. Ass. J. , 25,
62-84.
VOLKMAN, J.K., ALEXANDER, R., KAGI, R.I., NOBLE, R.A. &
WooDHOUSE, G.W. ( 1 983) A geochemical reconstruc­
tion of oil generation in the Barrow Sub-basin of
western Australia. Geochim. Cosmochim. Acta, 47,
209 1 -2 1 05.
WALDERHAUG, 0 . , BJ0RKUM, P.A. & BOLAS, H.G.M. ( 1 989)
Correlation of calcite-cemented layers in shallow­
marine sandstones of the Fensfjord Formation in the
Brage Field. In: Correlation in Hydrocarbon Explora­
tion, (Ed. Collinson, J.D.), pp. 367-3 7 5 . Graham &
Trotman, London.
WALL, V.J. ( 1 987) Hydrocarbon reservoir quality in the
Cooper!Eromanga Basins. Final report, NERDDC
Project No. 808 (ref. 84/4067). Department of Earth
Sciences, Monash University (unpublished).
WILLIAMS, N.Y., THEOLOGOU, P.N., BOULT, P.J. et a/.
( 1 994) Unusual lacustrine reservoirs and seals of the
Murteree Horst area, Eromanga Basin, Australia. Soc.
Petrol. Eng. , SPE 287 5 1 , 99- 1 1 2.
WooD, J.R. & BoLES , J.R. ( 1 99 1 ) Evidence for episodic
cementation and diagenetic recording of seismic pump­
ing events, North Coles Levee, California, USA. Appl.
Geochem., 6, 509-52 1 .
WooD, J.R. & HEWLETT, T.A. ( 1 982) Fluid convection and
mass transfer in porous sandstones-a theoretical
model. Geochim. Cosmochim. Acta, 46, 1 707- 1 7 1 3.
WooDs, E.P. ( 1 9 9 1 ) Eocene seismic velocity anomalies
from the Timor Sea. Explor. Geophys., 22, 46 1 -464.
WOODSIDE OFFSHORE PETROLEUM ( 1 97 1 ) Angel- l well com­
pletion report (unpublished).
WOODSIDE OFFSHORE PETROLEUM ( 1 972) ' Ange/-2 well com-
pletion report (unpublished).
WOODSIDE OFFSHORE PETROLEUM ( 1 988) A review of the
petroleum geology and hydrocarbon prospectivity of the
Barrow-Dampier Sub-basin and environs. In: The
North West Shelf, Australia (Eds Purcell, P.G. & Purcell,
R.R.), pp. 1 1 5- 1 28. Petroleum Exploration Society of
Australia, Perth.
WOODSHIDE OFFSHORE PETROLEUM ( 1 990) Ange{-4 well
completion report (unpublished).
Spec. Pubis int. Ass. Sediment. ( 1 998) 26, 363-393

Carbonate cementation-the key to reservoir properties

of four sandstone levels (Cretaceous) in the Hibernia Oilfield,
Jeanne d'Arc Basin, Newfoundland, Canada

R. H E S S E and I . A . A B I D
Department of Earth and Planetary Sciences, McGill University,
3450 University Street, Montreal, Quebec HJA 2A 7, Canada,
e-mail rein_h@geosci.lan. mcgill. ca

ABSTRACT

The effects of carbonate cementation on the diagenetic evolution of four sandstone reservoir levels in
the Hibernia Oilfield of the Jeanne d'Arc Basin, which occur between 2000 and 5000 m subsurface
depth, were investigated in a petrographic thin-section and SEM study corroborated by carbon and
oxygen isotopic analyses. After precipitation of minor chlorite coatings, siderite, quartz overgrowths and
pyrite, early ferroan calcite was the most important cement in the Hibernia Field. It formed after loss of
about I 0- 1 5% primary porosity by mechanical compaction. Quartz overgrowths continued in parts of
the reservoirs, mainly thick sandstone beds not reached by the calcite-precipitating fluids. A major
dissolution event predominantly affected calcite but also silicates (feldspar, chert grains, mud clasts,
heavy minerals). Subsequent recementation by late ferroan calcite, ferroan saddle dolomite, quartz
overgrowths, kaolinite and pyrite further reduced porosity before the emplacement of hydrocarbons.
Secondary porosity development as the main contributor to the present reservoir porosity in
Hibernia Field is closely related to the former presence of early calcite cements. The fraction of total
porosity which is secondary increases with depth, from 20% in Avalon/Ben Nevis Sandstone
(Hauterivian-Albian), to 60% in Catalina Sandstone (Lower Hauterivian), and to >80% in Hibernia
Formation (Berriasian to Mid-Valanginian). In the Avalon/Ben Nevis Sandstone the formation of
secondary porosity may have been caused by meteoric water influx. In the deeper reservoirs it was
caused by acidic pore fluids generated by organic-matter maturation. The present average geothermal
gradient of 26 "C/km suggests that the most deeply buried sandstone reservoirs in Hibernia (Tithonian
Jeanne d'Arc Formation) did not experience temperatures in excess of l30"C.

INTRODUCTION

The timing and extent of carbonate cementation in
sandstones is a major factor controlling porosity
evolution in oilfield reservoirs. Early diagenetic
carbonate cementation can protect substantial pro­
portions of the porosity from irreversible destruc­
tion during burial compaction by conserving it in
the form of pore-filling cement, until it is recovered
by carbonate dissolution at greater depth. This
secondary porosity generated from locked-in pri­
mary porosity tends to resist destruction longer
than does primary porosity (Schmidt & McDonald,
1 979a). The transfer of primary to secondary poros­
ity via intermediary carbonate cements thus has a
pore-stabilizing effect, rendering carbonate cemen­
tation, particularly early cementation, in potential
reservoir sandstones an economically important
diagenetic process.
The Hibernia Oilfield of the Jeanne d'Arc basin,
located on the Grand Banks, offshore Newfound­
land, was the first major oil discovery in 1 979 after
more than a decade of offshore exploration in the
Canadian east coast region and, together with the
adjacent Terra Nova Field, is the largest oil accu­
mulation on the eastern margin of North America
(Taylor et a!., 1 992). Since its discovery and first
description (Arthur et a!., 1 982), two dozen papers

Carbonate Cementation in Sandstones Edited by Sadoon Morad 363

© 1998 The International Association of Sedimentologists
ISBN: 978-0-632-04777-2
364 R. Hesse and I.A. Abid

have been published on all major aspects of the
geology of the Grand Banks, but very little on
diagenesis. Apart from a geochemical study of
carbonate cements in Avalon Sandstone (Hutcheon
et al., 1 98 5 ) and a sedimentological and petro­
graphic study of the Hibernia Sandstone (Brown et
al., 1 989), no comprehensive study of burial diage­
netic trends in the major reservoir sandstones of the
Jeanne d'Arc basin has been undertaken prior to
Abid ( 1 988). In this paper we present carbonate­
cementation and porosity-evolution trends through
four sandstone reservoirs of Hibernia Oilfield, as­
sessing the relative timing of events on the basis of
thin-section parageneses and a limited set of oxygen
and carbon isotope analyses.
GEOLOGICAL SETTING
Reviews of the geological evolution of the continen-
tal margin of Newfoundland and its hydrocarbon­
bearing basins have been published by Amoco
Canada Petroleum Company Ltd and Imperial Oil
Ltd ( 1 97 3), Sherwin ( 1 9 7 3), Jansa & Wade ( 1 975),
Arthur et al. ( 1 982), Grant et al. (I 986), Enachescu
( 1 987, 1 98 8 , 1 992), Tankard & Welsink ( 1 987,
1 98 8 , 1 989), Sinclair ( 1 988, 1 993), Grant &
McAlpine ( 1 990), and Sinclair et al. ( 1 992). The
Jeanne d'Arc basin of the central Grand Banks
(Fig. I) is a rift basin of the western North Atlantic
that formed in response to two major periods of
rifting and related thermal and isostatic subsidence
following the opening of the Atlantic Ocean. A first,
relatively short rifting episode along the present east
coast of North America during the Late Triassic
lasted 25 Myr and extended northward up to the
Jeanne d'Arc basin. It did not lead to the generation
of oceanic crust in the Grand Banks region north of
the Newfoundland fracture zone (Klitgord &
Schouten, 1 986; Tankard & Welsink, 1 9 87), but

52"W

,...

'
,/

.•.

_,..- Normal Fault

___.- Transfer Fault
tst Order Transfer
Fault Trend

__u...- Basin Edge

� San

C Central Gr. Banks

S Southern Gr. Banks

Fig. 1 . Sedimentary basins of the

Grand Banks after Tankard &
Welsink ( 1 987). For explanation
see text.
 Sandstone reservoir levels in the Hibernia Oilfield 365

mi
0

10

20

40 0 100

mil CENOZOIC fld MESOZOIC • CONTINENTAL CRUST D MANTLE

Fig. 2. Simple shear model as an explanation for the asymmetry in basin evolution on opposite conjugate margins of
Eastern Canada and Iberia, with wider and deeper basins on the Grand Banks and narrower and shallower basins on
the Galicia Bank (Tankard & Welsink, 1 98 8).

renewed extension during 50 Myr from the Callov­
ian to the Aptian led to separation of the Grand
Banks from its conjugate Galicia Bank margin
(Fig. 2) in the Iberian Peninsula (Masson & Miles,
1 984, 1 986; Mauffret & Montadert, 1 9 87). The
region is divided by major transform faults into
separate crustal blocks with varying amounts of
extension. The Newfoundland and Charlie Gibbs
fracture zones respectively separate the Grand
Banks from the Scotian Shelf in the south and the
Labrador Shelf in the north. The Grand Banks are
further subdivided into southern, central and north-
ern segments by relatively small transfer faults
(Fig. 1 ). In the central Grand Banks the amount of
extension was twice that in the southern segment.
As a result, the basins in the central Grand Banks,
such as the Jeanne d'Arc basin, are considerably
deeper ( - 1 6 km). Extension in the region is ex­
plained by a 'simple shear' model involving a
low-angle detachment fault, which gently dips to the
west and accounts for the asymmetry in basin
evolution on the opposite conjugate margins of
Canada and Iberia (Fig. 2). Below the Jeanne d'Arc
basin, the detachment is inferred to reach 26 km

A A'
�������4�
Central Ridge Complex
km
0
JEANNE D'ARC BASIN OILFIELDS

10

20

0 30 km
L..--....1
46°
0 km 10
30 V.E. = 1:1 49° 48
°
o mi 6

Fig. 3. NE-SW cross-section of the Jeanne d'Arc basin, showing the lystric Murre Fault which bounds the basin in the
west. ( ! ) Stable shelf (Bonavista Platform); (2) rollover anticline resulting from antithetic movement of the
hanging-wall block along the Murre Fault; (3) central Jeanne d'Arc basin; (4) Central Ridge. TR, interpreted Triassic
continental red beds at the base of the Mesozoic sedimentary succession overlain by Lower Jurassic evaporites
involved in diapirism (squared pattern). Insert map shows hydrocarbon fields of the Jeanne d'Arc basin mentioned in
text and location of section. From Tankard & Welsink ( 1 988). A, T, movement of fault block away from viewer and
towards viewer, respectively.
366 R. Hesse and !.A. Abid
48" 45'
a +46"50' b WEST
5000 m
5km
5 mi
Fig. 4. (a) Seismic travel-time structure map showing surface of Lower Cretaceous limestone marker in the Hibernia
Anticline and location of wells used in this study. From Arthur et a!. ( 1 982). (b) Schematic cross-section along line
A-A' (corresponding to zone 2 in Fig. 3), showing three main reservoir sandstones of Hibernia Field: Avalon, 'B'
(� Catalina) and Hibernia. From Benteau & Sheppard ( 1 982).
depth (Fig. 3), where the northeast trending listric
Murre Fault soles out into it. The Murre Fault
strikes parallel to the axis of the basin and bounds it
in the west. Entrapment of hydrocarbons in the
Hibernia Field occurred in a rollover anticline that
resulted from antithetic movement of the hanging­
wall block along the Murre Fault (Fig. 4) (En­
achescu, 1 987).
DEPOSITIONAL ENVIRONMENTS
AND STRATIGRAPHY
The post-Palaeozoic sedimentary sequence of the
Jeanne d'Arc basin starts with Upper Triassic con­
tinental red beds of the Eurydice Formation, uncon­
formably overlying metasedimentary rocks of
Middle to Late Devonian age (Fig. 5). The thin
evaporites and lagoonal and tidal-flat carbonates of
the Iroquois Formation (Lower Jurassic), which
include oolitic limestones, are followed by neritic
and subtidal fine-grained clastics of the Downing
Formation, with some intercalated carbonates
(Jansa & Wade, 1 975). The Lower to Middle Juras­
sic sequence records the transition to normal open
marine conditions following continued subsidence
of the region and westward transgressions.
The Upper Jurassic (Bathonian to Kimmeridg-
P-15 K-18 B-08 EAST
0
- ­
2
KILOMETRES
ian) Voyageur and Rankin Formations contain both
the source rocks and the oldest reservoir rocks of
the Hibernia Field. Organic matter rich shales (up
to 8% total organic carbon, TOC) of the Oxfordian­
Kimmeridgian Egret Member of the Rankin
Formation are the main source rocks for the hydro­
carbon accumulations in the basin (Swift &
Williams, 1 980; Powell, 1 9 85; Creaney & Allison,
1 987; Von der Dick et al., 1 989; Fowler & Brooks,
1 990). The reservoir rocks of the Tithonian Jeanne
d'Arc Formation comprise braided-river conglom­
erate and sandstone. Renewed Late Jurassic rifting
generated the rugged relief at the western basin
margin which produced the flood of synrift clastic
sediments (Meneley, 1 986). .
In the Lower Cretaceous (Berriasian-Albian) Hi­
bernia to Avalon/Ben Nevis formations, which con­
tain the main reservoir levels of the Hibernia Field
(the 200 m thick Hibernia Sandstone and the !50 m
thick Catalina Member of the Whiterose Formation,
as well as the up to 800 m thick Avalon/Ben Nevis
Sandstone), the shift to clastic-dominated marginal­
marine and marine sedimentation occurred, as rift­
ing continued in the Grand Banks region. Most of the
detrital material was derived from elevated areas to
the west (Bonavista Platform; Figs I and 3) and
southwest (Avalon Uplift; Jansa & Wade, 1 975) of
the Jeanne d'Arc basin, but also from uplifted and
 Sandstone reservoir levels in the Hibernia Oilfield 367

AGE FORMATIONS TECTONIC HISTORY

POST
(.) RIFT
6
N
0 SUBSIDENCE
z
UJ
(.)

Separation of
European
Greenland
Plate

Separation
Europe

Separation of
Iberia
RIFT
(.) PHASE
6 2
N
0
(/) Separation of
UJ
;::.'!; Africa

EPEIRIC
BASIN

RIFT
PHASE
1

Fig. 5. Generalized stratigraphy and tectonic history of Jeanne d'Arc basin. From Grant et a/. ( 1 986) and Tankard &
Welsink ( 1 98 7).

tilted blocks within the basin (i.e. Central Ridge;
Fig. 3). Sediment transport was dominantly north­
ward into deltaic-marine depositional environments
(Arthur et a!., 1 982; Brown et a!., 1 989). Continued
uplift in Aptian time is interpreted as isostatic re­
bound, which resulted from partial unloading of the
lithosphere afler doming and unroofing in the exten­
sional phases (Tankard & Welsink, 1 98 7).
The post-rift sedimentary history is recorded in
the transgressive Upper Cretaceous Dawson Can­
yon and Tertiary Banquereau Formations above the
mid-Cretaceous break-up unconformity. Shallow­
ing occurred in Oligocene time, and may have led to
subaerial exposure in the Miocene (Grant et a!.,
1 986).
368 R. Hesse and I.A. Abid
MATERIALS AND METHODS
About 450 m of logged drill core from the B-08
B-27, 0-3 5 , K- 1 8 and C-96 wells of the Hibernia
Field were used in this study, their simplified
lithology being shown in Fig. 6. Closely spaced
samples were taken in carbonate-rich sandstone
zones alternating with porous zones. Thin sections
impregnated with blue epoxy resin (70 samples)
were half stained with a mixture of alizarin red-S
and K-ferricyanide for identification of Fe-bearing
carbonates, and half with alizarin-S only for distinc­
tion of calcite and dolomite. A large number of
stained thin sections were made available on loan
from the Mobil Oil Company, Toronto. Out of 1 6 8
thin sections studied under the petrographic and
scanning electron microscopes, 36 were used for
quantitative estimates of mineralogical composi­
tion and porosity by point counting (500 points per
section). Carbon and oxygen isotopes were mea­
sured on sandstone microsamples at the University
of Michigan.
SANDSTONE PETROGRAPHY:
FRAMEWORK GRAINS
Based on Folk's ( 1 974) classification scheme, the
average sandstone of the Hibernia Field is a sublith­
arenite (Q 7.9 F .6 Rs.s). Individual samples are
8 3
either quartz-arenitic or sublitharenitic, and some
are subarkoses (Fig. 7; Table 1 ). If shale clasts,
which are probably predominantly rip-up clasts of
intrabasinal origin and do not carry provenance
information with respect to terrigenous source
areas, are excluded, then the majority of the sam­
ples would be classified as quartz arenites.
Detrital quartz averages 88% (range 78-99%) of
the framework grains and is predominantly mono­
crystalline. The small percentage (7%) of poly­
crystalline grains and monocrystalline grains with
undulous extinction (requiring > 5 stage rotation
• carbonate cement ( 1 3% on average) and the cur­
for complete extinction; 6% of all detrital quartz)
indicates a predominant plutonic source for the
detrital quartz. Some polycrystalline grains with
sutured grain boundaries appear to be metamor­
phic. Grains are rounded to subangular; roundness
generally improves with stratigraphical depth.
Feldspar averages 3 . 5% (range 0-8%) and is
predominantly plagioclase. Microcline is rare. Most
grains show signs of dissolution and diagenetic
alteration (replacement by carbonate).
Rock fragments ( 8 . 5% on average, range 0.5-
1 8%), predominantly of sedimentary origin (chert,
shale, sandstone, siltstone, limestone), represent the
second most abundant framework constituent.
Identifiable metamorphic (Fig. 8A), volcanic and
plutonic rock fragments are negligible. With respect
to their diagenetic behaviour, the rock fragments
were grouped into ductile, silicate, carbonate and
fossil fragments. Of these, the ductile fragments
(mud and shale clasts, average 1 . 3o/oo) are diagenet­
ically particularly important, because with increas­
ing burial depth they become squeezed between the
framework grains, thereby reducing porosity. The
silicate fragments comprise chert grains ( 1 . 5% on
average, range 0-5%), siliceous shale, sandstone,
siltstone, schist and volcanic and other igneous rock
fragments. Carbonate fragments (average 1 . 5%,
range 0-8%) display a great variety of lithologies,
including oolites. Fossil fragments (pelecypods, gas­
tropods, serpulid worm tubes, ostracods, Fora­
minifera) constitute 1 .2% on average, but may
contribute up to 30% of the total framework in
fossil-rich zones (Fig. 8C). Heavy minerals (average
1 .2%, range trace amounts to 7%) include, in order
of abundance, pyrite, tourmaline, zircon, epidote,
hornblende, pyroxene, fluorapatite and garnet.
SANDSTONE DIAGENESIS
Mechanical and chemical compaction
Before precipitation of the first early diagenetic
cements, 7-29% (average 1 7%) of the primary
porosity was lost by mechanical compaction. These
figures are calculated by subtracting the 'minus­
cement porosity', i.e. the porosity encountered by a
given cement at the time of precipitation, from the
initial porosity assumed to be, 40% (e.g. House­
knecht, 1 987). The minus-cement porosity (average
2 3%, range 1 1 -33%) is taken as the sum of the
rently preserved primary porosity ( 1 Oo/o on average),
as determined in thin sections (Fig. 9), assuming the
latter was not further reduced by postcementation
compaction. Mechanical compaction involves grain
rotation and rearrangement, grain fracturing and
deformation of ductile grains. Benthic organisms
initially play an important role in the process,
causing bioturbational mixing of sand and mud and
producing open holes and tubes, thereby either
decreasing or enhancing permeability. Both the
 Sandstone reservoir levels in the Hibernia Oilfield 369

Depth 8_08 B-27 K-18 Age

m
0-35 C-96

2000
Albian­
Hauterivian

3000

Tithonian

5000

D SANDSTONE c=J SILTSTONE E:i::9 LIMESTONE � WELL-CORE

�---� SHALE
SANDSTONE
AND SHALE
UNCONFORMITY

Fig. 6. Simplified lithology of the wells used in this study (available core intervals shown in black). Modified from
unpublished data of the Canadian Oil and Gas Land Administration (COGLA). T.D., total depth in metres.

Avalon/Ben Nevis and Hibernia Sandstones show
examples of extensive bioturbation (Fig. 8B).
Levels of mechanical compaction are least in the
Avalon/Ben Nevis Sandstone, as indicated by rela­
tively loose packing. Pressure solution contacts are
absent except between quartz grains and fossil or
limestone fragments (see Fig. 1 70). The presence of
a high percentage of sedimentary rock fragments
(average 1 0%), which include ductile shale clasts,
did not have a major porosity-reducing effect be­
cause of the relatively shallow burial levels. With
2 5-30% of calcite cement in the tight-sandstone
zones, a minimum of 1 0- 1 5% porosity was lost by
mechanical compaction prior to calcite precipita­
tion (assuming 40% original porosity), indicating
relatively early calcite cementation in the Avalon/
Ben Nevis Sandstone.
Significant porosity changes due to mechanical

Q
(I Avalon/Ben Nevis Sandstone
• Catalina Sandstone
6 Hibernia Sandstone
o Jeanne d'Arc Formation

Fig. 7. QFR triangular diagram of

sandstone petrography of Folk ( 1 974).
Q, quartz; F, feldspar; R, rock
fragments. I, quartz arenite; 2,
subarkose; 3, sublitharenite; 4,5,
arkose; 6, 7, litharenite.
370 R. Hesse and I.A. Abid

Table 1. Detrital and authigenic mineralogy and porosity of reservoir sandstones from the Hibernia Field

Depth (m) Q F ss SH CH LS Fos METI Vole M HM Qog Cal Dol Clay Py Sid $

Avalon Sandstone
2 1 84.6 55.3 3.7 2.4 5.7 2.5 1 .2 2.6 5.3 0.6 4.2 4.0 tr tr 1 2. 5
2 1 85.8 43.9 3.4 5.5 2.5 1 .2 1 .6 23.4 2.9 0.4 2.4 3.6 tr 9.2
2 1 90.3 59.8 4.3 0.3 3.1 1.5 0.2 1 1 .5 5.1 3.5 4.5 6.2
2 1 90.8 59. 1 5.1 1 .6 1 .6 0.4 1 0. 8 6.7 2.0 3.5 tr 9.2
2 1 97 . 5 6 1 .4 2.2 1 .0 3.1 0.4 0.8 8.0 1 .0 2.7 1 .4 tr 1 8.0
2 1 97.9 59. 1 3.6 1 .4 1 .5 1 .0 0.6 1 .0 0.4 0.6 30.2 0.6 0.0
2 1 97.9* 70.5 3.0 1.1 3.2 0.8 0.4 1 .0 2.2 5.4 0.6 1.6 1 0.2
257 1 . 5 6 1 .0 2.4 2.2 5.0 tr tr tr 1 .0 28.4 tr 0.0
2578.8 62.3 1.8 1 .4 0.4 tr 8.2 0.4 0.2 0.7 0.2 2.2 0.4 0.6 2 1 .2
2659.6 60.7 2.4 1.1 1 .4 1 .2 0.5 tr 0.7 7.8 0.6 9.8 1 .7 0.7 tr 1 1 .4
266 1 . 1 54.7 4.4 1 .2 1 .7 2. 1 2. 1 0.3 0.6 tr 2.1 0.5 0.4 28.9 1 .0 0.0
Average 58.9 3.3 1 .7 2.7 1 .0 1.3 3.2 0.2 tr 3.9 1 .6 2.9 8.4 0.0 2.0 tr tr 8.9

Catalina Sandstone
3 1 74.9 56.0 3.5 0.5 1 .4 0.5 0.5 5.2 0.8 3 1 .6 0.0
3 1 78.0 67.0 0.6 1 . 6 0.2 5 . 3 0.0 0.8 0.2 3.5 2.7 3.4 0.7 1 .4 1 2. 6
3 1 79 . 1 65.3 1 .6 0.5 0. 1 1 .9 0. 1 tr 0. 1 0. 1 5.0 0.3 9.8 3.1 0.3 1 .0 1 0. 8
3 1 80.3 57.9 6.0 0.2 0.4 1 .3 2.6 1.5 2.0 1 . 8 tr 2.7 23.2 0.4 0.0
3 1 85.9 57.2 4.5 0.4 1 .2 0.4 1 .4 1 .4 3.5 20.3 9.7
3 1 3 1 .5 6 1 .2 4.7 1 .6 1 .6 2.9 1 .9 tr 2 . 3 2. 1 8.4 2.4 tr 1.3 9.6
Average 60.8 3.5 0.7 0.5 2.3 0.9 0.4 0.4 tr 1.8 2. 1 4. 7 1 0. 6 3.5 0.7 7.1

Hibernia Sandstone
348 1 . 3 67.4 2. 1 2. 1 1 . 7
0.9 0.2 0.5 4. 1 0.4 1 1 .4 tr 0.3 8.9
3482.6 7 1 .9 2.6 0.3
1 .0 0.8 2.7 1 .3 5.8 0.3 1 .0 0.2 12. 1
3483.4 67.5 1 .4 1 .0 1 . 8
0.6 tr 7.9 0.8 1 0.4 tr 2.3 6.3
3555. 1 8 1 .5 1 .0 0.6 0.6 0.4 1 .6 0.2 1.6 1 2. 5
3606.2 79.0 0.3 0.8 0.2 1 3.6 1.1 1 .5 tr 0. 1 3.4
36 1 9.4 63.6 4.5 1.3
0.5 1 . 4 0. 1 0. 1 tr 4.5 0.2 9.0 0.6 0.2 1 .2 1 2. 8
3622.4 85.3 0.2 1.7 0.6 1 .2 1 .2 9.8
3624.2 66.8 7.4 0.9 0.2
0.9 1 .7 1 .4 1 1 .2 0.3 1 .5 7.7
3845 . 8 74.0 0.9 0.3 1 . 3
0.3 2.1 0.6 8.4 1 .4 tr 1 0.7
3846.5 75 . 1 0.4 1 .0 0. 1 4.2 tr 1 9.2
3849.75 69.8 0.4 1 .0 1 . 3
0.6 0.7 0. 1 9.2 tr 1 6.9
3 8 50.3 70.2 0.4 0.2 0.4 0.4 1 .5 0.2 4.3 0.3 0.2 tr 2 1 .9
3 8 50.7 7 1 .5 0.6 0.4 0.2 tr 2.0 0.3 25.0
3 8 54.5 76.6 1 .2 0.4 2.2 tr 2.9 4. 1 1 2.6
3860.8 36.2 2.1 0.4 3.8 7.6 0.4 1.8 1 5 .9 3 1 .8 0.0
3 879.0 71.1 0.2 0.4 0.6 0.7 tr 2.1 8.9 1 6.0
3895.2 83.3 0.2 0.4 1 .0 0.9 1 4.2
Average 7 1 .2 1 .4 0.5 0.4 0.9 0.3 0.5 0. 1 tr 2.7 0.3 4.9 1 .0 0. 1 0.5 0.5 2. 1 1 2.4

Jeanne D'Arc Member

45 3 3 . 5 65.9 2.2 0.2 0.5 1 . 1 3.5 0.2 1 0. 8 1 . 3 4. 5 3.4 3.3 3.1 0.0
4534.0 5 1 .8 1 .0 4.2 2.0 2 . 8 4.0 0.8 1 .0 2.4 tr 2.9 1 9. 5 0.6 0.0
Average 58.8 1 .6 2.2 1 .6 1 .9 3.7 0.5 0.5 tr 6.6 0.6 2.2 3.1 1 1 .4 1.8 0.0 0.0 0.0

Cal, calcite; CH, chert fragments; Dol, dolomite; F, feldspar; Fos, fossil fragments; HM, heavy minerals; LS, limestone
fragments; M, matrix; METI, metamorphic and igneous rock fragments; Py, pyrite; Q, quartz, Qog, quartz overgrowth;
SH, shale fragments; Sid, siderite; SS, sandstone/siltstone fragments; tr, trace; q>, total porosity; Vole, volcanic rock
fragments.
*Uncemented half of thin section.

compaction occur between 2 500 and 3000 m depth 3000 m the effect of overburden becomes visible by
in the Catalina Sandstone Member. Above 2 500 m, the development of long concavoconvex grain con-
sandstones show a fair degree of compaction; below tacts, flowage of argillaceous intraclasts around and
Fig. 8. (A) Photomicrograph of schist fragment containing silt-sized quartz, chlorite and other phyllosilicates and
displaying pressure-solution contacts with adjacent quartz grains. Jeanne d'Arc Member, 4534.82 m, 0-35 well.
(B) Large (arrow 2) and small mud-lined, sand-filled burrows in fine-grained Hibernia Sandstone of the B-27 well,
3862.42 m. The small sand-filled structures are probably casts of Chondrites sp. (arrow 3). Note also mud clasts
(arrow 1 ) and reworked carbonaceous material (arrow 4). (C) Fine-grained, oil-stained sandstone with basal lumachelle
and lack of carbonate cementation. Avalon Sandstone, 2 1 96.67 m, 0-35 well; bedding is subhorizontal. The scale bars
for B and C are in centimetres. (D) Advanced chemical compaction (pressure-solution) contact indicated by stylolite
seam (arrows) and concavoconvex contacts between quartz grains. Hibernia Sandstone, B-08 well, 3606 . 1 8 m, A,D:
plane-polarized light; bedding direction N-S; scale bar= 0. 1 mm.
372 R. Hesse and I.A. Abid

• Avalon TEMPERAnJRE oc
• Catalina
A Hlbemla 70 80 90 100 km
� 2
c
0 ••• •
(I) c
::J
0
CJ)
�
0 • •

�(.) • •

� ...
•
,
3

.9
A A A AA
� A
"'!
A AA A A AA
0
E A A A AA ... A
Fig. 9. Porosity/depth plot for the
A
AA
� Hibernia Oilfield. Geothermal
:c 4 gradient from Suie (personal
communication). Sandstone
0 5 10 15 20 20 40 60 80 20 40 60 80 diagenetic-maturity classification
after Schmidt & McDonald
% Total porosity % Primary porosity% Secondary porosity ( 1 979a).

fracturing of quartz grains. Some of the fractures
developed after calcite dissolution (see section on
secondary porosity development); many were sub­
sequently healed by calcite, dolomite or pyrite.
Advanced chemical compaction in the form of
pressure-solution contacts between quartz grains is
encountered in the Hibernia Sandstone below the
3 500 m present burial depth, particularly in hori­
zons spared by early siderite, quartz or calcite
cementation. In extreme cases stylolitic sutures
have formed between quartz grains (Fig, 80). On
the other hand, where early cements were precipi­
tated, quartz may be floating in a calcite cement.
The early cements, however, were largely dissolved
during secondary porosity development (see later),
subjecting grains to fracturing after framework col­
lapse, although unusually loosely packed, open fab­
rics are also preserved.
The highest level of compaction is encountered in
the Jeanne d'Arc Formation at 4500 m burial
depth, including enhanced pressure-solution con­
tacts, in particular between quartz grains and schist
(Fig. 8A) and limestone fragments.
Cement precipitation
Authigenic minerals make up between 1 . 6 and 48%
(average l5o/o) of the bulk volume of the reservoir
sandstones of the Hibernia Field. Of the l 0 authi­
genic minerals identified, only six (siderite, quartz,
calcite, dolomite, kaolinite and pyrite) are com­
monly present. They are described below in the
order of their paragenetic appearance.
Siderite
Siderite, the first major cement, ranging from trace
amounts to a maximum of 32 volo/o, includes four
different types: (i) 'wheat-seed' siderite, which con­
sists of lenticular crystals of wheat-seed shape
(Fig. l OA); (ii) euhedral to subhedral siderite, re­
placing early ferroan calcite cement (Fig. l OB) or
forming grain-rimming ceme11t (Fig. l OD) (iii)
spherulitic siderite, which may contain a pyrite
nucleus; and (iv) dense, sucrose siderite (Fig. l OC)
in fine-grained sandstones adjacent to (overlying)
dark grey shales. Siderite nodules also occur in shale
beds.
Quartz
Quartz formed syntaxial overgrowths on detrital
grains (Fig. l2A) during early, middle and late
stages in the diagenetic history of the Hibernia
Field, and the present depth of occurrence does not
necessarily indicate the depth of formation. Quartz
Fig. 10. (A) 'Wheat-seed' siderite crystals (average length about 40 �m), filling major part of available pore space.
Avalon Sandstone, 0-3 5 well, 2 1 90.8 1 m; scale bar= 0.5 mm. (B) Euhedral to subhedral siderite crystals (S), 40-60 �m
in diameter, replacing early ferroan calcite cement (C). Note corrosion of quartz grains by the calcite cement (arrows).
Avalon Sandstone, B-27 well, 2562.8 m, scale bar= 0.05 mm (C) Dense sucrose siderite cement extensively replacing
quartz and other framework grains. Hibernia Sandstone, B-27 well, 3 9 1 0.45 m, scale bar= 0. 1 mm. (D) Grain-rimming
authigenic siderite outlining boundary of detrital quartz grain (lower arrow) and preceding syntaxial quartz
overgrowths (Fig. 1 2D), making it the first major cement in Hibernia Field. Siderite precipitation continued after
quartz growth (upper arrow). Hibernia Sandstone, B-27 well, 3 8 54.45 m, scale bar= 0. 1 mm. All photomicrographs
plane-polarized light.
374 R. Hesse and I.A. Abid

is the second most abundant cement in Hibernia the immediate host sediments as a silica source.
reservoir sandstones (average 3. 7%; range: trace Many samples show advanced quartz overgrowths
amounts to I I o/o). The amount of quartz overgrowth without accompanying pressure-solution effects
increases from the Avalon/Ben Nevis Sandstone (Fig. 1 2A). The import of silica cement into sand­
(average 3%) through the Catalina and Hibernia stones from adjacent shales has been advocated in
sandstones (Fig. I I). Most samples with more than the literature (e.g. Bj 0rlykke, 1 979; Boles & Franks,
8o/o quartz cement come from depths exceeding 1 979). Many of the coarser-grained deep-seated
3000 m. Minor quartz overgrowths preceded early sandstones, on the other hand, display no signifi­
ferroan calcite. Where early ferroan calcite is ab­ cant quartz overgrowths, despite their low matrix
sent, precipitation of quartz may have continued to contents, the absence of clay coatings and available
the point of developing an interlocking mosaic of pore space. However, petrographic evidence (relict
quartz cements, drastically reducing intergranular cement, oversized pores, etc.) suggests that this pore
porosity (Fig. 1 2B). Domain overgrowth refers to space is secondary in origin and may have been
an initial stage in which several individual euhedral filled by an early calcite cement, which would have
crystals on the same host crystal have not yet fully prevented silica cements from forming. Feldspar
merged (Fig. 1 2C). Early to middle-stage quartz overgrowths have been observed but are much rarer
cement (at depths shallower than 3000 m) is cor­ than quartz overgrowths.
roded and partially replaced by calcite and dolo­
mite, whereas late quartz shows no effects of
Calcite
corrosion. The occurrence of late quartz over­
growths is also documented where it engulfs kaolin­ Calcite cement occurs as minor iron-free calcite,
ite. Authigenic kaolinite (see below) occurs only early ferroan and late ferroan calcite, and fills a
below 3 1 00 m. Other evidence for late quartz is the maximum 32% of pore space. Ferroan calcite is also
occurrence of overgrowths extending into second­ observed as fracture-filling cement. Iron-free calcite
ary pores (see below). cement forms syntaxial overgrowths on fossil frag­
The increase in quartz cement with depth (below ments, especially echinoderm fragments, and is
3000 m) is not dependent on pressure solution in probably the earliest calcite cement.
Early ferroan calcite succeeds early quartz over­
m growths and is the most abundant carbonate cement
Ql Reservoir
Stage
2000
in Hibernia reservoir sandstones, filling the pores
Zone
•t:
(])
rJl • • •• • • completely and usually forming poikilotopic ce­
ments (Fig. 1 3A; see also Plate I , facing p. 374). It
Avalon
Ss also forms sparry crystals or rim cements on detrital
Albian-
"'
:::J
•• • • micritic limestone fragments, and rarely fibrous
0
Hauterivian
OJ cements on fossil fragments (Fig. 1 4A). It replaces
u
E silicate framework grains and clay matrix. Replace­
3000
� • • • •
ment of plagioclase preferentially starts along albite
u ..... atalina ••
Valanginian Ss twin planes. Grain-margin corrosion by early fer­
Qj
3 • ••
roan calcite cements is fairly common (Fig. 1 3A).
.3 Jl' • •
p
Thick clay coatings appear to rotect grains from
Berriasian Hibernia ...
•• •• calcite replacement. In a few cases, early ferroan
Ss •
• calcite cement was itself partly replaced by euhedral
4000
• Avalon Ss
siderite (Fig. I OB).
u • Catalina Ss
Jeanne Early precipitation of calcite cement is indicated
"iii & Hlbemla Ss
2 llthonlan d'Alc * Jeanne d Nc Member by high (>25%) minus-cement porosity and the
:::J
...., Member absence of other cements, except minor siderite and
:::i quartz cements and clay-mineral coatings. Frame­
*

2 4 6 8 10 12 work grains enclosed by early calcite cement are

loosely packed and often appear to be floating as a
% silica cement result of marginal replacement. Precipitation of the
Fig. 1 1 . Silica cement in form of quartz overgrowths in early calcite cement occurred after certain burial
Hibernia reservoir sandstones. because some grain breakage had already taken
Fig. 12. (A) Broad rims of syntaxial quartz overgrowths (OV) leaving pore space open. Hibernia Sandstone, B-08 well,.
348 1 .28 m, scale bar= 0 . 1 mm. (B) Interlocking mosaic of quartz (Q) overgrowths (arrows) completely filling available
pores. Compromise growth boundaries between grains are difficult to differentiate from pressure-solution contacts.
Location same as in Fig. 1 3A, 3480. 3 3 m, crossed nicols; scale bar= 0.05 mm. (C) Quartz grain displaying domain
overgrowth (OV). Several individual euhedral (or subhedral) overgrowth crystals are in state of merging to form a
continuous overgrowth. Hibernia Sandstone, B-27 well, 3 8 50.27 m, scale bar= 0.05 mm. (D) Pressure-solution
contacts (arrow I) associated with quartz (Q) overgrowths (arrow 2) in pore of possible secondary origin (P) in deeply
buried Hibernia Sandstone, C-96 well, 3927.20 m; scale bar = 0 . 1 mm.
376 R . Hesse and I.A. Abid
 Sandstone reservoir levels in the Hibernia Oilfield 377

place (Fig. 1 4B). After precipitation of this cement,
further mechanical compaction was arrested. Tight­
sandstone zones, which are completely cemented by
early ferroan calcite, are common in the Avalon/
Ben Nevis and Catalina Sandstones, but less abun­
dant in the Hibernia Sandstone, and are the result
of locaJized early calcite cementation. No dissolu­
tion features were observed at contacts between
tight and porous zones of the Avalon Sandstone in
the 0-3 5 , C-96 and K- 1 8 wells (Fig. 1 3B), whereas
the B-2 7 and B-08 wells, which are located basin­
ward from the aforementioned wells, show dissolu­
tion features at this kind of contact.
The less common late ferroan calcite cement is
associated with a minus-cement porosity of < 1 5%
(Fig. 1 3C), indicating that it was precipitated after
significant mechanical compaction. It consists of
pervasive sparry calcite, which completely filled the
remaining reduced intergranular pore space.
Fracture-filling calcite cement occurs in grain frac­
tures (Fig. 1 3D) and rock fractures. It is common
below the Avalon Sandstone.
700 J.l.m in diameter. It shows the characteristics of
'saddle' dolomite, i.e. sweeping extinction and
rarely curved cleavage planes (Fig. 1 5C). In deeply
buried carbonates it is generally considen;d a late
cement (Radke & Mathis, 1 9 80; Machel, 1 98 7),
forming at temperatures between 60 and 1 500C,
although a low-temperature origin (near 2s·q has
also been suggested (Assereto & Folk, 1 980; Mor­
row et a!., 1 986). In Hibernia Field late ferroan
dolomite was precipitated after significant compac­
tion and dissolution of calcite at moderate to great
burial depths (below 2600 m). Precipitation after
carbonate dissolution is indicated by its occurrence
in oversized and irregular secondary pores, includ­
ing mouldic pores lined with heavy oil (Fig. lSD). It
occurs in the deeper reservoirs of the Catalina,
Hibernia and Jeanne d'Arc Sandstones.
Replacement dolomite forms euhedral to subhe­
dral rhombs up to 85 J.l.m in diameter in the clay
matrix of sandstones and in early ferroan calcite
cement.

Dolomite
Dolomite cements include minor non-ferroan, early
ferroan, late ferroan and replacement dolomite.
Early ferroan dolomite occurs in an 1 8 m thick
fine-grained unit of the Hibernia Sandstone in well
B-27, where in thin zones (<I m thick) it occluded
all available pore space and replaced detrital quartz.
The early diagenetic nature of this cement (crystal
size 30-200 J.l.m, average 50 J.l.m) is indicated by a
high minus-cement porosity (>30%) and the lack of
quartz overgrowths or other cements.
Late ferroan dolomite forms large 'cloudy' crys­
tals riddled with inclusions (Fig. 1 5A), patches of
sparry dolomite (Fig. 1 5B) and large individual
euhedral to subhedral rhombs (Fig. 1 5D) up to
Clay minerals
Authigenic clay minerals identified by SEM/EDS
include minor amounts of kaolinite, chlorite, illite
and mixed-layer illite/smectite, which form grain
coatings, platelets or fibrous cements.
Authigenic kaolinite amounts to less than I % on
average (maximum 6%). It occurs as small 'book­
lets' in primary and secondary pores, forming a
meshwork which usually occupies only a portion
of the pores (Fig. 1 4C), but occasionally may fill
them entirely. Authigenic kaolinite is absent in
carbonate-cemented sandstones. It is a late cement,
which-formed after feldspar dissolution (see discus­
sion on 'Origin of secondary porosity' , below), and
commonly occupies oversized, irregular or elon­
gated pores.

Fig. 13. (Opposite) (A) Medium-grained Avalon Sandstone completely cemented by early, poikilotopic ferroan calcite
cement (C), which corrodes quartz (Q) grains (arrows). 0-35 well, 2 1 9 1 . 5 m, scale bar= 0. 1 mm. (B) Boundary between
calcite-cemented and uncemented horizons in Avalon Sandstone, showing no evidence of dissolution. Straight crystal
faces of the poikilotopic calcite (left) at the boundary (centre) indicate the presence of a cementation front, rather than
a dissolution front. Note that framework grains in the porous zone are coated with thin clay rims, which are absent in
the cemented zone. The former is also slightly more compacted than the latter. Same locality as Fig. 1 4A, 2 1 97.9 m,
scale bar= 0. 1 mm, stratigraphical top left. (C) Late ferroan calcite cement associated with < 1 5 % minus-cement
porosity. Note concavoconvex grain contacts and corroded quartz grain boundaries (arrow). Hibernia Sandstone, B-27
well, 3905.0 m, scale bar= 0.05 mm. (D) Completely shattered quartz grain between two other larger grains cemented
with poikilotopic calcite (C). The widespread occurrence of healed microfractures in siliciclastic rocks has been largely
overlooked until very recently (Milliken, 1 994). Hibernia Sandstone, B-08 well, 3496.4 7 m, scale bar = 0.05 mm. All
photomicrographs plane-polarized light; A,B,D, alizarin-S stained. (See also colour Plate I, facing p. 3 7 4.)
Fig. 14. (A) Eady fibrous carbonate (f ) cement (aragonite?) on fossil fragment (FR) oriented perpendicular to the
cement crystals. Avalon Sandstone, 0-35 well, 2 1 97.9 m, scale bar= 0.05 mm. (B) Serpulid worm tube broken by
mechanical compaction before early calcite (arrows) cementation. Same locality as Fig. 1 4A, 2 1 97.9 m, scale
bar= 0.5 mm. (C) Two elongated pores filled to different degrees with stacks ('books') of pseudohexagonal authigenic
kaolinite platelets. There is still microporosity left, even in the kaolinite-filled pore, but the permeability is drastically
reduced. Hibernia Sandstone, B-08 well, 3483.42 m, scanning electron micrograph, scale bar= 1 0 Jlm. (D)
Grain-coating authigenic illite/smectite (IIS, arrows) followed by authigenic kaolinite (K) partly filling the remaining
pore space. The quartz overgrowth, which forms the substrate for the IIS, was probably not formed in situ but
imported with a detrital grain. Hibernia Sandstone, C-96 well, 3908.3 m, scale bar= 0.05 mm.
Fig. 15. (A) Large ferroan dolomite (D) crystals riddled with inclusions. Framework grains are extensively replaced by
the dolomite. Jeanne d'Arc Member, 0-35 well, 45 34.0 m, crossed nicols, scale bar� 0 . 1 mm. (B) Fractured feldspar
grain (F) partly filled by late sparry dolomite (D). Fractured feldspar grains were only observed below 3000 m
subsurface depth, suggesting that late dolomite precipitation occurred at relatively great depth. Catalina Sandstone,
0-35 well, 3 1 85.9 m, crossed nicols. (C) Late saddle-shaped ferroan dolomite (D) with curved cleavage planes (arrow)
and undulous (sweeping) extinction (dark fan-shaped area). Note small quartz overgrowth on grain at bottom. Same
locality as Fig. 1 5A, 4534.82 m, crossed nicols. (D) Partial dissolution of a calcite rock fragment (C) followed by
emplacement of hydrocarbon linings (black) and the precipitation of late dolomite cement (D), which largely occluded
the secondary porosity. Same locality as Fig. 1 58, 3 1 82.82 m, plane-polarized light. Scale bar for B-D � 0.05 mm.
380 R. Hesse and !.A. Abid

Authigenic chlorite occurs as thin grain coatings
and as fibrous cements, in both the Avalon/Ben
Nevis and Catalina Sandstones, but rarely in the
deeper reservoirs of Hibernia Field. It is one of the
earliest cements in Hibernia Field, preceding early
quartz cement.
Authigenic illite forms characteristic grain­
bridging cements, as seen by SEM. In very few
samples, rims of illite of uniform thickness were
found lining the pores (Fig. 1 4D) and contributing
substantial amo unts (2-5%) of cement, but in gen­
eral illite cement, like chlorite, is negligible in
Hibernia Field.
Pyrite
Pyrite occurs both as early and late cement in
Hibernia Field. Early pyrite is present in early
calcite cement, fossil fragments and siderite nod­
ules. Late pyrite is found in intergranular pores and
fractures. The association of pyrite with a porous
zone in Hibernia Sandstone and its absence in
adjacent ferroan-calcite cemented sandstone sug­
gests that this pyrite formed after the dissolution of
the ferroan calcite.
Sequence of cementing minerals
The relative sequence of diagenetic events (Fig. 1 6),
reflected by the various cementation episodes, was
Secondary porosity
based on fabric relationships as a starting point. It
cannot be stated in terms of absolute time or burial
depth. Absolute dating of the events is not possible,
nor do the present depths of the diagenetic phases
indicate their depth of origin. However, the conclu-
sions drawn from the microscopic study will be
corroborated with evidence from limited isotopic
and geochemical data.
The chemical diagenesis of the Hibernia reservoir
rocks commenced with the formation of minor
chlorite coatings and siderite cement. This was
followed by the earliest quartz overgrowths. Lo­
cally, however, siderite precipitation continued dur­
ing this early quartz cementation stage. Early
ferroan calcite occurred after loss of about 1 0- 1 5%
of the original porosity by mechanical compaction
(corresponding to a minus-cement porosity of 25-
30%). Its widespread distribution controlled later
secondary porosity development, underlining the
significance of carbonate cementation for reservoir
quality. Minor pyrite was coprecipitated with early
non-ferroan calcite or, more commonly, preceded
early ferroan calcite. Early calcite was not ubiqui­
tous, however, and where it was absent, quartz
overgrowth continued. A major dissolution phase
generating secondary porosity (see below) occurred
below 3200 m of burial. This was followed by
minor compaction, late ferroan calcite cement,
grain fracturing and late quartz overgrowths, which
in part reversed the effects of the porosity­
enhancing event. Kaolinite, late ferroan dolomite
and late pyrite also occur in secondary pores,
further reducing the porosity gain.
Secondary porosity is the main contributor to the
present reservoir porosity in Hibernia Field, but its
significance varies considerably between different
stratigraphical levels. In what follows, different ge-

..

Phases (listed in order Early Diagenesis Base of Middle Diagenesis

Increasing
of first appearance) (<1500 m) (>3600 m)
( Its with >70% S) burial depth (Its with <20% S)
1-Mechanical Compaction �
2-Giay coaling of framework - - - - --

grains including early

fibrous cement ____.. r-
- --

3-Siderite -- - -

4-Early quartz overgrowth ___... -

-- - -

5-Early pyrite ? -
&-Early Fe·Calcite --
-? -
-
-
?-Early Fe-dolomite - - -
- -

�ua:rtz overgrowth --
�Dissolution
- -
-?

1 0-Late Calcite - �-�

Fig. 16. Schematic paragenetic
1 1 -Minor grain fracturing - -- - -
sequences of diagenetic events in
1 2-0uartz overgrowth -- - - --- - - -

- - - --
Hibernia Field based on
13-Kaolinite
1 4-La1e Fe·dolomite -- - -
petrofabric evidence. Abundance
1 5-Fe·Calcite overgrowth -
- -
of cements and intensity of
1 5-Late pyrite
-
dissolution events indicated by
1 7-Hydrocarbon Migration - ? ....
width of the bar.
 Sandstone reservoir levels in the Hibernia Oilfield
netic types of secondary pores and their relationship
to carbonate diagenesis will be discussed, and poros­
ity evolution for individual reservoirs evaluated.
Secondary porosity may be difficult to differen­
tiate from primary porosity, if both occur in the
same thin section. Primary porosity, the intergran­
ular pore space retained during deposition, is recog­
nized by the more or less even and homogeneous
pore distribution, the lack of oversized, elongate
and mouldic pores, and the lack of grain-margin
corrosion and partially dissolved grains. Secondary
porosity is recognized by the presence of these
features, by floating, honeycombed and fractured
grains, by inhomogeneous grain packing and by
grain-rimming pores indicative of shrinking
(Schmidt & McDonald, 1 979b). Besides interparti­
cle (intergranular) porosity, primary porosity and
secondary porosity both include intraparticle poros­
ity in the form of pores and cavities in fossil shells,
and honeycomb grains. Secondary porosity may
either recover lost primary porosity or generate new
porosity by the dissolution of framework grains
and/or replacive cements and by grain shrinkage.
Most of the secondary porosity in Hibernia Field
is intergranular, and was formed by dissolution of
pore-filling and replacive cements (Fig. I 7 A; see
also Plate 2, facing p. 3 7 5). Mouldic porosity
demonstrates the dissolution of former framework
grains and is observed in all Hibernia reservoir
sandstones. Oversized and elongated pores and
irregularly distributed pores originated largely from
the dissolution of fossil fragments and the removal
of cements, including grain replacements, and are
common in the Catalina and Hibernia sandstones.
Intraparticle porosity of secondary origin is also
most common in the Catalina and Hibernia Sand­
stones (Fig. 1 7B). Fracture porosity is insignificant
volumetrically, but may have enhanced reservoir
permeability. Shrinkage porosity, another minor
type of secondary porosity, is typically found asso­
ciated with collophane (Fig. 1 8B) and glauconite.
Avalon/Ben Nevis Sandstone
The Avalon/Ben Nevis Sandstone is an example of
the depositional environment having affected sub­
sequent sandstone diagenesis. The dominantly fine­
grained, marine to marginal marine and coastal
sandstones (Sinclair, 1 993) and interbedded shales
are rich in calcareous fossils and probably provided
the source of much of the early calcite cement. Early
ferroan calcite cement in the Avalon/Ben Nevis
38 1
Sandstone amounts to 8. 5%, compared with a total
of 5% non-calcite cements (chlorite, siderite, quartz
overgrowths, pyrite). Late cements such as kaolinite
and dolomite are absent. Early ferroan calcite ce­
mentation affected sandstone beds less than 1 . 5 m
thick, completely occluding their porosity and
forming 'tight sandstones'. Based on five wells
studied, about 20% of the Avalon Sandstone drill
cores are tightly calcite cemented. The 'tight­
sandstone' zones are particularly abundant in wells
B-27 and C-96 (22 and 40% tightly calcite­
cemented zones, respectively). The early calcite
cement interrupted mechanical compaction and
prevented the precipitation of other cements, mak­
ing these sandstones excellent candidates for sec­
ondary porosity evolution at deeper levels. In the
present Avalon/Ben Nevis Sandstone, however, po­
rosity is predominantly primary in origin (9.6%), as
shown by the thin-section study, with the addition
of 2.6% secondary porosity derived from both
cement dissolution ( 1 .9%) and framework-grain
dissolution (0. 7%; Table 2; e.g. Fig. 1 7C). In the
vicinity of shale contacts, secondary porosity may
reach 20%. Dissolution is more widespread in wells
B-27 and B-08 than in 0-3 5 . This secondary poros­
ity mimics primary porosity and can easily be
overlooked. After the dissolution pulse, almost no
recementation and no significant compaction oc­
curred in Avalon/Ben Nevis Sandstone. The abun­
dance of primary porosity (Fig. 1 7D) and lack of
significant chemical compaction (pressure solution)
render the Avalon/Ben Nevis Sandstone immature
in the diagenetic maturity classification of Schmidt
& McDonald ( 1 979a).
Catalina Sandstone Member
Secondary porosity contributes more than half
(6. 7%) to the total porosity (average.of 1 1 %, exclud­
ing 'tight-sandstone' zones) of the Catalina Sand­
stone. Of the available drill cores from wells 0-35
and K- 1 8, 4 5 . 7 and 1 7. 5%, respectively, are com­
pletely cemented by early ferroan calcite. The dif­
ference between the two wells may be due to the fact
that the sandstone beds are thicker in well K- 1 8.
Where the early ferroan calcite is absent, mechani­
cal compaction and quartz overgrowths have re­
duced the primary porosity further. In K- 1 8, the
framework grains are largely coated by a micritic­
calcite rim, approximately 25 Jlm thick, which pre­
vented silica cementation. Dissolution of the early
ferroan calcite cement contributed most of the
Fig. 17. (A) Irregularly shaped secondary pores (P) resulting from dissolution of framework grains and pore-filling and
replacive cements. Hibernia Sandstone, B-27 well, 3 8 50.65 m. (B) Intraparticle microporosity of secondary origin
resulting from partial dissolution of calcite in recrystallized portion of fossil fragment. Arrows indicate uncorroded
crystal faces of calcite. Incipient pressure solution between quartz and calcite shell. Avalon Sandstone, 0-35 well,
2 1 8 5 . 8 m. (C) Large mouldic pore resulting from dissolution of carbonate fragment. The prismatic ferroan-calcite
cement crystals that had grown on the fragment are less affected by dissolution and relatively well preserved. Their
colour varies as a function of distally increasing iron content. Avalon Sandstone, B-27 well, 2578.84 m, scale
bar = 0.05 mm. (D) Fossil-rich fine-grained sandstone lacking carbonate cement despite pressure solution (arrows)
between quartz grains and fossil fragments. Porosity is primary. Same locality as Fig. 1 78, 2 1 95.4 m. All
photomicrographs plane-polarized light. A,B,D, scale bar= 0.5 mm. (See also colour Plate 2, facing p. 375.)
 Sandstone reservoir levels in the Hibernia Oilfield 383

Fig. 1 8 . (A) Beginning dissolution o f feldspar grain. Hibernia Sandstone, B-27 well, 3849.75 m. (B) Shrinkage porosity:
thin rim of open pore space (arrows) around collophane (fluorapatite) grain. Avalon Sandstone, 0-35 well, 2 1 90.8 1 m.
(C) Almost completely dissolved chert grain leaving secondary pore (P). Hibernia Sandstone, same locality as Fig. 1 8A,
3 850.27 m, scale bar= 0. 1 mm. (D) Incomplete dissolution of shale clast and rimming calcite cement generating
secondary pore (P). Catalina Sandstone, 0-35 well, 3 1 76.0 m. All photomicrographs plane-polarized light. A,B,D, scale
bars= 0.05 mm.
384 R. Hesse and !.A. Abid

Table 2. Point-count analyses of pore types

Depth (m) Tot.P SP pp DC DFG DRC F

Avalon Sandstone
2 1 84.6 12.5 1.8 1 0. 9 1 .8
2 1 85.8 9.2 1 .2 8.0 1 .2
2 1 90 . 3 6.2 0.2 6.0 0.2
2 1 90 . 8 9.2 0.8 8.4 0.8
2 1 97.5 1 8.0 3.8 1 4. 2 3.2 0.4 0.2
2 1 97.9 1 0. 2 1 .6 8.6 1 .6
2578.8 2 1 .2 6.0 1 5 .2 3.2 2.8
2659.6 1 1 .4 5.4 6.0 3.4 2.0

Average 1 2. 2 2.6 9.6 1 .9 0.7

Catalina Sandstone
3 1 78.0 1 2.6 1 0. 2 2.4 6.4 2.6 1 .2
3 1 7 9. 1 1 0. 8 6.7 4.7 3.0 1 .6 1 .5
3 1 85.9 9.7 5.3 4.4 4.0 0.5 0.8
3 1 3 1 .5 9.6 5.0 4.6

Average 1 0. 7 6.7 4.0 3.35 1 .2 0.9

Hibernia Sandstone
348 1 . 3 8.9 4. 1 4.8 2.6 1 .2 0.3
3482.6 1 2. 1 8.0 4. 1 4.2 2.9 0.8 0. 1
3483.4 6.3 3.6 2.7 1 .2 0.8 1 .6
3555. 1 1 2. 5 8.9 3.6 4.8 3.5 0.6
3606.2 3.4 1.8 1 .6 0. 1 1 .7 0.0
36 1 9. 4 1 2. 8 8. 4 4.4 4.3 3.0 1 .0 0. 1
3622.4 9.8 6.9 2.9 5.5 1 .0 0.4
3624.2 7.7 3.3 4.4 2.2 0 .5 0.4 0.2
3845.8 1 0. 7 8.8 1 .9 1 .7 1 .3 5.8
3846.5 1 9 .2 1 6. 6 2.6 8.8 3.4 4.4
3 8 49 . 7 5 1 6 .9 1 5 .0 1 .9 7.5 2.2 5.1 0.2
3 8 50.3 2 1 .9 2 1 .9 1 1 .2 9.0 1 .6
3 8 50.7 25.0 25.0 1 6.0 6.6 2. 4
3854.5 1 2.6 7.0 5.6 4.8 1 .2 1 .0
3879.0 1 6. 0 1 6 .0 1 0. 3 4.4 1.3
3895.2 1 4. 2 1 4.2 8.0 1 .4 4.8

Average 13.1 1 0.6 2.5 5.8 2.8 2.0

DC, dissolved cement; DFG, dissolved framework grains; DRC, dissolved replacive cements; F, fracture porosity; PP,
primary porosity; SP, secondary porosity; ToT.P, total (thin-section) porosity.

secondary porosity (Table 2). Secondary porosity is

highest in sandstones immediately above shale
beds. K- 1 8 shows much less dissolution than 0-3 5 ,
i n line with its much lower early ferroan calcite
cement. The porosity in K- 1 8 is still largely pri­
mary.
Late cements causing post-dissolution porosity
reduction include minor kaolinite and late ferroan
calcite. Late ferroan dolomite is an important ce­
ment in all porous sandstones of the 0-35 well,
where it reduced porosity by 4.2% on average
(ranging from trace amounts to 20%). In the diage­
netic maturity classification of Schmidt & MacDon­
ald ( 1 979a) the Catalina Sandstone is semimature.
Hibernia Sandstone
Early ferroan calcite cementation played a major
role in preventing permanent porosity loss in Hiber­
nia Sandstone. Although at present calcite cements
occupy only 1 % of the bulk volume, excluding cal­
cite-rich zones, which make up 4% of all sandstone
cores studied, most of the samples contain remnants
of calcite cements. Together with other indicators of
secondary porosity, such as extensive grain-margin
corrosion, oversized and mouldic pores, inhomo­
geneous framework packing and irregular porosity
distribution, these remnants suggest that calcite
cement was once much more pervasive and that its
 Sandstone reservoir levels in the Hibernia Oilfield 385

dissolution was the main source ( I I %) for the
present thin-section porosity, which totals 1 3%.
Based on these criteria, one may conclude that
thin sandstone beds (less than 3 m thick) were
generally completely cemented by the early calcite
cement, which prevented mechanical compaction
or irreversible quartz cementation. These medium­
grained sandstone beds today show a high propor­
tion of secondary porosity (up to 90% of the total
porosity). To argue that these rocks have never seen
carbonate cements would be to neglect significant
evidence. Remnant cements occur in 48% of the
thin sections studied. Many thick sandstones ( 6-
1 5 m thick), on the other hand, were not completely
cemented by early calcite and show evidence of
primary porosity loss by silica cementation (quartz
overgrowths) or mechanical compaction, including
squeezing of shale clasts into primary pores. Poros-

Table 3. C and 0 isotopic analyses of carbonate cements (the dissirriinated wheat-seed siderite (Fig. 1 OA) of the
Avalon Sandstone was not analysed isotopically)

Subsurface
Reservoir sandstone Well no. depth (m) 0 1 80 (PDB) o 1 3C (PDB) Petrography

Early calcite
Avalon Ss 0-35 2 1 47 . 7 -6.07 2. 1 5 fcal, poi
Catalina Ss 0-35 3 1 80.2 -8.38 -0.65 cal, crf
K- 1 8 3 1 35.6 -5.02 -0.94 feat, crf
K- 1 8 3 1 44.2 -4. 3 3 0.0 1 feat
Hibernia Ss K- 1 8 3 8 1 4.0 -6.83 -3.23 fcal, poi
K- 1 4 386 1 . 7 -8.08 -3.33 fcal, poi
K- 1 4 3872.2 -6.08 -1.31 fcal
K- 1 4 3932.5 -8.56 -5.89 fcal
Average -6.66 - 1 .64

Late calcite
Hibernia Ss B-27 3885.4 -9.53 -9.40 feat
B-27 3892.3 -9.20 - 1 2. 1 3 fcal
B-27 3892.4 -9.24 - 1 2.22 feat
B-27 3893.6 -9.03 - 1 2.30 feat
B-27 3905.0 -9. 2 1 - 1 1 .22 fcal, poi
B-27 3906.0 -8. 8 5 - 1 1 .68 feat
C-96 3939. 1 -6.92 -7.50 feat
Average -8.85 - 1 0.92

Early dolomite
Hibernia Ss B-27 3858.5 -3.62 -3.86 fdol, anhed
B-27 3858.7 -3.20 -4. 30 fdol, anhed
B-27 3862.0 -2. 5 1 -2. 3 8 fdol, anhed
Average -3. 1 1 -3.5 1

Late dolomite
Hibernia Ss B-08 3485.6 -9. 1 1 -6. 1 8 fdol
Jeanne d'Arc 0-35 4533.4 -6. 3 1 -4. 55 fsaddol
0-35 4534.8 -6.52 -4.45 fsaddol
Average -7. 3 1 -5.06

Siderite
Hibernia Ss B-27 3885.8 -6.49 -5.65 sri
B-27 3 9 1 0. 4 -3.90 - 1 0.20 sri
K- 1 8 3860.3 -5.22 -7. 1 3 sri
Average -5.20 -7.66

anhed, anhedral crystals; cal, non-ferroan calcite cement; crf, sample containing carbonate rock fragments; fcal, ferroan
calcite cement; fdol, ferroan dolomite cement; fsaddol, saddle dolomite; poi, poikilotopic cement; sri, siderite-rich
laminae; Ss, sandstone.
3 86 R. Hesse and I.A. Abid
ity loss by silica cementation averages 5% in the
Hibernia Sandstone. Sandstones with high second­
ary porosity generally show no or little quartz
overgrowth (Fig. 1 7 A), probably indicating exten­
sive early carbonate cementation, which prevented
silica cementation.
Porosity reduction due to recementation after the
dissolution event is generally small ( 1 -2%) and
involved kaolinite, quartz, pyrite and ferroan cal­
cite and dolomite cements. Locally, as much as 1 0%
of late cements have been precipitated. Kaolinite
usually reaches no more than trace amounts, al­
though locally up to 6% has been observed. Late
quartz overgrowths are negligible. Late pyrite gen­
erally does not exceed 1 %; however, locally it may
occlude all available pore space. Late ferroan calcite
cement has completely filled the available pores at a
few levels, but the total porosity loss by this cement
and by late ferroan dolomite is not significant.
Jeanne d'Arc Sandstone
Available drill cores from the 0-35 well show very
low thin-section porosity ( <5%). Advanced me­
chanical compaction, including shale-clast defor­
mation, together with ferroan dolomite cement and
quartz overgrowths, has occluded most of the po­
rosity.
INTERPRETATIONS: SOURCES AND
BURIAL ENVIRONMENTS OF
CARBONATE CEMENTS
IN LIGHT OF ISOTOPIC
AND GEOCHEMICAL DATA
The history of porosity development at the various
reservoir levels in Hibernia Field clearly under­
scores the role of carbonate cements in delaying
irreversible porosity loss in sandstone reservoirs.
Early non-ferroan and ferroan calcite cements, pre­
cipitated at burial depths shallower than 2000 m,
are widespread in oilfields around the world (e.g.
Lindquist, 1 977; Blatt, 1 979; Loucks et a!., 1 984;
Olaussen et a!., 1 984; Bj0rlykke et a!., 1 986; Imam
& Shaw, 1 987; Kantorowicz et a!., 1 987; Saigal &
Bj0rlykke, 1 987; and many others) and are typical
as first major cements for a group of cement
parageneses (Franks & Forester, 1 9 84). Four dif­
ferent sources of this early calcite cement have been
considered in the literature: (i) carbonate dissolu-
tion from fossil and limestone fragments and repre­
cipitation as sandstone cements (Almon & Davies,
1 979; Blatt, 1 979; Bj0rkum & Walderhaug, 1 990);
(ii) percolating meteoric water supersaturated with
respect to calcite (Longstaffe, 1 984; Walderhaug &
Bj0rkum, 1 992); (iii) decreasing calcite solubility in
upward advecting pore water due to decreasing
PC0 ; and (iv) increased carbonate alkalinity and
2
pH owing to early diagenetic sulphate reduction
and methane generation and associated Fe and Mn
reduction in marine sediments (Irwin et a!., 1 977).
i> 1 3 C measurements constrain the possible sources
of the carbonate cements and point towards a mix­
ture of carbonate from dissolved framework grains
of marine origin and carbonate derived from
organic-matter degradation in the sulphate-reduc­
tion and thermocatalytic reaction zones. Marine bio­
genic calcite averages +2%o D 1 3 Cp08. Bacterial oxi­
dation of organic matter in the zones of sulphate
reduction and thermocatalytic decarboxylation pro­
duces isotopically light bicarbonate (to values as low
as -25%o il 1 3 C}, whereas isotopic fractionation dur­
ing methanogenesis generates heavy bicarbonate
with il 1 3 C values up to + 1 5%o (Irwin et a!., 1 97 7).
D 1 8 0 measured in conjunction with D 1 3 C constrains
the temperature range of cement precipitation, and
may hint at meteoric-water influx.
Early ferroan calcite
The average i3 1 3 C value of this study for early calcite
( - 1 .6%o; Fig. 1 9) indicates dissolution of marine
carbonate as the predominant source. In Avalon/
Ben Nevis Sandstone the average is close to zero
(-0.9%o PDB; Hutcheon et a!., 1 98 5 ) (Fig. 1 9), show­
ing that this source could have provided the carbon­
ate almost exclusively, in line with the abundant
dissolution effects seen in the shallow-marine car­
bonate fossil fragments of the sandstones.
The oxygen isotopic values for these early calcite
cements vary between -0.5 and -8.5%o i3 1 8 0 (Table
3), too large a range to be caused by temperature
variations alone in the early burial diagenetic envi­
ronment of probably less than 500 m subsurface
depth (corresponding to minus-cement porosity
values of 25-30%). With a geothermal gradient of
2 6 o C for the Hibernia Field (Correia et a!., 1 990},
the permissible temperature range would be ap­
proximately 1 5 ° C, not > 3 5 o C as suggested by the
isotopic data. The wide range of negative values
therefore reflects modification of the original sea
 Sandstone reservoir levels in the Hibernia Oilfield
+3
+1
-1
-3
813 c -5
(PDB)
-7
-9
-11
-13
Fig. 1 9 . Covariation o f carbon and
oxygen isotopic composition of
carbonate cements from the Hibernia
Field. Average of Avalon Sandstone
from Hutcheon et a!. ( 1 985).
water isotopic signature of the pore fluids (assuming
Late ferroan calcite
a Cretaceous sea water SMOW of Oo/oo (Veizer, 1 98 3),
possibly by influx of and mixing with meteoric
water. The results of trace element analyses (Sr, Mn,
Mg, etc.) of Avalon calcite cements by Hutcheon et
al. ( 1 98 5 ) are compatible with mixing; however,
they are not conclusive and would allow mixing
either with meteoric waters or with rising deeper
(and hotter) basin fluids.
Dissolution of biogenic marine carbonate as the
source of the early cements does not exclude mod­
erate migration distances of the precipitating fluids.
The occurrence of sandstones rich in calcite cement
above fossil-rich calcareous shales points towards
import of the carbonate cements from adjacent
shales by upward-moving pore fluids expelled by
shale compaction. The presence of carbonate ce­
ments in sandstone zones free of fossil or carbonate
fragments (Fig. 1 3A), and, vice versa, their absence
in some of the fossil-rich zones (Fig. 1 70) also
supports carbonate redistribution or an external
(although possibly nearby) source.
387
North Sea and
0
•
0 Early calcite
* Early dolomite
• Siderite
• Late calcite
& Late dolomite
& Average of Avalon
o Average of early
calcite cement
-12 -10 -8 -6 -4 -2 - 1
Late calcite cements are considerably lighter isoto­
pically (average of - ! 0.9o/oo 0 1 3 CPDB range -7.5 tO
'
- 1 2 . 3o/oo) (Figs 1 9 and 20). Together with the more
negative 0 1 8QPDB va)UeS (average -8.8 5o/oo, range
-6.9 to -9. 5o/oo), which may indicate elevated tem­
peratures, and the fabric evidence for precipitation
at greater burial depth discussed earlier, the carbon
isotopic results imply a greater contribution of
carbonate from organic-matter decomposition in
the thermocatalytic decarboxylation zone.
Geochemical observations concerning the iron
content of the calcite cements conform with the
isotopic and petrofabric results. Both early and late
calcite cements contain variable concentrations of
iron, as indicated by our staining results and con­
firmed by the microprobe analyses of Hutcheon et
al. ( 1 985) for the Avalon Sandstone (range 1 . 5-
4.4 wto/o Fe). Possible sources are the reduction of
iron in Fe-oxide and hydroxide coatings for the
early ferroan calcite, the intrastratal solution of
388 R. Hesse and I.A. Abid

Reservoir m
'" Stage
.iii Zone 2000
<f)
•

<f)
:::J Avalon 0
0 0
0 Ss 0 0
(j) 0 0 (
0 AJbian-
0 0
+- Hauterivian
(j) o O
0
3000
(5
3 Catalina ...
_g Valanginian Ss
0

Berriasian Hibernia
- •
' •
Ss • •
4000

()
·u;
</) Jeanne
0 Tithonian
's d ' hc • Early calcite 4500
--, • Late calcite
Member
::i 0 Early calcite (Hutcheon et aL. 1 985)

Fig. 20. o 1 3C values of early and

-1 2 -8 -4 0 +4
late calcite cements versus
8 13 C (PDB) subsurface depth.

unstable heavy minerals for both the early and late
Fe-bearing cements, and the smectite-to-illite reac­
tion (Abid, 1 98 8 ) for late ferroan calcite cements.
Ferroan dolomite
The isotopic composition of ferroan dolomite fol­
lows similar trends to the calcite, with o 1 3 C values
decreasing from early to late cements (Fig. 1 9) but a
smaller range (-2.4 to -6.2o/oo). The less negative
average 8 1 3 C for the late dolomite (-5. 1 o/oo) com­
pared with the late calcite (- 1 0 . 9o/oo) probably indi­
cates a stronger contribution from carbonate
dissolution than from organic-matter decomposi­
tion, in line with fabric evidence for the association
of late dolomite with secondary pores (Fig. 1 5D).
The average o 1 8 0p08 for the early dolomite
(-3 . 1 o/oo) is less negative than that for the early
calcite (-6. 7o/oo) by about 3o/oo, which is the differ­
ence between co-precipitated calcite and dolomite
resulting from different equilibrium fractionation
factors (Land, 1 980). However, the observed differ­
ence may be fortuitous. A case has been made by
Morad et al. ( 1 992) for early dolomite precipitation
in the phreatic-vadose mixing zone, but these
dolomites are different isotopically and chemically
from the Hibernia cements.
Siderite
As for ferroan dolomite, the number of siderite
samples analysed isotopically is very limited and
can only give a hint of the diagenetic environment
of siderite precipitation. Siderite, like other ferroan
carbonates, can only be precipitated beneath the
sulphate reduction zone in the zone of methano­
genesis and deeper, where no , dissolved sulphide
competing for reduced iron exists. Also, the concen­
tration ratio of dissolved Fe and Ca [Fe 2 + ]/[Ca2 + ]
must b e relatively high, otherwise ferroan calcium
carbonates would form (Curtis, 1 967). On the o 1 3 C/
8 1 8 0 plot of Fig. 1 9, the three siderite samples de­
fine a trend. of increasing o 1 3 C values with decreas­
ing o 1 8 0. This might reflect precipitation in the
methane generation zone with increasing distance
from the bottom of the sulphate reduction zone. At
the transition from the sulphate reduction zone to
the methane generation zone, o 1 3 C values tend to
be negative but evolve towards being positive.
Alternatively, a trend of increasing o 1 3 C values with
 Sandstone reservoir levels in the Hibernia Oilfield
decreasing o 1 8 0 can originate during deeper burial
in the thermocatalytic reaction zone, if an increas­
ing proportion of the carbonate is derived from dis­
solution of solid carbonate rather than from organic
matter. Petrofabric evidence for the early appear­
ance of siderite supports the first possibility (pre­
cipitation in the methane zone), but a deeper
origin cannot be excluded for samples with lower
o 1 8 0 values. The sources of iron for the siderite may
be the same as those suggested for the early ferroan
calcite cement. The observed replacement of early
ferroan calcite by siderite (Fig. I OC) indicates a
change in the pore fluid composition towards ele­
vated Fe 2 + activity. The activity of Ca2 + was low­
ered relative to that of Fe2 + by calcite precipitation,
leading to subsequent siderite formation.
DISCUSSION: ORIGIN OF
SECONDARY POROSITY
Up to 90% of the porosity in individual reservoir
horizons of the Hibernia Field is interpreted to be
of secondary origin (Fig. 9). As stated in the intro­
duction, secondary porosity development is most
closely related to carbonate cementation, because
the main process responsible for the generation of
secondary pores is carbonate dissolution, followed
in importance by feldspar dissolution (Fig. I 8A)
(Schmidt & McDonald, I 979a; Surdam et at.,
I 984). Dissolution of chert grains (Fig. 1 8C) (Shan­
mugam & Higgins, I 98 8 ; Bloch et al., I 990), mud
or shale clasts (Fig. I 8D) and other minor or acces­
sory framework components (such as unstable
heavy minerals) contributed subordinate amounts.
Fluids that cause dissolution must be undersatu­
rated with respect to the solid phase. Such fluids can
either be meteoric waters (Bjorlykke, I 979;
Mathisen, I 984) or rich in acids generated during
diagenesis. The latter may comprise carbonic acid,
organic acids and hydrochloric acid formed by
bacterial or thermocatalytic organic-matter degra­
dation reactions and clay-diagenetic reactions
(equation 3 in Nesbitt, I 98 5 ) in shales interbedded
with sandstones.
Carbon dioxide generated during all stages of
bacterial and thermocatalytic organic-matter degra­
dation will only produce carbonic acid if the pH is
controlled by the carbonate system itself. In the
sulphate reduction and methane generation zones,
the reduction of metals (Fe3 + , Mn4 + ) raises the pH
and an increase in the partial pressure of C02 leads
389
to carbonate precipitation rather than dissolution
(Surdam et at., 1 9 84). Carbonate dissolution by
carbonic acid will therefore only occur at very
shallow burial depths in the oxidation zone, the
nitrate reduction and beginning sulphate reduction
zones, and at greater burial depths in the thermo­
catalytic decarboxylation zone, where the pH buffer
provided by organic acids may be overcome by the
carbonate system. Organic acids together with car­
bonic acid have a higher dissolution potential than
carbonic acid alone (Surdam et at., 1 984), as they
can also attack silicates, particularly feldspar. At
temperatures less than I 20 ° C, organic acids are
stable and do not undergo decarboxylation. In the
temperature range 70- I 20 ° C, organic-acid radicals
such as oxalate may complex aluminium and thus
greatly enhance AI mobility (Surdam et al., 1 984).
Low AI mobility in the absence of this complexing
reaction is a major obstacle for aluminosilicate
dissolution. Carboxylic-acid generation from or­
ganic matter preceding significant liquid hydrocar­
bon generation has been demonstrated both
experimentally (for a review see Fein, 1 994) and by
field studies (e.g. Carothers & Kharaka, 1 9 78; Lun­
degard & Kharaka, I 990). The development of
maximum acidity in organic matter rich shales is
broadly coincident with the onset of ordering in
illite/smectite (liS) mixed-layer clays at 80- 1 OO O C
(Dypvik, 1 98 3 ; Pearson et al., 1 98 3), which occurs
at a smectite content of about 35% in I/S (Perry &
Hower, 1 9 72). In Hibernia Field this level is
reached below 2500 m burial depth (Abid, 1 9 88,
1 996). The decrease of expandable smectite in liS
mixed layers is associated with the release of signif­
icant amounts of semi-bound water from the clays
to the pore space, which is the main source of deep
basinal fluid other than overpressure zones, essen­
tial for the transport of acids and to boost hydrocar­
bon migration (Burst, 1 969; Br.uce, 1 984). In
Hibernia Field, smectite decrease is most rapid
between 2000 and 2500 (Abid, 1 98 8 , 1 996).
Avalon/Ben Nevis Sandstone
The Avalon/Ben Nevis Sandstone, which at present
is buried to depths between 2 1 00 and 2600 m
(corresponding to 5 5-70 ° C subsurface tempera­
ture), not surprisingly shows little secondary poros­
ity. After what has been said above, acid release
would have just started at these temperatures. How­
ever, much of the acid generated in the shales would
have been neutralized locally by reaction with
390 R. Hesse and I.A. Abid
carbonate from the calcareous shales. The disappear­
ance of calcite from the shales below 3600 (Abid,
1 988) shows that carbonate dissolution is a continu­
ing process. The isotopic data (large spread of the
o 1 80 values) and trace element composition of
calcite cements indicate the involvement of mete­
oric waters (Hutcheon et al., 1 985), which may have
invaded the reservoir from the Mid-Cretaceous
unconformity during shallow burial. Some dissolu­
tion may have been associated with this event.
Catalina Member
Secondary porosity development at greater burial
depths is likely to be caused by mechanisms other
than meteoric-water influx, which becomes exceed­
ingly more difficult with depth and results in a
decrease of the dissolution potential of such waters
as they become saturated with the dissolved species
along the way. In the Catalina Sandstone, secondary
porosity is closely associated with sandstone-shale
contacts (Fig. 2 1 ). Feldspar grains and shale clasts
Ca l c i te Fe-do 1 omi te
cement cement
3173 5 10 10 20 5 10
% % % explain 1 -2% of secondary porosity, not the 1 0%
3175
3180
3185
Fig. 2 1 . Distribution of porosity and carbonate cements
in the Catalina Sandstone (='B' Sandstone), 0-35 well.
The sandstone beds are largely cemented with early (?)
ferroan calcite, except for a few centimetres above shale
interlayers, where cement dissolution has generated
significant secondary porosity.
(Figs 1 8A,D) are commonly corroded or dissolved,
compared with nearby calcite-cemented zones. The
present burial depth of about 3000 m provides the
conditions ( 7 5-8 5 o C) necessary for maximum acid
generation outlined above.
Hibernia Sandstone
Maximum secondary porosity values in Hibernia
Field have been identified with the deeply buried
Hibernia Sandstone, where they exceed the remain­
ing primary porosity. One of the problems previ­
ously encountered with secondary porosity
development was to find source rocks that can
generate sufficient quantities of acids to explain the
newly formed pores in the reservoir rock. The
problem is closely related to the source-rock prob­
lem for hydrocarbons in general. Where the latter
has been solved, it usually comprises the answer for
the origin of the secondary porosity. Lundegard et
al. ( 1 984) estimated that the present TOC content
in the Oligocene Frio Formation of the Texas Gulf
Coast of 0.28%, which they extrapolated to 0 . 5 wto/o
of original organic matter (of type III, with 25 wto/o
of oxygen, 25% of this oxygen being located in
carboxyl groups), could generate sufficient acids to
observed on average in the Frio. Contrary to the
affirmation by these authors that their initial
organic-matter estimate was high, it was probably
grossly underestimated. The present TOC in shale
interbeds of the Hibernia sandstone is 0.6% by
weight (Swift & Williams, 1 980) and is of terrestrial
origin (type III), which has the highest C0 2-
generating capacity (Tissot & Welte, 1 984). The
original TOC, however, may have been as high as
1 -3%. The shale/sandstone ratio is 2 in the region
of the K- 1 8 and B-27 wells (Brown et al., 1 9 89),
compared with 4.4 in the Frio Formation; however,
the ratio increases from the margins (Hibernia
Field) towards the centre of the Jeanne d'Arc basin.
Without going into a detailed calculation, it appears
that the secondary porosity generating capacity of
the Hibernia Sandstone from internal shale sources
would be marginal, if not insufficient. However, the
main hydrocarbon source rocks of the basin are
Kimmeridgian shales, with up to 8 wt% TOC, as
stated before. In the Avalon sub-basin they occur
less than 300 m beneath the Hibernia Sandstone,
and acidic fluids could have easily reached the
reservoir horizons by upward migration, as did the
hydrocarbons.
 Sandstone reservoir levels in the Hibernia Oilfield
CONCLUSIONS
The significance of the present diagenetic study is
its comprehensive nature, comparing different res­
ervoir levels of Hibernia Field at different burial
levels. It enabled us to demonstrate the interplay
between carbonate cementation and cement/grain
dissolution events as major factors preserving eco­
nomically interesting porosity levels at various
stages of burial. The results confirm earlier findings,
first summarized by Franks & Forester ( 1 984), that
the complex cement parageneses in carbonate­
bearing sandstones show an orderly sequence of the
various cementation phases and intervening disso­
lution events. If this sequence is taken as a norm,
deviations from it can be interpreted in terms of
local or temporal disturbances of the normal phys­
icochemical burial history.
ACKNOWLEDGEMENTS
Funding of this research by the Natural Sciences
and Engineering Research Council (NSERC) of
Canada through an operating grant to R.H., and by
the Government of Pakistan through a scholarship
to I.A., is gratefully acknowledged. Special thanks
go to Joe N. van Elsberg (formerly Mobil Oil,
Canada) and Volkmar Schmidt (Petroscan, Calgary,
Alberta) for suggesting this project and advice
during its execution, and to E.M. Leavitt (Mobil),
G. Campbell (COGLA, Ottawa), and D.F. Sherwin
(Canada-Newfoundland Offshore Petroleum
Board, St John's) for their assistance in making drill
core samples, well cuttings and thin sections avail­
able from the Hibernia and West Ben Nevis Fields;
D. Blair (Mobil, Calgary) and G. Karg (Dartmouth,
N.S.) for their help in drill core sampling; and to
Henning Dypvik for reviewing the manuscript.
REFERENCES
Asm, I. A. ( 1 988) Mineral diagenesis and porosity evolution
in the Hibernia Oil Field, Jurassic-Cretaceous Jeanne
D 'Arc rift graben, eastern Grand Banks of Newfound­
land, Canada. MSc thesis, McGill University, Mont­
real, Quebec, 280 pp.
Asm, I. A. ( 1 996) Mixed-layer illite/smectite diagenesis in
the rift and post-rift sediments ofthe Jeanne d'Arc Basin,
offshore Newfoundland, Canada. PhD thesis, Memorial
University, St John's, Newfoundland, 243 pp.
ALMON, W.R. & DAVIES, D.K. ( 1 979) Regional diagenetic
391
trends in the Lower Cretaceous Muddy Sandstone.
Powder River Basin. In: A spects of Diagenesis (Eds
Scholle, P.A. & Schluger, P.R.). Spec. Publ. Soc. Econ.
Paleont. Miner., Tulsa, 26, 379-400.
AMOCO CANADA PETROLEUM COMPANY LTD & IMPERIAL OIL
LTD ( 1 973) Regional geology of the Grand Banks. Bull.
Can. Petrol. Geol. , 2 1 , 479-503.
ARTHUR, K.R., COLE, D.R., HENDERSON, G.G.L. & KUSH­
NIR, D.W. ( 1 982) Geology of the Hibernia discovery. In:
The Deliberate Search for the Subtle Trap (Ed. Halbouty,
M.T.). Mem. Am. Ass. Petrol. Geol., Tulsa, 32. 1 8 1 -
1 95.
ASSERETO, R. & FOLK, R.L. ( 1 980) Diagenetic fabrics of
aragonite, calcite, and dolomite in an ancient peritidal­
spelean environment: Triassic Calcare Rosso, Lombar­
dia, Italy. J. sediment. Petrol. , 50, 3 7 1 -394.
BENTEAU, R.I. & SHEPPARD, M. G. ( 1 982) Hibernia-a
petrophysical and geological review. J. Can. Petrol.
Tech. , 2 1 , 59-72.
BJORKUM, P.A. & WALDERHAUG, 0. ( 1 990) Geometrical
arrangement of calcite cementation within shallow ma­
rine sandstones. Earth Sci. Rev. , 29, 1 45- 1 6 1 .
BJORLYKKE, K. ( 1 979) Discussion: Cementation of sand­
stones. J. sediment. Petrol. , 49, 1 35 8- 1 359.
BJORLYKKE, K., AAGAARD, P., DYPVIK, H., HASTINGS, D.S.
& HARPER, A.S. ( 1 986) Diagenesis and reservoir proper­
ties of Jurassic sandstones from the Haltenbanken area.
offshore mid-Norway. In: HabitaL of Hydrocarbons on
the Norwegian Continental She/f (Eds Spencer, A.M. e/
a/.), pp. 27 5-286. Graham & Trotman, London.
BLATT, H. ( 1 979) Diagenetic processes in sandstones. In:
A spects of Diagenesis (Eds Scholle, P.A. & Schluger.
P.R.). Spec. Publ. Soc. Econ. Paleont. Miner., Tulsa, 26.
1 4 1 - 1 57.
BLOCH, S., McGowEN, J.H. & DuNCAN, J.R. ( 1 990) Poros­
ity enhancement from chert dissolution beneath Neoco­
mian unconformity: Ivishak Formation, North Slope,
Alaska: Discussion. Bull. Am. Ass. Petrol. Geo/. , 74,
85-88.
BOLES, R.J. & FRANKS, S.G. ( 1 979) Clay diagenesis in
Wilcox Sandstones of southwest Texas: implications of
smectite diagenesis on sandstone cementation. J. sedi­
ment. Petrol. , 49 , 55-70.
BROWN, D.M., MCALPINE, K.D. & YOLE, R.W. ( 1 989)
Sedimentology and sandstone diagenesis of Hibernia
Formation in Hibernia Oil Field, Grand Banks of
Newfoundland. Bull. Am. Ass. Petrol. Geol. , 73, 55 7-
575.
BRUCE, C.H. ( 1 984) Smectite dehydration and its relation
to structural development and hydrocarbon accumula­
tion in Northern Gulf of Mexico Basin. Bull. Am. Ass.
Petrol. Geol. , 68, 673-683.
BuRST, J.F. ( 1 969) Diagenesis of Gulf Coast clayey sedi­
ments and its possible relation to petroleum migration.
Bull. Am. Ass. Petrol. Geol. , 53, 73-93.
CAROTHERS, W.W. & KHARAKA, Y.K. ( 1 978). Aliphatic
acid anions in oil-field waters: implication for origin of
natural gas. Bull. Am. Ass. Petrol. Geol. , 62, 244 1-245 3.
CORREIA, A., JONES, F.W. & FRICKER, A. ( 1 990) Terrestrial
heat flow density estimates for the Jeanne d'Arc Basin
offshore eastern Canada. Geophysics 59, 1 62 5-1 6 3 3 .
CREANEY, S . & ALLISON, B .H . ( 1 987) A n organic geochem­
ical model of oil generation in the Avalon/Flemish Pass
392 R. Hesse and I.A. Abid
sub-basins, East Coast Canada. Bull. Can. Petrol. Geol. ,
35, 1 2-23.
CuRTIS, C.D. ( 1 967) Diagenetic iron minerals in some
British Carboniferous sediments. Geochim. Cosmo­
chim. Acta 3 1 , 2 1 09-2 1 2 3 .
DYPVIK, H. ( 1 983) Clay mineral transformations i n Ter­
tiary and Mesozoic sediments from the North Sea. Bull.
Am. Ass. Petrol. Geol. , 67, 1 60- 1 65.
ENACHESCU, M.E. ( 1 987) Tectonic and structural frame­
work of the northeast Newfoundland continental mar­
gin. In: Sedimentary Basins and Basin-Forming Mecha­
nisms (Eds Beaumont, C. & Tankard, A.). Can. Soc.
Petrol. Geol. Mem., 1 2 , 1 1 7- 1 46 .
ENACHESCU, M.E. ( 1 988) Extended basement beneath the
intracratonic rifted basins of the Grand Banks of New­
foundland. Can. J. Explor. Geophys. , 24, 48-65.
ENACHESCU, M.E. ( 1 992) Basement extension on the New­
foundland continental margin (Canadian east coast). In:
Basement Tectonics 7 (Ed. Mason, R.), pp. 227-256.
Kluwer Academic, Dordrecht.
FEIN, J.B. ( 1 994) Porosity enhancement during clastic
diagenesis as a result of aqueous metal-carboxylate
complexation: experimental studies. Chern. Geol. , l i S,
263-279.
FOLK, R.L. ( 1 974) Petrology a/Sedimentary Rocks. Hemp­
hill, Austin, TX.
FOWLER, M.G. & BROOKS, P.W. ( 1 990) Organic geochem­
istry as an aid in the interpretation of the history of oil
migration into different reservoirs at the Hibernia K- 1 8
and Ben Nevis 1-45 wells, Jeanne d'Arc Basin, offshore
eastern Canada. Org. Geochem., 16, 46 1 -475.
FRANKS, S.G. & FORESTER, R.W. ( 1 984) Relationships
among secondary porosity, pore-fluid chemistry and
carbon dioxide, Texas Gulf Coast. In: Clastic Diagenesis
(Eds McDonald, D. A. & Surdam, R.). Mem. Am. Ass.
Petrol. Geol., Tulsa, 37, 63-80.
GRANT, A. C. & McALPINE, K.D. ( 1 990) The continental
margin around Newfoundland. In: Geology of the Con­
tinental Margin off Eastern Canada (Eds Keen, M.J. &
Williams, G.L.). Geol. Soc. Am. Geol. North Am. 1-1 ,
pp. 239-292.
GRANT, A.C., McALPINE, K.D. & WADE, J.A. ( 1 986) The
continental margin of eastern Canada-geological
framework and petroleum potential. In: Future Petro­
leum Provinces of the World (Ed. Halbouty, M.T.).
Mem. Am. Ass. Petrol. Geol. , Tulsa, 40, 1 77-205.
HouSEKNECHT, D.W. ( 1 98 7) Assessing the relative impor­
tance of compaction processes and cementation to
reduction of porosity in sandstones. Bull. Am. Ass.
Petrol. Geol. , 7 1 , 63 3-642.
HUTCHEON, 1., NAHNYBIDA, C.G. & KROUSE, H.R. ( 1 98 5 )
The geochemistry o f carbonate cements in the Avalon
sand, Grand Banks of Newfoundland. Miner. Mag. , 49,
457-467.
IMAM, M.B. & SHAW, H.F. ( 1 9 8 7) Diagenetic controls on
the reservoir properties of gas bearing Neogene Surma
Group sandstones in the Bengal Basin, Bangladesh.
Mar. Petrol. Geol. , 4, 1 03- 1 1 1 .
IRWIN, H., CoLEMAN, M. & CuRTIS, C.D. ( 1 977) Isotopic
evidence for source of diagenetic carbonates formed
during burial of organic-rich sediments. Nature, 269,
209-2 1 3.
JANSA, L.F. & WADE, J.A. ( 1 97 5 ) Geology of the continental
margin offNova Scotia and Newfoundland. Geol. Surv.
Canada, Open File Report 74-30 , 5 1 - 1 05.
KANTOROWICZ, J.D., BRYANT, J.D. & DAWANS, J.M. ( 1 987)
Controls on geometry and distribution of carbonate
cements in Jurassic sandstones: Bridport Sands, south­
ern England and the Viking Group, Troll Field, Norway.
In: Diagenesis of Sedimentary Sequences (Ed Marshall,
J.D.), pp. I 03- 1 1 8. Blackwell Scientific Publications,
Oxford.
KLITGORD, K.D. & SCHOUTEN, H. ( 1 986) Plate kinematics
of the central Atlantic. In: The Geology of North Amer­
ica (Eds Vogt, P.R. & Tucholke, B.E.). The Western
North Atlantic Region, Vol. M, pp. 3 5 1 -3 7 8 . Geological
Society of America, Boulder, CO.
LAND, L.S. ( 1 980) The isotopic and trace element
geochemistry of dolomite: the state of the art. In:
Concepts and Models of Dolomitization (Eds Zenger,
D.H. et a/. ). Spec. Pub!. Soc. Econ. Paleont. Miner.,
Tulsa, 28, 87- 1 1 0.
LINDQUIST, S.J. ( 1 977) Secondary porosity development
and subsequent reduction, overpressured Frio Forma­
tion sandstone (Oligocene), South Texas. Trans. Gulf
Coast Ass. Geol. Soc. 27, 99- 1 07 .
LONGSTAFFE, F.J. ( 1 984) The role o f meteoric water in
diagenesis of shallow sandstone: stable isotope studies
of the Milk River Aquifer and Gas Pool, southeastern
Alberta. In: Clastic Diagenesis (Eds McDonald, D. A. &
Surdam, R. C.). Mem. Am. Ass. Petrol. Geol. , Tulsa, 37 ,
8 1 -98.
LOUCKS, R.G., DODGE, M.M. & GALLOWAY, W.E. ( 1 984)
Regional controls on diagenesis and reservoir quality in
Lower Tertiary sandstones along the Texas Gulf Coast.
In: Clastic Diagenesis (Eds McDonald, D.A. & Surdam,
R.C.). Mem. Am. Ass. Petrol. Geol., Tulsa, 37, 1 5-45 .
LUNDEGARD, P.D. & KHARAKA, Y.K. ( 1 990) Geochemistry
of organic acids in subsurface waters. In: Chemical
Modeling ofAqueous Systems II (Eds Melchior, D.C. &
Bassett, R.L.). Am. Chern. Soc. Symp. Series 4 1 6,
1 69- 1 89.
LUNDEGARD, P.D., LAND, L.S. & GALLOWAY, W.E. ( 1 984).
Problem of secondary porosity: Frio Formation Texas
Gulf Coast. Geology, 1 2, 399-402.
MACHEL, H.-G. ( 1 987) Saddle dolomite as a by-product of
chemical compaction and thermochemical sulfate re­
duction. Geology, 15, 936-940.
MASSON, D.G. & MILES, P.R. ( 1 984) Mesozoic sea-floor
spreading between Iberia, Europe and North America.
Mar. Geol. , 56 , 279-287.
MASSON, D.G. & MILES, P.R. ( 1 986) Development and
hydrocarbon potential of Mesozoic sedimentary basins
around margins of North Atlantic. Bull. Am. Ass. Petrol.
Geol. , 70, 72 1 -729.
MATHISEN, M.E. ( 1 984) Diagenesis of Plio-Pleistocene
nonmarine sandstone, Cagayan Basin, Philippines:
early development of secondary porosity in volcanic
sandstone. In: Clastic Diagenesis (Eds McDonald, D.A.
& Surdam, R.C.). Mem. Am. Ass. Petrol. Geol., Tulsa,
37, 1 77 - 1 9 3 .
MAUFFRET, A . & MONTADERT, L . ( 1 987) Rift tectonics of
the passive continental margin off Galicia (Spain). Mar.
Petrol. Geol. , 4, 49-70.
MENELEY, R.A. ( 1 986) Oil and gas fields in the East Coast
and Arctic basins of Canada. In: Future Petroleum
 Sandstone reservoir levels in the Hibernia Oilfield
Provinces ofthe World (Ed. Halbouty, M.T.). Mem. Am.
Ass. Petrol. Geol., Tulsa, 40, 1 43- 1 76.
MILLIKEN, K.L. ( 1 994) The widespread occurrence of
healed microfractures in siliciclastic rocks: evidence
from scanned catholodoluminescence imaging. In: Rock
Mechanics: Models and Measurements-Challenges
from Industry (Eds Nelson, P.P. & Laubach, S.E.),
pp. 825-832. Balkema, Rotterdam.
MORAD, S., MARFIL, R., AL-ASAM, I.S. & GOMEZ-GRAZ, D.
( 1 992) The role of mixing-zone dolomitization in sand­
stone cementation: evidence from the Triassic Bunt­
sandstein, the Iberian Range, Spain. Sediment. Geo/. ,
80 , 5 3-65.
MORROW, D.W., CUMMING, G.L. & KOEPNICK, R.B. ( 1 986)
Manetoe facies-a gas bearing, megacrystalline, Devo­
nian dolomite, Yukon and Northwest Territories,
Canada. Bull. Am. Ass. Petrol. Geo/. , 70, 702-720.
NESBITT, H.W. ( 1 98 5 ) A chemical equilibrium model for
the Illinois Basin formation waters. Am. J. Sci. , 285,
436-458.
0LAUSSEN, S., DALLAND, A., GLOPPEN, T.G. & JOHANNES­
SEN, E. ( 1 984) Depositional environment and diagenesis
of Jurassic reservoir sandstones in the eastern part of
the Troms I area. In: Petroleum Geology of the Northern
European Margin (Ed. Spencer, A.M.), pp. 39-60. Gra­
ham & Trotman, London.
PEARSON, M.J., WATKINS, D., PITTION, J.-L., CASTON, D. &
SMALL, J.S. ( 1 98 3 ) Diagenesis, organic maturation and
paleothermal history of an area in the South Viking
Graben, North Sea. In: Petroleum Geochemistry and
Exploration of Europe (Ed. Brooks, J.). Spec. Pub!.
Geol. Soc. London, 1 2 , 1 6 1 - 1 74.
PERRY, E. A. & HOWER, J. ( 1 972) Late stage dehydration in
deeply buried pelitic sediments. Bull. Am. Ass. Petrol.
Geo/. , 56, 20 1 3-202 1 .
PowELL, T.G. ( 1 985) Paleogeographic implications for the
distribution of Upper Jurassic source beds, Offshore
Eastern Canada. Bull. Can. Petrol. Geo/. , 33, 1 1 6- 1 1 9.
RADKE, B.M. & MATHIS, R.L. ( 1 980) On the formation and
occurrence of saddle dolomite. J. sediment. Petrol. , 50 ,
1 1 49- 1 1 68.
SAIGAL, G.C. & BJ0RLYKKE, K. ( 1 987) Carbonate cements
in clastic reservoir rocks from offshore mid-Norway­
relationship between isotopic composition, textural de­
velopment, and burial depth. In: Diagenesis ofSedimen­
tary Sequences (Ed. Marshall, J.D.), pp. 3 1 3-324.
Blackwell Scientific Publications, Oxford.
ScHMIDT, V. & McDONALD, D.A. ( 1 979a) The role of
secondary porosity in the course of sandstone diagene­
sis. In: Aspects of Diagenesis (Eds Scholle, P.A. &
Schluger, P.R.). Spec. Pub!. Soc. Econ. Paleont. Miner.,
Tulsa, 26, 1 75-207.
ScHMIDT, V. & McDONALD, D.A. ( 1 979b) Texture and
recognition of secondary porosity in sandstones. In:
Aspects of Diagenesis (Eds Scholle, P.A. & Schluger,
P.R.). Spec. Pub!. Soc. Econ. Paleont. Miner., Tulsa, 26,
209-225.
SHANMUGAM, G. & HIGGINS, J.B. ( 1 988) Porosity enhance­
ment from chert dissolution beneath Neocomian un­
conformity: Ivishak Formation, North Slope, Alaska.
Bull. Am. Ass. Petrol. Geo/., 72, 523-5 3 5 .
393
SHERWIN, D.F. ( 1 973) Scotian Shelf and Grand Banks. In:
The Future Petroleum Provinces of Canada (Ed. Mc­
Crossan, R.G.). Mem. Can. Soc. Petrol. Geol., Tulsa, I ,
5 1 9-559.
SINCLAIR, I.K. ( 1 988) Evolution of Mesozoic-Cenozoic
sedimentary basins in the Grand Banks area of New­
foundland and comparison with Falvey's ( 1 974) rift
model Bull. Can. Petrol. Geo/. , 36, 255-273.
SINCLAIR, l.K. ( 1 993) Tectonism: the dominant factor in
mid-Cretaceous deposition in the Jeanne d'Arc Basin,
Grand Banks. Mar. Petrol. Geo/. , 10, 5 30-549.
SINCLAIR, l.K., McALPINE, K.D., SHERWIN, D.F. & Mc­
MILLAN, N.J. ( 1 992) Petroleum resources of the Jeanne
d'Arc Basin and environs, Grand Banks, Newfoundland.
Part 1: Geological framework. Geol. Surv. Canada
Paper, 92-8, 1 -3 8 .
SuRDAM, R.C., BoESE, S.W. & CROSSEY, L.J. ( 1 984) The
chemistry of secondary porosity. In: Clastic diagenesis
(Eds McDonald, D.A. & Surdam, R.C.). Mem. Am. Ass.
Petrol. Geol. , Tulsa, 37, 1 27- 1 5 1 .
SWIFT, J.H. & WILLIAMS, J.A. ( 1 980) Petroleum source
rocks, Grand Banks area. In: Facts and Principles of
World Petroleum Occurrence (Ed. Miall, A.D.). Can.
Soc. Petrol. Geol. Mem., 6, 1 67- 1 88 .
TANKARD, A.J. & WELSINK, H.J. ( 1 987) Extensional tecton­
ics and stratigraphy of Hibernia Oil Field, Grand Banks,
Newfoundland. Bull. Am. Ass. Petrol. Geo/. , 7 1 , 1 2 1 0-
1 232.
TANKARD, A.J. & WELSINK, H.J. ( 1 988) Extensional tecton­
ics, structural styles and stratigraphy of the Mesozoic
Grand Banks of Newfoundland. In: Triassic-Jurassic
Rifting: Continental Breakup and the Origin of the
Atlantic Ocean and Passive Margins, Part A (Ed.
Manspeizer, W.). Devel. Geotectonics 22, 1 29- 1 65 .
Elsevier, Amsterdam.
TANKARD, A.J. & WELSINK, H.J. ( 1 989) Mesozoic extension
and styles of basin formation in Atlantic Canada. In:
Extensional Tectonics and Stratigraphy of the North
Atlantic Ocean Margins (Eds Tankard, A.J. & Balkwill,
H.R.). Mem. Am. Ass. Petrol. Geol. , Tulsa, 46, 1 75-
1 95.
TAYLOR, G.C., BEST, M.E., CAMPBELL, G.R. et a/. ( 1 992)
Petroleum resources of the Jeanne d'Arc Basin and
environs, Grand Banks, Newfoundland. Part II: Hydro­
carbon Potential. Geol. Surv. Can. Paper 92-8, 39-48.
TISSOT, B.P. & WELTE, D.H. ( 1 984) Petroleum Formation
and Occurrence. Springer-Verlag, New York, 699 pp.
VEIZER, J. ( 1 983) Trace elements and isotopes in sedimen­
tary carbonates. In: Carbonates: Mineralogy and Chem­
istry (Ed. Reeder, R.J.). Rev. Miner., Miner. Soc. Am.,
1 1 , 265-299.
VON DER DICK, H., MELOCHE, J.D., DWYER, J. & GUNTHER,
P. ( 1 989) Source-rock geochemistry and hydrocarbon
generation in the Jeanne d'Arc Basin, Grand Banks,
Offshore Eastern Canada. J. Petrol. Geol. , 12, 5 1 -68.
WALDERHAUG, 0. & BJ0RKUM, P.A. ( 1 992) Effect of mete­
oric water flow on calcite cementation in the Middle
Jurassic Oseberg Formation, well 30/3-2, Veslefrikk
field, Norwegian North Sea. Mar. Petrol. Geo/. , 9,
308-3 1 8.
Spec. Pubis int. A ss. Sediment. (1998) 26, 395-408

The significance of ()13C of carbonate cements

in reservoir sandstones: a regional perspective
from the Jurassic of the northern North Sea

C.I. M A C AULAY*, A. E. FALLIC K*, O. M. McLAU GHLINt1 ,

R. S. H A SZEL D INEt a n d M.J. P E A R S ONt
*Isotope Geosciences Unit, Scottish Universities Research and Reactor C entre,
East Kilbride, Glasgow G75 OQF, Scotland, e-mail c.macaula}@surrc.gla.ac.uk;
tDepartment of Geology and Applied Geology, University of Glasgow, Glasgow G 12 8QQ, Scotland,
e-mail s.haszeldine@geology.gla.ac.uk; and
tDepartment of Geology and Petroleum Geology, University ofAberdeen, Aberdeen AB9 2UE, Scotland,
e-mail m.pearson@geol.abdn.ac.uk

ABSTRACT

Diagenetic carbonate minerals in Jurassic reservoir sandstones from 13 oilfields of the northern North
Sea record predominantly negative carbon isotopic (o13C PDB) compositions. o13C values range
from +15.8%o to -28.8%o (358 analyses), but show a strong mode at -9 to -l l%o. This observation
indicates a very significant contribution of isotopically negative carbon derived from organic material,
most likely the regional source rocks of the Kimmeridge Clay Formation. The strong mode in the o13C
data is significant because these diagenetic carbonate minerals could have grown in the reservoir
sandstones from mixtures of carbon from many sources as different fluids passed through the
sandstones during burial, and as hydrocarbons were generated during thermal maturation of the
adjacent organic-rich mudrocks.
Upper Jurassic marine sands enclosed within organic-rich mudstones contain diagenetic carbonates
with a much smaller range of o13C than that observed for Middle Jurassic deltaic sands, although both
data sets have strong modes around -IO%o. We suggest that this -IO%o mode in carbon isotopic
compositions may represent carbon derived from the decomposition of organic acids, rather than from
the mixing of two or more isotopically different sources. The wider range in o13C seen in carbonates
from the Middle Jurassic deltaic sands reflects the wider range in fluid and C02 sources available in
such settings.

INT R ODUCTION

The northern North Sea, with its organic-rich sedi­
ments, remains an area of significant economic
importance. As an oil province the area is now
mature, and having been studied geologically in
considerable detail, geochemical data exist in quan­
tities which allow regional perspectives to be ap­
proached. Over the past 10-15 years the diagenesis
of reservoir sandstones has been studied from many
individual North Sea oilfields. In each of these
studies the authors have observed a suite of diage-
1 Present address: Exxon Production Research Company,
PO Box 2189, Houston, TX 77252-2189, USA.
netic dissolution and precipitation events, and have
constructed models involving factors such as burial
temperatures, pore fluid compositions, petrogra­
phy, sedimentology and stratigraphy, organic and
inorganic geochemistry and isotope geochemistry to
explain the diagenetic sequences (see references in
Table 1 ). Isotope geochemistry has been increas­
ingly used to constrain mineral growth tempera­
tures, fluid origins and the origins of chemical
components in diagenetic minerals.
In this chapter we discuss the carbonate minerals
that have grown in Jurassic reservoir sandstones in
the northern North Sea, and in particular their

Carbonate Cementation in Sandstones Edited by Sadoon Morad 395

© 1998 The International Association of Sedimentologists
ISBN: 978-0-632-04777-2
396 C. I. Macaulay et al.

Table I. References to oilfield diagenesis studies from which data in Figs 3-8 were compiled (see also Abbotts. 1991)

Formation Bottom hole

Formation Depositional depth temperature
Oilfield References sampled environment Age (m) ('C)

Magnus Macaulay et a/. ( 1993) Kimmeridge Submarine U. Jurassic 2990 116

Emery et a/. (1993) Clay fan sst.
Formation
Murchison Prosser et a/. (1993) Brent Gp Deltaic M. Jurassic 2940 110
Thistle Brint (1989) Brent Gp Deltaic M. Jurassic 2745 104
Dunlin Brint (1989) Brent Gp Deltaic M. Jurassic 2715 100
Cormorant Brint (1989) Brent Gp Deltaic M. Jurassic 2745 96
Heather Glasmann et a/. (1989) Brent Gp Deltaic M. Jurassic 3200 110
Brent Hamilton et a/. (1987) Brent Gp Deltaic M. Jurassic 2590 96
Ninian Kantorowicz (1985) Brent Gp Deltaic M. Jurassic 2960 102
Emerald Osborne (1993) Emerald Sst Sheet sst M. Jurassic 1615 60
Veslefrikk Walderhaug & Bj0rkum (1992) Brent Gp Deltaic M. Jurassic 2900 125
South Alwyn Hogg (1989) Brent Gp Deltaic M. Jurassic 3690 127
Bruce McBride (1992) Brent Gp Deltaic M. Jurassic 3900 107
S. Brae McLaughlin ( 1992), Brae Sst Submarine U. Jurassic 3870 121
McLaughlin et a/. (1994) fan sst.

carbon isotopic compositions and what these might
imply about the origins of the carbon in these
carbonates. Volumetrically, diagenetic carbonate
minerals are one of the most significant cements
found in sandstone oilfield reservoirs, with major
implications for porosity and permeability.
Diagenetic carbonates documented from the Ju­
rassic in the northern North Sea are dominated by
calcite, but also include dolomite, ankerite and
siderite. The occurrence of diagenetic carbonates in
this region has been described as ranging from early
through, more commonly, to later diagenesis, incor­
porating a range in growth temperature estimates
from 40 to 14o·c (see references in Table 1).
The carbonate o 13C data we have used are taken
both from the literature and from unpublished PhD
theses (Table I). We have compiled all the available
carbonate o13 C data at our disposal to identify the
processes influencing carbonate precipitation on the
regional scale which might not be obvious on the
field scale. Many of the reservoir sandstones in the
northern North Sea share some similar characteris­
tics, such as source rocks in the Kimmeridge Clay
Formation and reservoir sandstones from the Mid­
dle Jurassic Brent Group, or from Upper Jurassic
submarine fan sandstones. Regionally extensive
organic-rich source rocks are an obvious source of
carbon, and in this synthesis we examine their
influence on the isotopic composition of diagenetic
carbonate minerals.
A larger-scale understanding of the diagenetic
processes affecting reservoirs in the North Sea
should contribute to the development of more
accurate diagenetic models, which might then be
applicable during the development of some less
mature oil provinces.
SOUR CES OF DATA
The data used in this synthesis have been compiled
from 13 oilfield studies of diagenesis in Jurassic
reservoir sandstones from the northern North Sea
(Fig. I). Of these studies, eight are published in the
literature (Kantorowicz, 1985; Hamilton et a/.,
1987; Glasmann et a!., 1989; Walderhaug & BjeJr­
kum, 1992; Emery et al., 1993; Macaulay et a/.,
1993; Prosser et al., 1993; McLaughlin et a!., 1994)
and five are unpublished PhD theses (Brint, I989;
Hogg, 1989; McBride, 1992; McLaughlin, 1992;
Osborne, 1993).
These studies combined contain a total of 35 8
carbon and oxygen isotope analyses (o 13 C and
8180) of diagenetic carbonate minerals from the
Jurassic of the northern North Sea.
Of the oilfields involved (Fig. I), the majority
have as reservoirs deltaic sandstones of the Middle
Jurassic Brent Group and equivalents, although the
furthest north and south, Magnus and South Brae,
have Upper Jurassic submarine fan sandstone res­
ervoirs. The general Jurassic stratigraphy of the
northern North Sea is shown in Fig. 2.
 Carbonate cements in reservoir sandstones 397

N 0'
Ryazanian
�
l
1
,
3�· 2
s I" Volgian
_]_QQ_Ign_ v
, Kimmeridge Clay Fm
�·

�g
.. . �
10 c. Kimmeridgian
11 0.
;:J
fll'

_gL-----1 Oxfordian
12 E
"

�
'il:o
l: Heather Fm
Callovian

r"

1. Magnus � Bathonian
:0
2.Murchisor. 8.Ninicm ""

3. Thistle 9.
�
Emerald Bajocian
u

4. Dunlin 10. Veslefrikk

·�
5. Cormorant 11. S. Alwy n 2, Aalenian

6. Heather 12. Bruce

7. Brent 13. Central & 5. Brne Toarcian

Fig. 1. Locations of the 13 oilfields from which

Pliensbachian
diagenetic carbonate carbon isotope data were available.
:;
�
0
..J
Sinemurian

REPRESENTATIVENESS OF Statfjord Fm
Hettangian

SAMPLIN G
The reservoir sandstones in this synthesis (Table l)
encompass a variety of differences in depositional
settings, detrital compositions and burial histories.
In each of the 13 field-scale studies (Table l ) a
model has been developed by the relevant author(s)
to explain the sequence of diagenetic events ob­
served. These diagenetic models, although contain­
ing differences in detail, are surprisingly similar in
the broad sequence of the timing and types of
quartz, clay and carbonate diagenetic cements pre­
cipitated in the sandstones they describe. It is worth
remarking that many features of these models, and
in particular petrographic textures, are rather
loosely constrained and depend to a large extent on
the interpretative preferences of the observer. Sum­
maries of Brent Group sandstone diagenetic se­
quences can be found in the volume edited by
Morton et al. ( 1992).
Despite the variety of geological differences in
these oilfields, however, the Jurassic sandstones of
the northern North Sea share common features,
such as organic-rich Upper Jurassic source rocks
and feldspar-rich reservoir sandstones, which may
help explain some of their diagenetic similarities.
These common threads also hint that large-scale
diagenetic processes, which might be obscure or
unrecognizable on the field scale, may become
�
�
Rhaetian
Cormorant Fm
Fig. 2. General Jurassic stratigraphy of the East Shetland
basin in the northern North Sea (simplified after Brown,
1986). Deposition of the Brae sands and conglomerates
in the South Viking Graben overlaps the depositional
timeframe of the Magnus sands.
apparent from a broader, more regional view.
The value of the interpretation of a regional
database such as this depends on the data collated
being representative. Using the available database,
which is unavoidably constrained to core samples,
we here attempt to model the majority of the
diagenetic carbonates developed in the Jurassic
sandstones of the northern North Sea in terms of
o13 C. Carbonate minerals with obvious effects on
reservoir porosity and permeability, such as the
calcite 'doggers' in some Brent Group fields, are
likely to be more extensively sampled than dissem­
inated cements. However, previous studies have
shown that it is possible to develop an understand­
ing of the overall diagenetic picture from limited
data sets. For example, the range and average of a
very large (222 samples) isotope data set were used
by Lundegard ( 1994) to describe the source of
carbon, and the fractionation processes affecting its
isotopic composition, incorporated into diagenetic
398 C.!. Macaulay et al.

20 �------,

tO 0

++

+ calcite

o dolomite

.6. ankerite

D siderite

-20
++ + -t
+

-304----,--,_:_---j Fig. 3. Plot of ot3C vs ot80 for all

-20 -!5 -tO
-5
the diagenetic carbonate data
81�0 ('!I.. ) PDB collected (n 358).
�

calcite in the Veslefrikk Field. The origin of diage­ 8180 for the different minerals calcite, dolomite,
netic ankerite in the Ninian Field was described by ankerite and siderite. Mineral separates were de­
Kantorowicz (1985) using only four isotopic analy­ rived by heavy liquid, magnetic and hand-picking
ses. We have gathered 358 data from the sources separation techniques. Carbon isotopic composi­
shown in Table 1, which include samples of calcite, tions for the data as a whole range in 813 C from
dolomite, ankerite and siderite. The regional data­ + 10.3o/oo to -28.8o/oo, with a mean of -9.3% PDB,
base we have collected here therefore offers the best and oxygen isotopic compositions range from 8180
opportunity of revealing regional diagenetic pro­ -16. 7o/oo to -2.2o/oo, with a mean of -9.Oo/oo PDB.
cesses in the northern North Sea, at least in sand­ The 813 C data for diagenetic carbonate minerals
stones of economic interest. in Jurassic reservoir sandstones in the northern
North Sea are shown in Fig. 4. The histogram is
plotted with 1o/oo divisions against frequency of
RESULTS OF COMPILATION occurrence. The analytical precision quoted by
most authors for their isotope data is typically
In Fig. 3 carbonate 813 C data are plotted against 0.2o/oo. The overall shape of Fig. 4 is not signifi-

60�------,

so +-----

40 +---
c

30 4---------·-··-....____.._______.._......--···-·---
0

Fig. 4. Histogram of o13C for all

-30 · 26 -22 -18 -14 -10 -6 -2 0 2 10 14 the carbonate data available
(n � 358) from Jurassic oilfields in
the northern North Sea.
 Carbon ate cemen ts in reservoir san dston es
Table 2. Summary of o13C data presented
o 1 3C (%o)
Mineral Max. Min. Mean
Calcite 5.6 -28.8 -10.6
Dolomite 1.3 -12.9 -7.0
Ankerite* 2.1 -13.6 -9.5
Siderite 10.3 -18.4 -9.7
All 10.3 -28.8 -9.3
*Ankerite data l 5.8, -13.6, -6.8 if very early
methanogenic ankerites from Emerald field are included.
cantly affected by using a smaller bin interval.
Figure 4 again displays the range in values of o 13C,
but highlights the fact that the great majority of the
data are distinctly negative, with a very strong
mode at -9 to I I%o.
The ranges and means in o 13 C of the different
carbonate minerals, are listed in Table 2 and plotted
respectively in Figs SA, 6A, 7A and 8A.
SOUR CES OF CARBON
The carbon isotopic composition (o 13 C) of diage­
netic carbonate minerals reflects the origin of the
carbon from which they are composed. In a rift­
setting sedimentary basin such as the North Sea
many possibilities exist for the source of carbon.
I Marine bicarbonate with o13 C around O%o (Keith
& Weber, 1964). Variations in this value through
geological time have been generally minor (Schid­
lowski, 1988).
2 Dissolved inorganic carbon in meteoric waters.
The o13 C of total dissolved carbon (TDC) in mete­
oric waters can range from positive to very negative,
and depends on a combination of factors, such as
C0 2 from organic carbon (o13 C -25%o), carbon
� across the region. Some of the res,ervoir rocks are
from dissolution of carbonates (o13C +2%o), at­
� marine deposits containing shell debris or evidence
2
mospheric C02 (o 13 C �- 7%o) and uptake of 1 C
into organic carbon during photosynthesis, which
leads to more positive o13 C TDC (Hoefs, 1987).
3 Dissolution of shell debris in sediments. Ana­
lysed unaltered shell material from marine sand­
stones in the northern North Sea has o 13 C values
close to O%o (e.g. Macaulay et a!. 1993).
. pretation. Using noble gas abundances and isotopes
4 Isotopically negative organic carbon (o13C -20 to
-30%o) from plant and animal remains in sedi­
ments. particularly mudrocks. Irwin et a!. ( 1977)
proposed a general model relating the o13C of C02
generated in mudstones at shallow burial depths to
bacterial processes. In their model, with increasing
399
depth bacterial oxidation, bacterial sulphate reduc­
tion and then bacterial fermentation produce C02
with o 13C values of -25%o, -25%o and + 15o/oo,
respectively. At greater depths and higher tempera­
tures abiotic reactions produce C02 with o13 C -10
to -25%o. This model has been used extensively in
the interpretation of diagenetic carbonate mineral
o13C data. Carbonate minerals with o13C values
intermediate to the model zone values have been
interpreted as forming from mixtures of C02 from
various of these zones, or mixtures of organic­
derived C02 and carbon from other sources, such as
marine bicarbonate. Kerogen, and petroleum gen­
erated from its maturation during deeper burial,
typically have o13C between -20 and -30o/oo. In a
review, Emery & Robinson (1993) quote textbook
isotopic compositions of low molecular weight or­
ganic acids derived from kerogen as typically -10 to
-20o/oo. Biodegradation of petroleum is a source of
C02 which can result in carbonate cements with a
wide range in o13C, from +14o/oo to -20%o (Dimi­
trakopolous & Muehlenbachs, 1987).
5 Magmatic C02 with o C
13 � 7%o (see Taylor,
1986; Hoefs, 1987).
Unequivocal identification of carbon sources in
diagenetic carbonates is difficult. Longstaffe (1989)
summarized the complexities involved and empha­
sized the need for a combination of analytical ap­
proaches. In stating that a dash of intuition is often
also required, Longstaffe (1989) offered a reminder
that many uncertainties still exist in the understand­
ing of isotope geochemistry and diagenesis. A wide
range of potential carbon sources exist in the north­
ern North Sea which could have contributed to car­
bonate cementation in Jurassic reservoir sandstones.
The Jurassic source-rock mudstones of the Kim­
meridge Clay Formation (KCF) are now buried to
depths of between 2 and 4 km (7000-13000 ft)
of its past presence, some are fluvial and some del­
taic. Many have experienced post-depositional uplift
and subaerial erosion, with the possibility of flushing
by meteoric waters.
Of the possible carbon sources listed above, only
magmatic carbon can be eliminated from our inter­
to assess the possibility of mantle-source volatiles in
North Sea hydrocarbon accumulations, Hooker et
a!. (1985) and Turner et a!. (1993) concluded that
deep-source contributions are negligible, even in
fields close to deep graben faults. This situation can
be contrasted with the sediments and basalts of the
400 C.!. Macaulay et al.
Norwegian Sea, where carbonates with low 8 13 C
values were tentatively attributed by Lawrence &
Taviani (1988) to oxidation of emanating CH4 of
mantle or deep-seated origin.
INTERPRETATION
General
A wide spread in both 813C and 8'80 is displayed
by the carbonate isotopic compositions shown
in Fig. 3. The shape of the data distribution is
roughly triangular, with a large range in 813C at
higher 8 180 values and a small range in 813 C at
lower 8180 values. In very general terms this distri­
bution pattern could be interpreted as the result of
mixing of carbon from several sources of variable
8 13 C at low temperatures (see Irwin et al., 1977) to
give a large spread, with increasing homogenization
of carbonate 813C values, perhaps involving car­
bonate dissolution/reprecipitation and recrystalli­
zation, with increasing temperature and related
lower 8180.
However, this first-order interpretation disguises
trends in the data that provide clues to processes
which may be influencing carbonate 813 C; also, in
sandstones where early and late carbonate cements
can be distinguished texturally, in CL and by
composition, the different generations still retain
their original stable and radiogenic isotopic compo­
sitions (e.g. McLaughlin et al., 1994). These clues to
larger-scale processes become more apparent in the
following section, when the data are examined in
the stratigraphically divided subsets of the Upper
and Middle Jurassic.
Carbon isotopic compositions
A striking point which can be drawn from the
histogram of all the carbonate carbon isotope data
compiled in Fig. 4 is that isotopically negative
carbon derived from organic material has contrib­
uted significantly to the majority of the diagenetic
carbonates represented. Less than 8% of the carbon­
ate cements could be interpreted as having wholly
marine bicarbonate components with 813C of
around Oo/oo, but over 80% have 813C of -5%o and
lower, and approximately 40% of the data have
813 C of between -9 and - I Io/oo.
On division of the data into Middle and Upper
Jurassic sandstones (Figs 5-8), a division which
reflects broad sedimentological differences, more
complex and geologically constrained interpreta­
tion becomes necessary. In general terms, the Mid­
dle Jurassic sandstones for which we have data are
deltaic deposits, and the Upper Jurassic are sub­
marine fan sandstones. These broad depositional
differences influence the availability of carbon from
different sources during carbonate mineral precipi­
tation.
The reservoirs of Middle Jurassic age are deltaic
sandstones of the Brent Group and equivalents. In
the East Shetland basin, the Brent Group is approx­
imately 300 m thick (Richards, 1992). The deltaic
sediments include coal and soil horizons, and have
been subjected to significant meteoric water incur­
sion. As a result, the potential range in carbonate
precipitation mechanisms during earlier diagenesis
is wide in these deltaic sediments (e.g. meteoric/
marine mixing zones, bioclastic-rich horizons), as is
the range in carbon isotopic compositions that
results during near-surface methane oxidation, bac­
terial fermentation and sulphate reduction (Irwin et
al., 1977). During burial, carbon sources from
thermal maturation of source-rock organic material
become more significant. The largest organic car­
bon reservoirs close to the Brent Group sandstones
are the mudstones of the underlying Lower Jurassic
Dunlin Group and the overlying Upper Jurassic
Kimmeridge Clay Formation (Fig. 2).
A wide spread in carbon isotopic compositions is
observed in calcite, dolomite, ankerite and siderite
in the Middle Jurassic sandstones (Figs 5B, 6B, 7B
and 8B), reflecting the various bacterial processes
operating during early diagenesis. Calcite ranges
from a small group of data up to +6o/oo (probably
derived from C02 of bacterial fermentation) to a
larger group close to Oo/oo (probably reflecting ma­
rine bicarbonate), down to some sparse methane­
related data between -20 and -?Oo/oo. The majority
lie in a bimodal distribution between -6 and -20o/oo,
with modes at -9 and -15o/oo (Fig. 5B). Lundegard
(1994) found an average 8 13C of -15. 7o/oo in analy­
ses of over 200 early calcites from the Veslefrikk
Field, which he interpreted as being strongly influ­
enced by C02 produced from bacterial sulphate
reduction of early biogenic methane. The -9o/oo
mode is similar to that seen in calcite data from
Upper Jurassic sandstones, and may reflect pro­
cesses of organic maturation (see submarine fan
section). Ankerite data mostly group between -4
and -14o/oo, again with a mode at -1Oo/oo (Fig. 7B),
but also include an unusual shallow burial group of
 Carbonate cements in reservoir sandstones 401

All calcite

20 +------
c

-30 -26 -22 -18 -14 -10 -6 -2 0 2 6 10 14

(A) 813C ('X,) PDB

14 �-
---,
Middle Jurassic

12 +----ll---ll----�-l

10 1··········•·"""'"""'"""'''""'""""''"""""'""""'11..---·

-26 -22 -18 - 14 -10 -6 -2 0 2 10 14

(B) 813C ('X,) PDB

20

Upper Jurassic

1 5

c
0
u 10
n
I

Fig. 5. Histograms of calcite

carbon isotopic compositions: (A)
Middle and Upper Jurassic
reservoir calcite data combined;
(B) Middle Jurassic reservoir 0
-3 0 -26 -22 - 18 - 14 -I 0 -6 -2 0 2 10 I4
calcite data; (C) Upper Jurassic
reservoir calcite data. (C) 813C ('X.. ) PDB
402 C.!. Macaulay et al.

10 ,-----,
All dolomite

c
0

o+;��rTT+�-r����
-30 - 26 -22 -18 -14 -10 -6 -2 0 2 10 14

(A) .SDC (X
' .. ) ron

Middle Jurassic

2 .

-30 ·26 -22 -18 -14 -10 -6 -2 0 2 10 14

(B) .SDC ('X.. ) ron

Upper Jurilssic

Fig. 6. Histograms of dolomite

0��-rrTT+���rT�,_
-30 -26 -22 -18 -14 -10
(C)
carbon isotopic compositions: (A)
Middle and Upper Jurassic
reservoir dolomite data combined;
-6 -2 0 2 (B) Middle Jurassic reservoir
10 14
dolomite data; (C) Upper Jurassic
.S13C ('X.. ) ron reservoir dolomite data.
 Carbonate cements in reservoir sandstones 403

20
,-------�
All ankerite

�..........................................................................-..·--·
0
1
0

-30 -26 -22 -18 -14 -1 0 .6 .2 0 2 10 14

1i13C ('X,.) PDB

(A)

Middle Jurassic

c
0

-3 0 -26 -22 -18 -14 -1 0 •6 .2 0 2 10 14

1i13C ('X.. ) PDB

(B)

2 0 �-------,

Upper Junssic

4 ·················································································-·····
0
10

Fig. 7. Histograms of ankerite

carbon isotopic compositions: (A)
Middle and Upper Jurassic
reservoir ankerite data combined;
o+;-rrrTT��rr+++;­
(B) Middle Jurassic reservoir •3 0 • 26 -2 2 . 18 -14 ·1 0 .6 .2 0 2 10 14

ankerite data; (C) Upper Jurassic

reservoir ankerite data. (C) 813C ('X.. ) PDB
404 C.!. Macaulay eta!.

All siderite

c
0
u
n
I

-30 ·26 -22 -18 14 . 10 -6 -2 10 14

1)13C ('X.. ) PDB
(A)

.6,-
----,
Middle Jurassic

0 -hH-t-t--t-t-rt-t-t-
·30 -26 -22 -18 -14 ·10 -6 -2 0 2 10 14
/j13C ('X.. ) PD1l
(B)

Upper Jurassic

c
0

Fig. 8. Histograms of siderite

carbon isotopic compositions: (A)
Middle and Upper Jurassic
reservoir siderite data combined;
o+;�H-t-t-TT�-t-t-t-�� (B) Middle Jurassic reservoir
-30 -2 6 -22 -18 -14 -10 -6 -2 0 2 10 14
siderite data; (C) Upper Jurassic
(C) li13C ('X.,) PDB reservoir siderite data.
 Carbonate cements in reservoir sandstones
positive o13 C ankerites from the Emerald Field
which were produced through bacterial fermenta­
tion reactions (Osborne, 1993). Siderite data in­
clude one soil-related value at +1Oo/oo, a sulphate­
reduction related group between -16 and -20o/oo,
and a largest group between -5 and -8o/oo (Fig. 8B).
The data in this largest group are difficult to assign
to any particular carbon source, and may indeed
reflect mixing. Even in a regional study it is worth
remembering that carbonate precipitation can also
be influenced by very localized chemical condi­
tions, as described by Boles & Johnson (1983) in
their study of the promotion of siderite growth by
raised pH between the layers of detrital micas.
Variable local influences may affect the validity of
stretching the interpretation of chemical data across
large distances. Dolomite in Middle Jurassic sand­
stones shows no clear trend, with a range in o 13C
from slightly positive values down to around -13o/oo
(Fig. 6B). A range such as this is compatible with
dolomite growth in mixing zones between marine
or basinal and meteoric waters in the deltaic Brent
Group sandstones.
In comparison, the submarine fan sandstones of
the Magnus and Brae Fields are much more localized
deposits, and are enclosed by their oil source rocks of
the Kimmeridge Clay Formation. Carbonate ce­
ments in the submarine fan sandstones have narrow
ranges of o13C. Calcite, dolomite, ankerite and sider­
ite (Figs 5C, 6C, 7C and 8C) all show strong modes at
between -9 and -12o/oo. Different growth tempera­
tures, spanning 40-14o·c overall, have been pro­
posed for the different carbonates using minus
cement porosities, diagenetic sequences and oxygen
isotope compositions (McLaughlin, 1992; Macaulay
et al., 1993; McLaughlin et al., 1994), yet all four
carbonate types have very similar o13C signatures
and much smaller ranges than are seen in the Middle
Jurassic sands. This o13 C similarity suggests that in
these Upper Jurassic sands, where organic-rich
mudrocks are in close contact with the sandstone
reservoir, the o13 C of diagenetic carbonates is
strongly controlled by the o13 C of organic com­
pounds released into the sandstones during organic
maturation with burial. Furthermore, either the
o13C of organic compounds released into the sand­
stones remains relatively constant over a range of
temperatures, or one particular group of compounds
of appropriate carbon isotopic composition must
decompose over a range of temperatures, and pro­
mote the formation of carbonate cements. The
chemistry of the resulting carbonate cements must
405
then be locally controlled by the amounts of Ca, Mg
and Fe available in the pore waters in the sand­
stone.
An alternative to the standard model of mixing of
carbon from organic and inorganic sources, and one
which allows for the provision of carbon with a
relatively constant o13C over a range of diagenetic
temperatures, is the hypothesis of carbon derivation
from the breakdown of organic acids. Of the many
aqueous organic species released into sandstone
pore waters from organic-rich mudstones, carboxy­
lic acid concentrations have been quantified in most
detail. Contrary to early reports of very low organic
acid concentrations in sedimentary basin forma­
tion waters below 8o·c (Carothers & Kharaka,
1980), the concentration of carboxylic acids in
sedimentary basin pore waters is now believed to be
independent of temperature (Shock, 1988). Shock
( 1988) points out that at diagenetic temperatures
decarboxylation reaction kinetics are very slow,
even on a geological timescale. Oxidation reactions
between organic acids and C02, however, are much
faster. An example of carbonate which may record
such a mechanism is the siderite with o13C -1Oo/oo in
banded iron formations reported by Walker (1984).
The banded iron formation sediments contained
pore waters with very low oxygen fugacity, making
diagenetic oxidation of organic carbon, as described
for Mesozoic sediments by Irwin et al. (1977),
unlikely. An alternative explanation for the light
carbon isotope compositions found in the siderite is
provided by Shock (1988), who suggests a metasta­
ble equilibrium between carboxylic acids, C02 and
siderite. Of course, no direct relationship between
the carbon isotopic composition of carboxylic acids
released from organic matter in Precambrian mud­
stones and Mesozoic mudstones is implied here.
A spatial relationship is observed between
organic-rich mudstones and carbonate cements in
some cases, for example in the Magnus Sandstone
Macaulay et al., 1993). In the Magnus Field, anker­
ite cements are found in sandstones only adjacent to
boundaries with the Kimmeridge Clay Formation
mudstones and in thin sandstones within the mud­
stones. The o13C of this ankerite is grouped very
tightly around its mean of -9.6o/oo, which must
strongly reflect the isotopic composition of the
carbon released from these mudstones during dia­
genesis. Siderite, however, is distributed throughout
the Magnus Sandstone. In most studied oilfield
sandstones, however, carbon of organic derivation
has been transported out from mudstones to be
406 C.!. Macaulay eta!.
incorporated in diagenetic carbonates throughout
sandstone units.
Oxygen isotopic compositions
The oxygen isotopic composition of the carbonates
ranges from -16.7 to 2.2%o PDB (Fig. 3), reflecting
carbonate growth over a range in temperatures and
from waters which, because of different origins such
as meteoric, marine and basinal, had different
oxygen isotopic compositions. Estimates of growth
temperatures range from about 4o·c for many of
the early calcites (e.g. McLaughlin eta!., 1994) to
around 140•c for late ankerites (Kantorowicz,
1985). The range in temperatures at which 13C­
depleted carbonates can grow in sandstones is
therefore wide, and not limited to low-temperature
carbonates formed close to the sea floor, such as
those described by Lundegard (1994). Pore water
composition estimates are o180 -7%o for Jurassic
meteoric water (Hamilton et a!., 1987), -1o/oo for
Jurassic sea water (Shackleton & Kennett, 1975),
and higher values for basinal waters that evolved
through water-rock interactions. Unravelling re­
gional trends from a data set with so many possible
combinations of variables is difficult, although in
individual fields trends have been observed. For
example, Lundegard (1994) found a tendency for
lower o180 to correlate with higher o13C in calcite­
cemented zones of Middle Jurassic sandstones in
the Veslefrikk Field in the Norwegian sector, in
contrast to the more usual trend of low o 180/low
o13 C correlation expected through increased or­
ganic carbon input with increasing temperature. In
explanation of this trend, Lundegard (1994) sug­
gested that the calcite had precipitated in a mixing
zone, with isotopically lower organic carbon associ­
ated not with a meteoric water (low o180) but with
an 180-enriched intrabasinal water. Glasmann eta!.
(1989) described a similar trend from the Middle
Jurassic Brent Group in the Heather Field, which
contains calcite cements isotopically similar to
those in Veslefrikk. Mixing of meteoric water with
fluids transported up faults from depth and organic
carbon from the Dunlin and Ness Formations was
proposed by Glasmann eta!. (1989) to explain the
carbon and oxygen isotopic trend in the Heather
Field.
In contrast to these localized studies, no singular
trend or correlation can be drawn from the northern
North Sea data. However, as discussed above,
across the wide range of o180 values observed a
striking majority of the carbonates have negative
o13 C values close to -10o/oo (Fig. 3).
DISCUSSION
Comparison with other sedimentary basins
There have been very few basin-scale studies of
diagenetic carbonate minerals in which o13C could
be related to particular carbon sources or C02
generation events. However, where a single domi­
nant carbon source has been identified for a partic­
ular carbonate precipitation event, the o13 C of that
carbonate phase has been shown to be consistent on
a large scale. Baker et a!. ( 1995) described consis­
tent o13 C values of -4.0 to +4.1%o for dawsonite
(NaAlC03(0Hh) throughout the Bowen-Gun­
nedah-Sydney basin system in Australia, where the
dawsonite is considered to have formed from mag­
matic C02 linked to widespread major igneous
activity.
In the thick Tertiary sandstones of the Texas Gulf
Coast oil province, carbon derived from organic
matter has had a minor influence on the carbon
isotopic composition of diagenetic carbonates. Cal­
cite in the Frio Formation averages o13C -4.1o/oo
and ankerite in the Wilcox Formation -6.3%o (Lun­
degard eta!., 1984). Dissolved inorganic carbon in
the Frio and Wilcox Formations has o13 C values
which are mostly greater than -1Oo/oo, and C02 from
natural gas produced from Frio Formation reser­
voirs has o13 C averaging -5.4%o (Lundegard eta!.,
1984). Lundegard & Land (1986) suggest that car­
bon sources such as skeletal carbonate with o13C
near Oo/oo must have dominated the isotopic compo­
sition of these carbonate cements, and that C02
may have migrated upwards from considerably
deeper in the basin and promote.d carbonate precip­
itation. Estimates, from oxygen isotope geother­
mometry, of growth temperatures of 80-1oo·c for
calcite and 95-120"C for ankerite (Lundegard &
Land, 1986) may be above the temperatures at
which significant kerogen decarboxylation occurred
in the Frio Formation (Lundegard et a!., 1984). In
fact, Fig. 9 of Land ( 1984) shows a clear trend from
shallow carbonates with a wide range in o13C from
-1 to -1Oo/oo to deeper carbonates with a very much
narrower range in o13C, at around -3o/oo. Overall, the
range in carbonate isotopic compositions seen in the
Texas Gulf Coast reservoirs is restricted compared
with those in the northern North Sea.
 Carbonate cements in reservoir sandstones
In perhaps the largest-scale previous study of
North Sea diagenetic carbonates, Lundegard (1994)
summarized that diagenetic calcite in the Middle
Jurassic Brent Group sandstones is similar in both
the Norwegian and UK sectors of the North Sea,
citing Veslefrikk Field and Heather Field (Glas­
mann et a!., 1989) as examples. A variable influence
of meteoric water is invoked to explain differences
in oxygen isotopic composition between the calcites
in the two fields, because calcite appears to have
formed at similar times in the diagenetic sequence
in both sandstones (Lundegard et a!., 1984). Carbon
isotopic compositions in both fields are highly
variable, from around 0 to -30o/oo.
In Upper Jurassic shelf sandstones in the Norwe­
gian sector, calcite doggers have been studied in
detail (Lundegard, 1994, and references therein).
The calcites are commonly associated with bioclas­
tic accumulations and consequently have relatively
heavy 813C values. As has been discussed earlier,
Upper Jurassic submarine fan sandstones contain
diagenetic carbonates which are distinctly different
from those in Middle Jurassic sandstones, in having
a narrow range in carbon isotopic composition. The
submarine fan sandstones contain very little bio­
clastic carbonate, and 13 C-depleted organic carbon
has been a major influence on diagenetic carbonate
compositions. This same strong organic influence is
also present, but is less obvious, among the wider
ranges in 813C from the Brent Group sandstones of
the Middle Jurassic. Local field-scale complexities
such as those above mask, in the Middle Jurassic
sandstones, the large-scale carbon mass-balance re­
quirement for a regional source of carbon with 813C
of -1O%o in Jurassic reservoir sandstones across the
northern North Sea.
CON CLUSIONS
The carbon isotopic compositions of diagenetic
carbonate minerals in Jurassic reservoir sandstones
of the northern North Sea vary from 813C -30
to +1O%o, but show a very strong mode around
-10%o.
The range in compositions is much narrower in
Upper Jurassic submarine fan sandstones than in
deltaic Middle Jurassic sandstones, although both
have major modes at 813 C -1Oo/oo. This difference is
due to the exposure of the Middle Jurassic deltaic
sediments to a wider range of carbon sources and
bacterial isotope-fractionation processes than the
407
submarine sands. Carbonates in the Upper Jurassic
submarine fan sandstones contain carbon derived
predominantly from organic matter in the mud­
stones of the Kimmeridge Clay Formation.
The mode in 813C at -1Oo/oo is particularly strong
for all the different types of carbonate mineral in the
Upper Jurassic sands enclosed by the Kimmeridge
Clay Formation. The Kimmeridge Clay Formation
is the major regional source rock in the northern
North Sea. This observation suggests that in this
region the breakdown of organic acids released
from mudstones as a result of kerogen maturation is
an important regional carbon source for carbonate
growth in reservoir sandstones during burial dia­
genesis. From a mass-balance perspective, the ma­
jority of the carbon derived by this process and
precipitated as carbonate cements has 813C close to
-10%o.
ACKNOWLEDGEMENTS
Past students are thanked for the use of data,
generated at SURRC, from unpublished PhD the­
ses. O.M.M. is grateful to Marathon Oil (UK) Ltd
and the Brae Group partners for funding and
support. The authors' interpretations presented
here are not necessarily those of Marathon Oil (UK)
Ltd and the Brae Group partners. Douglas Maclean
at Glasgow University processed the photographs.
Thanks are also due to Jorg Schulz-Rojahn and
Kitty Milliken for their constructive reviews. The
SURRC is supported by NERC and a consortium
of Scottish Universities.
REFEREN CES
ABBOTS, I.L. (1991) United Kingdom Oil and Gas Fields, 25
Years Commemorative Volume. Mein. Geol. Soc. Lon­
don, 14.
BAKER, J.C., Guo, P.B., HAMILTON, P.J., GoLDINos, S.D. &
KEENE, J.B. (1995) Continental-scale magmatic carbon
dioxide seepage recorded by Dawsonite in the Bowen­
Gunnedah-Sydney basin system, Eastern Australia. J.
sediment. Res., 65, 522-530.
BOLES, J.R. & JOHNSON, K.S. (1983) Influence of mica
surfaces on pore-water pH. Chern. Geol. , 43, 303-317.
BRINT, J.F. (1989) Isotope diagenesis and palaeofluid
movement: Middle Jurassic Brent Sandstones, North
Sea. PhD thesis, University of Strathclyde.
BROWN, S. (1986) Jurassic. In: Introduction to the Petro­
leum Geology of the North Sea (Ed. Glennie, K.W.),
pp. 133-160. Blackwell Scientific Publications, Oxford.
CAROTHERS, W.W. & KHARAKA, Y.K. (1980) Stable carbon
408 C.! Macaulay et a/.
isotopes of HC03- in oil-field waters-implication for
the origin of C02. Geochim. Cosmochim. Acta, 44,
323-332.
DIMITRAKOPOLOUS, R. & MUEHLENBACHS, K. (1987) Bio­
degradation of petroleum as a source of 1 3C-enriched
carbon dioxide in the formation of carbonate cement.
Chern. Geol. (IGS), 65, 283-29 1 .
EMERY, D. & ROBINSON, A. (1993) Inorganic Geochemistry:
Applications to Petroleum Geology. Blackwell Scientific
Publications, London, 254 pp.
EMERY, D., SMALLEY, P.C. & OxTOBY, N.H. (1993) Syn­
chronous oil migration and cementation in sandstone
reservoirs demonstrated by quantitative description of
diagenesis. Phil. Trans. Roy. Soc. Lond. , 344, 115-125.
GLASMANN, J.R., LUNDEGARD, P.D., CLARK, R.A., PENNY,
B.K. & COLLINS, J.D. ( 1 989) Geochemical evidence for
the history of diagenesis and fluid migration, Brent
Sandstone, Heather Field, North Sea. Clay Miner. , 24,
255-284.
HAMILTON, P.J., FALLICK, A.E., MACINTYRE, R.M. &
ELLIOT, S. ( 1 987) Isotopic tracing of the provenance and
diagenesis of Lower Brent Group Sandstones, North
Sea. In: Petroleum Geology of North West Europe (Eds
Brooks, J. & Glennie, K.W.), pp. 939-949. Graham &
Trotman, London.
HoEFS, J. (1987) Stable Isotope Geochemistry. Springer­
Verlag, Berlin, 24 1 pp.
HOGG, A.J.C. ( 1 989) Petrographic and isotopic constraints
on the diagenesis and reservoir properties of the Brent
Group Sandstones, Alwyn South, northern UK North
Sea. PhD thesis, University of Aberdeen.
HOOKER, P.J., O'NIONS, R.K. & OXBURGH, E.R. ( 1 985)
Helium isotopes in North Sea gas fields and the Rhine
rift. Nature, 318, 273-275.
IRWIN, H., CuRTIS, C. D. & CoLEMAN, M. ( 1 977) Isotopic
evidence for source of diagenetic carbonates formed
during burial of organic-rich sediments. Nature, 269,
209-2 1 3.
KANTOROWICZ, J.D. (1985) The origin of authigenic anker­
ite from the Ninian Field, UK North Sea. Nature, 3 1 5,
2 1 4-2 1 6.
KEITH, M.L. & WEBER, J.N. ( 1 964) Isotopic composition
and environmental classification of selected limestones
and fossils. Geochim. Cosmochim. Acta, 28, 1787-1816.
LAND, L.S. ( 1 984) Frio Sandstone diagenesis: a regional
isotopic study. In: Clastic Diagenesis (Eds McDonald,
D.A. & Surdam, R.C.). Mem. Am. Assoc. petrol. Geol.,
Tulsa, 37, 47-62.
LAWRENCE, J.R. & TAVIANI, M. (1988) Extreme hydrogen,
oxygen and carbon isotope anomalies in the pore waters
and carbonates of the sediments and basalts from the
Norwegian Sea: methane and hydrogen from the man­
tle? Geochim. Cosmochim. Acta, 52, 2077-2083.
LoNGSTAFFE, F.J. (1989) Stable isotopes as tracers in clastic
diagenesis. In: Burial Diagenesis (Ed. Hutcheon, I.E.).
Min. Assoc. Can. Short Course Series, 15, 201-277.
LUNDEGARD, P.D. (1994) Mixing zone origin of 1 3C­
depleted calcite cement: Oseberg Formation sandstones
(Middle Jurassic), Veslefrikk Field, Norway. Geochim.
Cosmochim. Acta, 58, 266 1 -2675.
LUNDEGARD, P.D. & LAND, L.S. (1986) Carbon dioxide
and organic acids: their role in porosity enhancement
and cementation, Paleogene of the Texas Gulf Coast. In:
Roles of Organic Matter in Sediment Diagenesis (Ed.
Gautier, D.L.). Spec. Pub!. Soc. econ. Paleont. Miner.,
Tulsa, 38, 129-146.
LUNDEGARD, P.D., LAND, L.S. & GALLOWAY, W.E. (1984)
Problem of secondary porosity: Frio Formation (Oli­
gocene), Texas Gulf Coast. Geology, 12, 399-402.
MACAULAY, C.!., HASZELDINE, R.S. & FALLICK, A.E. (1993)
Distribution, chemistry, isotopic composition and ori­
gin of diagenetic carbonates: Magnus Sandstone, North
Sea. J. sediment. Petrol. , 63, 3 3-43.
McBRIDE, J.J. (1992) The diagenesis of Middle Jurassic
reservoir sandstones of Bruce Field, UK North Sea. PhD
thesis, University of Aberdeen.
McLAUGHLIN, O.M. (1992) Isotopic and textural evidence
for diagenetic fluid mixing in the South Brae oil field,
North Sea. PhD thesis, University of Glasgow.
MCLAUGHLIN, O.M., HASZELDINE, R.S., FALLICK, A.E. &
RoGERS, G. (1994) The case of the missing clay, alumin­
ium loss and secondary porosity, South Brae oil field,
North Sea. Clay Miner. , 29, 651-664.
MORTON, A.C., HASZELDINE, R.S., GILES, M.R. & BROWN,
S. (1992) Geology of the Brent Group. Spec. Pub!. Geol.
Soc. London, 6 1 , 506 pp.
OsBORNE, M.J. ( 1993) The effect of differing hydrogeologi­
cal regimes on sandstone diagenesis: Brent Group oil
fields, North Sea. PhD thesis, University of Glasgow.
PROSSER, D.J., DAWS, J.A., FALLICK, A.E. & WILLIAMS,
B.P.J. (1993) The geochemistry and diagenesis of stra­
tabound calcite cement layers within the Rannoch
Formation of the Brent Group, Murchison Field, North
Viking Graben (northern North Sea). Sediment. Geol. ,
87, 139-1 64.
RICHARDS, P.C. (1992) An introduction to the Brent
Group: a literature review. In: Geology of the Brent
Group (Eds Morton, A. C., Haszeldine, R.S., Giles, M.R.
& Brown, S.). Spec. Pub!. Geol. Soc. London, 6 1 , 15-26.
SCHIDLOWSKJ, M. (1988) A 3,800-million-year isotopic
record of life from carbon in sedimentary rocks. Nature,
333, 313-3 1 8.
SHACKLETON, N.J. & KENNETT, .J.P. (1975) Paleotempera­
ture history of the Cenozoic and the initiation of
Antarctic glaciation: oxygen and carbon analyses in
DSDP sites 277, 279, 281. In: Initial Reports ofthe Deep
Sea Drilling Project (Eds Kennett, J.P. & Howtz, R.E.),
24, 743-755, US Government Printing Office, Wash­
ington, DC.
SHOCK, E. L. (1988) Organic acid metastability in sedimen­
tary basins. Geology, 16, 886-890.'
TAYLOR, B.E. ( 1 986) Magmatic volatiles: isotopic varia­
tion of C, H and S. In: Stable Isotopes in High Temper­
ature Geological Processes (Eds Valley, J.W., Taylor,
H.P. Jr. & O'Neil, J.R.). Rev. Mineral., Mineral. Soc.
Am., 16, 185-226.
TURNER, G., BURNARD, P., FORD, J.L. et a/. ( 1993) Tracing
fluid sources and interactions. Phil. Trans. Roy. Soc.
Land. , 344, 127-140.
WALDERHAUG, 0. & BJ0RKUM, P.A. (1992) Effect of mete­
oric water flow on calcite cementation in the Middle
Jurassic Oseberg Formation, well 30/3-2, Veslefrikk
Field, Norwegian North Sea. Mar. Petrol. Geol. , 9,
308-3 1 8.
WALKER, J.C.G. (1984) Suboxic diagenesis in banded iron
formations. Nature, 309, 340-342.
Spec. Pubis int. Ass. Sediment. (1998) 26, 409-435

Origin and significance of fracture-related dolomite in porous sandstones:

an example from the Carboniferous of County Antrim, Northern Ireland

R. E V A N S*, J . P. H E N D RY*, J. P A R N ELL* a n d R. M . K A L I Nt

*School of Geosciences, The Queen's University of Belfast, Belfast BT7 INN, UK,
e-mail richard.evans@qub.ac.uk; j. hendrJ@qub.ac.uk; j.parnell@qub.ac. uk; and
tDepartment of Civil Engineering, The Queen's University of Belfast,
Belfast BT7 INN, UK, e-mail r. kalin@qub.ac.uk

ABSTRACT

Dinantian fluvio-deltaic sandstones at Ballycastle, County Antrim, Northern Ireland, provide a record
of the palaeoflu id hydraulics in operation during the structural and diagenetic evolution of the area.
Dolomitized fractures cause a reduction in reservoir quality by structural compartmentalization, which
may be significant for subsurface analogues in the prospective Rathlin basin. An integrated field,
petrographic and stable isotopic study has elucidated the physical and diagenetic origin of the cemented
fractures and provides clues to their timing in the context of the regional tectonic evolution of northeast
Antrim. The dolomite is highly ferroan but with near-stoichiometric Ca2+ content (50-53 mol%
CaC03; 32-42 mol% MgC03; 5-18 mol% FeC03), and has oxygen and carbon isotopic compositions
of -3.8 to -0.9%o PDB, and -9.5 to -4 2%o PDB, respectively. Planar crystal fabrics and a preponderance
.

of monophase aqueous fluid inclusions indicate a relatively low cementation temperature and negate
the involvement of hydrothermal fluids.
The results of the study demonstrate that dolomite was precipitated during multiple episodes of
dilatational reactivation of cataclastic slip bands, in response to elevated pore fluid pressures associated
with tectonic enhancement of subsurface fluid flow. Such hydraulic fracturing has rarely been recognized
in porous sandstones, and a minor dextral shear component recorded in the fractures concurs with a late
Carboniferous origin. Dolomite was sourced either from compactional dewatering of basinal pro-delta
shales augmented by strain cycling, or from local mudrocks via transfer across active faults. An
alternative interpretation is of Tertiary fracturing and fluid input from venting of overpressured
mudrocks during catagenesis in the deep Rathlin basin. This would also fit the data, but is considered
less realistic from structural considerations. The tightly constrained orientation of cemented fractures
suggests that a reservoir sand body of similar nature would be strongly heterogeneous rather than
completely ineffective, and further integration of field and petrological data is required to assess the
regional importance of structural-diagenetic compartmentalization in potential reservoirs.

INTRODUCTION

Fracturing and fault compartmentalization of sand­
stones fundamentally affects reservoir properties
and may significantly influence the fluid migration
pathways in a basin (Knipe, 1993). Open fractures
may form high-permeability conduits, whereas
cement-sealed fractures form barriers to fluid flow.
Seismic, petrophysical and reservoir performance
data allow regional (field-scale) effects of faulting on
fluid flow to be constrained. However, much fractur­
ing and associated cementation may occur at sub-
seismic scale. Outcrop analogues provide the best
means of both describing and measuring such fea­
tures in three dimensions, and of understanding the
interrelationship of fracturing, fluid flow and cemen­
tation. Data from outcrop case studies can thus pro­
vide cost-effective empirical input for reservoir en­
gineers to simulate hydrocarbon production, from
reservoirs in which fracture-related cementation is
identified in core or cuttings.
This chapter presents results from an investiga-

Carbonate Cementation in Sandstones Edited by Sadoon Morad 409

© 1998 The International Association of Sedimentologists
ISBN: 978-0-632-04777-2
410
tion of fracture-related dolomite in Carboniferous
sandstones from the north Antrim coast, Northern
Ireland (Fig. 1 ). These sandstones were deposited
on the southern margins of the Rathlin basin, a
region of active exploration interest and a northern
continuation of the economically important Irish
Sea hydrocarbons province. Ferroan dolomite is the
principal cementing phase. The field relationships
and textural characteristics of dolomite-cemented
fractures will be presented, with emphasis on an·
outcrop at Ballycastle (Fig. 1 ). The timing of frac­
turing and cementation will be considered in the
context of the structural history of the Rathlin
basin. Dolomite stable isotopic data will also be
discussed in terms of potential fluid sources and
palaeohydrological interpretations. Finally, the im­
plications of the fracture-related dolomite in terms
of permeability heterogeneity and reservoir com­
partmentalization will be considered.
Terminology
Internally complex, ferroan dolomite-cemented
fractures at Ballycastle are closely associated with
regional normal faulting and are genetically linked
with cataclastic textures typical of brittle deforma­
tion in porous sandstones. 'Cemented fractures'
described in this paper are distinguished from the
principal slip planes ('faults') on the basis of mini­
mal displacement (centimetre scale at most), and
the definition encompasses the tectonodiagenetic
products of initial cataclasis, cementation of the
LEGEND
f?
'I!
- Fault
Y=
(for sectiOn X-Y see figure 2)
Northern Murlou�h Bay
South
Ireland
HBR= Highland Border Ridge
Londonderry
R. Evans et a!.
adjacent sandstone, and subsequent dilatation and
cementation episodes.
BASIN EVOLUTION AND
STRUCTURE
The Rathlin basin is a Carboniferous-Cretaceous
transtensional sedimentary basin that extends from
onshore Northern Ireland to offshore Scotland, and
is bounded by the NE-SW (Caledonian) trending
Tow Valley and Foyle Faults (Fig. 1 ). The basin-fill
comprises up to 2.2 km of Carboniferous and
Permo-Triassic siliciclastic sedimentary rocks,
which are overlain by a cover ( 300-500 m) of
Jurassic calcareous shales, Cretaceous pelagic lime­
stone and Tertiary basalt (Fig. 2). The basement
consists of Dalradian (Upper Precambrian) meta­
sediments. Carboniferous exposures of North Ant­
rim occur on wave-cut platforms and cliff sections
between Ballycastle and Murlough Bay (Fig. 1 ), on
the southeast margin of the basin. Deep seismic
reflection profiles, borehole information and re­
gional geology suggest as much as 1 000 m of Car­
boniferous strata immediately to the west, beneath
the southern part of the Rathlin basin (Evans et a!.,
1 980; McCann, 1 988). Carboniferous organic-rich
shales and coals within the study area have vitrinite
reflectance (R0) values between 0. 5 and 0.6 (Parnell,
1 99 1 ).
The Rathlin basin has undergone a complex
tectonosedimentary evolution, and this is reflected
STUDY AREA
Carboniferous
Fig. 1. Tectonic map of northeast
• Outcrop
Ireland-southwest Scotland, showing the
[] Basin
Rathlin basin and the onshore area
investigated near Ballycastle (arrowed).
Carboniferous outcrops in black.
Cross-section X-Y is shown in Fig. 2.
 Fracture-related dolomite cements in porous sandstones
1Dkm
1 km
X �west
Fig. 2. Simplified subsurface
LEGEND
section across the southern Rathlin
basin to the Ballycastle study area [2] Dalradian
(arrowed); see Fig. I for location.
Adapted from McCann ( 1 988).
in a four-stage subsidence history of the basin
margin in the Ballycastle area (Fig. 3) (Kerr, 1987;
Parnell, 1992; Anderson et al., 1995).
1 Rapid Carboniferous subsidence was followed by
Late Carboniferous-Permian uplift which was part
of the regional Late Variscan inversion event. Al­
though a dominant N-S to NNW-SSE compres­
sional stress field is recognized through NW Europe
at this time (Coward, 1995), Kerr (1987) demon­
strated evidence for local E-W compression in the
North Antrim region. This resulted in E-W strike­
slip faults and transtensional N-S reactivation of
I o
15.
""
0 1.0
" and thicker deposits accumulated in the basin
1.0
:[ 2.0
a Karner, 1987). In North Antrim, this caused re­
0
"
3.0
300 200 100
Time(Ma)
Fig. 3. Reconstructed burial histories for the
Carboniferous succession onshore (top) and in the
southern Rathlin basin (bottom). The dark line in the
top diagram represents the approximate position of the
studied sand body, assuming 400 m post-Carboniferous
erosion estimated from regional geology. Rathlin basin
data from Parnell ( 1 992). Burial temperatures at
positions A-C are discussed in the text.
411
East-> Y
D Carboniferous LJ Permo-Triassic 12::3 Lower Lias -Tertiary
(SL= Sea level)
older (Caledonian) NE-SW structures. A similar
stress regime operated in the Midland Valley of
Scotland (Mykura, 1967; Coward, 1990; Francis,
1991).
2 ENE-WSW extension in the late Permian and
Triassic was accommodated on NNW-SSE to N-S
normal faults and reactivation of existing struc­
tures. Contemporaneous faults of similar orienta­
tion are present in other Permo-Triassic basins in
the northwestern British Isles (Anderson et al.,
1995). Major subsidence of the Rathlin basin ac­
companied reactivation of the Tow Valley and
Foyle Faults. More gradual subsidence on the basin
margins permitted the deposition of a thin conti­
nental Permo-Triassic and transgressive marine Ju­
rassic sequence on the margin of the 'Highland
Border Ridge' Precambrian basement high (Fig. 1),
depocentres.
3 A change in the regional stress field to ESE­
WNW compression in the late Jurassic was proba­
bly related to plate rearrangements associated with
the onset of rifting in the Bay of Biscay and North
Atlantic margins (Dewey, 1982; Kerr, 1987; Lake &
newed uplift, and by the early Cretaceous all post­
Liassic deposits had been eroded from the study
area. Late Cretaceous eustatic sea level rise (Haq,
1987), coupled with renewed regional subsidence,
resulted in northward-directed transgression. Slow
deposition of pelagic carbonate produced approxi­
mately 50 m of chalk across the North Antrim
region (McCaffery & McCann, 1992).
4 Crustal doming associated with Latest Cretace­
ous-Paleocene rifting in the North Atlantic termi­
nated marine deposition in northeast Ireland.
Subsequently, Paleocene extrusive igneous activity
deposited at least 240 m of plateau basalts over the
412 R. Evans et al.
North Antrim region (McCann, 1988). This was
associated with several episodes of fault initiation
and reactivation of older trends in response to
general ENE-WSW extension, punctuated by a
period of NNE-SSW compression ( Kerr, 1987).
The northwest limit of the study area is defined
by the Caledonian Tow Valley Fault (TVF), which
was the major southeasterly bounding fault of the
Rathlin basin during the Mesozoic (Fig. 1 ). The
Great Gaw Fault (GGF), a probable splay of the
TVF, has a maximum normal throw of 400 m to the
west of Ballycastle (Wilson & Robbie, 1966), with
the displacement decreasing eastwards (about
200 m in the study area). Stratigraphical and struc­
tural relationships to the east of the study area
suggest that movement on the GGF was predomi­
nantly Late Carboniferous to pre-Triassic, with
some reactivation in the Late Jurassic-Early Creta­
ceous (Wilson & Robbie, 1966; Kerr, 1987). The
TVF has been reactivated several times during the
evolution of the basin, and has a complex
Caledonian-Late Tertiary history. Subsidiary faults
in the Ballycastle area are predominantly NNW­
SSE trending, with normal displacement and
decametre-scale downthrow to the west. The timing
of movement on these faults is difficult to ascertain
because preserved Mesozoic-Cenozoic cover is lim­
ited to the eastern margin of the Carboniferous
outcrop, within the southernmost Rathlin basin
(Fig. 2), and because of Tertiary reactivation.
Carboniferous strata in the offshore Rathlin basin
experienced a similar burial history to the onshore
succession until the late Permian. Thereafter, pro­
gressive and relatively rapid subsidence in the
Triassic was followed by slower subsidence in the
Early-Middle Jurassic, and a renewed period of
very rapid subsidence in the Paleogene (Fig. 3). In
terms of potential hydrocarbon generation, the
entire Carboniferous section reached maturity by
the end of the Paleogene, although hydrocarbon
generation may have begun in the Late Jurassic.
SEDIMENTOLOGY
Over 600 m of Namurian-Visean interbedded
sandstones, shales, coals and limestones are ex­
posed in cliffs and foreshore between Ballycastle
and Murlough Bay (Fig. 4). Sandstones dominate
the upper part of the succession at Ballycastle and
consist of stacked, 6-8 m thick, upward-fining,
fluvio-deltaic and nearshore-marine sandstones
with intervening shale beds. Some of these can be
laterally correlated to Murlough Bay, where the
lower part of the succession is also exposed. Here,
Asbian-Brigantian basaltic lavas overlie conglom­
erates, thin nodular limestones and limestone brec­
cias, which rest unconformably on top of Dalradian
metasediments (Wilson & Robbie, 1966). Four
prominent and laterally continuous marine shales/
limestones ('marine bands' of earlier workers) in the
Ballycastle-Murlough Bay outcrop contain an
abundant shallow-marine fauna and probably rep­
resent highstand deposits. The thickest 'marine
band' is the Main Limestone, which is a 1.1- 1.6 m
thick, sandy and partially dolomitized bioclastic
limestone (Fig. 4). Sandstones immediately under­
lying the 'marine bands' commonly display evi­
dence of tidal and/or estuarine processes.
At Ballycastle, multistorey fluviatile sand bodies
up to 30 m thick are partitioned by overbank
mudstones and siltstones into channelized packages
averaging 4-6 m in thickness. Individual channel
fills contain conglomerate and gravel lags and/or
siltstone rip-up clasts, and show abrupt or gradual
upward-fining successions. Five facies associations
can be recognized (Fig. 4) (Evans, 1995).
1 Marine shelffacies association. Laterally contin­
uous, dark grey to black fissile shales, with interbed­
ded siltstones and shales and infrequent thin
fossiliferous limestones.
2 Shallow-marine facies association. Erosionally
based, very fine- to medium-grained sandstones
which exhibit wave ripple, swaley and occasional
hummocky cross-stratification.
3 Tidally influenced/estuarine facies association.
Fine- to medium-grained sandstones with clay
drapes interbedded with bioturbated siltstones and
mudstones containing pyrite nodules.
4 Fluvio-deltaic facies association. Coarse- to
medium-grained, upward-fining sandstones with
gravel lags, trough cross-bedding, and sharp or
erosive bases. Palaeocurrents are oriented in a
SW-W direction.
5 Floodplain facies association. Fine- to medium­
grained sandstones interbedded with siltstones,
brackish-freshwater shales, seat-earths and coals.
METHODS
The fractured sandstones that are the subject of this
study occur at top of the Dinantian (Fig. 4), mostly
within stacked fluvial sand bodies exposed on the
 Fracture-related dolomite cements in porous sandstones 4 13

(A) STRATIGRAPHY (B) SEDIMENTARY LOG

z
Facies Associations

..: 4/5/1
a:
Bath lodge Coal Main Coal

::::>
3/2/1 50
Hawk's Nest Coal

400
:::;

}
McGildowney's Marine
..:
z
Band
4/5
=
5/4
Main Coal

3
2 LOG (Fig. 4 B)
"""'WAA<

1 40
Main Limestone

2 '
300
Common

4
Reddened
Sandstones

Carrickmore -r
z 30
Marine Band

..:
i=
z
..:
z 200
0

5/1
���stone 20

100
10

(/)
Basalt lava

Fig. 4. (A) Stratigraphy of �

�
� 5 Q)
Carboniferous rocks at Ballycastle, :::;
0
based on outcrop logging and data
from Wilson & Robbie ( 1966). ?
� 0
mu&si{lt� f

I
m c

D. I �
Sandstones continue for
approximately 200 m below the
base of the log but are poorly
� sand

gravel I
exposed. (B) Sedimentary log of
ss
� �����i���� n - c;8 Mu d-drapes
the study horizon labelled with LEGEND
sampling sites: ( I ) cemented
� Conglomerate s
fractures and host sandstone; (2)
{EJLimestone � If��M�it\g� ­ G ����lion
the Main Limestone; (3) sandstone D Sandston e
Q Lava D Shale � �r��s���;xt;���Y B Seatearths
of similar facies but lacking
dolomite-cemented f ractures; ( 4) [Z]Tuff -Coal � �!g������sts/ = Convoluted
L:o.J beddi ng
dolomite 'beef' vein. See text for
Lti hology Sedimen tary Structures
description of facies associations.

foreshore, from Pan's Rock to beyond Marconi's
Cottage (Fig. 5). A section of foreshore located
beneath Corrymeela (about 3 km east of Ballycastle
town) was selected for detailed study, on the basis of
3D outcrop quality, abundant cemented fractures,
and proximity to a NNW-SSE-trending normal
fault. This fault dips to the southeast and runs
beneath the low-water mark just at the western end
of the outcrop (Figs 5 and 6). In an adjacent inland
cliff exposure it juxtaposes McGildowney's marine
band against the shales of the Main Limestone,
giving a throw of about 60 m (Fig. 4). The distribu­
tion, dimensions, morphology and internal fabrics
of dolomite-cemented fractures were characterized
by mapping a I 00 m long foreshore transect away
from the fault (Fig. 6). Spatial relationships between
faulting and fracture distribution could thus be
recorded.
Hand specimens were collected from several ce­
mented fractures along the transect. Sawn and
polished surfaces were stained with acidified ali­
zarin red-S and potassium ferricyanide for carbon­
ate identification (Dickson, 1966). The petrography
of the fractures, including cataclastic slip bands,
cemented and uncemented host sandstones, were
examined in thin section. Several porous sand-
414 R. Evans et a!.
_ � _ Main Coal � Great Gaw Fault
Fig. 5. Location and geology of the study area, including dominant structural trends. Note distinct NNW-SSE to N-S
normal fault trend.
stones of the same facies but unaffected by ce­
mented fractures and located above the overlying
marine band were selected for comparison. Point
counting (300 counts per section) was carried out to
quantify the mineralogy and porosity of porous
sandstones, and to calculate intergranular volumes
(IGV) from the dolomite-cemented sandstone
within selected fractures (Table I ). The latter in­
volved counting along centimetre-scale transects
within regions unaffected by cataclasis or veining.
IGV values were used to distinguish shallow versus
deep burial cementation. Several porous sandstone
chips were examined with scanning electron mi­
croscopy (SEM), and polished thin sections of
dolomite-cemented samples were viewed with back­
scattered SEM (BSEM) and cathodoluminescence
(CL).
Dolomite samples of 50-250 mg were micro­
drilled from the polished stained slabs, care being
taken to discriminate between cementation epi­
sodes. Several samples of disseminated dolomite
were also extracted from porous sandstones adja­
cent to the pervasively cemented fractures. Sample
powders were digested in 1 00% H 3 P04 at 9o·c
using a Micromass MultiprepS (single acid reac­
tion) system, and purified C0 was analysed on a
2
Micromass-VG Prism III mass spectrometer. The
reproducibility of multiple NBS- 1 9 runs with the
Multi prep system is ± 0.05 for 81 3C and ± 0.09 for
8 1 80. All data are reported as o/oo deviation from
the PDB international standard, and are listed in
Table 2.
Scale 500 m
./\...../\...
./\...../\...
Atlantic Ocean
./\...../\...
./\...../\...
Faults
(1)
Marconi's
Cottage
Corrymeela
SANDSTONE PETROLOGY
AND DIAGENESIS
The host sandstones are medium- to coarse-grained
and moderately to well-sorted quartzarenites (Table
I ). Some thin lags of poorly-sorted, very coarse to
gravelly sandstone are also present. Surface weath­
ering has obliterated most of the sedimentary struc­
tures except where preserved by intergranular
dolomite cementation (see below). Detrital grains
are dominated by mono- and polycrystalline quartz,
subordinate lithic fragments, minor K-feldspars
and rare micas. Grains are mostly subangular to
sub-rounded and of moderate sphericity.
Other than the fracture-related cementation, dia­
genesis was dominated by moderate mechanical
compaction, partial feldspar dissolution, and minor
quartz overgrowth and clay mineral (chlorite and
kaolinite) authigenesis. Although grain packing is
relatively tight, no sutured grain contacts or grain
interpenetration is seen. Burial depths were proba­
bly insufficient for pressure dissolution. No wide­
spread carbonate cementation is seen, and detrital
quartz grains examined in SEM display no textural
evidence for corrosion related to former carbonate
cements. In contrast, quartz grains within dolomite­
cemented fractures have pronounced embayed and
notched margins (see Burley & Kantorowicz, 1986).
No differences in feldspar abundance were detected
between the dolomitized fractures and porous sand­
stones, although feldspars are slightly less corroded
where enclosed in dolomite. Consequently, the bulk
 Fracture-related dolomite cements in porous sandstones 415

A. c.

N = 125
x = 169°

Fracture orientations
100 metres

D.
20 20 25 25

20 20
15 15

Qi Qi 15 15
.0
..0
� 10 10 E
:::J
z
z 10 10

5 5
5 5

0 0 0
10 20 30 40 50 0 10 20 30 40 50 60 70 80 90 100 110
Length (m) Thickness (em)

Fig. 6. (A) Location of the mapped foreshore transect (see Fig. 5); the Main Limestone is exposed above in a 15-20 m
cliff section 17 5 m south of the foreshore. (B) Map of dolomite-cemented fractures on a 100 m x 20 m transect: some
fractures extend landward of the transect. (C) Summary of fracture orientations measured across the transect. (D)
Summary of fracture dimensions measured across the transect.

of intergranular porosity within the sandstones is uted the red coloration to deep oxidative weather­
interpreted as primary. This is significant because ing and the dissolution of a pervasive ferroan
Wang ( 1 992) examined reddened sandstones from dolomite cement during Late Carboniferous uplift
lower in the succession at Ballycastle, and attrib- and telogenesis. Wang proposed that intergranular
416 R. Evans et a!.

Table 1. Modal analysis data for cemented fractures and porous sandstones examined during this study: (a)
dolomite-cemented sandstones adjacent to fractures; (b) sandstones immediately surrounding the cemented fractures;
and (c) sandstones above the Main Limestone without cemented fractures. Quartz overgrowths (OG) could only be
clearly distinguished in one sample, although they have been observed more widely in SEM

(a) 2 3 4 5 6 7 8 9 10 Mean

Qtz mono. 49.3 41.3 54.3 57 52 47.6 47.6 52 56.6 51 50.9

Qtz poly. 3 1 3.3 5.6 7.6 5.6 1 4. 6 1 3.6 5.6 2.3 6.6 7.8
Feldspar 1 1.6 1.9 1.3 3.3 1.3 2.6 3.3 2 0 1.8
Mica 0 0 0 0 0 0 0 0 0 0 0
Lithic frag. I 0.6 0.3 5.3 7 6.3 3 7 3 2.3 3.6
Clays 0 I 1.3 0 0 0 0 0 13 1 0.6 2.6
Dolomite 44.3 38.8 33.3 25.6 24.9 25.6 3 1 .6 24.9 1 9. 3 27.6 2 9.6
Calcite 0 0 0 0 0 0.6 I 0 0 0.6 0.2
Qtz OG 0 0 2.3 0 0 0 0 0 0 0 0.2
Iron oxides I 1 .3 0 0 I 0 0.3 I 0 1 0.6
Organics 0.3 1 .6 0 0 0 2.3 0 0 0.3 0 0.5
Porosity 0 0.6 0 3 6 1.3 0 6 3.3 0 2.0
IGV 45.3 4 1 .7 36.9 28.6 3 1 .9 27.5 32.9 3 1. 9 35.6 3 9. 8 35.2

(b) 2 3 Mean (c) 2 3 4 Mean

Qtz mono. 41 .6 53. 1 59 5 1 .2 Qtz Mono. 53 49.4 63.3 65 57.7

Qtz poly. 6.3 4.5 6.3 5.7 Qtz poly. 5.6 11 . 3 4 3 6.0
Feldspar 5.6 2.2 1.9 3.2 Feldspar 0 2.2 1.3 1 .5 1 .3
Mica 0 0 0 0 Mica 1.6 0 0 0.5 0.5
Lithic frag. 1 .6 0.5 1.3 1.1 Lithic frag. 1 .6 0 0.3 0 0.5
Clays 8.3 0 0.6 3.0 Clays 2.6 6.8 16 8 8.4
Dolomite 5 6.8 1.3 4.4 Dolomite 6.6 0 0 0 1.7
Calcite 0 0 0 0 Calcite 0 0 0 0 0
Qtz OG 0 0 0 0 Qtz OG 0 0 0 0 0
Iron oxides o· 1.1 0 0.4 Iron oxides 1 0.5 0 0 0.4
Organics 0.6 0.5 0 0.4 Organics 0.3 2.8 0 5 2.0
Porosity 30.6 30.8 2 9. 3 30.2 Porosity 27.3 26.7 15 17 2 1 .5
IGV 43.9 38.7 31.2 37.9 IGV 37.5 34.0 3 1 .0 25.0 3 1 .9

Frag., fragments; IGV, intergranular volume; mono., monocrystalline; poly., polycrystalline; Qtz, quartz.

porosity in the reddened sandstones was mostly
secondary. The sand bodies in our study are not
reddened and we infer that they were never perva­
sively dolomite cemented.
DOLOMITE-CEMENTED FRACTURES
Morphology and field relationships
Dolomite-cemented fractures display a unimodel
NNW-SSE orientation and variable distribution,
with several zones of frequent, closely spaced exam­
ples (Fig. 6). Wave action has eroded the relatively
friable surrounding sandstones below the high-tide
mark, to leave the cemented fractures standing
proud (Fig. 7 A, B). Above the tidal limit the dolo-
mite is weathered in, but the features can be readily
traced across the outcrop. Cemented fractures con­
sist of extensive subvertical, planar to internally
meandering or conjugate sheets of well-lithified,
red-brown-weathering, yellowish sandstone. They
are surrounded by friable, buff-coloured sandstones
which show evidence of increased resistance to
erosion over several centimetres either side of thick
cemented fractures (Fig. 7A,F). The exhumed
dolomite-cemented fractures display highly irregu­
lar and digitate walls, resulting from antiaxial and
fabric-selective cementation of adjacent thickly
laminated sandstones. The margins of the cemented
fractures thereby form a cast of cross-bedding and
soft-sedimentary deformation features (Fig. 7B,C),
where the intervening friable sandstones have been
eroded away. In vertical sections through extant
 Fracture-related dolomite cements in porous sandstones 417

Table 2. Stable isotope data and sample descriptions

Stratigraphical
position
No. Sample Description (Fig. 4B) o18Q (%o PDB) o13C (%o PDB)
l A Intergranular dolomite - 1 .6 -6.2
2 Poikilotopic dolomite -1.7 -4.2
3 B Vein-fill dolomite -2. 9 -9.5
4 Intergranular dolomite -0. 9 -8.4
5 c Vein-fill dolomite -2.1 -8.0
6 Intergranular dolomite -3.1 -7. 9
7 D Vein-fill dolomite - 1 .8 -7.2
8 Intergranular dolomite -1.5 -6. 7
9 Poikilotopic dolomite -2. 1 -6.6
10 E Vein-fill dolomite -2.6 -8.7
ll Vein-fill dolomite -3.8 -6.8
12 Vein-fill dolomite -2.4 -7.2
13 Vein-fill dolomite -2.4 -8.3
14 Vein-fill dolomite -2.4 -8.3
-
15 Vein-fill dolomite 2 5. -8.3
16 Intergranular dolomite -2.7 -7.5
17 Intergranular dolomite - 1 .6 -7.7
18 Intergranular dolomite -2.8 -8.3
19 Intergranular dolomite -3.0 -9. 1
20 Intergranular dolomite -2. 9 -7.4
21 F Main limestone 2 -2.7 -0. 5
22 (gastropod shell - 1.8 -0. 1
23 pseudomorph) -3.7 -0.3
24 G Dolomite 'beef' vein 4 0.2 -8.6
25 Dolomite 'beef' vein -0.3 -9.4
26 Dolomite 'beef' vein -1.5 -10.1
27 Dolomite 'beef' vein -3. 1 -11.1

Mean Mean
Vein-fill dolomite -2.5 -8.0
Intergranular dolomite -2.2 -7.7
Poikilotopic dolomite -1.9 -5.4
Gastropod shell -2.7 -0.3
pseudomorph
Dolomite 'beef' vein - 1 .2 -9.8

sandstone beds, fracture-related cementation can
be seen to extend for up to 75 em laterally along
sedimentary laminae (Fig. 7E). More commonly,
the lateral 'relief' of the cemented fracture margins
is between I and I 0 em. Edwards et a!. ( 1 993)
documented analogous features where fracture­
related cements preferentially impregnated rela­
tively coarse-grained grain flow laminae of Permian
aeolian dunes in the Hopeman Sandstone Forma­
tion in northeast Scotland.
The majority of cemented fractures at Ballycastle
contain one or more medial veins of 0.5-3 mm
width, filled with a yellow-brown 'paste' of finely
crystalline dolomite. These features possess sharp
and parallel, albeit gently undu}atory, margins.
Anastomosing multiple veins are sometimes
present, particularly in relatively wide cemented
fractures. These may display high-angle branching
terminations or en echelon cross-linkages (e.g. Fig.
7F). The veins may be centrally disposed with re­
spect to the cemented fractures, but are more com­
monly closer to one or other lateral margin.
The overall width of the cemented fractures
(defined as the extent of complete dolomite cemen­
tation) ranges from 0.5 em to I m (Fig. 60), and
shows no relationship to position on the transect
(Fig. 8A). The thickest examples are termed com­
posite, consisting of two or more closely spaced,
Fig. 7. Dolomite-cemented fractures in the Ballycastle outcrop. (A) Exhumed, parallel-trending fractures exposed
below high-tide limit. Wave erosion has removed intervening friable sandstones except where partially cemented
immediately adjacent to the fractures. Field of view is approximately 4 m. (B) Cross-bedding (arrow) preserved as a
'cast' on the margins of an exhumed dolomite-cemented fracture; palaeocurrent towards the southeast. Field of view is
approximately 2 . 5 m. (C) Detail of convolute lamination picked out by dolomite cementation on the margins of an
exhumed fracture. Camera lens cap is 5 . 5 em wide. (D) Ribs of dolomite-cemented sandstone indicate preferential
pore fluid expulsion laterally from the fractures into the most permeable sedimentary laminae. Lens cap is 5 . 5 em
wide. (E) Vertical section through fractured sandstone bed exhibits dolomite cementation along depositional laminae
(arrow). This is an example of a composite fracture with en echelon cross-linkages. Field of view is about 1 . 5 m. (F)
Composite cemented fracture with cross-linkages in vertical plane. Note the resistance of the sandstone to erosion up
to l 0 em on either side of the tightly cemented zone. Lens cap is 5 . 5 em wide.
 Fracture-related dolomite cements in porous sandstones 419

(A boo-
E 250
�
en
�
:::> 200
t5
jg
0 150
en
en
Ql 100
c
.
<.>
-"'
:c
1- 50 ••

.. .
• ..
0
• •

0 20 40 60 80 100
Distance along transoct (m)

( 8) 7
•

6 •

I
en
� 5 . .
:::>
t5
4
jg
c
Ql 3
Ql
�
Ql
..
.0 2 . •

0>
•
..
c
·c:;
• •

"' ... ..
•
•
.
c. \
(/) I +
•
•• ,
•
• ..
Fig. 8. Scatter plots of cemented fracture 0
thickness (A) and spacing (B) with respect to 0 20 40 60 80 100
distance along the SW-NE along transect. Distance along transoct (m)

parallel or anastomosing cemented fractures (Fig.
7F). In addition, metre-scale zones of multiple
composite cemented fractures are present in two
parts of the transect (Fig. 6). A full geometrical
classification is given in Table 3 .
The majority o f cemented fractures trend NNW­
SSE, with a subordinate N-S to NNE-SSW set (Fig.
6). The mean strike direction of 349° is very close
to that of 347° for faults along the Ballycastle­
Murlough Bay outcrop (Fig. 5), although individual
fractures can display up to 50° local variation
around this mean direction along their lengths. The
principal NNW-SSE set are vertical to very steeply
dipping (70-90°), mostly westwards towards the
adjacent fault plane. The subordinate N-S to NNE­
SSW fractures are steeply inclined towards the west
(�65 ) and form conjugate or stepover fracture
o transect varies from 1 mm to 6.3 m, and appears
networks with the main set (e.g. Fig. 7B). Preserva­
tion of sedimentary structures by dolomite cemen­
tation reveals a lack of significant vertical
displacement across the fractures.
Many dolomite-cemented fractures can be fol-
lowed for > 1 0 m (up to 50 m in some cases), but
others taper into uncemented joint planes. Others
display straight, branching or forked terminations
and en echelon or stepped (continuous or discontin­
uous) linkages, with a minor dextral shear compo­
nent. 'Ramp' and 'eye' structures (see Antonellini &
Aydin, 1 995) are occasionally seen. The vertical
extent of individual fractures could not be estab­
lished, although the two zones of iptense fracturing
can be directly correlated with localized zones of
closely spaced cemented fractures and uncemented
joints in sandstone cliffs cropping out across a road
and vegetated cover from the foreshore (Fig. 6). On
this basis, these zones extend by at least 30 m
vertically and 1 7 5 m laterally.
The spacing of cemented fractures along the
relatively random. No statistical relationship be­
tween fracture width and spacing could be deter­
mined (Fig. 8B). However, one of the multiple
composite zones of intense fracturing is developed
at 0- 1 5 m, at the western end of the transect and
420 R. Evans et a!.

Table 3. Geometrical and size characteristics of cemented fractures on the mapped transect

Geometry Single Composite Width (em) Length (m) Field sketch

a. Straight j j 0.4-2 5-200

b. Forked j j 4.10 5 . 5- 1 6 �
c. Meandering j j 0.5- 1 0 1 -1 0 �
d. Anastomosing j 1-15 3- 1 5 �
e. Branching j 1 -2 5 1 . 1 5- 1 5 �
f. Zigzag j j 1 -20 5- 1 0

�
g. Offset j j 3-7 0.5-4
( 1 0 total)

h. En echelon j 1 2- 1 00
i. Conjugate j approx.
1 0-50
therefore close to the NNW-SSE-trending fault
plane. The second zone at 45-55 m does not appear
to be spatially associated with any significant slip
plane (Figs 6 and 8B).
Petrography and geochemistry
The cemented fractures are internally complex. On
polished surfaces and in thin section the region of
pervasive intergranular dolomite cementation is
seen to enclose a central sub-millimetre band of
cataclasis and (in many cases) an adjacent vein of
finely crystalline dolomite with very few detrital
grains (Figs 9A-D and 1 0). Intergranular dolomite
cement consists of a tightly interlocking mosaic of
equigranular, planar-s crystals (Gregg & Sibley,
1 984; Sibley & Gregg, 1 98 7), between 50 and
1 50 11m in size and with unit extinction. Intergran­
ular volumes range from about 29 to 45% (Table 1),
which is close to the porosity values from the
uncemented sandstones (bearing in mind the pe­
ripheral replacement of quartz clasts by dolomite).
The region ofintergranular dolomite cementation is
4-40
�
approx.
1 5-70 �
non-porous and relatively sharply defined (Fig. 1 0).
Immediately adj acent friable sandstone contains
only sporadic poikilotopic dolomite crystals, albeit
sufficient to give increased resistance to erosion
(Fig. 7A,F).
Dolomite cementation is developed around cata­
clastic slip bands (CSBs) (sensu Fowles & Burley,
1 994). In common with CSBs and granulation
seams described from other porous sandstones that
have undergone brittle deformation (Underhill &
Woodcock, 1 98 7; Edwards et ai., 1 993; Fowles &
Burley, 1 994), the detrital grains are both finer­
grained and more angular than those in the sur­
rounding sandstone. They may also display some
evidence of grain rotation towards a sub-vertical
alignment (Fig. 9A). A central seam of very tightly
packed angular quartz silt and sand fragments plus
Fe-oxides is sometimes present. There is no evi­
dence for crystal plastic deformation associated
with the CSBs, in the form of an increase in strained
or sub-grained clasts (e.g. Jamison & Steams, 1 982).
Sandstone surrounding the CSBs shows no evidence
of mechanical grain diminution.
 Fracture-related dolomite cements in porous sandstones 42 1

Fig. 9. Thin-section photomicrographs from dolomite-cemented fractures; all scale bars are 0.5 mm. (A)
Well-developed CSB (arrow), showing diminution and rotation of detrital quartz. Central dark seam is rich in Fe-oxide
and the entire field of view is dolomite cemented. (B) CSB (highlighted) and surrounding undeforrned sandstone are
both cemented by dolomite. (C) Very finely crystalline dolomite fills dilatational vein in the centre of a cemented
fracture; note the paucity of quartz clasts within the vein. (D) Complex, multiply banded and cross-cutting dolomite
vein fills. Note that the vein dilation initially reactivated a CSB, the sharp margin of the veining against the
dolomite-cemented wall-rock (black arrow) and the slightly coarser-grained dolomite cement in the final episode of
veining (e.g. white arrow). (E) Complex dolomite-cemented veins displaying high-angle branching and cross-cutting
relationships. In this example there appears to be more than one CSB. (F) Multistage dolomite-filled vein with
high-angle branch which fissures a large quartz clast. This implies that intergranular dolomite cement had fully lithified
the sandstone prior to vein formation. Note quartz silt from earlier cataclasis visible on the margin of the thick vein.
422 R. Evans et a/.
Cemented fracture
Cataclastic slip-band (CSB)
\

Quartz
grain

Porous
poorly
lithified
-10mm
sandstone

10mm

Fig. 10. Hand specimen,

photograph and simplified
drawings of a complex
dolomite-cemented fracture,
illustrating the adopted
terminology and the mutual
relationships of CSBs,
%
intergranular dolomite cement
(pale grey) and subsequent
dolomite-filled dilatational veins
(dark grey-black). The graph
0 5 10 15 20 shows the lateral variation of
Distance from central fracture (mm) point-counted intergranular
dolomite cement and porosity
em
Detrital Quartz Dolomite Porosity from the centre to the margin of
- ------ the fracture.

The intergranular carbonate cement is always
iron-rich dolomite (or ankerite). Electron probe
microanalysis gives between 7 and 18 mol% FeC03
(and 0.2-0.8 mol% MnC03) (Fig. 1 1). The calcium
content of the dolomites is relatively stoichiomet­
ric, with an average of 5 1 mol% CaC03 (range
50.0-5 3 . 3 mol%). The cement is generally dull to
very dull red-brown in CL, with subtle concentric
zonation. It is relatively homogeneous when exam­
ined in BSEM, suggesting that intracrystalline
chemistry is reasonably consistent within and be­
tween individual fractures. The dolomite crystals
have a variable fluid inclusion density, with a vast
preponderance of monophase aqueous inclusions.
Any two-phase inclusions identified in thin section
possess very small vapour bubbles. Neither host
dolomite nor fluid inclusions qisplay any fluores­
cence in ultraviolet light. Dolomite crystals on the
margins of some exhumed cemented fractures have
been partially calcitized, in some cases along intra­
crystalline concentric zones. The resultant porosity
in these areas contains Fe-oxides, which probably,
account for a prominent rusty red-brown surface
colour observed in outcrop.
The finely crystalline dolomite filling dilatational
central veins is slightly less ferroan than the sur­
rounding intergranular spar (5-9 mol% FeC03)
(Fig. 1 1), and dull red to red-brown luminescent.
Crystals are mostly less than 1 0 Jlm in size and
predominantly planar-s. Multiple-stage vein fills are
 Fracture-related dolomite cements in porous sandstones 423

MgC03 0180
-4.0 -3.0 -2.0 -1.0 0 1.0
� - X
�vx--- -r-----�:x----�----_,--�
--

• Vein-fill
c lntergranular
-2.0
& Poikilotopic
x Gastropod -4.0
eBeef

...
-6.0
•
oo • .�
D
D
D
•
D * • -8.0
. •
�
20 • - .
100
•

C aC03 _______ 5 0______-.FeC03+MnC03

_

.-
A lntergranular dolomite
(edge of cemented fracture) Fig. 12. cross-plot of stable isotope data from the
e lntergranular dolomite
dolomite cement types discussed in the text.
(adjacent to CSB)
Gastropod-fill and 'beef' samples are not from the
fractured sand body (see Fig. 4).
0 Dolomite in dilatational veins

Fig. 11. Ternary plot of ferroan dolomite compositions

measured by electron probe microanalysis.
symmetrically banded, although these bands may
exhibit complex cross-cutting relationships (e.g.
Fig. 9D-F). They also dissect relatively large detri­
tal grains (Fig. 9E,F). The final stage of multiple
vein fills is usually coarser than earlier stages or than
single-generation vein fills (e.g. Fig. 9D). Banded
finely crystalline dolomite is also occasionally over­
printed by coarser (50-200 Jlm) planar-e rhombs
with subtle CL zonation. Where discernible, all
crystal types have unit extinction. The dilatational
veins are commonly juxtaposed with pre-existing
CSBs (Fig. 9D,E), suggesting that (despite inter­
granular dolomite cementation) the latter repre­
sented planes of relative weakness during extension.
There appears to have been no cataclasis of the
intergranular dolomite, although it would be dif­
ficult to distinguish from the finely crystalline vein­
fill cement.
Stable isotope analysis of intergranular, vein-fill
and poikilotopic dolomite cements produced very
similar results (Fig. 1 2). Overall, dolomite o 1 8 0
values fall between -3.8 and 0.9%o PDB and o 1 3 C
-
Main Limestone gave similar o 1 80 values to those
of the cemented fractures (- 1 .8 to -3.7%o PDB) but
significantly more positive o 1 3C values (-0. 1 to
-0.5%o PDB). A fibrous dolomite ('beef') vein in a
mudrock exposed higher in the succession (Fig. 4)
yielded a range of o 1 3C (-8.6 to - 1 1 . 1 %o PDB) and
o 1 8 0 ( +0.2 to -3. 1 %o PDB) values; in both cases the
values become progressively lighter from the centre
to the edge of the vein.
DISCUSSION
Several important factors need to be considered in
appraising the origin and significance of the ce­
mented fractures at Ballycastle:
1 the physical and diagenetic mechanisms recorded
in the cemented fracture fabrics; .
2 the palaeohydrology and controls. on cement pre-
cipitation; ,
3 the origin of the pore fluids and ,cement compo­
nents;
4 the timing of the fracturing and cementation in
relation to regional tectonics.

values between -4.2 and -9. 5%o PDB. The bulk of

the data display a crude positive covariance, albeit
with several outlying points. However, no con­
sistent trends were detected across individual
cemented fractures. Three samples of ferroan dolo­
mite pseudomorphing gastropods in the overlying
Genesis of d olomite-cemented fractures
The consistent orientation of cemented fractures in
outcrop, and its coincidence with the modal fault
trend between Ballycastle and Murlough Bay (Figs 5
and 6), suggests that the cemented fractures formed
424 R. Evans et a!.
in response to a tectonic rather than synsedimen­
tary event. This is supported by the extrapolation of
individual dolomite-cemented fractures to unce­
mented joint planes on the foreshore and inland
cliff section. Furthermore, the lack of displacement
across conjugate fracture networks (NNW-SSE/
NE-SW), and their identical cementation fabrics,
suggests structural concurrence.
Sandstones that host the dolomite-cemented frac­
tures at Ballycastle are highly porous and friable.
Published studies of brittle deformation in
Palaeozoic-Tertiary-age porous sandstones docu­
ment a common development of CSBs with charac­
teristic microfabrics involving grain fracturing,
diminution and rotation. Most authors agree that
the CSBs are sites of severe porosity and permeabil­
ity loss, owing to reduction in grain size and sorting
(e.g. Pittman, 1 98 1 ; Underhill & Woodcock, 1 987;
Edwards et a!., 1 993; Hippler, 1 993; Fowles &
Burley, 1 994; Antonellini & Aydin, 1 995). The
petrophysical characteristics of CSBs suggest that
they should be effective barriers to regional fluid
flow. However, the digitate external morphology of
the cemented fractures at Ballycastle unequivocally
demonstrates that fluid flowed outward from a
central conduit, displacing the in situ pore fluids
from the immediately adjacent sandstone. This
indicates that the pore pressure was higher in the
fracture conduits than in host porous sandstone.
Furthermore, because of the relatively low solubil­
ity of dolomite (e.g. Tucker & Wright, 1 990) it is
likely that a high fluid flux was necessary to provide
sufficient pore volumes of fluid for the observed
cementation. The concentration of dolomite ce­
mentation around CSBs therefore suggests the
possibility of a genetic link between brittle deforma­
tion, fluid flow and cementation.
Several authors have suggested that the mechan­
ics of cataclasis involve initial dilatancy, followed
by grain crushing and compaction. Underhill &
Woodcock ( 1 987) proposed that pulsed fluid flow
may result from high pore pressures during the
permeability collapse events. Other workers favour
the hypothesis that initial dilatant episodes may
localize fluid flow, thereby promoting mineral au­
thigenesis or dissolution (e.g. Edwards et a!., 1 993;
Hippler, 1 993; Antonellini et a!. , 1 994). Edwards et
a!. ( 1 99 3) described silica and fluorite cementation
of sandstones adjacent to CSBs. However, they saw
no evidence that dilation and cataclasis episodes
were mutually related, proposing instead that fluid
flow accompanied dilatational reactivation on the
margins of the CSBs some time after their forma­
tion. In contrast, Fowles & Burley ( 1 994) specifi­
cally recorded quartz cementation and leaching of
Fe-oxides within CSBs in the Permian Penrith
Sandstone Formation from northwest England, and
inferred that fluid flow was focused in the incipient
CSBs during regional faulting. They also recorded a
remnant millimetre-centimetre wide zone of en­
hanced porosity and permeability adjacent to CSBs,
which they suggest may have formed in response to
the increased pore fluid pressures during subse­
quent compaction and grain disintegration. Mozley
& Goodwin ( 1 995) described preferential calcite
cementation associated with CSBs in a Cainozoic
extensional fault zone, but did not discuss the
implications for permeability evolution during de­
formation. Nevertheless, there is a striking similar­
ity with the situation at Ballycastle, except that the
subvertical cemented zones described by Mozley &
Goodwin lacked the lateral 'fingering' morphology
which we describe.
Cataclastic slip bands are formed during single
slip events, with subsequent deformation accom­
modated either incrementaily on new, parallel and
adjacent CSBs, or on slip planes (i.e. faults) (e.g.
Aydin & Johnson, 1 983; Antonellini et a!., 1 994;
Fowles & Burley, 1 994). This can be explained in
terms of strain hardening, whereby the cohesive
shear strength of CSBs is greater than that of the
adjacent undeformed sandstones (Aydin &
Johnson, 1 983; Underhill & Woodcock, 1 987). It
arises because of the close packing and grain break­
age into angular fragments, which increases friction
between grains. The production of new CSBs versus
slip planes (which represent a loss of shear strength)
may be controlled by relative strain rates (Morrow
& Byerlee, 1 989); consequently the two might not
be strictly coeval, despite having formed as part of a
deformation sequence in the S(\me tectonic stress
field. The composite and multicomposite sets of
cemented fractures at Ballycastle may be analogous
to cataclastic slip zones and, as discussed below, the
initial dolomite cementation may pre-date the slip
episode on the associated fault. This could explain
why the frequency of cemented fractures does not
steadily increase towards the fault at the western
end of the measured transect (Figs 6 and 8) (see
Fowles & Burley, 1 994).
The strain hardening phenomenon also means
that any extensional reactivation of CSBs tends to
occur by dilatational opening of the margins to
form potential fluid flow conduits and sites for the
 Fracture-related dolomite cements in porous sandstones
precipitation of vein cements. Edwards et al. ( 1 993)
suggested that multiple episodes of reactivation and
fluid flow along former CSBs could be linked to a
seismic pumping mechanism (Sibson, 1 98 1 ). In
their study, Edwards et a!. ( 1 993) recorded a tran­
sition from pervasive fluorite cement adjacent to
the CSBs to dispersed poikilotopic crystals in adja­
cent sandstones, a similar pattern to that observed
with dolomite cements at Ballycastle. They also
found examples of cementation restricted to one
side of a fracture, and not within the CSB itself.
This differs from the situation at Ballycastle, al­
though asymmetric cementation around CSBs is
observed.
The characteristics of dolomite-filled dilatational
veins within the cemented fractures at Ballycastle
have further implications for the relationship be­
tween deformation and fluid flow. Similarity in
isotopic and chemical composition between inter­
granular and vein-fill dolomite implies near­
contemporaneous precipitation (see Figs 1 1 and
1 3). Following the hypothesis of Edwards et a!.
( 1 99 3 ), rheological contrast between the CSBs and
surrounding friable sandstone would have favoured
vein opening immediately adjacent to the CSB
under renewed tensile stress. The Ballycastle frac­
tures show a close j uxtaposition of the dilatational
veins and the CSBs (Fig. 9D,E). As long as the CSB
retained some permeability at this time, dolomite­
supersaturated fluids focused in the dilatant zone
would have dissipated into the surrounding sand­
stone, forming the intergranular cement. The offset
of some dilatational veins from the centre of ce­
mented fractures then indicates a lateral permeabil­
ity heterogeneity resulting from the presence of the
CSB on one side of the vein. These veins would
have remained planes of relative weakness in the
cemented fractures, and would therefore have been
prone to tectonic reactivation by a stress transfer
mechanism. Because the vein walls would have
been tightly cemented at this stage, dilation re­
placed cataclasis (Antonellini et al., 1 994), detrital
quartz clasts were dissected rather than circumnav­
igated, and cementation took the form of a passive
infill (Fig. 9C-F).
In contrast, if the intergranular dolomite precip­
itated during pore fluid flow associated with catacla­
sis (Hippler, 1 993; Fowles & Burley, 1 994), it
becomes difficult to account for the medial position
of the subsequent veins. The greatest rheological
contrast would be expected at the margins of the
dolomite cementation rather than within the ce-
425
mented CSB. Another argument in favour of post­
cataclasis cementation is the time-scale required.
Complete calcite cementation of sandstone under
constant pressure and temperature requires
through-flow of more than 3 x 1 06 equivalent pore
volumes of supersaturated fluid (Berner, 1 980;
Bj0rkum & Walderhaug, 1 990), and the value is
likely to be higher for dolomite because of kinetic
inhibition of precipitation. The average volume of a
single cemented fracture at Ballycastle is approxi­
mately 1 m 3 , and the average dolomite cement
proportion is about 30% (Table 1 ). Cementation
would therefore require more than 9 x 1 05 m3 of
dolomite-supersaturated fluid to have discharged
through the exposed fracture, and considerably
more through the entire fault zone, although this
does not account for the effects of C02 degassing on
decreasing dolomite solubility. Assuming that indi­
vidual brittle deformation events are short-lived
(e.g. a similar time-scale to earthquakes), a truly
vast and geologically instantaneous fluid flux would
be required to precipitate the cement. In contrast,
vein dilation can be a repetitive event depending on
the regional tectonic stress regime. Fluid flow could
thereby be more prolonged and episodic, the latter
being compatible with the observation of concentric
zonation in the dolomite under CL.
Pore f lu id pressu re and contro ls o n
c em ent ation
The external morphology of dolomite-cemented
fractures at Ballycastle implies that the cementing
pore fluids migrated from the central vein into
surrounding sandstones under elevated pressures.
Such pressures could have been generated in several
ways, but all are intimately associated with faulting
(e.g. Sibson et a!., 1 97 5 ; Sibson, 1 98 1 ; Carter et al.
1 990; Knipe, 1 993; Sample et a!., 1 9,93). Individual
faults behave as fluid conduits during coseismic
strain release, and as fluid barriers between high
strain rate events (Knipe, 1 99 3). Large volumes of
fluid may be transmitted along normal faults due to
the equilibration of subsurface pressure compart­
ments during slip (seismic valving) (Sibson, 1 98 1 ,
1 990; Burley et a!., 1 989; Wood & Boles, 1 99 1 ).
The flow vectors will depend upon the relative
distribution of overpressured geocompartments
and hydrostatically pressured permeable conduits
with respect to the finite extent of the fault. During
active faulting in non-overpressured basins, strain
partitioning within the failure envelope may trans-
426 R.
mit fluids between hanging wall and footwall aqui­
fers (Knipe, 1 993).
On a more regional scale, normal faulting can
influence fluid flow via strain cycling (Muir-Wood,
1 994). Dilation of high-angle microfractures in
crustal rocks accompanies regional interseismic ex­
tension. Fluid is drawn into these fractures until
they catastrophically collapse as slip on normal fault
planes causes coseismic elastic rebound. The conse­
quent pore pressure change will focus fluid flow into
permeability fairways such as mesofractures. Gross
fluid volumes of the order of 0.5 km 3 can be
expelled to the surface during the main rupture and
subsequent aftershock period of major (magnitude
7) earthquakes (Sibson, 1 98 1 ; Wood, 1 985; Muir­
Wood, 1 994). A prolonged episode of extensional
tectonism will therefore result in a cycle of hydrau­
lic fracturing and net upward flow, with dissipation
of the fluid pressure into permeable strata or even­
tually to the surface. Furthermore, just as the slip on
major faults can be incrementally recorded on
parasitic faults and splays, dispersed fracturing and
veining would be expected to dissipate fluid flow
from the principal slip surfaces. From cement dis­
tributions in the Hopeman Sandstone, Edwards et
a!. ( 1 993) concluded that fluid ingress took place via
large cross-formational slip zones, but that dispersal
was via smaller-scale faults centred on CSBs. In the
same manner, activity on NNW-SSE normal faults
in the Ballycastle region is inferred to have pro­
vided episodes of heightened pore fluid pressure
that may both have initiated fault-parallel cataclasis
and subsequently reactivated the CSBs as dilata­
tional veins. High pore fluid pressures were proba­
bly generated by the fault activity, and once the
veins opened, fluid flow exploited relatively perme­
able zones in the surrounding sandstone until hy­
drostatic pressure was reattained. Cementation
therefore took place in the sandstone immediately
surrounding the CSBs.
Dolomite precipitation probably took place at
relatively low temperatures (< I OO OC), on the basis
of planar crystal fabrics (Sibley & Gregg, 1 987) and
the dominance of monophase aqueous fluid inclu­
sions (Goldstein & Reynolds, 1 994). Under such
conditions, the precipitation of well-ordered dolo­
mite is favoured by high (Mg2 + + Fe2 + )/Ca2 + ra­
tios, low salinities and high carbonate alkalinity in
pore fluids (Folk & Land, 1 97 5 ; Machel &
Mountjoy, 1 986). Low salinity reduces ion pairing,
and high co�- activities facilitate dehydration of
Mg2 + ions. Dolomite exhibits retrograde solubility,
Evans et a!.
but fluids expelled up faults from more deeply
buried strata are unlikely to have increased in
temperature. Precipitation is therefore most likely
to have accompanied a fall in pore fluid pressure
and associated C0 degassing (Morse & Mackenzie,
2
1 990; Tucker & Wright, 1 990). In particular, dissi­
pation of flow (and therefore of PC0 ) may be
2
recorded in the lateral transition from pervasive
cements into dispersed poikilotopic crystals away
from the central conduits.
Multistage reactivation of the veins produced
complex, symmetrically banded cement fills. The
nature of these, including the presence of forking
and tapering high-angle terminations, strongly re­
sembles hydraulic fractures described in the litera­
ture (e.g. Sibson et a!., 1 97 5 ; Beach, 1 980; Carter et
a!., 1 990). Assuming that each episode of reopening
was in response to a transient rise in pore fluid
pressure, subsequent PC0 release during dilation
2
was probably sufficient to drive precipitation from
dolomite-oversaturated fluids. Each pressure re­
lease thereby triggered a rapid precipitation event,
which in tum recreated the stress conditions for
rupture (Phillips, 1 972). Nevertheless, it is surpris­
ing that the dolomite in the central veins is much
finer grained than the preceding intergranular ce­
ment. It may be that the supersaturation increased
with time, or that the pore pressure release during
repeated failure of the lithified fracture zones was
more rapid than when fluid could dissipate into the
surrounding sandstone. Alternatively, the fine to
medium crystalline intergranular dolomite cement
at Ballycastle could represent the pre-slip dilatant
episode, and the multiple microcrystalline dolomite
veining be associated with progressive and intermit­
tent strain release via slip on the parallel faults. The
change in deformational style from cataclasis to
dilation thereby reflects the influence of diagenesis
(specifically porosity reducti9n), rather than a
change in the tectonic regime (see Antonellini et a!.
1 994). A possible analogy is an example of veining
associated with normal faulting in the Neogene
sandstones of Iran, described by Sibson ( 1 98 1 ).
Within about I 00 m of these faults, and parallel to
them, numerous extensional fractures are filled with
gypsum and locally cut by subparallel neptunian
dykes. The veins are interpreted to have formed
hydraulically during pre-slip fracture dilatancy, and
once slip took place on the faults the collapse of
dilatant fractures increased local pore pressures and
mobilized the unlithified sandstone into neptunian
dykes.
 Fracture-related dolomite cements in porous sandstones
Theoretical considerations suggest that hydraulic
fracturing is characteristic of the uppermost (near­
surface) regions of normal faults. At greater depths
(e.g. > 1000 m) pore fluid pressures must exceed the
hydrostatic head (Sibson, 198 1). This may occur in
overpressured zones, and is also facilitated during
tectonic uplift. It is difficult to envisage overpressur­
ing in the high-porosity sandstones at Ballycastle,
and it is probable that the cemented fractures
formed at relatively shallow burial and/or during
uplift. This is supported by relatively high maxi­
mum IGV values of the dolomite cemented sand­
stone (Table 1). Fluid flux through fractures is
limited by the continuity of flow (input at the base
versus outflow at the top), and is potentially greatest
at shallow depths or where fractures are continuous
to the surface (Bj0rlykke, 1993). However, the pore
pressure increase resulting from coseismic fluid
focusing into arterial fractures is likely to be consid­
erable, and in the Ballycastle case it is clear that
fluids could readily dissipate into the surrounding
high-porosity sandstones.
Origin of pore f lui d s and sourc e of
d olomite cement
Calcite cements are widespread in sandstones, but
dolomites are considerably less common. In general
this reflects the requirement for major sources of
magnesium, plus the relatively greater kinetic inhi­
bition of dolomite nucleation at low temperatures
(Tucker & Wright, 1990). Consequently, sandstone­
hosted dolomite cements are generally associated
with arid climates ( dolocrete; Spot! & Wright,
1992), cross-formational flow from buried carbon­
ates and evaporites undergoing pressure dissolution
(e.g. Sullivan et al., 1990; Purvis, 1992; Turner et
al., 1993), hydrothermal circulation of sea water
and fluid-rock interaction with mudrocks or volca­
nics (Searl & Fallick, 1990; Searl, 1991; Morad et
al., 1996), or mass transfer from mudrocks where
clay mineral transformations provide Ca2 + , Mg2 +
and Fe2 + (Boles, 1978; McHargue & Price, 1982;
Gawthorpe, 1987; Land et al., 1987; Macaulay et
al., 1993; de Souza et al., 1995).
The facies relationships, spatial distribution and
petrographic characteristics of dolomite cement in
the Ballycastle foreshore are incompatible with a
dolocrete origin. Formation from evaporite-related
fluids can be discounted if the dolomites are pre­
Late Permian in age (and no Permian evaporites
have been proven in the Rathlin basin; e.g. Me-
427
Cann, 1988). A hydrothermal origin for the dolo­
mites can also be be ruled out. The oxygen isotopic
values and lack of non-planar crystal fabrics indi­
cate relatively low precipitation temperatures (see
below), and Carboniferous volcanics at Ballycastle
pre-date deposition of the sandstones. The nearest
Permian volcanics recorded are 70 km north of
Ballycastle, between I slay and Jura (Fig. 1) (see
Upton et al., 1987; Hitchen et al., 1995) and 60 km
to the southeast, where >600 m of lavas are re­
corded in the Lame borehole (McCann, 1990). It is
generally accepted that there was little hydrother­
mal activity associated with the Tertiary igneous
centres of western Scotland (Hudson & Andrews,
1987; Searl, 1994), and there is no evidence that the
situation in northeast Ireland was any different.
The quartzose fluvial sandstones hosting the ce­
mented fractures are unlikely to have contained
significant detrital carbonate, and contain no evi­
dence of detrital or eogenetic ferriferous phases.
Consequently, import of dissolved Mg2 + , Fe2 + and
carbonate must have taken place along the fractures
from an external source. Ferromagnesian carbonate
cements are most commonly sourced from
mudrocks (Machel & Mountjoy, 1986; Taylor &
Sibley, 1986). Dolomite o 1 3C data (mean : -6.8%o)
(Fig. 13) indicate a significant component of orga­
nogenic carbon, and thereby support the involve­
ment of organic-rich mudrocks in providing the
dolomitizing fluid (Hudson, 1977; Irwin et a!.,
1977). Although some Ca2 + may have been
sourced from plagioclase dissolution in the sand­
stones, a simple mass-balance calculation suggests
that complete dissolution of 25% modal plagioclase
(Na/Ca ratio 1) would be required to account for
:
the observed volumes of dolomite. Furthermore,
such a source is incompatible with the evidence of
fluid expulsion from the fractures, which would
have counteracted any inward diffusion of solutes
from the surrounding sand body (see Bj0rlykke,
1993). Hence Ca2 + was also imported through the
fractures during dolomite cementation.
On the basis of equilibrium fractionation rela­
tionships (Land, 1983), dolomite 8 1 8 0 values can
be interpreted in terms of a range of temperatures
and corresponding pore fluid o 1 8 0 values (Fig. 13).
Pore fluids sourced from within the local succession
would be expected to range from marine to fresh
water (Fig. 4). Assuming a surface temperature of
20 °C, and non-glacial Carboniferous sea water of
Oo/oo SMOW (Popp et a!., 1986), precipitation from
any mixture of marine and meteoric fluids would
428 R. Evans et al.

have taken place at less than about 5 5 " C. This
temperature is reasonable in terms of the published
vitrinite reflectance values from the area, but is
slightly higher than that extrapolated from the
regional geology. The studied sand body is near the
top of the Carboniferous succession in north Ant­
rim (Fig. 4), and would have reached a maximum
Late Carboniferous burial depth of about 570 m,
with uplift to about I 70 m by the early Permian
(Fig. 3). Given a standard continental 3 0 " C km- 1
geothermal gradient, the corresponding tempera­
ture range for pre-Triassic dolomite precipitation is
3 7-25 " C. On this basis, the corresponding pore
fluid would have been brackish in nature (�0 to
-3. 5%o SMOW) (Fig. 1 3). It is unlikely to have been
·
entirely meteoric, as low Mg2 + /Ca2 + ratios (e.g.
Drever, 1 982) would have favoured calcite rather
than dolomite cementation.
Although sandstone dominated, the Ballycastle­
Murlough Bay section contains subordinate marine
to brackish shales rich in organic detritus (Fig. 4).
Microbial reactions in shallow-buried organic-rich
mudrocks can produce solutes that, upon diffusion
into adjacent sandstones, may precipitate small
amounts (c. I 0 vol%) of carbonate cement (McMa-
hon et a!., 1 992). However, mass transfer is ex­
tremely localized and in many cases pH changes in
the paralic setting would most likely favour the
reincorporation of Mg2 + , Fe2 + and Ca2 + into in
situ diagenetic carbonates (e.g. Curtis et a!., 1 986):
A dolomitic fibrous calcite ('beef') vein within
shales stratigraphically above the studied sandstone
has a more negative oxygen isotopic composition
than most of the fracture-related dolomites, indicat­
ing precipitation from local isotopically lighter pore
fluids and/or at deeper burial (Fig. 1 3). Likewise,
the 'marine' o 1 3C signature of dolomite filling gas­
tropod moulds in the Main Limestone suggests a
distinct, and probably localized, origin.
The mass transfer problem may be resolved if
eogenetic pore fluids were transmitted locally across
the GGF in response to fault-related changes in
pore pressure and near-surface hydrology (e.g.
Knipe, 1 993). The cemented fractures at Ballycastle
show clear evidence for elevated pore fluid pres­
sures and hydraulic fracturing, and the pulsed
nature of dolomite cementation implies a tectonic
drive. Kerr ( 1 987) commented on the likelihood
that the GGF was active in the Late Carboniferous/
Permian. Argillaceous coastal plain sediments asso-

Temperature eC)
1 00 Water-rock interaction due to

-
shale dehydration reactions

80

Estimated
maximum Depositional temperature
(warm enough for coals)

20°C -

-1 0 -8 -6 -4 -2 0 2 4 6
(surface temperature 20°C) =

Fig. 13. Equilibrium fractionation plot showing possible combinations of water o 1 80 values and temperatures which
are compatible with the intergranular and vein-fill dolomites. Curves plotted using the equation from Land ( 1 983).
Estimated O%o SMOW for Carboniferous sea water from Popp et a/. ( 1 986); low-latitude meteoric water from Anderson
& Arthur (1 983). Burial temperatures calculated from subsidence history (Fig. 3), 2 0 " C surface temperature and
30"C km- 1 geothermal gradient.
 Fracture-related dolomite cements in porous sandstones
ciated with the Murlough Bay Coals contain
abundant authigenic ankerite and dolomite that
may represent shallow burial diagenetic products of
organic matter oxidation and iron reduction (see
Matsumoto & Iij ima, 1 98 1 ). In the Ballycastle area
the GGF downthrows the studied sandstones
against these deposits (Fig. 5), and some of the
NNW-SSE-trending faults parallel to the cemented
fractures intersect the GGF. Brackish, Fe2 + - and
Mg2 + -rich pore fluids from the Murlough Bay
Coals might thus have been transmitted into the
NNW -SSE fractures at Ballycastle during faulting.
In contrast, at Murlough Bay itself the GGF juxta­
poses two sandstone units, and although CSBs and
microfaults are common they are not dolomitized.
Alternatively, dolomite-precipitating fluids could
have been derived from a more distal source.
Carboniferous deltas prograded from east to west
across north Antrim (Evans, 1 99 5) and the offshore
Rathlin basin is expected to contain substantial
thicknesses of pro-delta mudrocks. These would
have been buried to little more than the onshore
succession in the Late Carboniferous (Fig. 3), but
could have supplied considerably greater volumes
of pore fluid. During moderate burial (� 1 000 m)
compactional dewatering expels almost 50% of the
interstitial pore fluids of mudrocks into overlying
strata and laterally towards basin margins, along
permeability fairways such as sandstone lobes. The
resultant fluid flow rates are commonly believed to
be insufficient to account for significant carbonate
cementation of basin-margin sandstone beds (e.g.
Bj0rkum & Walderhaug, 1 990; Bj0rlykke, 1 993).
However, thermobaric fluid flux can be dramati­
cally enhanced where it is focused through fractures
or other high-permeability conduits, particularly in
association with faulting-related changes in pore
fluid pressure.
Strain cycling during episodes of extensional
tectonism may have driven basinal fluids up the
major basin-bounding faults. Thus, if the NNW­
SSE-trending faults at Ballycastle are structurally
linked to the TVF/GGF, the opportunity would
have existed for dolomitizing fluid to be transmit­
ted up the fault and fracture array until it inter­
sected the porous basin-margin sandstones at
shallow levels. A pro-delta depositional setting
means that basin-derived pore fluids would have
been predominantly marine, and consequently had
higher Mg2 + /Ca2 + ratios and bicarbonate activities
than onshore brackish shales. Precipitation of inter­
granular dolomite may have been enhanced by
429
mixing of marine-derived thermobaric fluids with
residual meteoric waters in the fluvial sand body
(see Taberner & Santisteban, 1 98 7 ; Morad et a!. ,
1 992). Such mixing could also account for the
slightly more negative dolomite il 1 8 0 values than
predicted for fully marine pore fluids and shallow
Late Carboniferous burial (Fig. 1 3).
Basinal fluids suitable for dolomite precipitation
may also have been produced during the Mesozoic­
Tertiary. The Rathlin basin underwent major sub­
sidence in the Triassic-Jurassic, and again in the
Palaeogene, with Carboniferous strata entering the
oil window at about 2700 m depth (Parnell, 1 992)
(Fig. 3). This would have been sufficiently deep for
the dehydration and partial illitization of smectite
in mudrocks (see Hower et a!., 1 976; Pearson &
Small, 1 98 8), potentially increasing the Mg2 + ,
Fe2 + (and Ca2 + ) activity of the pore fluids, as well
as pore fluid pressure (Burst, 1 976; Boles & Franks,
1 9 79). Further augmentation of pore pressure and
the release of organic acids and C0 into the pore
2
fluids would have accompanied kerogen maturation
(e.g. Burley, 1 986). Significant overpressuring can
result where this is combined with compactional
disequilibrium (e.g. Luo et a!., 1 994; Snowdon,
1 99 5). Overpressure might therefore have been
generated in the Carboniferous mudrocks of the
Rathlin basin during the Paleogene phase of rapid
burial. The generation of organic acids from kero­
gen would have mobilized any detrital carbonate in
the muds, providing a mixed organic-marine dis­
solved carbon reservoir to source carbonate ce­
ments in the sandstones (e.g. Milliken & Land,
1 99 3). Seismic valving (Cathles, 1 990; Sibson,
1 990) associated with Tertiary fault activity is a
plausible mechanism for transmitting pulses of
dolomitizing fluid into the shallow basin margins at
this time. Clay mineral dehydration reactions can
increase pore fluid il 1 8 0 values .by up to 5%o
(Wilkinson et a!., 1 992) and the Ballycastle dolo­
mite could have precipitated from a moderately
1 80-enriched marine-derived pore fluid (e.g. + l
to +3o/oo SMOW) at about 40-70"C. This is slightly
higher than a predicted temperature of 43 ·c at an
estim<�ted 750 m Early Tertiary burial depth (Fig.
3), although no account is taken of any enhanced
geothermal gradient due to Paleogene igneous
activity.
Ti ming of f ractu ring and c ementatio n
A genetic link has been established between
430 R. Evans et al.
dolomite-cemented fractures and normal faulting in
the Ballycastle region, with the similar orientation
of cemented fractures and local NNW-SSE­
trending normal faults suggesting formation in the
same tectonic episode. Unfortunately, the displace­
ment on these faults is poorly dated, being pre­
Tertiary with local Tertiary reactivation. It has been
shown that potential dolomitizing fluid sources
existed in the Late Carboniferous, Triassic-Jurassic
and Early Tertiary. Other circumstantial evidence
therefore has to be used in attempting to date the
cementation. For example, isotopic data suggest
that intergranular and vein-fill dolomite formed at a
similar time, and the presence of en echelon cross­
linkages in some cases demonstrates a component
of dextral shear associated with vein dilation (e.g.
Figs 6, 7F and I 0), and suggests a predominantly
N-S to NNE-SSW transtensional stress regime.
The three episodes of extensional tectonics that
have been recognized in the north Antrim region
are Late Carboniferous-Early Permian, Triassic­
Jurassic, and Late Cretaceous-Tertiary (Fig. 3).
Kerr ( 1 98 7) ascribed the formation of NNW-SSE
normal faults at Ballycastle to Triassic-Jurassic
ENE-WSW to E-W extension accompanying sub­
sidence of the Rathlin basin. He proposed that the
formation of CSBs in the Carboniferous sandstones
accompanied Late Tertiary rej uvenation of these
structures, and that their development reflected a
facies control owing to the textural maturity of the
rocks compared with surrounding less well-sorted
and porous units. However, it is likely that CSBs
formed during the first tectonic episode to affect the
rocks after deposition, rather than after more than
200 Myr and several episodes of faulting and fault
reactivation. Although it is impossible to unequiv­
ocally discount a Tertiary age for the dolomite
cementation, the balance of evidence favours the
Late Carboniferous for both the formation of CSBs
and their reactivation to form cemented fractures.
1 Triassic or Tertiary NNW-SSE fractures formed
in an ambient E-W to ENE-WSW extensional
stress regime would be expected to show evidence of
sinistral rather than dextral transtension. Kerr
( 1 987) noted the evidence for late-stage dextral
shear reactivation of the NNW-SSE faults, which
he ascribed to a short-lived change in the Tertiary
stress field. However, Tertiary veins in the Creta­
ceous carbonates near Ballycastle are dominated by
NE-SW to ENE-WSW orientations, with a subor­
dinate NNW-SSE to N-S set. Both are filled by
multiple calcite cements with no evidence of fer-
roan dolomite (Kerr, 1 987). Furthermore, although
several joint trends are present in the llallycastle­
Murlough Bay region, only those with a NNW-SSE
to N-S orientation and steep inclination feature
carbonate cementation (Roberts, 1 97 6).
2 The Late Carboniferous tectonic regime in north­
east Ireland was one of N-S to NNE-SSW exten­
sion. Although Kerr ( 1 987) did not recognize the
initiation of NNW-SSE faults at this stage, the
stress field would permit transtensional reactivation
of NNW-SSE faults and CSBs with a component of
dextral shear, as observed in the vein geometries
within the cemented fractures.
3 Regional uplift accompanying faulting in the late
Carboniferous would have favoured hydraulic frac­
turing, and the sandstones would have been at
relatively shallow burial depths (Fig. 3). This con­
curs both with the petrographic characteristics of
intergranular dolomite cementation and with the
isotopic data in terms of potential fluid sources.
4 Triassic (re)activation of NNW-SSE-trending
faults at Waterfoot (east Antrim) was accompanied
by injection of sandstone neptunian dykes but
without associated dolomite-cemented fractures
(Kerr, 1 98 7). In contrast, no neptunian dykes are
seen at Ballycastle.
5 The fractured sand body is not reddened and did
not experience widespread ferroan dolomite cemen­
tation as described by Wang ( 1 992) from several
other beds in the Ballycastle-Murlough Bay succes­
sion. Although Wang did not speculate on the
petrogenesis of the disseminated dolomitization, its
presence in thick fluvial distributary sandstones is
incompatible with an in situ origin. As such, it is
plausibly related to the same episodes of fluid flow
that produced the dolomite-cemented fractures,
and would indicate a Late Carboniferous age for the
latter. The much more pervasive cementation in the
reddened sandstones would p�;esumably have re­
sulted from their intersection by larger-scale or
more long-lived conduits, or a change in the tec­
tonic 'plumbing' network. Further isotopic work
and mapping will be required to confirm this hy­
pothesis.
1M PLICATIONS FOR
RESERVOIR QUALITY
The accurate modelling of reservoir performance in
siliciclastic systems requires an understanding of
structural and diagenetic permeability heterogene-
 Fracture-related dolomite cements in porous sandstones
ities as well as quantitative geological/sedimento­
logical input from outcrop analogues (e.g. Dranfield
et a!., 1 98 7 ; Hurst, 1 98 7 ; Miall, 1 98 8 ; Dreyer et a!.,
1 990; Weber & van Geuns, 1 990; Bryant & Flint,
1 99 3). Sedimentological and petrophysical data
form the primary inputs for reservoir characteriza­
tion and stochastic modelling of fluid flow in fluvial
sandstones (e.g. North & Taylor, 1 996). Diagenetic
data are frequently ignored or 'averaged' over indi­
vidual sand bodies in a reservoir model. Although
this may be an acceptable approach for dissemi­
nated or strongly facies-related cements, it is inap­
propriate to fracture-related diagenesis, which can
produce highly localized conduits or barriers to
fluid flow. Furthermore, the role of fractures may
have evolved through geological time. Because dia­
genetic fluids also change during burial and uplift
episodes, the behaviour of fractures may have
profound consequences for the non-uniform distri­
bution of enhanced or reduced poroperm.
The sand body examined in this study has been
compartmentalized by fracture-related dolomite ce­
mentation. The degree of connectivity remaining is
determined by fracture size (thickness, length,
height), intensity (spacing per m3) and the degree of
permeability and porosity reduction from cataclasis
and cementation. Because cemented fractures are
effectively parallel to the dominant local fault trend,
fluid flow would be hampered in an E-W/NE-SW
direction (at least). Selective cementation of depo­
sitional laminae adjacent to the fractures indicates a
significant pre-existing permeability heterogeneity
within the sand body, albeit on a scale an order of
magnitude smaller than that imposed by the frac­
tures. Both effects would need to be incorporated
into a realistic reservoir model.
It is noteworthy that there is very little evidence
of dedolomitization in the studied sand body at
Ballycastle, in marked contrast to that described by
Wang ( 1 992) from reddened sandstones in the same
succession. Given the interpretation that fracture­
related cementation was Late Carboniferous in age,
a possible implication is that flushing of uncon­
formity-sourced oxidizing meteoric fluids in the
Latest Carboniferous/Early Permian was impeded
by the structural-diagenetic compartmentalization
of the sand body, compared with those in which
ferroan dolomite cement was more homogeneously
distributed.
Dolomite cementation at Ballycastle has also
reduced the net sand within the sand body. On the
basis of a thickness of 6 m and maximum foreshore
43 1
width of about 50 m, the total volume of the
exposed sand body is approximately 3 x 1 04 m3•
The cumulative volume of cemented fractures,
calculated from the measured transect, is 1 1 1 3 m 3,
giving a net/gross ratio of 0.96. This is clearly
negligible compared with the effects of compart­
mentalization on the effective net sand, particularly
where pockets of the sand body are enclosed by
anastomosing and conjugate dolomite-cemented
fracture sets.
The wider significance of the Ballycastle study is
that more deeply buried Carboniferous sandstones
in the Rathlin basin and North Channel regions are
potential hydrocarbon reservoirs (Parnell, 1 992).
Carboniferous mudrocks in the Rathlin basin
reached peak maturity in the Early Tertiary (Fig. 3),
after the inferred timing of cataclasis and dolomite
cementation. If analogous features are widespread,
they may have a large impact on reservoir quality,
particularly in terms of incomplete filling, low
recovery efficiencies and permeability anisotropy.
The scale of the structures is below even the highest­
resolution 3-D seismic currently available, and
further outcrop examination is probably the best
way to assess the regional significance of cemented
fractures and their relationship to the principal
(seismically resolvable) fault trends. Such data will
aid the planning of production strategies, such as
directional drilling, hydrofracturing or acidization
(North, 1 98 5).
CONCLUSIONS
Internally complex, ferroan dolomite-cemented
fractures are a prominent feature within a fluvial
sand body at Ballycastle, on the margins of the
Rathlin basin in northeast Ireland. The cemented
fractures display a tight modal or.ientation that is
coincident with the dominant local normal faulting
trend, and are interpreted to have formed in the
same tectonic regime. However, there is no clear
trend in the spatial distribution or width of the
fractures with respect to a fault plane situated at one
end of the outcrop.
Detailed examination of the external morphology
and internal fabrics of the cemented fractures re­
veals that they formed by reactivation of cataclastic
slip bands in an extensional to dextral transten­
sional regime. Cementing fluids were transmitted
through the fractures under high pore pressures,
initially invading the surrounding sandstones along
432 R.
the most permeable sedimentary laminae. Dolo­
mite cement precipitated in response to dissipation
of PC0 as the injected fluids mixed with and
2
displaced the residual pore waters of the fluvial
sand body. This is reflected in a sharp lateral
transition from pervasive fine to medium crystal­
line cement to dispersed poikilotopic crystals away
from the medial fracture.
Despite initial dolomite cementation, the central
fractures remained planes of relative weakness and
were subsequently reopened a number of times.
Because of the surrounding intergranular cement,
each dilation episode was followed by passive infill
by finely crystalline dolomite, producing a complex
symmetrically banded pattern. Precipitation is as­
cribed to the rapid diminution of pore pressure as
dilation took place, and the resultant vein morphol­
ogies are typical of hydraulic fracture mechanisms.
Dolomite cement components were all sourced
from outside the sand body, most probably from
local or basinal mudrocks. Stable isotope data
indicate a mixed organogenic-marine carbonate
source, and precipitation at relatively low tempera­
tures (�70'C, if pore fluids were sourced from clay
mineral dehydration reactions during deep burial of
Carboniferous mudrocks in the Rathlin basin;
�5 5 ' C if they were locally sourced). Thermobaric
mass transfer was enhanced by tectonic pulsing and
dolomite precipitation was driven by C0 degas­
2
sing.
The timing of cementation cannot be indisput­
ably proved, but the weight of evidence suggests a
Late Carboniferous age. Although the associated
normal faults have been ascribed to Triassic exten­
sion, the prevalent stress regime would not have
produced the dextral shear sense observed in some
of the cemented fractures. In contrast, these would
be compatible with the local tectonic regime in the
Late Carboniferous. A Tertiary origin for the initial
cataclastic slip bands (Kerr, 1 987) is rejected on the
basis that such features would be expected to have
formed in one of the several earlier tectonic epi­
sodes that affected the region. In addition, the
singular orientation of cemented fractures and the
mineralogy of the fill are inconsistent with proven
Tertiary veins in overlying Cretaceous deposits.
Dolomite cementation has had a minuscule im­
pact on the net/gross ratio in the studied sand body,
but will have severely partitioned it in terms of
potential fluid throughflow. Preservation of the
dolomite and lack of reddening, in contrast to other
formefly ferroan dolomite-cemented sandstones in
Evans et al.
the local area (Wang, 1 992), may directly attest to
this. The tectonodiagenetic fabrics observed in the
Ballycastle sand body are of subseismic resolution.
Equivalent strata offshore in the Rathlin basin are
potential hydrocarbon reservoirs (Parnell, 1 992)
and, if similarly affected, may contain limited re­
serves and/or exhibit low recovery efficiencies.
However, analogous cemented fractures do not
appear to be present in the majority of sandstones
in the north Antrim succession, and further work is
therefore needed to constrain the timing and struc­
tural framework of dolomite cementation in order
to make accurate predictions and extrapolations
into the subsurface.
ACKNOWLEDGEMENTS
R.E. is indebted to the Department of Economic
Development (Northern Ireland) and the Geologi­
cal Survey of Northern Ireland for funding this
work, which forms part of a doctoral research
project on the hydrocarbon potential of the Car­
boniferous of the north of Ireland. Paul Carey, Pat
McBride and Paddy Gaffikin provided much­
appreciated technical assistance during this study.
Discussions with Bernard Anderson, Alastair
Ruffell and Richard Worden helped us to clarify
several of the interpretations (and to purge some of
the more exotic ones). We particularly thank
Mogens Ramm, Ihsan Al-Aasm and Sadoon Morad
for conscientious and constructive criticism of an
earlier draft of the paper.
REFERENCES
ANDERSON, T.B., PARNELL, J. & RUFFELL, A. H. ( 1 995)
I nfluence of basement on the geometry of Permo­
Triassic basins in the north-west British Isles. In: Per­
mian and Triassic Rifling in Northwest Europe (Ed.
Boldy, S.A.R.). Spec. Pub!. Geol. Soc. London, 9 1 ,
1 03- 1 22.
ANDERSON, T.F. & ARTHUR, M.A. ( 1 983) Stable isotopes of
oxygen and carbon and their application to sedimento­
logic and palaeoenvironmental problems. In: Stable
Isotopes in Sedimentary Geology (Eds Arthur, M.A.,
Anderson, T.F., Kaplan, I.R., Veizer, J. & Land, L.S.).
Soc. Econ. Paleont. Mineral. Short Course, 10, 1 . 1 -
1 .1 5 1 .
ANTONELLINI, M. & AYDIN, A ( 1 995) Effect of faulting on
fluid flow in porous sandstones: geometry and spatial
distribution. Bull. Am. Ass. Petrol. Geol. , 79, 642-67 1 .
ANTONELLINI, M.A., AYDIN, A & POLLARD, D.O. ( 1 994)
Microstructures of deformation bands in porous sand-
 Fracture-related dolomite cements in porous sandstones
stones at Arches National Park. J. Struct. Geol., 16,
941-959.
AYDIN, A. & JOHNSON, A.M. ( 1 983) Analysis of faulting in
porous sandstones. J. Struct. Geol. , 5, 1 9-3 1 .
BEACH, A. ( 1 9 80) Numerical models of hydraulic fractur­
ing and the interpretation of syntectonic veins. J. Struct.
Geol. , 2, 425-438.
BERNER, R.A. ( 1 980) Early Diagenesis: A Theoretical Ap­
proach. Princeton University Press, Princeton, NJ.
BJ0RKUM, P.A. & WALDERHAUG, 0. ( 1 990) Geometrical
arrangement of calcite cementation within shallow ma­
rine sandstones. Earth-Sci. Rev. , 29, 145- 1 6 1 .
BJ0RLYKKE, K. ( 1 993) Fluid flow i n sedimentary basins.
Sediment. Geol. , 86, 1 3 7-15 8 .
BoLES, J.R. ( 1 978) Active ankerite cementation in the
subsurface Eocene of Southwest Texas. Contrib. Min­
eral. Petrol. , 68, 1 3-22.
BOLES, J.R. & FRANKS, S.G. ( 1 979) Clay diagenesis in
Wilcox Sandstones of Southwest Texas: implications of
smectite diagenesis on sandstone cementation. J. sedi­
ment. Petrol. , 49, 55-70.
BRYANT, l.D. & FLINT, S.S. ( 1 993) Quantitative clastic
reservoir geological modelling: problems and perspec­
tives. In: The Geological Modelling of Hydrocarbon
Reservoirs and Outcrop Analogues (Eds Flint, S.S. &
Bryant, I.D.). Spec. Pub!. Int. Ass. Sediment., 15, 3-20.
BURLEY, S.D. ( 1 986) The development and destruction of
porosity within Upper Jurassic reservoir sandstones of
the Piper and Tartan fields, Outer Moray Firth, North
Sea. Clay Miner. , 21, 649-694.
BURLEY, S.D. & KANTOROWICZ, J.D. ( 1 986) Thin section
and SEM textural criteria for the recognition of cement­
dissolution porosity in sandstones. Sedimentology, 33,
5 8 7-604.
BURLEY, S.D., MULLIS, J. & MATTER, A. ( 1 989) Timing
diagenesis in the Tartan Reservoir (UK North Sea):
constraints from combined cathodoluminescence mi­
croscopy and fluid inclusion studies. Mar. Petrol. Geol. ,
6, 98-120.
BuRST, J.F. ( 1976) Argillaceous sediment dewatering. Ann.
Rev. Earth Planet. Sci. , 4, 293-3 1 8.
CARTER, N.L., KRONENBURG, A.K., Ross, J.V. &
WILTSCHKO, D.V. ( 1 990) Controls of fluids on deforma­
tion of rocks. In: Deformation Mechanisms. Rheology
and Tectonics (Eds Knipe, R.J. & Rutter, E.H.). Spec.
Pub!. Geol. Soc. London, 54, 1 - 1 3 .
CATHLES, L.M. ( 1990) Scales and effects of fluid flow in the
upper crust. Science, 248, 323-328.
COWARD, M. ( 1 990) The Precambrian, Caledonian and
Variscan framework to NW Europe. In: Tectonic Events
Responsible for Britain 's Oil and Gas Reserves (Eds
Hardman, R.F.P. & Brooks, J.). Spec. Pub!. Geol. Soc.
London, 55, 1 -34.
CowARD, M. ( 1 995) Permo-Triassic basins ofNW Europe.
In: Permian and Triassic R ifting in Northwest Europe
(Ed. Boldy, S.A.R.). Spec. Pub!. Geol. Soc. London, 9 1 ,
7-39.
CURTIS, C.D., COLEMAN, M.L. & LOVE, L.G. ( 1 986) Pore
water evolution during sediment burial from isotopic
and mineral chemistry of calcite, dolomite and siderite
concretions. Geochim. Cosmochim. Acta, 50, 2 3 2 1 -
2 3 34.
DE SouzA, R.S., DE Ros, L.F. & MoRAD, S. ( 1 995)
433
Dolomite diagenesis and porosity preservation in lithic
reservoirs: Carm6polis Member, Sergipe-Aiagoas Basin,
northeastern Brazil. Bull. Am. Ass. Petrol. Geol. , 79,
725-748.
DEWEY, J.F. ( 1 9 82) Plate tectonics and the evolution of the
British Isles. J. Geol. Soc. London, 139, 3 7 1 -4 1 2.
DICKSON, J.A.D. ( 1 966) Carbonate identification and
genesis as revealed by staining. J. sediment. Petrol. , 36,
49 1 -505.
DRANFIELD, P., BEGG, S.H. & CARTER, R.R. ( 1 987) Wytch
Farm oil-field: reservoir characterisation of the Triassic
Sherwood Sandstone for input into reservoir simulation
studies. In: Petroleum Geology of North- West Europe
(Eds Brooks, J. & Glennie, K.W.), pp. 149- 1 60. Gra­
ham & Trotman, London.
DREYER, J.I. ( 1 982) The Geochemistry of Natural Waters.
Prentice-Hall, Englewood Cliffs, NJ.
DREYER, T., SCHEIE, A. & WALDERHAUG, 0. ( 1 990)
Minipermeameter-based study of permeability trends in
channel sand bodies. Bull. Am. Ass. Petrol. Geol. , 74,
359-374.
EDWARDS, H.E., BECKER, A.D. & HOWELL, J.A. ( 1 993)
Compartmentalisation of an aeolian sandstone by struc­
tural heterogeneities: Permo-Triassic Hopeman Sand­
stone, Moray Firth, Scotland. In: Characterisation of
Fluvial and Aeolian Reservoirs (Eds North, C.P. &
Prosser, D.J.). Spec. Pub!. Geol. Soc. London, 73,
3 3 9-365.
EVANS, D., KENOLTY, N., DOBSON, M.R. & WHITTINGTON,
R.J. ( 1 980) The Geology of the Malin Sea. Inst. Geol.
Sci. Rep., 79/15.
EVANS, R. ( 1 995) Stratigraphy, fluid flow and reservoir
characteristics of the Carboniferous ofBallycastle. Geol.
Surv. Northern Ireland Tech. Rep., 95/5.
FOLK, R.L. & LAND, L.S. ( 1 975) Mg/Ca ratio and salinity:
two controls on crystallisation of dolomite. Bull. Am.
Ass. Petrol. Geol. , 59, 60-68.
FOWLES, J. & BURLEY, S. ( 1 994) Textural and permeability
characteristics of faulted, high porosity sandstones.
Mar. Petrol. Geol. , 1 1 , 608-623.
FRANCIS, E. H. ( 1 99 1 ) Carboniferous-Permian igneous
rocks. In: Geology ofScotland(Ed. Craig, G.Y.), pp. 349-
420. Geological Society Publishing House, Bath.
GAWTHORPE, R.L. ( 1 987) Burial dolomitization and poros­
ity development in a mixed carbonate-clastic sequence:
an example from the Bowland Basin, northern England.
Sedimentology, 34, 5 3 3-55 8 .
GOLDSTEIN, R.H. & REYNOLDS, T.J. ( 1 994) Systematics of
Fluid Inclusions in Diagenetic Minerals. Soc. Econ.
Paleont. Mineral. Short Course, 31.
GREGG, J.M. & SIBLEY, D.F. ( 1 984) Epigenetic dolomitiza­
tion and the origin of xenotopic dolomite texture. J.
sediment. Petrol. , 54, 908-93 1 .
HAQ, B.U., HARDENBOL, J . & VAIL, P.R. ( 1 987) Chronology
of fluctuating sea levels since the Triassic. Science, 235,
1 1 56-1 1 6 7.
HIPPLER, S.J. ( 1 993) Deformation microstructures and
diagenesis in sandstone adjacent to an extensional fault:
implications for the flow and entrapment of hydrocar­
bons. Bull. Am. Ass. Petrol. Geol. 77, 625-637.
.
HITCHEN, K., STOKER, M.S., EVANS, D. & BEDDOE­
STEPHENS, B. ( 1 995) Permo-Triassic sedimentary and
volcanic rocks in basins to the north and west of
434 R.
Scotland. In: Permian and Triassic Rifting in Northwest
Europe (Ed. Baldy, S.A.R.). Spec. Pub!. Geol. Soc.
London, 9 1 , 88-1 02.
HOWER, J., ESLINGER, E.V., HOWER, M.E. & PERRY, E.A.
( 1 976) Mechanism of burial metamorphism of argilla­
ceous sediment: I . Mineralogical and chemical evi­
dence. Geol. Soc. Am. Bull. , 87, 725-737.
HuDSON, J.D. ( 1 977) Stable isotopes and limestone lithifi­
cation. J. Geol. Soc. London, 133, 63 7-660.
HUDSON, J.D. & ANDREWS, W.J. ( 1 987) The diagenesis of
the Great Estuarine Group, Middle Jurassic, Inner
Hebrides, Scotland. In: Diagenesis of Sedimentary Se­
quences (Ed. Marshall, J.D.). Spec. Pub!. Geol. Soc.
London, 36, 256-276.
HuRST, A. ( 1 987) Problems of reservoir characterization
in some North Sea sandstone reservoirs solved by the
application of microscale geological data. In: North Sea
Oil and Gas Reservoirs (Eds Kleppe, J., Berg, E.W.,
Buller, A.T., Hjelmeland, 0. & Torsaeter, 0.), pp. 1 5 3-
1 67 . Graham & Trotman, London.
IRWIN, H., CuRTIS, C. & CoLEMAN, M. ( 1 977) Isotopic
evidence for source of diagenetic carbonates formed
during burial of organic-rich sediments. Nature, 269,
209-2 1 3.
JAMISON, W.R. & STEARNS, D.W. ( 1 982) Tectonic deforma­
tion of Wingate Sandstone, Colorado National Monu­
ment. Bull. Am. Ass. Petrol. Geol. , 66, 25 84-2608.
KERR, I.D.V. ( 1 987) Basement/cover structural relation­
shps in the north Antrim area, Ireland. PhD Thesis, The
Queen's University of Belfast, Belfast.
KNIPE, R.J. ( 1 993) The influence of fault zone processes
and diagenesis on fluid flow. In: Diagenesis and Basin
Development (Eds Horbury, A.D. & Robinson, A.G.).
Am. Ass. Petrol. Geol. Studies in Geology, 36, 1 3 5- 1 5 1 .
LAKE, S.D. & KARNER, G.D. ( 1 987) The structure and
evolution of the Wessex Basin, southern England: an
example of inversion tectonics. Tectonophysics, 137,
347-378.
LAND, L.S. ( 1 9 83) The application of stable isotopes
studies of the origin of dolomite and problems of
diagenesis of clastic sediments. In: Stable Isotopes in
Sedimentary Geology (Eds Arthur, M.A., Anderson,
T.F., Kaplan, I.R., Veizer, J. & Land, L.S.). Soc. Econ.
Paleont. Mineral. Short Course, 10, 4. 1 -4.22.
LAND, S.L., MILLIKEN, K.L. & McBRIDE, E. F. ( 1 987)
Diagenetic evolution of Cenozoic sandstones, Gulf of
Mexico Sedimentary Basin. Sediment. Geol. , 50, 1 95-
225.
Luo, M., BAKER, M.R. & LEMONE, D.V. ( 1 994) Distribu­
tion and generation of the overpressure system, eastern
Delaware basin, western Texas and southern New Mex­
ico. Bull. Am. Ass. Petrol. Geol. , 78, 1 3 86-1 405.
MACAULAY, C.!., HASZELDINE, R.S. & FALLICK, A.E. ( 1 993)
Distribution, chemistry, isotopic composition and ori­
gin of diagenetic carbonates: Magnus Sandstone, North
Sea. J. sediment. Petrol. , 63, 33-43.
MACHEL, H.G. & MOUNTJOY, E.W. ( 1 986) Chemistry and
environments of dolomitization-a reappraisal. Earth­
Sci. Rev., 23, 1 75-222.
MATSUMOTO, R. & IIJIMA, A. ( 1 98 1 ) Origin and diagenetic
evolution ofCa-Mg-Fe carbonates in some coalfields of
Japan. Sedimentology, 28, 2 39-259.
McCAFFERY, R.J. & McCANN, N. ( 1 992) Post-Permian
Evans et al.
basin history of NE Ireland. In: Basins on the A tlantic
Seaboard: Petroleum Geology, Sedimentology and Basin
Evolution (Ed. Parnell, J.). Spec. Pub!. Geol. Soc.
London, 62, 277-290.
McCANN, N. ( 1 988) An assessment of the subsurface
geology between Magilligan Point and Fair Head,
Northern Ireland. Irish J. Earth Sci. , 9, 7 1 -78.
McCANN, N. ( 1 990) The subsurface geology between
Belfast and Lame, Northern Ireland. Irish J. Earth Sci. ,
10, 1 5 7- 1 73 .
McHARGUE, T.R. & PRICE, R.C. ( 1 982) Dolomite from
clay in argillaceous marine associated carbonates. J.
sediment. Petrol. , 52, 873-886.
McMAHON, P.B., CHAPELLE, F.H., FALLS, W.F. & BRADLEY,
P.M. ( 1 992) Role of microbial processes in linking
sandstone diagenesis with organic-rich clays. J. sedi­
ment. Petrol., 62, 1 - 1 0.
MIALL, A.D. ( 1 988) Reservoir heterogeneities in fluvial
sandstones: lessons from outcrop studies. Bull. Am. Ass.
Petrol. Geol. , 72, 682-697.
MILLIKEN, K.L. & LAND, L.S. ( 1 993) The origin and fate of
silt sized carbonate in subsurface Miocene-Oligocene
mudstones, south Texas Gulf Coast. Sedimentology, 40,
1 07-1 24.
MORAD, S., MARFIL, R., AL-AASM, I.S. & GOMEZ-GRAS, D.
( 1 992) The role of mixing-zone dolomitization in sand­
stone cementation: evidence from the Triassic Bunt­
sandstein, the Iberian Range, Spain. Sediment. Geol. ,
80, 5 3-65.
MORAD, S., DE Ros, L.F. & AL-AASM, I.S. ( 1 996) Origin of
low 8 '80, pre-compactional ferroan carbonates in the
marine St0 Formation (Middle Jurassic), offshore NW
Norway. Mar. Petrol. Geol. , 13, 263-276.
MORROW, C.A. & BYERLEE, J.D. ( 1 989) Experimental
studies of compaction and dilatancy during frictional
sliding on faults containing gouge. J. Struct. Geol. , 1 1 ,
8 1 5-825.
MORSE, J.W. & MACKENZIE, F.T. ( 1 990) Geochemistry of
Sedimentary Carbonates. Elsevier, Amsterdam.
MOZLEY, P.S. & GOODWIN, LB. ( 1 995) Patterns of cemen­
tation along a Cenozoic normal fault: a record of palaeo­
flow orientations. Geology, 23, 53 9-542.
MUIR-WooD, R. ( 1 994) Earthquakes, strain-cycling and
the mobilisation of fluids. In: Geofluids; Origin, Migra­
tion and Evolution ofFluids in Sedimentary Basins (Ed.
Parnell, J.). Spec. Pub!. Geol. Soc. London, 78, 85-98.
MYKURA, W. ( 1 967) The Upper Carboniferous rocks of
south-west Ayrshire. Bull. Geol. Surv. GB, 26, 23-98.
NORTH, C.P. & TAYLOR, K.S. ( 1 996) Ephemeral-fluvial
deposits: integrated outcrop and simulation studies
reveal complexity. Bull. Am. Ass. Petrol. Geol. , 80,
8 8 1 -883.
NORTH, F.K. ( 1 985) Petroleum Geology. Unwin Hyman,
Winchester, MA.
PARNELL, J. ( 1 99 1 ) Hydrocarbon potential of Northern
Ireland: Part II. Reservoir potential of the Carbonifer­
ous. J. Petrol. Geol. , 14, 1 43 - 1 60.
PARNELL, J. ( 1 992) Burial histories and hydrocarbon
source rocks on the North West Seaboard. In: Basins on
the Atlantic Seaboard: Petroleum Geology, Sedimentol­
ogy and Basin Evolution (Ed. Parnell, J.). Spec. Pub!.
Geol. Soc. London, 62, 3-1 6.
PEARSON, M.J. & SMALL, J.S. ( 1 988) Illite-smectite diagen-
 Fracture-related dolomite cements in porous sandstones
esis and palaeotemperatures in northern North Sea
Quaternary to Mesozoic shale sequences. Clay Miner. ,
23, 1 09-1 32.
PHILLIPS, W.J. (1972) Hydraulic fracturing and minerali­
sation. J. Geo/. Soc. London, 1 28, 3 3 7-359.
PITTMAN, E. D. ( 1 98 1 ) Effect of fault-related granulation on
porosity and permeability of quartz sandstones, Simp­
son Group (Ordovician), Oklahoma. Bull. Am. Ass.
Petrol. Geol. , 65, 2 3 8 1 -2387.
POPP, B.N., PODOSEK, F.A., BRANNON, J.C., ANDERSON,
T.F. & PIER, J. ( 1 986) 8 7Sr/ 86 Sr ratios in Permo­
Carboniferous seawaters from the analysis of well pre­
served brachiopod shells. Geochim. Cosmochim. Acta,
50, 1 32 1 - 1 328.
PURVIS, K. ( 1992) Lower Permian Rotliegend sandstones,
southern North Sea: a case study of sandstone diagene­
sis in evaporite-associated sequences. Sediment. Geo/. ,
77, 1 55-1 7 1 .
RoBERTS, J.C. ( 1 976) The joint and fault patterns of the
north coast of counties Antrim and Londonderry be­
tween Murlough Bay and Castlerock. Proc. Roy. Irish
Acad. , 76, 6 1 9-628.
SAMPLE, J.C., REID, M.R., TOBIN, H.J. & MOORE, J.C.
( 1 993) Carbonate cements indicate channeled fluid flow
along a zone of vertical faults at the deformation front
of the Cascadia accretionary wedge (north-west U.S.
coast). Geology, 21, 507-510.
SEARL, A. ( 1 9 9 1 ) Early Dinantian dolomites from East
Fife: hydrothermal overprinting of early diagenetic
fabrics? J. Geo/. Soc. London, 148, 73 7-747.
SEARL, A. ( 1 994) Diagenetic destruction of reservoir po­
tential in shallow marine sandstones of the Broadford
Beds (Lower Jurassic), north-west Scotland: deposi­
tional versus burial and thermal history controls on
porosity destruction. Mar. Petrol. Geol. , 1 1 , 1 3 1 - 1 47.
SEARL, A. & FALLICK, A. E. ( 1 990) Geochemistry of some
Dinantian dolomites from East Fife: hydrothermal
overprinting of early mixing-zone stable isotopic and
Fe/Mn compositions. J. Geol. Soc. London, 147 , 623-
638.
SIBLEY, D.F. & GREGG, J.M. ( 1 9 87) Classification of
dolomite rock textures. J. sediment. Petrol. . 57, 967-
975.
SIBSON, R.H. ( 1 9 8 1 ) Fluid flow accompanying faulting:
field evidence and models. In: Earthquake Prediction:
an International Review (Eds Simpson, D.W. & Rich­
ards, P.G.). Am. Geophys. Union, Maurice Ewing
Series, 4, 593-603.
SIBSON, R.H. ( 1 990) Conditions for fault-valve behaviour.
In: Deformation Mechanisms, Rheology and Tectonics
(Eds Knipe, R.J. & Rutter, E.H.). Spec. Publ. Geol. Soc.
London, 54, 1 5-28.
SIBSON, R.H., MOORE, J.M. & RANKIN, A. H. ( 1 975) Seismic
pumping-hydrothermal fluid transport mechanism. J.
Geol. Soc. London, 1 3 1 , 653-659.
SNOWDON, L.R. ( 1 995) Multiphase systems, overpressure
and migration. Bull. Can. Petrol. Geol. , 43, 3 1 5-3 1 9.
SP6TL, C. & WRIGHT, V.P. ( 1 992) Groundwater dolocretes
435
from the Upper Triassic of the Paris Basin, France: a
case study of an arid, continental diagenetic facies.
Sedimentology, 39, 1 1 1 9- 1 1 37.
SULLIVAN, M.D., HASZELDINE, R.S. & FALLICK, A.E. ( 1 990)
Linear coupling of carbon and strontium isotopes in
Rotliegend Sandstone, North Sea: evidence for cross­
formational fluid flow. Geology, 18, 1 2 1 5- 1 2 1 8 .
TABERNER, C. & SANTISTEBAN, C. ( 1 987) Mixed water
dolomitization in a transgressive beach-ridge system,
Eocene Catalan Basin, NE Spain. In: Diagenesis of
Sedimentary Sequences (Ed. Marshall, J.D.). Spec. Publ.
Geol. Soc. London, 36, 1 2 3- 1 39.
TAYLOR, T.R. & SIBLEY, D.F. ( 1 9 86) Petrographic and
geochemical characteristics of dolomite types and the
origin of ferroan dolomite in the Trenton Formation,
Ordovician, Michigan Basin, USA. Sedimentology, 33,
6 1 -86.
TuCKER, M.E. & WRIGHT, V.P. ( 1 990) Carbonate Sedimen­
tology. Blackwell Scientific Publications, Oxford.
TURNER, P., JONES, M., PROSSER, D.J., WILLIAMS, G.D. &
SEARL, A. ( 1 9 93) Structural and sedimentological con­
trols on diagenesis in the Ravenspur North gas reser­
voir, UK southern North Sea. In: Petroleum Geology of
Northwest Europe: Proceedings of the 4th Conference
(Ed. Parker, J.R.), pp. 7 7 1 -785. Geological Society of
London Publishing House, Bath.
UNDERHILL, J.R. & WOODCOCK, N.H. ( 1 987). Fault mech­
anisms in high porosity sandstones; New Red Sand­
stone, Arran, Scotland. In: Deformation of Sediments
and Sedimentary Rocks (Eds Jones, M.E. & Preston,
R.M.). Spec. Publ. Geol. Soc. London, 29, 9 1 - 1 05 .
UPTON, J.C., FITTON, J.G. & MACINTYRE, R.M. ( 1 9 87) The
Glas Eilean Lavas: evidence of a lower Permian
volcano-tectonic basin between !slay and Jura, Inner
Hebrides. Trans. Roy. Soc. Edin. , 77, 289-293.
WANG, W.H. ( 1 992) Origin of reddening and secondary
porosity in Carboniferous sandstones, Northern Ire­
land. In: Basins on the Atlantic Seaboard: 'Petroleum
Geology, Sedimentology and Basin Evolution (Ed. Par­
nell, J.). Spec. Publ. Geol. Soc. London, 62, 243-254.
WEBER, K.J. & VAN GEUNS, L.C. ( 1 990) Framework for
constructing clastic reservoir simulation models. J.
Petrol. Techno/., 34, 1 248-1297.
WILKINSON, M., CROWLEY, S.F. & MARSHALL, J.D. ( 1 992)
Model for the evolution of oxygen isotopic ratios in the
pore fluids of mudrocks during burial. Mar. Petrol.
Geol. , 9, 98-l 05.
WILSON, H.E. & ROBBIE, J.A. ( 1 966) Geology ofthe Country
around Bal/ycast/e. Mem. Geol. Surv. N. Ireland, 8.
HMSO, Belfast.
WooD, J.R. & BOLES, J.R. ( 1 99 1 ) Evidence for episodic
cementation and diagenetic recording of seismic pump­
ing events, North Coles Levee, California, USA. Appl.
·
Geochem. , 6, 509-52 1 .
WooD, S.H. ( 1 985) Regional increase in groundwater
discharge after the 1 983 Idaho earthquake: coseismic
strain release, tectonic and natural hydraulic fracturing.
USGS Open-File Report, 85-290, 5 7 3-592.
Spec. Pubis int. Ass. Sediment. ( 1998) 26, 437-460

Saddle (baroque) dolomite in carbonates and sandstones:

a reappraisal of a burial-diagenetic concept

C. SPOTL* a n d J . K . PITMANt
*lnstitut fiir Geologie und Palaontologie, Universitat lnnsbruck,
lnnrain 52, 6020 lnnsbruck, Austria,
e-mail christoph.spoetl@uibk.ac.at; and
tVS Geological Survey, 939 [)enver Federal Center,
Lak(!wood, CO 80225, USA, e-mail jpitman@usgs.gov

AB S TR A C T

Saddle(baroque) dolomite, defined as coarse-crystalline dolospar with regularly to irregularly curved

crystal boundaries and sweeping extinction, has been described from numerous diagenetically altered
carbon!ltes and sandstones in hydrocarbon reservoirs, palaeoaquifers and Mi�sissippi-Valley-type
(MVT) ore deposits. This chapter reviews petrographic, geochemical and tiuid inclusion d,ata on saddle
dolomite from carbonate rocks and sandstones published since 1 980, in order to reassess the original
interpretation of this type of dolomite as a potential high-temperature diagenetic geotqermometer.
The compilation shows t\lat saddle dolomite from various sedimenrary Qasins has the following
characteristics: (i) variable Fe + Mn and Ca enrichment of saddle dolomite hosted in sandstones
relative to saddle dolomite hosted in c;;arbonate rocks; (ii) carbon isotopic compositions ranging from
slightly positive (in carbonate rocks) to moderately negative values (in sandstones); (iii) moderately
negative oxygen isotope values in sandstones and carbonate rocks; and (iv) strontium isotope ratios
commonly more radiogenic than Phanerozoic seawater(> 0. 708). Fluid inclusions in saddle dolomite
homogenize at temperatures;;;;. 60-80'C, with a maximum between 90 and 1 60'C. T\le salinity of the
palaeofluids is uniformly greater than that of seawater ("" 1 8-25 wt% NaCI eq.). Low eutectic
temperatures suggest a complex aqueous so)ution dominated by NaCI + CaC12 ± MgC12 ± KCI.
Geochemical and fluid inclusion data, in conjunction with mineral paragenetic information,
demonstrate that saddle dolomite is a reliable indicator of rock-brine interactions at temperatures that
coincide largely with the liquid hydrocarbon 'window' and extend well into the dry gas zone.

I N TR OD U C T I O N

I n 1980 Radke and Mathis published an influential
paper which documented the occurrence of saddle­
shaped dolomite crystals in diagenetically altered
sedimentary rocks. Although this petrographic do­
lomite type had been known for many decades,
their work established the genetic relationship be­
tween deep-burial settings and the occurrence of
saddle dolomite. At that time surprisingly little
documented evidence was available to support this
concept. Reliable fluid inclusion data were scarce
and the understanding of deep-burial diagenesis
was in its infancy. Since 1980, many additional
occurrences of saddle dolomite have been reported
and a number of detailed petrographic and
g�ochemical studies are available that permit a
better understanding of the formation of saddle
dolomite. The purpose of this chapter is to critically
examine these post-1980 studies in order to re­
address two main questions: (i) is saddle dolomite a
valid qualitative indicator of interactions between
rocks and reducing, warm to hot basinal brines, as
originally suggested by Radke and Mathis, and (ii)
did specific physicochemical conditions prevail dur­
ing saddle dolomite precipitation?

Carbonate Cementation in Sandstones Edited by Sadoon Morad 437

© 1998 The International Association of Sedimentologists
ISBN: 978-0-632-04777-2
438 C. Spot! and JK Pitman
DEFINITIONS
Following the widely used petrographic classifica­
tion of Sibley & Gregg ( 1 987), two main types of
dolomite texture can be differentiated, namely pla­
nar and non-planar. The former refers to euhedral
and subhedral dolomite crystals showing straight
crystal boundaries and pinpoint extinction in cross­
polarized light, whereas the latter describes irregu­
lar, curved or anhedral crystals, often showing
undulatory extinction. Saddle dolomite, regarded as
a subcategory of non-planar dolomite (Gregg &
Sibley, 1 984; Sibley & Gregg, 1 987), is identified in
thin section by a combination of features, including
coarse crystal size(> 1 00 Jlm up to several millime­
tres in diameter), curved (saddle-shaped) crystal
boundaries when viewed three-dimensionally, and
sweeping extinction i.e. a peculiar form of undula­
tory extinction whereby the zone of total extinction
in crossed polarized light migrates systematically
across an individual crystal as the microscope stage
is rotated (Pedone, 1 978; Radke & Mathis, 1 980;
Searl, 1 989).
The term saddle dolomite (Radke & Mathis,
1 g8o, their p. 1 1 50) is essentially equivalent to the
baroque dolomite first used in an abstract by Folk &
Assereto ( 1 974) and subsequently introduced by
Folk into the Glossary of Geology(Bates & Jackson,
1 987). There has been some dispute in the past over
which term deserves priority(e.g. Radke & Mathis,
1 98 1 ; Zenger, 1 981). Currently, the term saddle
dolomite appears to be more widespread interna­
tionally, whereas baroque dolomite is still popular
among North American petrographers. In the liter­
ature some saddle dolomites were described as pearl
spar(a mineral collector's term), white sparry dolo­
mite (Beales, 1 97 1 ), late ferroan dolomite cement
(Choquette, 1 97 1 ), late dolomite cement (Leach et
a!., 1 99 1 ), gangue dolomite (Ebers & Kopp, 1 979)
and hydrothermal dolomite (Goldberg & Bogoch,
1 978; Rowan & Leach, 1 989).
TH E D ATAB A S E
Table 1 gives an overview o f the published work
that was analysed for the present review. Only
studies that provide detailed information on the
petrographic and/or compositional characteristics
of saddle dolomite were included in the database.
We also scanned many more research articles in the
recent literature that mention saddle dolomite
(even without using this term). These papers were
not included in the database because they either
lacked analytical data (see below), and/or the au­
thors did not clearly separate saddle dolomite data
from those of other dolomite types. Naturally any
compilation will be far from complete, but we have
done our best to survey the post- 1 980 literature on
this subject (particularly those studies published in
major international journals and indexed in biblio­
graphic databases, including the Science Citation
Index, GeoRef and Chemical Abstracts), and we
apologize for any inadvertent omissions.
Studies involving saddle dolomite are grouped
into three categories based on their host lithology:
carbonate, sandstone and carbonate-sandstone
mixtures. Most saddle dolomite has been reported
from carbonate rocks (Table 1 ), particularly units
associated with MVT ore deposits, whereas only a
few studies provide detailed information on saddle
dolomite in siliciclastic rocks.
The amount and quality of information in these
studies varies greatly and only a few of the more
recent papers applied a combination of several
different methods to decipher the origin of saddle
dolomite. We scanned published work for the fol­
lowing information on saddle dolomite occur­
rences: optical mineralogy, cathodoluminescence
(CL) characteristics, elemental composition, car­
bon, oxygen and strontium isotopic compositions,
and fluid inclusion analysis. In the following sec­
tions we shall address these types of data in detail.
Petrographic features
Saddle dolomite has distinct petrographic charac­
teristics that can be distinguished from other types
of dolospar, and in many samples this dolomite
type can be identified macroscopically. The petro­
graphic features make saddle dolomite a potentially
useful 'index mineral', an aspect that will be ad­
dressed below.
There are two modes of occurrence that charac­
terize saddle dolomite: cement and replacement
(Radke & Mathis, 1 980; Tucker, 1 990). Saddle
dolomite distributed as a pore-lining, void-filling or
fracture-filling cement is typically coarse crystalline,
strongly undulatory(sweeping extinction), and may
be overgrown by either late-diagenetic calcite or
anhydrite (Fig. 1 B-D; see also Plate 1 , facing p.
446). This type of saddle dolomite (xenotopic-C
variety of Gregg & Sibley, 1 984) usually lacks solid
inclusions but may contain abundant fluid inclu-
Table 1. Sources of petrographic and geochemical data for saddle dolomite occurrences{post-1980 publications)

Elemental Stable Strontium Fluid

Reference Unit Age Location composition isotopesa isotopesa inclusionsb

Carbonate rocks
Buchbinder et at. {1984) Zohar and Shderot Fms Dogger, Maim Coastal plain of Israel - 37
Graf{ l 984) Bonneterre Fm. Cambrian SE Missouri Yes
Stueber et at. ( 1 984) Smackover Fm. U. Jurassic GCSB, Louisiana and - - 2
Arkansas
Woronick & Land (1985) Pearsall and L. Glen Rose L. Cretaceous S Texas Yes 30d 5
Fms
Aulstead & Spencer{1985) Keg River Fm. M. Devonian NW Alberta - - - 58/26
Morrow et at. ( 1 986) Manetoe facies Devonian Yukon and Northwest Yes 162 2 rtr•
Territories, Canada �
Henrich & Zankl (1986) Wetterstein Fm. M./U. Triassic S Germany Yes 3 - -
Taylor & Sibley (1986) Trenton Fm. Ordovician Michigan Basin, Ontario Yes 10 - -
�
1i;'
Alsharan & Williams ( 1 987) Shuaiba Fm. L. Cretaceous Abu Dhabi - 6 - -
Lee & Friedman{1987)
McNutt et at. ( 1 987)
Ellenburger Group
Trenton Fm.
L. Ordovician
Ordovician
Texas, New Mexico
Michigan basin, Ontario
-
-
23
-
-
3
35/r
-
�3
2f
-·
Machel (1987) Nisku reef U. Devonian Alberta Yes 13 r/r 1il
Moore et at. (1988) Smackover Fm. U. Jurassic GCSB, Texas and Arkansas Yes ·5 2 -
;::;·
Aulstead et at. {1988) Presqu'ile and Manetoe Devonian W Canada - - l39!15S
facies 2
Searl ( 1988) Oolite Group L. Carboniferous S Wales Yes0 - - - C3-
0
Zenger & Dunham ( 1 988) Siluro-Devonian units Silurian- New Mexico Yes0 I - 9/r ;:s
�
Devonian �
Searl{ 1 989) No information given L. Carboniferous Fife/Scotland, Wales, Yes - - - �
N England ;:s
$:)...
Rowan & Leach {1989) Bonneterre Fm. Cambrian SE Missouri - - - 1 51171 '""'
- �
Sellwood et at. (1989) Great Oolite Dogger S England 16 4 9/4 ;:s
Friedman {1989, 1 994) Lockport Fm. Silurian N Appalachian basin, Yesc !Oh - 2 1 16 a-
New York C)
;:s
Qing & Mountjoy ( 1 989) Keg River Fm. M. Devonian Alberta Yes 20 - r/-i �
Gregg & Shelton (1990) Bonneterre and Davis Fms Cambrian SE Missouri Yes 10 - 1 1 /8
Moldovanyi et at. ( 1 990) Smackover Fm. U. Jurassic GCSB, Arkansas - I
Searl & Fallick (1990) L. Limestone Group and Devonian E Fife, England Yes 2
Calciferous Sandstone
Measures
Wallace (1990) Various units of the Devonian Canning Basin, Yesc 6
Barbwire Terrace W Australia
Kaufman et at. (1990, 1 9 9 1 ) Swan Hills Fm. Devonian Alberta Yes 29 5 22/22
Zeeh{1990) Wetterstein Fm etc. M. and U. S Austria Yes II - r/r
Triassic
Kupecz & Land (1991) Ellenburger Group L. Ordovician W Texas - 6 9
Tobin (1991) Bonneterre Fm. U. Cambrian SE Missouri - - - 26/rj
Schneider et at. (1991) Various units Devonian SW China Yes 4 - r/- �
w
--- \0
Continued
 �
�
0
Table 1. (Cont inued)
Elemental Stable Strontium Fluid
Reference Unit Age Location composition isotopes• isotopes• inclusionsb

Carbonate Rocks (Cont inued)

Mountjoy & Wabamun Group U. Devonian N Alberta - 7 2 69/r
Halim-Dihardja( 1 991)
Muchez et a!. ( 1 99 1 ) Visean carbonates L. Carboniferous N Belgium Yes 4 - 91-
Gao & Land ( 1 99 1 ) Middle Arbuckle Group L. Ordovician SW Oklahoma Yes 6 6
Gao et a!. (1992) Upper Arbuckle Group L. Ordovician SW Oklahoma Yes 8 6
Amthor & Friedman ( 1 992) Ellenburger Group L. Ordovician Permian Basin, W Texas Yes 23k -
Coniglio & Williams-Jones Trenton Fm. etc. M./L. Ordovician Michigan Basin, Ontario - 20 - 39/ 1 7
(1992) and Coniglio
et a!. ( 1 994)
Shelton et a!. (1992) Bonneterre Fm. Cambrian SE Missouri - - -
r/r1
Mountjoy et a!. ( 1 992) Various units of the M. Devonian WCSB, Alberta - 16 20 r/rm
n
Presqu'ile Barrier
Qing & Mountjoy (1992, Various units of the M. Devonian WCSB, Alberta Yes 58 15 14/ 1 3 �
1 994a,b) Presqu'ile Barrier g_
Marchand et a!. ( 1 994) No information given L. Carboniferous E Belgium Yes 10 I r/r �
;::
Nesbitt & Muehlenbachs No information given Cambrian WCSB, Alberta - 36 2n rlr !':>...
(1994) �
Kuhlemann ( 1 995) Wettersteih Fm. etc. M. and U. S Austria and Slovenia - 15 - r/r0 �
Triassic �
Zeeh et a!. (1995; see also Wetterstein Fm. etc. M. and U. Austria -
42 - r/r
§'
Zeeh & Bechstiidt, 1994) Triassic �
;::
Moss & Tucker( 1 99 5) Limestone Fm. L. Cretaceous SE France -
3 - -

Geng & Zeeh (1995) Muschelkalk M. Triassic SW Germany - 3 - r/r

Riciputi et a!. (1996) Nisku Fm. U. Devonian WCSB, Alberta - - - 22/22P
Moritz et a!.( 1 996) Chambani Group U. Triassic Pucara basin, central Peru - I 1q r/r
Suchy et a!. (1996) Pridoli Fm. U. Silurian Barrandian Basin, Czech - 6
Republic

Carbonate-sandstone mixtures
Gawthorpe (1987) Pendleside limestone and L. Carboniferous Bowland Basin, England Yesc 2
sandstone
Spot! et a!. (1993) Sandstones and dolocretes U. Triassic Paris Basin, France Yes 20 16
Drzewiecki et a!. ( 1 994) Glenwood Fm. M. Ordovician Michigan Basin Yes 23 15 691-
and Simo et a!. (1994)
Wojcik et a!. (1992, 1994) Limestones and sandstones U. Carboniferous Cherokee Basin, Kansas Yes 64 - 349/200
and Walton et a!. (1995)

Continued
Table I. (Cont inued)
Elemental Stable Strontium Fluid
Reference Unit Age Location composition isotopes• isotopes• inclusionsb

Sandstones
Rieken & Gaupp (1991) Various units U. Carboniferous Lower Saxony Basin, - 4 - r/r
to L. Cretaceous Germany
Barnes et a/. (1992) St Peter Sandstone M. Ordovician Michigan Basin Yes 14 - r/r
De Craen & Swennen Zechstein conglomerate U. Permian Campine Basin, Yes 27 - r/r �
(1992) NE Belgium �
- - -
<il
Girard & Barnes (1995) St Peter Sandstone M. Ordovician Michigan Basin 67/-
Pitman & Spot!( 1996) St Peter Sandstone M. Ordovician Illinois Basin Yes 7 - r/r �
Spot! et a/. (1996) Spiro Sandstone U. Carboniferous Arkoma Basin, Oklahoma Yes 29 a-
and Arkansas �­
�
WCSB, Western Canada Sedimentary Basin; GCSB, US Gulf Coast Sedimentary Basin. s·
•Numbers given in this column are total number of individual isotope analyses. 8
bThe first number is the total number of homogenization temperatures and the number following the slash gives the number of final melting temperatures.
Sources which only reported the range of fluid inclusion temperatures are indicated by 'r'. g.
;::s
<only very few measurements. !::)
ctworonick & Land ( 1985) reported 38 oxygen isotope and 28 carbon isotope values in graphical form. fi
eMorrow et a/. (1986) quoted data from a fluid inclusion study by Aulstead & Spencer(in press), probably Aulstead & Spencer (1985). !::)
;::s
rData from Anderson(PhD thesis, 1985) cited in Machel(1987). 1:<.
SAlthough the authors only talk about white coarsely crystalline dolomite it can be inferred from sources cited in this paper and from Qing & Mountjoy(1989) 1:)
that most of Aulstead et al.'s(1988) samples are saddle dolomites. ;::s
h No carbon isotope values reported. �
'Quoting data from Aulstead & Spencer ( 1985) and Aulstead et a/. (1988).
0
;::s
j Excluding data from secondary fluid inclusions. �
kSame data as reported earlier by Lee & Friedman (1987).
1Shelton et a/. (1992) reported data from 4 76 fluid inclusions from various types of epigenetic dolomites. Because it was unclear which data points refer to
saddle dolomites only the range of values was regarded here.
mData from Qing(PhD thesis, 1991) in Mountjoy et a/. (1992).
nData from Koffyberg(MS thesis, 1993) in Nesbitt & Muehlenbachs(1994).
0Quoting data from Zeeh & Bechstadt(-1994).
PData from Nahnybida et a/. (1982), Anderson(PhD thesis, 1985) and Machel(PhD thesis, 1985).
qThis value was not used because it apparently is a dolomite-calcite mixture.

""'"
""'"
442 C. Spot! and J.K. Pitman

Fig. 1. Thin-section photomicrographs in cross-polarized light showing characteristic features of saddle dolomite and
ankerite in sandstones and carbonate rock. (A) Ferroan saddle dolomite crystal marginally replacing quartz framework
grains(arrow). Intergranular quartz cement postdated the formation of saddle dolomite. Upper Carboniferous Spiro
Sandstone, Arkoma basin, USA. Scale bar= 200 Jlm. (B) Quartzarenite pervasively cemented by saddle ankerite and
minor pyrobitumen(opaque material). Individual dolomite crystal outlines are poorly developed but sweeping
extinction is widespread. Same unit as(A). Scale bar= 500 Jlm. (C) Fracture-fill saddle ankerite composed of crystals
with well-developed boundaries and strong curvature. Note authigenic quartz (arrows) intergrown with saddle ankerite.
Upper Carboniferous carbonate rock, Ouachita Mountains, USA. Scale bar= 500 Jlm. (D) Void-filling coarse-crystalline
saddle dolomite postdating authigenic quartz. Triassic limestone with epigenetic dolomite and sphalerite-galena­
fluorite mineralization, Eastern Alps(Wanneck area, Tyrol). Scale bar= 500 Jlm.
Continued
 Saddle dolomite in carbonates and sandstones 443

sions, thereby creating a turbid appearance (e.g.
Radke & Mathis, 1 980; Wojcik et a!., 1 994). Re­
placement saddle dolomite (xenotopic-A and -P
varieties of Gregg & Sibley ( 1 984)) shows similar
features. In carbonate rock this mineral replaces
precursor carbonate phases, forming a tight, inter­
locking fabric, and is commonly characterized by
abundant micrometre-sized solid inclusions, mostly
of organic origin (Fig. 1 E,F). Individual crystals
show undulatory to sweeping extinction, the latter
more pronounced in larger crystals (Fig. 1 F). Saddle
dolomite in sandstones commonly replaces adja­
cent framework grains (e.g. quartz; Fig. 1 A,B).
Crystal boundaries are typically slightly curved, but
irregular compromise boundaries are also common
(Fig. 1 B,C). Whereas the identification of saddle
dolomite cement is straightforward in both carbon­
ates and sandstones, transitions exist between sad­
dle dolomite and other non-planar replacement
dolomites that render a clear-cut identification
problematic. For example, non-planar dolomite
crystals may show undulatory extinction yet lack
regularly curved boundaries. Likewise, fine­
crystalline replacement dolomite may display incip­
ient sweeping extinction that is commonly poorly
developed (Gregg & Sibley, 1 984; Lee & Friedman,
1 987; Gregg & Shelton, 1 990; Wallace, 1 990;
Amthor & Friedman, 1 992; Drzewiecki et a!., 1 994;
Simo et a!., 1 994; Pitman & Spot!, 1 996). Accord­
ing to the defin ition given above, we suggest that the
term saddle dolomite be restricted to dolospar
fabrics that show the following petrographic fea­
tures: coarse crystal size (a few hundred microine­
tres up to several millimetres), irregularly to
regularly curved crystal boundaries and sweeping
extinction. Because curved crystal boundaries are
generally poorly developed in the replacement
mode we regard the sweeping extinction pattern as
the most diagnostic petrographic feature of repla­
cive saddle dolomite crystals.
Many authors reported paragenetic relationships
that allow the relative timing of saddle dolomite
precipitation within a given rock unit to be deter­
mined. There is general agreement among these
studies that precipitation occurred after major me­
chanical compaction of the rock and coincided with
chemical compaction, e.g. stylolitization in carbon­
ate rocks. In most basins saddle dolomite precipita­
tion was a late diagenetic event that occurred during
deep burial (at least 1 -2 km). Saddle dolomite is
highly uncommon in sediments at shallow burial
depths, although there is one extreme example
where very fine-crystalline, saddle-shaped proto­
dolomite has been reported in shallow-buried Ho­
locene sediments of Western Australia (Rosen &
Coshell, 1 992). However, it is important to note
that the mineralogical and geochemical characteris­
tics of this protodolomite are unlike those of most
burial dolospar.
CL characteristics
Many authors reported on the qualitative CL char­
acteristics (emission intensity and colour) of saddle
dolomite. These fall into two broad categories, one
group showing moderately bright orange-red CL,
and another group that displays dull brown to non­
luminescence. Luminescent saddle dolomite is
more common in carbonate host rocks (e.g. MVTs),
whereas the non-luminescent type is more typical of
sandstones. Compositional analyses show that the
change in CL intensity in sandstones parallels a
general trend towards ferroan dolomite and anker­
ite (see below), which in turn reflects an increase in
the Fe2+ /Mn 2+ ratio in the dolomite crystal, i.e.
Fe2+ quenching. Although most Fe-poor saddle
dolomites lack CL zoning and show rather uniform
bright or dull emission, saddle dolomites from some
basins, such as those associated with MVT miner­
alizations, exhibit concentric CL zonation (e.g.
Ebers & Kopp, 1 979; Gregg & Hagni, 1 987; Voss et
a!., 1 989; Farr, 1 992); not all .of the dolomite
described in these studies, however, classify as
saddle dolomite. Only a few spectrometric analyses
of CL emission of saddle dolomite have been
performed. In one study (Blanc et a!., 1 994) a CL
spectrum of Mn 2+ -activated saddle dolomite also

Fig. I. (Continued) (E) Coarse-crystalline saddle dolomite replacing host dolomicrospar. Note optical zoning of
dolospar due to variable inclusion densities. Triassic limestone with epigenetic dolomite and sphalerite-galena-fluorite
mineralization, Eastern Alps (Wanneck area, Tyrol). Scale bar� 500 j.lm. (F) Host carbonate rock completely replaced
by non-planar dolospar. Note that about half of the crystals in the field of view show regularly curved crystal
boundaries and/or incipient sweeping extinction, i.e. saddle dolomite. Triassic limestone with epigenetic dolomite and
sphalerite-galena-fluorite mineralization, Eastern Alps(Wanneck area, Tyrol). Scale bar� 500 j.lm. (See also colour
Plate l , facing p. 446.)
444 C. Spot/ and J.K. Pitman
displayed a prominent emission band in the near­
ultraviolet region. Another study (Habermann et
al., 1 996) showed an example of CL in saddle
dolomite activated by rare earth elements.
The ultraviolet fluorescence of 'hydrothermal
saddle dolomite' was briefly reported for samples
from the Ozark region of the USA (Hayes et al.,
1 989). In this investigation, variations in fluores­
cence were found to correlate with the mineral
paragenesis and CL variations. Changes in fluores­
cence were interpreted to be due to an interplay of
activators(Pb, Mn) and quenchers (Fe).
Elemental composition
Figures 2-4 summarize the currently available in­
formation on the stoichiometry and minor or trace
element concentration of saddle dolomite in car­
bonates, sandstones and carbonate-sandstone mix­
tures. Saddle dolomite shows a variable elemental
composition, from near-stoichiometric to calcian
dolomite and/or ankerite. The bulk of published
major, minor and trace element data on saddle
dolomite are based on electron microprobe analy­
ses, although some authors used powder X-ray
diffraction analysis to determine the dolomite stoi­
chiometry. Others used spectrometric techniques
(e.g. atomic absorption) to analyse for trace ele­
ments. We found it difficult to evaluate this pub­
lished information because some authors report
bulk analyses of several crystals, whereas others
report intracrystalline elemental variations. More­
over, analyses are often incomplete(e.g. wto/o Fe and
Mn; no data on Ca and Mg) and cannot be directly
recalculated in terms of mol%. Further, some au­
thors report a single measurement whereas others
list hundreds of data points.
The stoichiometry of saddle dolomite varies
widely, from 48 to 64 mol% CaC03 . Most contains
,;:; 54 mol%, which is consistent with data from
Radke & Mathis ( 1 980). The mol% CaC03 in
saddle dolomite from sandstones overlaps the com­
positional range reported for carbonate-hosted sad­
dle dolomite, although that in carbonate rocks tends
to show slightly less excess Ca (Fig. 2).
The values of Fe and Mn reported for saddle
dolomite are presented together in Fig. 3 because
the geochemical properties of these elements are
similar. Fe+ Mn concentrations in saddle dolomite
show wide variability, from < 1 0 ppm to 30 mol%
(Fig. 3), reflecting in part the compositional zoning
in the saddle dolomite structure. In sandstones,
saddle dolomite commonly displays an Fe enrich­
ment approaching that of Fe-dolomite and ankerite
(i.e. > 20% FeC03+ MnC03 ; Fig. 3), whereas most
saddle dolomite in carbonate rocks shows signifi­
cantly lower Fe+ Mn concentrations (commonly
less than 1 mol% (Fe,Mn)C03). Several studies
show that increasing Fe+ Mn concentrations in
saddle dolomite correlate with decreasing amounts
of excess Ca and reflect the effects of increased
temperature (e.g. Wojcik, 1 99 1 ; Spot! et a!., 1 996).
This substitutional trend has also been reported for
other ankerites (e.g. Boles, 1 978; McDowell &
Paces, 1 985; Land & Fisher, 1 987; Dutton & Land,
1 98 8), but the large scatter of the data implies that
factors other than temperature also play a role, e.g.
Ca/Mg ratio and Fe2+ activity of the pore fluid.
Several authors have reported Sr values of saddle
dolomite in carbonate rock that range from < 1 0 to
1 52 ppm (Fig. 4). Currently no Sr data on saddle
dolomite from sandstones are available, partly be­
cause individual crystals are difficult to obtain, but
also because Sr concentrations are typically below
instrument (microprobe) detection limits. A few
data on other trace elements in saddle dolomite are
available. Concentrations of Na, K, Rb and rare
earth elements(Graf, 1 984; Henrich & Zankl, 1 986;
Moore et al., 1 988; Zenger & Dunham, 1 988; Qing
& Mountjoy, 1 989, 1 994b; Kaufman et al., 1 990;
Wallace, 1 990; Zeeh, 1 990; Schneider et a!. , 1 99 1 ;
Sima et al., 1 994) are typically low and similar to
values reported for other dolomite varieties (e.g.
Veizer, 1 983). Qing & Mountjoy ( 1 994b) did a
detailed analysis of rare earth element contents in
various dolomite types in regional dolostones from
the Western Canada Sedimentary Basin including
saddle dolomite, and concluded that elemental
traces provide important constraints on the fluid/
rock ratio during regional dolomitization(and asso­
ciated MVT mineralization).
Compositional zoning is common in saddle dolo­
mite crystals, particularly when they are enriched in
Fe. For example, Wojcik ( 1 9 9 1 ) and Walton et a!.
( 1 99 5) documented multiple zones separated by
resorption boundaries using back-scattered electron
microscopy, and a peculiar type of compositional
zoning in saddle dolomite was reported by Searl
( 1 9 89). Based on back-scattered electron micros­
copy Searl identified an oscillatory pattern along the
edges of saddle dolomite crystals consisting of zones
depleted in Ca and Fe relative to Mg alternating
with zones of 'normal' composition. This chevron­
like zonation pattern, which to our knowledge has
 Saddle dolomite in carbonates and sandstones 445

CARBONATES
Woronick & Land (1985)
Taylor & Sibley (1986)
Zenger & Dunham (1988)
Searl (1988)
0 Friedman (1989)
Searl (1989)
0 Wallace (1990)
Gregg & Shelton (1990)
Gao & Land (1991)
while poie-filling saddle dolomite

: ...... Gao & Land (1991)

nodular sadc/le dolomite
Amthor & Friedman (1992)
$lldd/e dolOmite ciJfTI6nt

Amthor & Friedman (1992)

replacive saddle dolomite

- Gao et al. ( 1992)

'
Wojcik et al. (1992)
saddle dolomite in limestone

CARBONATE­
SANDSTONE
MIXTURES
0 Gawthorpe (1987)
Spall et al. (1993; and
unpublished data)

SANDSTONES
Wojcik et al. (1992)
saddle dolomilfl in sandstone

• De Craen & Swennen (1992)

Pitman & Spall (1996)

Spall et al. (1996)

48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

CaC03 (mol%)

o Single measurement
__. Mean and total range
1----1 Mean ± 2 standard deviations

Fig. 2. Chart summarizing the stoichiometry of saddle dolomite. The dashed vertical line indicates stoichiometric
dolomite. Data from Wojcik et a!. (1992) on saddle dolomite and ankerite are plotted in the limestone and sandstone
categories, rather than the mixed category (see Table I). Data published by Searl & Fallick (1990) could not be
included because of ambiguous data identification.

not been reported by other workers, has been
suspected to cause the presumed lattice distortion
in saddle dolomite crystals (Searl, 1989).
An interesting feature of some saddle dolomite is
the presence of anhedral inclusions of calcite rang­
ing in diameter from < 1 llm to a few tens of
micrometres. Using electron microprobe tech­
niques, Radke & Mathis ( 1 980) documented the
presence of discrete calcite. Such inclusions can also
be readily identified using back-scattered electron
microscopy and CL (Gawthorpe, 1987; Spot! et al. ,
1993, 1 996). The origin of this calcite is poorly
understood, but one is reminded of the generally
smaller domains of calcitic material observed in
some calcian saddle dolomite using transmission
electron microscopy techniques. According to
446 C. Spot! and iK. Pitman

CARBONATES
Woronick & Land (1985)
Morrow et al. (1986) •

Henrich & Zankl (1986)"

Taylor & Sibley (1986)
Moore et al. (1988)"
Searl (1988)
0 Zenger & Dunham (1988)
Searl (1989)
Qing & Mountjoy (1989)"
Gregg & Shelton (1990)
0 .Wallace (1990)
Kaufman et al. (1990)"
Zeeh (1990)"
Schneider et al. (1991)"
Muchez et al. (1991)"
- Gao & Land (1991)
white pore-filling saddle dolomite
- Gao & Land (1991)
nodular saddle dolomite

Wojcik et al. (1992)

saddle do/omile in limestone

... Gao et al. (1992)

Amthor & Friedman (1992)
saddle dolomite cement

Amthor & Friedman (1992)

rep/acive saddle dolomite

Qing & Mountjoy (1994b)"

Marchand et al. (1994)"

CARBONATE­
SANDSTONE
MIXTURES
-- : Gawthorpe (1987)
Spall et al. (1993; and
unpublished data)
Sima et al. (1994)"

SANDSTONES
Wojcik et al. (1992)
saddle dolomite in sandstone

Barnes et al. (1992)

•• De Craen & Swennen (1992)
.
Pitman & Spall (1996)
- Spall et al. (1996)
'

r-�--����--�,-���--���r-�--+-�rTT �
0.001 0.01 0.1 10 100·

(Fe, Mn)C03 (mol%)

Fig. 3. Compilation of iron and manganese concentrations in saddle dolomite (note logarithmic scale). Same key as in
Fig. 2. The dashed line marks the boundary between Fe-dolomite and ankerite according to Deer et a!. (1992). Some
studies provide Fe and Mn concentrations (reported as wt% or ppm) but lack Ca and Mg data. To achieve consistent
presentation, Fe and Mn values were recalculated as mol% (Fe, Mn)C03, assuming a dolomite composition of
50 mol% CaC03 (marked by an asterisk). Excess Ca results in systematically low (Fe,Mn)C03 values, but the error is
negligible. Data from Wojcik et a!. (1992) on saddle dolomite and ankerite are plotted in the limestone and sandstone
categories, rather than the mixed category (see Table 1). Data published by Searl & Fallick (1990) could not be
included because of ambiguous data identification.
 Saddle dolomite in carbonates and sandstones 447

CARBONATES
Morrow et al. (1986)
--------• Henrich & Zankl (1986)

Moore et al. (1988)

0 Zenger & Dunham (1988)
Qing & Mountjoy (1989)
0 Wallace (1990)
Kaufman et al. (1990)
Schneider et al. (1991)
- Muchez et al. (1991)
Gao & Land (1991)
while pt:Xf1·filling s11ddle dolomile

Gao & Land (1991)

nodular saddle dolomiriJ

• Gao et al. (1992)

Amthor & Friedman (1992)
saddle do/omit& cemant
Amthor & Friedman (1992)
replacive ssdd/a dolomite
Marchand et al. (1994)

CARBONATE­
SANDSTONE
MIXTURES
Sima et al. (1994)

0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150

Sr (ppm)
Fig. 4. Strontium concentrations in saddle dolomite. Same key as in Fig. 2. Note that there are no published data on
Sr contents in saddle dolomite from sandstones.

Barber et al. ( 1 98 5), discrete calcite in saddle dolo­
mite occurs as crystallographically coherent planar
defects, which in turn cause some distortion of the
crystal lattice. Lattice distortion, however, is not
necessarily the sole cause of the gross distortion of
the overall saddle dolomite crystals, although both
processes appear to be related (Barber et al., 1 985).
Stable isotopic composition
There is a fairly extensive stable isotope data set
available on saddle dolomite which is summarized
in Figs 5-7. For the purpose of comparison in
this study, 8180 values reported relative to SMOW
were converted to PDB using the relationship
0 180PDB 0. 97QQ2o 180sMOW - 29 . 9 8 (Coplen et
= tion temperature ( 1 27-146 oC) were reported for a
al., 1983). In some studies, 8180 values were cor­
rected for dolomite-phosphoric acid fractionation
effects (following Sharma & Clayton, 196 5 ; more
recently Rosenbaum & Sheppard, 1 98 6). In order to
achieve a consistent dataset, these data are reported
uncorrected. In addition, in studies where it is
unclear whether a fractionation factor was applied,
the 8 180 values are assumed to be uncorrected. The
current data set is slightly biased toward the Devo­
nian of the Western Canada Sedimentary Basin
( 3 6% of all data points), where saddle dolomite is
widespread and well characterized (see below).
A few studies reported considerable intracrystal­
line isotopic variations within millimetre-sized
saddle dolomite crystals. For ex,ample, a 1.4o/oo
variation in o180 and a 0. 7o/oo variation in o 13C
were found in isotopically zoned saddle dolomite
crystals from Cretaceous carbonates of south Texas
(Woronick & Land, 1985). In another study, a
5.2o/oo decrease in o 180 (-6 �o -11.2o/oo) and a
systematic increase in fluid inclusion homogeniza­
single saddle dolomite crystal in Devonian rocks of
Canada (Kaufman et al., 1990). Further, Spangen­
berg et al. ( 1 995) reported preliminary stable iso­
tope results for intracrystalline heterogeneities in
white sparry dolomite (largely equivalent to saddle
dolomite) in samples from a Peruvian MVT
448 C. Spot! and J.K. Pitman

° Carbonate rocks
o Carbonate-sandstone mixtures

5 • Sandstones

m-
e
c..
0
� 0
- Fig. 5. Stable isotopic
(.) -1 compositions of saddle dolomite.
(') Note the tendency toward negative
� -2
carbon isotope values from
-3 siliciclastic settings. Data from
Friedman (1989) could not be
-4 included because the author did
not report 813C values, and data
-5
from Woronick & Land (1985)
-6 are not included because of
• 0
• 0
inadequate data presentation. Data
from Wojcik et a!. ( 1992) on
saddle dolomite and ankerite are
-21 -20 -19 -18 -17 -16 -15 -14 -13 -12 -11 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 plotted in the limestone and
sandstone categories, rather than
()1 so (%o, PDB) the mixed category (see Table I).

140
N=818
D Carbonate rocks
12 0
• Carbonate-
"C sandstone mixtures
Q)
N • Sandstones
>-
co 100
!:
ctS
(/J
Q) 80
0..
E
ctS
(/J 60
-
0
....
Q)
.0 40

E Fig. 6. Frequency histogram of

::J
z 81 3C values of saddle dolomite.
20 Note that a significant portion of
the samples, particularly saddle
dolomite in carbonate host rocks,
overlaps the carbon isotopic
-13 -12 -11 -10 -9 -8 -7 -6 -5 -4 -3 ·2 -1 0 3 5 6 7 composition of Phanerozoic
seawater (see Lohmann & Walker,
81 3C (%o, PDB) 1989).
 Saddle dolomite in carbonates and sandstones 449

N=838 Most high-temperature dolomite

I
1 60
'C
D Carbonate rocks
� 1 40
• Carbonate­
Most/ow­
temperature
(ij sandstone mixtures dolomite
c: (-2.510 + 9%.)
ca 1 20 • Sandstones

(/)
Cl)
c. 100

E
ca
(/) 80
-
0
Fig. 7. Frequency distribution of ...
Cl)
60
8 1 80 of saddle dolomite. This .Q
diagram also includes data from E 40
Friedman (1989) which could not ::::J
be included in Figs 5 and 6 z
because no 81 3C values were 20
reported. Most saddle dolomite
shows oxygen isotopic
compositions that fall into the
range of high-temperature -22 -21 -20 - 1 9 -1 8 - 1 7 - 1 6 - 1 5 - 1 4 - 1 3 - 1 2 - 1 1 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0
dolomite according to Allan &
Wiggins (1993). 81 so ('roo, PDB)

deposit. To our knowledge, however, no systematic
study of intracrystalline stable isotope variations in
multiple samples of saddle dolomite has been con­
ducted.
Carbon isotope values of saddle dolomite in
carbonate and siliciclastic rocks range from - 1 1 . 1 to
+7.0%o, with a mean of -0.4 ±2.6%o (n 818;
Figs 5 and 6). The tendency of saddle dolomite in
carbonates to consistently show more positive o13C
values (-1 to +5%o; Fig. 6) indicates a marine
carbon source, probably reflecting the buffering of
carbon by interaction with the carbonate host rock.
In contrast, the dominance of negative o13C values
in sandstone-hosted saddle dolomites (-7 to -3%o;
Fig. 5) is attributed to an input of light carbon from
thermally decarboxylized organic matter; o13C val­
ues are thus dominated by organic carbon reser­
voirs. Extreme positive and negative o13C values,
diagnostic of early diagenetic methanogenesis and
methane oxidation, respectively, are missing in this
dataset.
The o180 values of saddle dolomite range from
-20. 1 to -1.6%o, with a mean of -9.4±2.9%o
(n 8 3 8 ; Fig. 7). No relationship exists between the
= Cambrian and Devonian carbonate strata increase,
oxygen isotopic composition of saddle dolomite
and host-rock lithology, suggesting that saddle do­
lomite precipitation occurred over a similar temper-
ature range and/or from fluids of similar oxygen
isotopic composition. Based on an extensive litera­
ture survey of both low- and high-temperature
dolomite, including some saddle dolomite occur­
rences, most high-temperature dolomite is charac­
terized by o180 < -6.5%o, whereas most low­
= temperature dolomite has o180 values > -2.5%o
(Allan & Wiggins, 1993). The majority of saddle
dolomite values fall within a range of -6 to - 1 2%o,
which suggests precipitation either from low­
temperature highly 8180-depleted pore waters or
from high-temperature pore fluids of variable o 180
composition (see Allan & Wiggins, 1993), or some
combination of both. There is general agreement
that the second option is most likely, though there is
little direct information on the oxygen isotopic
composition of the precipitating pore water.
One of the most extensively studied regions
where saddle dolomite is widespread is the Western
Canada Sedimentary Basin and adjacent Rocky
Mountain thrust belt. Along an E-W traverse from
the deeply buried thrust portion of the basin on to
the foreland, the o180 values of saddle dolomite in
whereas fluid inclusion homogenization tempera­
tures and 87Sr/86Sr ratios decrease (Qing &
Mountjoy, 1992, 1 994a) (Fig. 8). According to a
450 C. Spot/ and J.K. Pitman

Foreland <
Updip fluid flow Orogen (Table I; Fig. 9). In order to compare results from
(Canadian shield) (Cordilleran orogen)
different studies, 87Sr/86Sr ratios were adjusted to a
-6 ,-
----�
-
value of 0. 7 1 024 relative to the Sr isotopic standard

��
m SRM 987 (formerly NBS 9 87).
c
a. -10 The Sr isotope database for saddle dolomite
0 0 co consists of I 06 analyses plus a few minimum and
� oo 0 0 0
�
- maximum values for datasets that were not pub­
-14
s
0 lished in full. Note that none of the sandstone
co
� 8 studies reported Sr isotope data, and only four
-18
studies of carbonate-sandstone mixtures contain
such data (Table I). As a consequence, the current
zoo
dataset is of limited significance. Nevertheless, the
-
() 0 data depicted in Fig. 9 reveal some interesting
� 0 0 aspects. With a single exception, all of the saddle
J::
160 0
1- 0 0 dolomites are more radiogenic than 0. 708 and
1:
ctl
0 about one-third of the dolomites show ratios
Q) 120
Oef)
0 > 0. 7 1 0. Values in excess of 0. 7092 reflect precipi­
:E 0 tation from 87Sr-enriched pore waters rather than
8
80
initial (syndepositional) seawater (e.g. Smalley et
al. , 1994). The current Sr isotope database is biased
0.711
toward samples from lower Palaeozoic formations
0
(Table I), a period characterized by relatively high
.... 0.710 marine 87Sr/86Sr ratios ( 0. 7082-0. 7092).
�

CJ)
"' There is no clear-cut relationship between the Sr
1:: 8
CJ) 0 0 isotopic composition of saddle dolomite and its
r-. 0.709 0 host-rock type (Fig. 9). It could be argued that pore
co

oo 0 waters in siliciclastic rocks are generally enriched in

0.708 ]__R
114 116 118 120
Longitude (E-W)
Fig. 8. Isotope and fluid inclusion parameters for saddle
dolomite hosted in Devonian carbonates of western
Canada (modified from Qing & Mountjoy, 1994a, their
Fig. 13). Data show systematic changes from the deeply
buried portion of the basin to the shallower foreland.
palaeohydrological model of the basin (Qing &
Mountjoy, 1 994a), saddle dolomite precipitated
from hot basinal fluids moving eastward, updip on
to the Canadian shield. Gradual cooling resulted in
the formation of saddle dolomite with successively
more positive 8180 values. The ultimate source of
these fluids and the role of topographically driven
flow, however, remains, unresolved (e.g. Nesbitt &
Muehlenbachs, 1 994, 1 995; Qing & Mountjoy,
1 995; Yang et al., 1 995).
Strontium isotopic composition
The Sr isotopic composition of saddle dolomite has
been reported in only a small number of studies
87Sr relative to pore waters in marine carbonate
124 rocks because of the higher abundance of unstable,
highly radiogenic detrital silicate minerals. The fact
that saddle dolomite from both types of host rocks
is commonly enriched in 87Sr relative to the marine
ratios suggests a significant import of radiogenic,
shale-derived allochthonous fluids into permeable
carbonates during burial.
Saddle dolomite is among the most 87Sr-enriched
diagenetic carbonates, and most studies invoke a
model whereby saddle dolomite precipitated from
highly evolved basinal brines. during deep burial
(e.g. Kaufman et al., 1 990; Kupecz & Land, 199 1 ;
Mountjoy & Amthor, 1994). Fluids acquired their
high 87Sr/86Sr ratios during advection and concom­
itant interaction with siliciclastic rocks (breakdown
of detrital feldspars, alteration of micas, and diage­
netic transformation of clay minerals, e.g. Chaud­
huri & Clauer, 1 993). When these 87Sr-enriched
fluids migrated, either due to initiation of a topo­
graphically driven flow regime or as a result of
tectonic compression, they mixed with less radio­
genic formation waters. The Canadian saddle dolo­
mites described in the previous section would seem
to have formed in such a mixing system (Qing &
 Saddle dolomite in carbonates and sandstones
20 ,-------�
"C
� 1
5
iii
c
as
U)
Cl)
0..
E 10
as
U)
Fig. 9. Compilation of Sr isotope -
0
ratios of saddle dolomite. ...
Horizontal arrows indicate studies Cl)
that reported only a range of ..Q 5
value.s for saddle dolomite
E
:::::1
occurrences in carbonate rocks. z
The total range of Phanerozoic N=106
seawater 87Sr/86Sr ratios is shown
for comparison (Smalley et at., 0 -t--�...-+-+��-t-
1994). No Sr isotope data are 0.706
available for saddle dolomite in
sandstones.
Mountjoy, 1 992, 1 994a) (Fig. 8).
There are two notable examples where the 87Sr/
86Sr ratio of saddle dolomite is demonstrably lower
than the ratio of the presumed syndepositional
seawater. Saddle dolomite from the middle Ar­
buckle Group of Oklahoma (Lower Ordovician)
shows ratios between 0.7085 and 0.7087, which are
lower than the Ordovician seawater ratio of
0.7089-0.7092 (Gao & Land, 1 99 1 ; adjusted to an
SRM 9 8 7 value of 0. 7 1 024). The authors dismissed
the possibility that this saddle dolomite formed
from basinal brines derived from the Anadarko
basin, but instead advocated precipitation either
from descending meteoric water that dissolved
dolomite at unconformities, or by 'meteoric
solution-compaction' of pre-existing host dolomite.
Similarly, in a companion study (Gao et al. , 1 992),
87Sr/86Sr ratios in saddle dolomite of the upper
Arbuckle Group were reported to be lower than
those of late diagenetic dolomite and to partially
overlap the ratios of early diagenetic dolomite. The
authors also favoured a meteoric origin for ihe sad­
dle dolomite, and ruled out the involvement of
basinal brines except to explain the radiogenic na­
ture of the late-stage dolomite. Thus, in both studies,
low o 180 values of saddle dolomite are cited in sup­
port of a meteoric origin of the fluids, but direct
evidence for such fluids (e.g. by isotopic analysis of
fluid inclusions in saddle dolomite), is lacking.
45 1
Phanerozoic
seawater
D Carbonale rocks
II �:n����=-mixtures
Nesbitt & Muehlenbachs (1994)
Morrowe1al. (1986)
Mountjoy & Halim-Dihardja (1991)
0.707 0.708 0.709 0.710 0.7 1 1 0.712 0.7 1 3 0.714
8 7Srf8 6Sr
Fluid inclusions
Analyses of fluid inclusions in saddle dolomite were
first reported for the Pine Point lead-zinc deposit in
western Canada (Roedder, 1 96 8), but it was not
until the late 1 980s that fluid-inclusion analysis was
widely applied to the study of carbonate cements.
Carbonate is generally regarded as a less reliable
host mineral for fluid inclusions than quartz, in
particular because of its perfect cleavage. When
heated beyond their entrapment temperature, fluid
inclusions in carbonate minerals are prone to re­
equilibration by plastic deformation (stretching) or
to decrepitation (e.g. Barker & Goldstein, 1 990;
Prezbindowski & Tapp, 1 99 1 ). Although re­
equilibration due to overheating has· been suggested
(e.g. Girard & Barnes, 1 99 5), most studies listed in
Table I argue against such alteration. In addition, a
few studies documented a systematic increase in
homogenization temperature ( Th) from crystal core
to rim (Aulstead & Spencer, 1 985; Kaufman et al.,
1 990), which would not be expected if re­
equilibration had occurred.
Previous workers studying fluid inclusions in
saddle dolomite used mainly standard microther­
mometric heating-cooling methods. They reported
Th values and, less commonly, salinity estimates
and major cation ratios. More sophisticated tech­
niques, such as direct analysis of the major cations,
452 C. Spot! and J.K Pitman

anions, gases and organic compounds, as well as nolds, 1 994 ). However, liquid inclusions lacking a
isotopic analysis, have only rarely been performed water vapour bubble at room temperature have
on saddle dolomite (e.g. Buchbinder et al., 1984; been reported (Girard & Barnes, 1 995), but were
Nesbitt & Muehlenbachs, 1 994; Yang et al., 1 995; attributed to metastability phenomena. Some au­
Edon et al., 1 996; Kesler et al., 1 996). thors reported Th values of saddle dolomite ap­
Published Th measurements on primary fluid proaching and exceeding 200 " C (Morrow et al.,
inclusions in saddle dolomite (Fig. I 0; Table I ) are 1 986; Lee & Friedman, 1987; Zenger & Dunham,
primarily from three basins in North America, 1988; Mountjoy & Halim-Dihardja, 1 99 1 ; Rieken
namely SE Missouri ( 1 7%), the Western Canada & Gaupp, 1 991; Tobin, 1 99 1 ; Nesbitt & Muehlen­
Sedimentary Basin (29%) and the Cherokee basin in bachs, 1 994). Although some of these data may
Kansas (32%). The following trends are recognized indicate true minimum precipitation temperatures
in the data: (i) most inclusions homogenized be­ (e.g. saddle ankerite from hydrothermally altered
tween 90 and 1 60 " C (statistical mean 1 36±37 " C, sandstones in the deep German basin; Rieken &
median 1 27 • C); (ii) virtually no inclusions homog­ Gaupp, 1 99 1 ), it seems more likely that most of
enized at temperatures below 60-8o·c; and (iii) a these high temperatures are due to re-equilibration
few measurements indicate Th values in excess of as a consequence of increasing burial temperatures,
2oo · c. Provided that most inclusions are of pri­ leakage due to deformation, and/or in situ re­
mary origin (see Roedder, 1 984; secondary and equilibration during measurement (heating) of the
pseudosecondary inclusions are not included in the inclusions. Ideally, such high Th values should be
present review), these data indicate saddle dolomite verified by analysing fluid inclusions in coexisting
precipitation from high-temperature flu ids. This mineral cements with a higher resistance to re­
interpretation is supported by the lack of all-liquid equilibration, such as quartz.
inclusions, which are diagnostic of entrapment at Defining accurate trapping temperatures of indi­
temperatures below 40-5o · c (Goldstein & Rey- vidual saddle dolomite samples leads to questions

300
N= 1 1 1 9 Rieken & Gaupp (1991)

Nesbitt & Muehlenbachs (1 994)

'0
Q) Zeeh et al. (1 995)
N Shelton et al. (1 992)
>
(ij 250 Machel (1 987)

1: Coniglio & Williams-Jones ( 1992)

ca Zeeh (1 990)

1/) Moritz et at. (1 996)

1: Marchard et al. (1 994)
0
·c;;
-- Geng & Zeeh (1995)
200
-- Schneider et at. {1991)
::I
(j r-
1: -

'0
150
::I
;:
> r-
...
ca t-- Fig. 10. Frequency histogram of
E -
primary fluid inclusion
·;: 100 homogenization temperatures
a.
reported for saddle dolomite.
- -
0 Horizontal arrows indicate
... -...,. literature sources that reported a
-
Q)
.c 50
range of values only. Data from
E . ; r-r- Coniglio & Williams-Jones (1992)
::I are mostly from inclusions of

�
z secondary or indeterminate origin

0 f---.. 1 I�
and are shown for comparison.
The range of values in Shelton et
50 80 1 00 120 140 1 60 180 200 220 240 260 280 300 a/. ( 1 992) include data from saddle
dolomite and from other
Homogenization temperature (0C) epigenetic dolomite varieties.
 Saddle dolomite in carbonates and sandstones 453

about pressure correction. Some authors corrected with a well-defined maximum between - 1 4 and
the Th values of saddle dolomite inclusions (e.g. -24 " C (statistical mean - 1 8. 8 ± 4 . 9 " C, median
Zeeh et a!., 1 99 5), whereas others assumed that the - 1 9. 3 C). None of the inclusions showed salinities
•

fluid was close to saturation with respect to meth­ less than seawater. T values were converted to
m

ane, and that such a correction would therefore wto/o NaCl equivalents using the relationship of
result in erroneously high entrapment temperatures Bodnar ( 1 99 3 ), which is only a model for T m

(following Hanor, 1 9 80). Although the common temperatures below the eutectic temperature ( Te) of
association of saddle dolomite and bitumen sug­ the H 2 0-NaCl system (at 23.3 wto/o NaCl). The
gests the latter interpretation, identification of results suggest that most saddle dolomite formed
methane by standard decrepitation methods and its from brines of - 1 8-25 wt% NaCl eq., which is
quantification using a laser Raman microprobe are -5-7 times higher than modern seawater salinity.
clearly the only adequate approaches to address this Measurements of Te are sparse and highly vari­
question. Unfortunately, such information is lack­ able, largely because of the difficulty in observing
ing for most of the studies shown in Table I . small amounts of liquid in tiny frozen inclusions
Salinity estimates based on final melting temper­ within a highly birefringent mineral such as dolo­
atures ( T ) of frozen inclusions in saddle dolomite
m mite. Available Te values are low-typically -45 to
are not significantly affected by re-equilibration -55 " C-suggesting a multicomponent system com­
processes. T values are less frequently reported
m posed predominantly of NaCl + CaC1 2 ±MgC1 2±
than Th measurements (cf. 1 1 1 9 Th versus 404 T m KCl (Aulstead & Spencer, 1 98 5 ; Aulstead et a!.,
values; Figs 1 0 and 1 1 ), and most studies do not 1 98 8 ; Rieken & Gaupp, 1 99 1 ; Coniglio et a!. , 1 994;
specify whether the final melting phase was ice or Qing & Mountjoy, 1 994a; Wojcik et a!., 1 994; Zeeh
hydrohalite. T values range from -4. 9 to -42. 5 " C,
m et a!., 1 995). A few studies reported first melting

Salinity (wt.% NaCI eq.)

40 35 30 25 (To) 20 15 10 5 0

Rlekon & Gaupp (1991) ------

"C 1 40 Shohon ot al. (1992) -------

�
------ Coniglio & Williams.Jonos (1992)
Zooh (1990) ------
1ii Zooh otal. (1995) ------
� 120 Gang & Zooh (1995) --

Ctl Leo & Friedman (1 967) ---

1: Moritzot al. (1996,)------
0 Mountjoy & Halim·Dihardja (1991) ---­
·c;; 100 Nesbitt & Muohlonbachs (1994),-----
Fig. I I . Frequency histogram of ::::J
Marchard ot aL (1994)----
final melting temperatures of u
primary fluid inclusions in saddle 1: ._ ,
"' '
- ·
dolomite. Data from Coniglio & :2 80 "' '
;.: .
Williams-Jones ( 1 992) are mostly ::::J "' '
from inclusions of secondary or ., ,
(/) :
indeterminate origin and are �
shown for comparison. Horizontal ca 60

arrows indicate literature sources E

that reported a range of values ·;::
0..
only. The upper x axis gives an -
expression of the measured 0 40

temperatures as wt% NaCI eq., ...

Q)
based on a pure H20-NaCl system .0
(after Bodnar, 1 993). The E 20
composition of seawater is ::::J
z
indicated by the vertical dashed
line (Tm � - 2. 1 · q. The range of
values in Shelton et a!. (1992)
include data from saddle dolomite ·48 -44 -40 ·36 ·32 ·28 ·24 ·20 ·16 ·12 ·8 0

and from other epigenetic

dolomite varieties. Final melting temperature (°C)
454 C. Spot! and J.K. Pitman
temperatures as low as -70 "C, which have been
attributed to metastable melting phenomena (e.g.
Conglio et al. , 1994). The high salinities and Na­
Ca-Cl composition of these fluids in saddle dolo­
mite are unlike the water compositions that prevail
in shallow burial environments, but similar to the
brines encountered in deep sedimentary basins (e.g.
Hanor, 1 994a,b; Land, 199 5 ) and to the brines re­
sponsible for the massive epigenetic dolomite alter­
ation associated with MVT deposits worldwide (e.g.
Roedder, 1 984; Viets & Leach, 1990). These evolved
fluids acquired their high salt content by rock-water
interactions involving siliciclastic and/or evaporitic
sediments.
Basin-scale fluid inclusion studies of saddle dolo­
mite and related mineral cements have played a
crucial role in improving our understanding of
large-scale palaeohydrological processes during the
past decade. One of the most extensively studied
areas is the midcontinent of the USA, where region­
ally extensive saddle dolomite is genetically and
spatially related to localized MVT lead-zinc depos­
its in Palaeozoic carbonate rocks. Fluid-inclusion
temperatures in saddle dolomite across the Ozark
region, in conjunction with an increase in the K/Cl
ratio in fluid inclusions and widespread correlatable
CL microstratigraphy, record a regional hydrother­
mal brine-flow event in Late Pennsylvanian-Early
Permian time (Leach & Rowan, 1986; Rowan,
1986; Rowan & Leach, 1 989; Viets & Leach, 1 990).
According to palaeohydrological models (Bethke &
Marshak, 1 990; Garven et al., 1 993), topographi­
cally driven hydrothermal fluids recharged from the
Ouachita fold belt migrated updip on to the craton,
precipitating hydrothermal (saddle) dolomite along
their flow paths (Leach et al., I 991 ).
SADDLE DOLOMITE:
A H I GH-TEM PERATURE D I A G E N E T I C
' I N D EX M I N ERAL'
This question i s o f great importance for petro­
graphic investigations of dolomite-bearing sedi­
mentary rocks, because it is notoriously difficult to
quantify pressure, temperature and chemical com­
position (PTX) during the burial history of a given
sample set without detailed fluid inclusion or isoto­
pic data. Most physicochemical processes in the
diagenetic regime are controlled by kinetic rather
than thermodynamic reactions; minerals diagnostic
of specific PTX conditions as have been recognized
in metamorphic rocks are thus uncommon, al­
though a small number of diagenetic phases (e.g.
saddle dolomite) have been identified.
Saddle dolomite, unlike most low-temperature
dolomite, which is fine crystalline, poorly ordered
and shows 8180 values close to O%o (e.g. Land,
1985; Allan & Wiggins, 199 3), is coarsely crystal­
line, associated with hydrocarbons, megaquartz and
MVT-ore minerals, depleted in 1 80 relative to
low-temperature dolomite, and contains medium­
to high-temperature saline fluid inclusions. Virtu­
ally all petrographic and geochemical information
evaluated in the previous sections argues in favour
of a high-temperature origin of saddle dolomite.
However, there are a few exceptions, where data on
saddle dolomite have been interpreted to indicate
formation at near-surface conditions. For example,
two anomalously heavy saddle dolomite o 180 val­
ues (-2 . 8 and -2. 1 %o) were reported for Devonian
carbonates in Great Britain, although the majority
of 'late-stage' dolomites (apparently including sad­
dle dolomite) showed depleted 8180 values between
- I 0.6 and -6. 1 %o (Searl & Fallick, 1 990). Based on
equilibrium isotopic relationships and a pore water
8180 value of Oo/oo SMOW, the isotopically heavy do­
lomite formed at 'about 40 " C'. However, tempera­
ture probably was a less important factor in the for­
mation of saddle dolomite than the 'abundant
supply of dolomite-supersaturated fluid through vig­
orous fluid circulation' (Searl & Fallick, 1 990, their
p. 6 3 6). Despite the fairly heavy o 180 values of the
saddle dolomites (see Fig. 5), their estimated low
crystallization temperature is not convincing and
this dolomite could equally have formed at signifi­
cantly higher temperatures from 180-enriched pore
fluids. In another study, Sr isotope data of saddle
dolomite in Ordovician carbonates of Oklahoma
(USA) were interpreted to indicate that this cement
did not form from basinal brines (Gao & Land, 1 99 1 ;
Gao et al., 1992; see also section on Sr isotopes
above). Rather, meteoric water infiltration, appar­
ently in a shallow burial setting, was invoked to ex­
plain the data, but again no direct temperature con­
straints were given.
Q U A N T I TA T I V E C O N S T R A I N T S O N
SADDLE DOLOMITE FORMATION
While there is surprisingly unanimous agreement
among authors regarding the general genetic inter­
pretation of saddle dolomite as a high-temperature
 Saddle dolomite in carbonates and sandstones
product, the physicochemical conditions of its for­
mation are still poorly constrained.
Direct information on precipitation temperature
is only possible to obtain using fluid inclusion
observations, so these data are clearly of prime
importance. Figure I 0 shows the vast majority of
two-phase liquid-vapour inclusions in saddle dolo­
mite homogenized between 90 and 160 ' C. Al­
though pressure correction would increase these Th
values by several tens of degrees (not to mention the
problem of estimating the type and magnitude of
palaeopressures), we concur with most authors who
regard Th values as approximate entrapment tem­
peratures. The temperature range of saddle dolo­
mite precipitation therefore coincides largely with
the liquid hydrocarbon 'window', and some occur­
rences extend well into the dry gas zone of hydro­
carbon maturation (e.g. Tissot & Welte, 1984). As
the laser Raman microprobe becomes more widely
used in future studies, we anticipate that methane
will be identified in aqueous inclusions in saddle
dolomite and associated mineral cements. The best
current estimate of the minimum temperature re­
quired for saddle dolomite precipitation is 60-
80'C, but Fig. I 0 suggests that most saddle
dolomite did not form below 80-90 ' C. The temper­
ature range of saddle dolomite formation, as indi­
cated by the post- 1 980 studies (90-160' C),
therefore, is in good agreement with the original
temperatures (60-150 ' C) of Radke & Mathis
( 1 980), although some of the recent studies suggest
even higher temperatures.
Several parameters can be used to constrain the
types of fluids involved in saddle dolomite precipi­
tation, yet many of these are semiquantitative at
best. Microthermometric data indicate that saddle
dolomite formed from Na-Ca-Cl-type waters at
salinities in excess of the marine value (3. 5%), with
most samples suggesting 5-7 times seawater salin­
ity. It is therefore not surprising to find anhydrite
coexisting and/or postdating saddle dolomite in a
number of sandstone and carbonate rocks (e.g.
Heydari & Moore, 1 989; Kaufman et a!., 1990;
Barnes et a!., 1992; Spot! et a!., 1 993). Petrographic
observations also indicate that these saline pore
waters were commonly close to equilibrium with a
number of other minerals, including fluorite, pyrite,
sphalerite, quartz and K-feldspar, which provide
important petrographic constraints for quantitative
reaction-path modelling (e.g. Plumlee et a!. , 1 994).
The oxygen isotopic composition of saddle dolo­
mite can be used to calculate the o 1 80 value of the
455
precipitating pore waters, which in turn provides
valuable insights into the palaeohydrological regime
of a basin. Determining the 8 1 80 composition of
the water requires knowledge of (i) the precipitation
temperature, (ii) the rock/water ratio, and (iii)
whether or not precipitation occurred in isotopic
equilibrium. Late-diagenetic (deep burial) carbon­
ate precipitation is commonly assumed to be an
equilibrium process at high rock/water ratios (e.g.
Banner & Hanson, 1 990; Montanez & Read, 1 992),
validating the use of one of the dolomite-water
isotope fractionation equations. Most authors used
fluid inclusion data to constrain the precipitation
temperature and reported positive 8 1 80 values for
the palaeofluid. A rough calculation using the mean
of all o 1 80 values (-9 .4o/oo) and the average Th value
( 1 27'C; see above) yields an equilibrium 8 1 80
value for the pore fluid of � +4.3o/oo SMOW. Depend­
ing on the palaeogeographical setting of the basin,
these values are interpreted to reflect either evolved
basinal fluids or, more realistically, mixing of
basinal brines and meteorically derived advective
fluids (Nesbitt & Muehlenbachs, 1 994; Yang et a!.,
1995).
The pH and Eh of palaeoflu ids cannot be mea­
sured directly, and only a few attempts have been
made to constrain these parameters using geochem­
ical models (e.g. Plumlee et a!., 1994). Qualita­
tively, saddle dolomite-precipitating fluids were
moderately to strongly reducing because divalent Fe
and Mn are incorporated into the crystal lattice in
large concentrations. In addition, saddle dolomite is
frequently associated with late-stage pyrite and
other sulphide minerals (see MVTs), which require
strongly reducing conditions. The pH of these pore
waters is difficult to constrain, but thermodynamic
principles suggest that most burial diagenetic and
hydrothermal systems are probably carbonate buff­
ered, i.e. any surplus of acid or base would quickly
be neutralized by reaction with carbonates (see
Krauskopf & Bird, 1 995, their p. 5 1 8).
A final question that needs to be addressed is the
peculiar and name-giving crystal distortion of sad­
dle dolomite, which ranges from curved crystal
surfaces to (rare) complete saddle forms. This fea­
ture is not unique to dolomite and has been ob­
served in other minerals, including calcite, siderite,
rhodochrosite and galena (Pedone, 1978; Gonzalez
et a!. , 1 992; Kostecka, 1 993; Spot!, unpublished
data). Although a few investigations present
thought-provoking crystallographic models (Searl,
1 989; Kretz, 1992; Kostecka, 1 995), little genetic
456 C. Spot/ and JK. Pitman
information can be extracted from these studies or
the original model advanced by Radke & Mathis
( 1 980). Thus, the reason why planar dolomite
precipitates at one place and saddle dolomite forms
at another is still unknown. Temperature and the
degree of supersaturation are factors that could
explain this phenomenon (Sibley & Gregg, 1 98 7),
but compelling evidence is lacking. Saddle-shaped
dolomite crystals have been produced experimen­
tally (Bullen, 1 98 3 ; Gregg, 1 98 3 , and personal
communication 1 996), but we are unaware of any
detailed experimental investigation aimed at resolv­
ing this question.
CONCLUSIONS
A summary o f paragenetic, geochemical and fluid­
inclusion information on saddle dolomite supports
the suggestion advanced by Radke & Mathis ( 1 980)
that this peculiar carbonate mineral may be a
reliable semiquantitative geothermometer in burial­
diagenetic settings. The formation of saddle dolo­
mite as a cement or replacement mineral requires
minimum temperatures of 60-8o · c, although most
occurrences appear to have formed between 90 and
1 6o · c. The parent palaeofluids were highly saline,
low Eh, pH buffered to near-neutral values, had
variable radiogenic Sr isotope ratios and tended
toward 1 80 enrichment relative to SMOW. The car­
bon isotopic composition of these pore fluids was
controlled by the proportion of organic to inorganic
carbon in the reservoir. This reassessment under­
scores the need for experimental work to better
understand the crystal growth dynamics of dolo­
mite at deep burial diagenetic conditions.
A C K N O W L E D G E M E N TS
C.S. acknowledges the support of an Austrian Acad­
emy of Sciences APART research fellowship and
the Austrian Science Fonds (project P I 0 1 90-GEO)
during completion of this study. We thank Diethard
Sanders for providing samples from saddle dolo­
mite in alpine carbonate rocks, Reinhard Gaupp for
providing some references and David A. Budd and
Jay M. Gregg for helpful comments on an earlier
version of this paper.
REFERENCES
ALLAN, J.R. & WIGGINS, W.D. (1993) Dolomite Reservoirs.
Geochemical Techniques for Evaluating Origin and Dis­
tribution. Am. Ass. Petrol. Geol. Continuing Education
Course Note Series, 36, 129 pp.
ALSHARAN, A.S. & WILLIAMS, D.F. (1987) Petrography and
stable isotope composition of baroque dolomite from
the Shuaiba Formation (Lower Cretaceous), Abu
Dhabi, United Arab Emirates. J. African Earth Sci. , 6,
881-890.
AMTHOR, J.E. & FRIEDMAN, G.E. (1992) Early- to late­
diagenetic dolomitization of platform carbonates: lower
Ordovician Ellenburger Group, Permian Basin, West
Texas. J. sediment. Petrol. , 62, 131-144.
AULSTEAD, K.L. & SPENCER, R.J. (1985) Diagehesis of the
Keg River Formation, northwestern Alberta: fluid inclu­
sion evidence. Bull. Can. Petrol. Geol. , 33, 167 -183.
AULSTEAD, K.L., SPENCER, R.J. & KROUSE, H.R. (1988)
Fluid inclusion and isotopic evidence on dolomitiza­
tion, Devonian of western Canada. Geochim. Cosmo­
chim. Acta, 52, 1027-1035.
BANNER, J.L. & HANSON, G.N. (1990) Calculation of
simultaneous isotopic and trace element variations dur­
ing water-rock interaction with applications to carbon­
ate diagenesis. Geochim. Cosmochim. Acta, 54, 3123-
3137.
BARBER, D.J., REEDER, R.J. & SMITH, D.J. (1985) A TEM
microstructural study of dolomite with curved faces
(saddle dolomite). Contrib. Mineral. Petrol. , 9 1 , 82-
92.
BARKER, C. E. & GOLDSTEIN, R.H. (1990) Fluid-inclusion
technique for determining maximum temperature in
calcite and its comparison to the vitrinite reflectance
geothermometer. Geology, 18, I 003-1006.
BARNES, D.A., LUNDGREN, C. E. & LONGMAN, M.W. (1992)
Sedimentology and diagenesis of the St. Peter Sand­
stone, central Michigan- Basin, United States. Bull. Am.
Ass. Petrol. Geol. , 76, 1 507 -1532.
BATES, R.L. & JACKSON, J.A. (1987) Glossary of Geology,
3rd edn. American Geological Institute, Alexandria,
VA, 788 pp.
BEALES, F.W. (1971) Cementation by white sparry dolo­
mite. In: Carbonate Cements (Ed. Bricker, O.P.). Johns
Hopkins University Studies in Geology, 19, 330-338.
BETHKE, C.M. & MARSHAK, S. (1990) Brine migrations
across North America: the plate tectonics of ground­
water. Ann. Rev. Earth Planet. Sci. , 18, 287 -315.
BLANC, P., ARBEY, F., CROS, P., CESBRON, F. & 0HNENSTET­
TER, D. (1994) Applications de Ia microscopic electron­
ique a balayage et de Ia cathodoluminescence a des
materiaux geologiques (sulfates, carbonates, silicates).
Bull. Soc. geol. France, 165, 341-352.
BODNAR, R.J. (1993) Revised equation and table for
determining the freezing point depression of H20-NaCI
solutions. Geochim. Cosmochim. Acta, 57, 683- 684.
BOLES, J.R. (1978) Active ankerite cementation in the
subsurface Eocene of southwest Texas. Contrib. Min­
eral. Petrol. , 68, 13-22.
BUCHBINDER, L.G., MAGARITZ, M. & GOLDBERG, M. (1984)
Stable isotope study of karstic-related dolomitization:
Jurassic rocks from the coastal plain, Israel. J. sediment.
Petrol. , 54, 236-256.
 Saddle dolomite in carbonates and sandstones
BuLLEN, S.B. (1983) Synthetic dolomite textures. MS the­
sis, Michigan State University, East Lansing, 153 pp.
CHAUDHURI, S. & CLAUER, N. (1993) Strontium isotopic
compositions and potassium and rubidium contents of
formation waters in sedimentary basins: clues to the
origin of the solutes. Geochim. Cosmochim. Acta, 57,
429-437.
CHOQUETTE, P.W. ( 1 971) Late ferroan dolomite cement,
Mississippian carbonates, Illinois basin, USA. In: Car­
bonate Cements (Ed. Bricker, O.P.). Johns Hopkins
University Studies in Geology, 19, 339-346.
CONIGLIO, M. & WILLIAMS-JONES, A.E. (1992) Diagenesis
of Ordovician carbonates from the north-east Michigan
Basin, Manitoulin Island area, Ontario: evidence from
petrography, stable isotopes and fluid inclusions. Sedi­
mentology, 39, 813-836.
CONIGLIO, M., SHERLOCK, R., WILLIAMS-JONES, A.E., MID­
DLETON, K. & FRAPE, S.K. ( 1 994) Burial and hydrother­
mal diagenesis of Ordovician carbonates from the
Michigan Basin, Ontario, Canada. In: Dolomites. A
Volume in Honour ofDo/omieu (Eds Purser, B., Tucker,
M. & Zenger, D.). Spec. Pub!. hit. Ass. Sediment., 2 1 ,
231-254.
COPLEN, T.B., KENDALL, C. & HOPPLE, J. (1983) Compari­
son of stable isotope reference samples. Nature, 302,
236 -238.
DE CRAEN, M. & SwENNEN, R. (1992) Sedimentology and
diagenesis of the ankeritized basal Zechstein conglom­
erate in the Campine Basin (Bree borehole, NE Bel­
gium). Geo/. Mijnbouw, 71 , 145-160.
DEER, W.A., HOWIE, R.A. & ZUSSMAN, J. (1992) An
Introduction to the Rock-forming Minerals, 2nd edn.
Longman, Harlow, 696 pp.
DRZEWIECKI, P.A., SIMO, J.A., BROWN, P.E. et a/. (1994)
Diagenesis, diagenetic banding, and porosity evolution
of the middle Ordovician St. Peter sandstone and
Glenwood Formation in the Michigan Basin. In: Basin
Compartments and Seals (Ed. Ortoleva, P.J.). Mem.
Am. Ass. Petrol. Geol., Tulsa, 6 1 , 179-199.
DuTTON, S.P. & LAND, L.S. (1988) Cementation and burial
history of a low-permeability quartzarenite, Lower Cre­
taceous Travis Peak Formation, East Texas. Geo/. Soc.
Am. Bull., 100, 1271-1282.
EBERS, M.L. & KOPP, O.C. (1979) Cathodoluminescent
microstratigraphy in gangue dolomite, the Mascot­
Jefferson City district, Tennessee. Econ. Geo/. , 74, 908-
918.
EDON, M., RAMBOZ, C., VOLFINGER, M., CHOI, C.G. &
ISABELLE, D. ( 1996) Apport de metaux par des fluides
profonds de source evaporitique dans l'Oxfordien du
bassin SE (France). Contribution de Ia microanalyse
PIXE. CR Acad. Sci. Paris, 322, 633- 640.
FARR, M.R. (1992) Geochemical variation of dolomite
cement within the Cambrian Bonneterre Formation,
Missouri: evidence for fluid mixing. J. sediment. Petrol. ,
62, 636-651.
FOLK, R.L. & ASSERETO, R. (1974) Giant aragonite rays
and baroque white dolomite in tepee-fillings, Triassic of
Lombardy. Ann. Mtg Am. Ass. Petrol. Geo/. , San Anto­
nio, Program with Abstracts, 1 , pp. 34-35.
FRIEDMAN, G.M. (1989) Case history of deep-burial sulfide
mineralization in the northern Appalachian basin.
Carb. Evap. , 4, 231-241.
457
FRIEDMAN, G.M. (1994) Dolostone-hosted sulfide occur­
rences in Silurian strata, Appalachian basin of New
York. In: Sediment-Hosted Zn-Pb Ores (Eds Fontbote,
L. & Boni, M.). Soc. Geol. Appl. Miner. Deposits Spec.
Pub!., 10, 77 -88.
GAo, G. & LA ND, L.S. ( 1 991) Early Ordovician Cool Creek
dolomite, middle Arbuckle Group, Slick Hills, SW
Oklahoma, USA: origin and modification. J. sediment.
Petrol. , 6 1 , 161- 1 73.
GAo, G., LAND, L.S. & FOLK, R.L. (1992) Meteoric modi­
fication of early dolomite and late dolomitization by
basinal fluids, Upper Arbuckle Group, Slick Hills,
southwestern Oklahoma. Bull. Am. Ass. Petrol. Geo/. ,
76, 1649-1664.
GARYEN, G., GE, S., PERSON, M.A. & SYERJENSKY, D. A.
(1993) Genesis of stratabound ore deposits in the
midcontinent basins of North America. I. The role of
regional groundwater flow. Am. J. Sci. , 293, 497-568.
GAWTHORPE, R.L. (1987) Burial dolomitization and poros­
ity development in a mixed carbonate-clastic sequence:
an example from the Bowland Basin, northern England.
Sedimentology, 34, 533- 558.
GENG, A. & ZEEH, S. (1995) Diagenesis of Middle Triassic
Muschelkalk carbonate rocks of the Rhine graben and
adjacent areas (southwest Germany). Z. dt. geol. Ges. ,
146, 372-385.
GIRARD, J.P. & BARNES, D.A. ( 1 995) Illitization and
paleothermal regimes in the Middle Ordovician St.
Peter sandstone, central Michigan Basin: K-Ar,. oxygen
isotope, and fluid inclusion data. Bull. Am. Ass. Petrol.
Geo/. , 79, 49- 69.
GoLDBERG, M. & BoGOCH, R. (1978). Dolomitization and
hydrothermal mineralization in the Brur Calcarenite
(Jurassic), southern coastal plain, Israel. Israel J. Earth
Sci. , 27, 36 -41.
GOLDSTEIN, R.H. & REYNOLDS, T.J. (1994) Systematics of
Fluid Inclusions in Diagenetic Minerals. Soc. Econ.
Paleont. Mineral. Short Course, 3 1 , 198 pp.
GONZALEZ, L.A., CARPENTER, S.J. & LOHMANN, K.C. ( 1992)
Inorganic calcite morphology: roles of fluid chemistry
and fluid flow. J. sediment. Petrol. , 62, 382-399.
GRAF, J.L.J. (1984) Effects of Mississippi Valley-type
mineralization on REE patterns of carbonate rocks and
minerals, Viburnum trend, southeast Missouri. J. Geo/. ,
92, 307-324.
GREGG, J.M. (1983) On the formation and occurrence of
saddle dolomite-discussion. J sediment. Petrol. , 53,
1025-1033. .
GREGG, J.M. & HAGNI, R.D. (1987) Irregular cathodolu­
minescent banding in late dolomite cements: evidence
for complex faceting and metalliferous brines. Geol.
Soc. Am. Bull. , 98, 86-91.
GREGG, J.M. & SHELTON, K.L. (1990) Dolomitization and
dolomite neomorphism in the back reef of the Bon­
neterre and Davis Formations (Cambrian), southeast­
ern Missouri. J. sediment. Petrol. , 60, 549- 562.
GREGQ, J.M. & SIBLEY, D.F. (1984) Epigenetic dolomitiza­
tion and the origin of xenotopic dolomite texture. J.
sediment. Petrol., 54, 908-931.
HABERMANN, D., NEUSER, R.D. & RICHTER, D.K. (1996)
REE-activated cathodoluminescence of calcite and do­
lomite: high-resolution spectrometric analysis of CL
emissions (HRS-CL). Sediment. Geo/. , 1 0 1 , 1-7.
458 C. Spot! and J.K. Pitman
HANOR, J.S. (1980) Dissolved methane in sedimentary
brines: potential effect on PVT properties of fluid
inclusions. Econ. Geol. , 75, 603-617.
HANOR, J.S. (1994a) Physical and chemical controls on the
composition of waters in sedimentary basins. Mar.
Petrol. Geol. , 1 1, 31-45.
HANOR, J.S. (1994b) Origin of saline fluids in sedimentary
basins. In: Geojluids: Origin, Migration and Evolution of
Fluids in Sedimentary Basins (Ed. Parnell, J.). Spec.
Publ. Geol. Soc. London, 78, 151-174.
HAYES, T.S., B URRUSS, R.C., PALMER, J.R. & ROWAN, E.L.
(1989) UV-fluorescence of hydrothermal dolomite of
the Ozark regional Mississippi valley-type ore system.
Geol. Soc. Am. Ann. Mig, Abstracts with Programs, 2 1 ,
A3.
HENRICH, R. & ZANKL, H. (1986) Diagenesis of upper
Triassic Wetterstein reefs of the Bavarian Alps. In: Reef
Diagenesis (Eds Schroder, J.H. & Purser, B.H.),
pp. 245-268. Springer-Verlag, Berlin.
HEYDARI, E. & MOORE, C.H. (1989) Burial diagenesis and
thermochemical sulfate reduction, Smackover Forma­
tion, southeastern Mississippi salt basin. Geology, 1 7 ,
1080-1084.
KAUFMAN, J., MEYERS, W.J. & HANSON, G.N. (1990) Burial
cementation in the Swan Hills Formation (Devonian),
Rosevear field, Alberta, Canada. J. sedimenl. Petrol. , 60,
918-939.
KAUFMAN, J., HANSON, G.N. & MEYERS, W.J. (1991)
Dolomitization of the Devonian Swan Hills Formation,
Rosevear Field, Alberta, Canada. Sedimentology, 38,
41-66.
KESLER, S.E., MARTINI, A.M., APPOLD, M.S. el a/. (1996)
Na-Cl-Br systematics of fluid inclusions from Missis­
sippi Valley-type deposits, Appalachian Basin: con­
straints on solute origin and migration paths. Geochim.
Cosmochim. Acta, 60, 225-233.
KosTECKA, A. ( 1 993) Calcite from the Quaternary spring
waters at Tylicz, Krynica, Polish Carpathians. Sedimen­
tology, 40, 27 -39.
KoSTECKA, A. (1995) A model of orientation of subcrystals
in saddle dolomite. J. sediment. Res. , A65, 332-336.
KRAUSKOPF, K.B. & BIRD, D.K. (1995) Introduction to
Geochemistry, 3rd edn. McGraw-Hill, New York,
' tology and Geochemistry of Dolostones (Eds Shukla,
647 pp.
KRETZ, R. (1992) Carousel model for the crystallization of
saddle dolomite. J. sediment. Petrol. , 62, 190-195.
KUHLEMANN, J. (1995) Zur Diagenese des Karawanken­
Nordstammes (Osterreich/Slowenien): Spattriassische,
epigenetische Blei -Zink-Vererzung und mitteltertiare,
hydrothermale Karbonatzementation. Arch. Lagersltit­
Lenforsch. Geol. Bundesanstalt, 18, 57-116.
KUPECZ, J.A. & LAND, L.S. (1991) Late-stage dolomitiza­
tion of the lower Ordovician Ellenburger Group, West
Texas. J. sedimenl. Petrol. , 6 1 , 551-574.
LAND, L.S. (1985) The origin of massive dolomite. J. Geol.
Educ., 33, 112-125.
LAND, L.S. ( 1 995) Na-Ca-Cl saline formation waters, Frio
Formation (Oligocene), south Texas, USA: products of
diagenesis. Geochim. Cosmochim. Acta, 59, 2163-2174.
LAND, L.S. & FISHER, R.S. (1987) Wilcox sandstone dia­
genesis, Texas Gulf Coast: a regional isotopic compari­
son with the Frio Formation. In: Diagenesis of
Sedimentary Sequences (Ed. Marshall, J.D.). Spec. Pub!.
Geol. Soc. London, 36, 219-235.
LEACH, D.L. & ROWAN, E.L. (1986) Genetic link between
Ouachita foldbelt tectonism and the Mississippi Valley­
type lead zinc deposits of the Ozarks. Geology, 14,
931-935.
LEACH, D.L., PLUMLEE, G.S., HoFSTRA, A. H. et a/. (1991)
Origin of late-dolomite cement by C02-saturated deep­
basin brines: evidence from Ozark Region, Central
United States. Geology, 19, 348-351.
LEE, Y.I. & FRIEDMAN, G.M. ( 1987) Deep-burial dolomiti­
zation in the Ordovician Ellenburger Group carbonates,
west Texas and southeastern New Mexico. J. sediment.
Petrol. , 57, 544- 557.
LOHMANN, K.C. & WALKER, J.C.G. (1989) The 0 1 80 record
of Phanerozoic abiotic marine calcite cements. Geo­
phys. Res. Lett., 16, 319-322.
MACHEL, H.G. (1987) Saddle dolomite as a by-product of
chemical compaction and thermochemical sulfate re­
duction. Geology, 15, 936-940.
MARCHAND, B., SwENNEN, R. & MuCHEZ, P. (1994) Zebra­
dolomite formation: dolomitization and porosity devel­
opment in the burial realm. 14th Int. Sedim. Congress.
Recife 1 994, Abstract volume, H 16-H18.
McDowELL, S.D. & PACES, J.B. (1985) Carbonate alter­
ation minerals in the Salton Sea geothermal system,
California, USA. Miner. Mag. , 49, 469-479.
McNuTT, R.H., FRAPE, S.K. & DoLLAR, P. (1987) A
strontium, oxygen and hydrogen isotopic composition
of brines, Michigan and Appalachian basins, Ontario
and Michigan. Appl. Geochem., 2, 495-505.
MOLDOVANYI, E.P., WALTER, L.M., BRANNON, J.C. & Po­
DOSEK, F. A. (1990) New constraints on carbonate di­
agenesis from integrated Sr and S isotopic and rare earth
element data, Jurassic Smackover Formation, US Gulf
Coast. Appl. Geochem., 5, 449- 4 70.
MONTANEZ, J.P. & READ, J.F. (1992) Fluid-rock interac­
tion history during stabilization of early dolomites,
upper Knox Group (Lower Ordovician), US Appala­
chians. J. sediment. Petrol. , 62, 753-778.
MOORE, C.H., CHOWDHURY, A. & CHAN, L. (1988) Upper
Jurassic Smackover platform dolomitization northwest­
ern Gulf of Mexico: a tale of two waters. In: Sedimen­
V. & Baker, P.A.). Spec. Publ. Soc. Econ. Paleont.
Mineral., Tulsa, 43, 17 5-189.
MORITZ, R., FONTBOTio, L., SPANGENBERG, J. el a/. (1996)
Sr, C and 0 isotope systematics in the Pucara Basin,
central Peru. Mineral. Deposita, 3 1 , 147 -162.
MORROW, D.W., CUMMING, G.L. & KOEPNICK, R.B. (1986)
Manetoe facies-a gas-bearing, megacrystalline, Devo­
nian dolomite, Yukon and Northwest Territories. Can­
ada. Bull. A m. Ass. Petrol. Geol. , 70, 702-720.
Moss, S. & TuCKER, M.E. (1995) Diagenesis of
Barremian-Aptian platform carbonates (the Urgonian
Limestone Formation of SE France): near-surface and
shallow-burial diagenesis. Sedimentology, 42 , 853-874.
MouNTJOY, E.W. & AMTHOR, J.E. (1994) Has burial
dolomitization come of age? Some answers from the
Western Canada Sedimentary Basin. In: Dolomites. A
Volume in Honour ofDolomieu (Eds Purser, B., Tucker,
M. & Zenger, D.). Spec. Publ. Int. Ass. Sediment., 2 1 ,
203- 229.
MOUNTJOY, E.W. & HALIM-DIHARDJA, M.K. (1991) Multi-
 Saddle dolomite in carbonates and sandstones
ple phase fracture and fault-controlled burial dolomiti­
zation, upper Devonian Wabamun Group, Alberta. J.
sediment. Petrol. , 6 1 , 590-6 12.
MouNTJOY, E.W., Q ING, H. & McN uTT, R.H. ( 1992)
Strontium isotopic composition of Devonian dolomites,
Western Canada Sedimentary Basin: significance of
sources of dolomitizing fluids. Appl. Geochem., 7, 59- 7 5.
MUCHEZ, P., VIAENE, W.A., KEPPENS, E., MARSHALL, J.D. &
VANDENBERGHE, N. ( 199 1 ) Vein cements and the
geochemical evolution of subsurface fluids in the Visean
of the Campine basin (Poederlee borehole, Belgium). J.
Geol. Soc. Lond., 148, 1005- 1 1 17.
NAHNYBIDA, C., HUTCHEON, I. & KIRKER, J. ( 1982) Diagen­
esis of the Nisku Formation and the origin of late-stage
cements. Can. Mineral. , 20, 129- 140.
NESBITT, B. E. & MUEHLENBACHS, K. ( 1994) Paleohydrology
of the Canadian Rockies and origins of brines, Pb-Zn
deposits and dolomitization in the Western Canada
Sedimentary Basin. Geology, 22, 243-246.
NESBITT, B.E. & MUEHLENBACHS, K. ( 1995) Paleohydrology
of the Canadian Rockies and origins of brines, Pb-Zn
deposits and dolomitization in the Western Canada
Sedimentary Basin: reply. Geology, 23, 190.
PEDONE, V.A. ( 1978) Petrography, Chemistry, and Crystal­
lography of Baroque Dolomite, Kingsport, Tennessee.
MS Thesis, University of Texas at Austin, 143 pp.
PITMAN, J.K. & SP6TL, C. ( 1996) Origin and timing of
carbonate cements in the St. Peter Sandstone, Illinois
Basin: evidence for a genetic link to Mississippi Valley­
type mineralization. In: Siliciclastic Diagenesis and
Fluid Flow: Concepts and Applications (Eds Crossey,
L.J., Loucks, R. & Totten, M.W.). Spec. Pub!. Soc. Econ.
Palaeont. Mineral. , Tulsa, 55, 187 -203.
PLUMLEE, G.S., LEACH, D.L., HOFSTRA, A.H. et a/. ( 1994)
Chemical reaction path modeling of ore deposition in
Mississippi Valley-type Pb-Zn deposits of the Ozark
region, US midcontinent. Econ. Geol. , 89, 1 36 1- 1383.
PREZBINDOWSKI, D.R. & TAPP, J.B. ( 1991) Dynamics of
fluid inclusion alteration in sedimentary rocks: a review
and discussion. Org. Geochem . , 17, 13 1- 142.
Q ING, H. & MouNTJOY, E. ( 1992) Large-scale fluid flow in
the middle Devonian Presqu'ile barrier, western Cana­
dian sedimentary basin. Geology, 20, 903-906.
Q ING, H. & MOUNTJOY, E.W. ( 1989) Multistage dolomiti­
zation in rainbow buildups, middle Devonian Keg
River Formation, Alberta, Canada. J. sediment. Petrol. ,
59, 1 14- 126.
Q ING, H. & MOUNTJOY, E.W. ( 1994a) Formation of
coarsely crystalline, hydrothermal dolomite reservoirs
in the Presqu'ile barrier, Western Canada Sedimentary
Basin. Bull. Am. Ass. Petrol. Geol. , 78, 55-77.
Q ING, H. & MouNTJOY, E.W. ( 1994b) Rare earth element
geochemistry of dolomites in the Middle Devonian
Presqu'ile barrier, Western Canada Sedimentary Basin:
implications for fluid- rock ratios during dolomitiza­
tion. Sedimentology, 4 1 , 787-804.
Q ING, H. & MOUNTJOY, E.W. ( 1995) Paleohydrology of the
Canadian Rockies and origins of brines, Pb-Zn deposits
and dolomitization in the Western Canada Sedimentary
Basin: comment. Geology, 23, 189- 190.
RADKE, B.M. & MATHIS, R.L. ( 1980) On the formation and
occurrence of saddle dolomite. J. sediment. Petrol. , 50,
1 149- 1 168.
459
RADKE, B.M. & MATHIS, R.L. ( 198 1) On the formation and
occurrence of saddle dolomite-reply. J. sediment.
Petrol. , 5 1 , 1352- 1353.
RICIPUTI, L.R., COLE, D.R. & MACHEL, H.G. ( 1996) Sulfide
formation in reservoir carbonates of the Devonian
Nisku Formation, Alberta, Canada: an ion microprobe
study. Geochim. Cosmochim. Acta, 60, 325-336.
RIEKEN, R. & GAUPP, R. ( 1991) Fluideinschluss­
Untersuchungen an Sandsteinen des Gasfeldes Thonse.
Veroff. Niederstichsische Akad. Geowiss. , 6, 68 -99.
RoEDDER, E. ( 1 968) Temperature, salinity, and origin of
the ore-forming fluids at Pine Point, Northwest Territo­
ries, Canada, from fluid inclusion studies. Econ. Geol. ,
63, 439-450.
ROEDDER, E. ( 1984) Fluid Inclusions. Rev. Mineral.,
Mineral. Soc. Am., 12, 646 pp.
ROSEN, M.R. & COSH ELL, L. ( 1992) A new location of
Holocene dolomite formation, Lake Hayward, Western
Australia. Sedimentology, 39, 16 1- 166.
ROSENBAUM, J. & SHEPPARD, S.M.F. ( 1986) An isotopic
study of siderites, dolomites and ankerites at high
temperatures. Geochim. Cosmochim. Acta, 50, 1 147-
1 150.
RowAN, E.L. ( 1986) Cathodoluminescent zonation in
hydrothermal dolomite cements: relationship to Missis­
sippi Valley-type Pb-Zn mineralization in southern
Missouri and northern Arkansas. In: Process Mineral­
ogy VI (Ed. Hagni, R.D.), pp. 69- 87. Society of Mining
Engineers of the American Institute of Mining Metallur­
gical and Petroleum Engineers. New Orleans.
RowAN, E.L. & LEACH, D.L. ( 1 989) Constraints from fluid
inclusions on sulfide precipitation mechanisms and ore
fluid migration in the Viburnum trend Lead District.
Missouri. Econ. Geol. , 84, 1948 - 1965.
ScHNEIDER, W., GENG, A. & LIU , X. ( 1991) Diagenesis and
mineralization processes in Devonian carbonate rocks
of the Siding-Gudan lead-zinc mineral subdistrict,
Guangxi, southwest China. Carb. Evap. , 6, 53-67.
SEARL, A. ( 1988) Pedogenic dolomites from the oolite
group (lower Carboniferous) South Wales. Geol. J., 23.
157- 169.
SEARL, A. ( 1989) Saddle dolomite: a new view of its nature
and origin. Mineral. Mag. , 53, 547 -555.
SEARL, A. & FALUCK, A.E. ( 1990) Dinantian dolomites
from East Fife: hydrothermal overprinting of early
mixing-zone stable isotopic and Fe/Mn compositions. J.
Geol. Soc. Lond. , 147, 623- 638.
SELLWOOD, B.W., SHEPERD, T.J., EVANS', M.R. & J AME S B. ,
( 1989) Origin of late cements in oolitic reservoir facies:
a fluid inclusion and isotopic study (Mid-Jurassic,
southern England). Sediment. Geol. , 6 1 , 223-237.
SHARMA, T. & CLAYTON, R.N. ( 1965) Measurement of
1 80/ 1 60 ratios of total oxygen of carbonates. Geochim.
Cosmochim. Acta, 29, 1347- 1353.
SHELTON, K.L., BAUER, R.M. & GREGG, J.M. ( 1992) Fluid
inclusion studies of regionally extensive epigenetic do­
lomites, Bonneterre Dolomite (Cambrian), southeast
Missouri: evidence of multiple fluids during dolomiti­
zation and lead- zinc mineralization. Geol. Soc. Am.
Bull. , 104, 675-683.
SIBLEY, D.F. & GREGG, J.M. ( 1987) Classification of
dolomite rock textures. J. sediment. Petrol. , 57, 967 -
975.
460 C. Spot! and J.K Pitman
SIMO, J.A., JoHNSON, C.M., VANDREY, M.R. et a/. (1994)
Burial dolomitization of the Middle Ordovician Glen•
wood Formation by evaporitic brines, Michigan Basin.
In: Dolomites. A Volume in Honour of Dolomieu (Eds
Purser, B., Tucker, M. & Zenger, D.). Spec. Pub!. I nt.
Ass. Sediment., 21, 169-186.
SMALLEY, P.C., HIGGINS, A.C., HowARTH, R.J. et a/. (1994)
Seawater Sr isotope variations through time: a proce­
dure for constructing a reference curve to date and
correlate marine sedimentary rocks. Geology, 22, 431-
434.
SPANGENBERG, J., SHARP, Z.D. & FONTBOTE, L. (1995)
Apparent stable isotope heterogeneities in gangue car­
b onates of the Mississippi Valley-type Zn-Pb deposit of
San Vicente, centrai Peru. Mineral. Deposita, 30, 67 -
74.
Sr6TL, C., MATTER, A. & BREVARt, 0. ( 1 993) Diagenesis
and pore water evolution in the Keuper reservoir, Paris
Basin (France). J. sediment. Petrol. , 63, 909- 928.
SP6TL, C., HOUSEKNECHT, D.W. & BURNS, S.J. (1996)
Diagenesis of an 'overmature' gas reservoir: the Spiro
sand of the Arkoma Basin, USA. Mar. Petrol. Geo/. , 1 3,
2 5-40.
STUEBER, A.M., PIJSHKAR, P. & HETHERINGTON, E.A. (1984)
A strontium isotopic study of Smackover brines and
associated solids, southern Arkansas. Geochim. Cosmo­
chiin. Acta, 48, 1 637 -1649,
SUCHY, V., ROZKOSNY, 1., ZAK, K. & F RANCU, J. (1996)
Epigenetic dolomitization of the Pridoli formation (Up­
·per Silurian), the Barrandian basin, Czech Republic:
implications for burial history of Lower Paleozoic
strata. Geo/. Rundschau, 85, 264-277.
TAYLOR1 T.R. & SIBLEY, D.F. ( 1 986) Petrographic and
:geochemical characteristics of dolomite types and the
origin of ferroan dolomite in the Trenton Formation,
Ordovician, Michigan Basin, USA. Sedimentology, 33,
61-86.
Tisso'r, B.P. & WELTE, D.H. (1984) Petroleum Formation
and Occurrence. Springer-Verlag, New York, 699 pp.
TOBIN, R.C. (199 1 ) Diagenesis, thermal maturation and
burial history of the upper Cambrian Bonneterre dolo­
mite, southeastern Missouri; an interpretation of ther­
mal history from petrographic and fluid inclusion
evidence. Org. Geochem., 17, 143-151.
TucKER, M.E. (1990) Dolomites and dolomitization mod­
els. In: Carbonate Sedimentology (Eds Tucker, M.E. &
Wright, V.P.), pp. 365-400. Blackwell Scientific Publi­
cations, Oxford.
YE IZER, J. (1983) Chemical diagenesis of carbonates:
theory and application of trace element technique. In:
Stable Isotopes in Sedimentary Geology (Eds Arthur,
M.A., Anderson, T.F., Kaplan, I.R., Veizer, J. & Land,
L.S.). Soc. Econ. Paleont. Mineral. Short Course, 10,
1-100.
VIETS, J.G. & LEACH , D.L. (1990) Genetic implications of
regional and temporal trends in ore fluid geochemistry
of Mississippi Valley-type deposits in the Ozark region.
Econ. Geol. , 85, 842-861.
Voss, R.L., HAGNI, R.D. & GREGG, fM. (1989) Sequential
deposition of zoned dolomite and its relationship to
sulfide mineral paragenetic sequence in the Viburnum
trend, southeast Missouri. Carb. Evap. , 4, 195 -209.
WALLACE, M,W. (1990) Origin of dolomitization on the
Barbwire Terrace, Canning Basin, western Australia.
Sedimentology, 37, 105-122.
WALTON, A. W., WOJCIK, K.M., GOLDSTEIN, R.H. &
B ARKER C. E. (1995) Diagenesis of Upper Carbonifer­
,
ous rocks in the Ouachita foreland shelf in mid­
continent USA: an overview of widespread effects of a
Variscan-equivalent orogeny. Geol. Rundschau, 84,
535 - 551.
WoJCIK, K.M. ( 1991) Diagenesis of Pennsylvanian sand­
stones and limestones, Cherokee basin, southeastern
Kansas: importance of regional fluid flow. PhD Disser­
tation, University of Kansas, 349 pp.
WoJCIK, K.M., McKIBBEN, M.E., GoLDSTEIN, R.H. & WAL­
TON, A.W. (1992) Diagenesis, thermal history, and fluid
migration, Middle and Upper Pennsylvanian rocks,
southeastern Kansas. In: Source Rocks in the Southern
Midcontinent, 1 990 Symposium, Norman/OK (Eds
Johnson, K.S. & Cardott, B.J.). Oklahoma Geol. Surv.
Circ., 93, 144-159.
WOJCIK, K.M., GOLDSTEIN, R.H. & WALTON , A.W. (1994)
History of diagenetic fiuids in a distant foreland area,
Middle and Upper Pennsylvanian, Cherokee basin,
Kansas, USA: fluid inclusion evidence. Geochim. Cos­
mochim. Acta, 58, 11 7 5-1 1 91.
WORONICK, R.E. & LAND, L.S. ( 1985) Late burial diagene­
sis, Lower Cretaceous Pearsall and lower Glen Rose
Formations, south Texas. In: Carbonate Cements (Eds
Schneidermann, N. & Harris, P.M.). Spec. Pub!. Soc.
Econ. Paleont. Mineral., Tulsa, 36, 265-275.
YANG, W., SPENCER, R.J. & KROUSE, H.R. (1995) Stable
isotope and major element compositions of fluid inclu­
sions in Devonian and Cambrian dolomite cements,
western Canada. Geochim. Cosmochim. Acta, 59, 3159-
3172.
ZEEH, S. (1990) Fazies und Diagenese des obersten Wet­
tersteinkalkes (Karn) der Gailtaler Alpen (Drauzug,
Osterreich). Freiburger Geowiss. Beitr. , 1, 1 - 210.
ZEEH, S. &. BECHSTADT, T. (1994) Carbonate-hosted Pb-Zn
mineralization at Bleiberg-Kreuth (Austria): compila­
tion of data and new aspects. In: Sediment-Hosted
Zn-Pb Ores (Eds Fontbote, L. & Boni, M.). Soc. Geol.
Appl. Mineral. Deposits Spec. Pub!., 10, 271- 296.
ZEEH, S., BECHSTADT, T., McKENZI E, J. & RICHTER, D.K.
(1995) Diagenetic evolution of the Carnian Wetterstein
platforms of the Eastern Alps. Sedimentology, 42, I99-
222.
ZENGER, D.H. (1981) On the formation and occurrence of
saddle dolomite-discussion. J. sediment. Petrol. , 5 1 ,
1350-1352.
ZENGER, D.H. & DUNHAM , J.B. (1988) Dolomitization of
Siluro-Devonian limestones in a deep core (5350 m),
southeastern New Mexico. In: Sedimentology and
Geochemislry of Dolostones (Eds Shukla, Y. & Baker,
P.A.). Spec. Pub!. Soc. Econ. Paleont. Mineral., Tulsa,
43, 161-173.
Spec. Pubis int. Ass. Sediment. (1998) 26, 461-481

Application of quantitative back-scattered electron image analysis

in isotope interpretation of siderite cement:
Tirrawarra Sandstone, Cooper basin, Australia

M. R. R EZ A E E an d J. P . SC H U LZ - R OJ A HN1
Australian Petroleum Cooperative Research Centre (APCRC).
National Centre for Petroleum Geology and Geophysics (NCPGG). Thebarton Ca,mpus.
University of Adelaide. SA 5005. Australia

A BSTRACT

A new method i s presented t o improve the interpretation o f bulk-rock oxygen and carbon isotope data
in isotopically heterogeneous samples. In the tluvio-deltaic Tirrawarra Sandstone of the Fly Lake and
Moorari Fields, Cooper basin, Australia, volumetric estimation of individual cement generations of
siderite is accomplished using image analysis techniques in conjunction with electron microprobe
data. Results show that bulk-rock isotope values are controlled by the relative proportions of
three main siderite cement generations. The variation in 8180 can be expressed by the equation
0180(bulkl ( V5 1 x 01805 ) + ( V52 x 018052) + ( V5 x 01805 ). The variation in 813C can be expressed by
=
1 3 3
the same type of equation, but the correlation coefficient is lower (0.64) than the one for 8180 (0.82).
The 8180 and 813C end-member values for each cement generation are characterized by a narrow
range, allowing a precise definition of the conditions under which individual generations formed.
Integration of petrographic and fluid inclusion results has led to the identification of the following
siderite cementation events: (i) an early, homogeneous Fe-rich siderite with a 013C signature of +1.45%o,
indicative of low-temperature methanogenic processes ("'3o·c); (ii) an Mg-rich i.nhomogeneOl.!S
siderite characterized by a complex zoning, with a 813C signature of -8.5%o produced mainly by the
decarboxylation of organic matter at temperatures between 64 and 76•C; (iii) an Mg-rich. relatively
homogeneous pore-filling siderite with a 813·C character of -ll%o that was produced during kerogen
maturation, at more elevated temperatures(98-11o·q. Both the first and second generation of siderite
cement were followed by a period of cement dissolution. The technique presented here has particular
applications in cases where pure or nearly pure samples of end-member siderite cement generations are
not available for isotope analysis, provided the various cement generations have different chemical
compositions.

INTR O D UCTI O N

Siderite can form i n a wide variety of depositional
and oiagenetic environments, including in marine,
brackish and fresh-water settings (e.g. Gould &
Smith, 1 979; Matsumoto & Iijima, 1 981; Gautier,
1 982; Curtis & Coleman, 1 986; Carpenter et a!.,
1 988; Bahrig, 1 989; Mozley, 1 989; Mozley &
1Present address: Shell Development (Australia) Pty Ltd,
I Spring Street, Melbourne, Victoria 3000, Australia,
e-mail jorg.p.schulzrojahn@shell.com.au, WWW http:/
www.ncpgg.adelaide.edu.au/jorg.htm
Carothers, 1 992; Spiro et a!., 1 993; Morad et a/.,
1 994). The mineral is most suitable for the study of
pore-water evolution during sediment subsidence
because, unlike other carbonate minerals, siderite
probably does not undergo recrystallization and
isotope re-equilibration during burial diagenesis, as
it has no unstable precursors or polymorphs (Curtis
et a!., 1 975; Gautier, 1 9 82; Pearson, 1 98 5 ; Curtis &
Coleman, 1 9 86). Therefore, stable isotope data of
siderite cements can provide a powerful tool for the
interpretation of diagenetic events in geologic�!

Carbonate Cementation in Sandstones Edited by Sadoon Morad 46 1

© 1998 The International Association of Sedimentologists
ISBN: 978-0-632-04777-2
462 M.R. Rezaee and J.P. Schulz-Rojahn

provinces. However, siderite rarely represents the

only carbonate material in sedimentary rocks, and
mechanical and/or chemical methods must be em­
ployed to separate the siderite from other carbonate
minerals in order to obtain an isotopically homoge­
neous sample (see Al-Aasm et a!., 1 990). However,
even rocks containing only one carbonate cement
type can be characterized by widely varying bulk
isotope compositions in the same geological prov­
ince. The bulk-rock isotope compositional varia­
tions for rocks with only one carbonate type can be
caused by the presence of multiple carbonate ce­
ment generations which cannot be chemically sepa­
rated from each other for separate isotope analysis.
In recent years, advancements in laser ablation and
ion microprobe technology have led to improve­
ments in spatially resolved isotope analysis (e.g.
Dickson et a!., 1 990; Smalley et a!., 1 992; Riciputi
& Paterson, 1 994). However, these technologies are
not yet widely available in the scientific community,
highlighting the need for alternative methods to
determine the isotope signatures of single-phase,
multigenerational carbonate cements.
The present investigation quantitatively exam­
ines the importance of varying amounts of different Fig. 1 . Map of the southern Cooper basin showing
siderite cement generations on carbon and oxygen major structural elements and the location of the
isotope signatures in the Lower Permian Tirrawarra Moorari and Fly Lake Fields in the Patchawarra
syncline, Cooper basin. Modified from Stuart et a/.
Sandstone of the Moorari and Fly Lake Fields,
(1988).
Cooper basin, South Australia (Fig. 1 ). The data are
based on the integration of petrographic, electron
microprobe and fluid inclusion techniques to derive Battersby, 1 976; Stuart, 1 976; Thornton, 1 979;
a new procedure for isotope interpretations using Williams et a!., 1 98 5 ; Fairburn, 1 989). The basin is
back-scattered electron (BSE) image analysis. They unconformably underlain by the Early Palaeozoic
show that variations in the isotopic character of marine and volcanic rocks of the Warburton basin
siderite cement that generally appears homoge­ (Gatehouse, 1 986) and unconformably overlain by
neous under the optical microscope are a function the Jurassic-Cretaceous sediments of the Eromanga
of the relative proportion and composition of each basin (Exon & Senior, 1 976; Senior et a!., 1 978;
generation of siderite cement, which are best char­ Armstrong & Barr, 1 9 86). Each of these basins has a
acterized using BSE imaging techniques in the study different areal extent, and the various depocentres
area. shifted over geological time. Cooper basin sediment
deposition terminated at the end of the Early-Mid­
Triassic, when widespread compressional folding,
G E O L O GI C A L SETTI N G regional uplift and erosion occurred (Battersby,
1 976). Rejuvenation of pre-Permian faults along the
The Permo-Triassic Cooper basin of central Austra­ flanks of many structures occurred contemporane­
lia (Fig. 1 ) is Australia's largest onshore hydro­ ously with Cooper basin deposition (Battersby,
carbon province, containing about 6 TCF of 1 976; Stuart, 1 976; Apak et a!., 1 993). The basin is
recoverable gas and 300 MMSTB of oil and gas characterized by three major synclinal areas, namely
liquids (Heath, 1 989; Laws, 1 989). The basin con­ the Patchawarra, Nappamerri and Tennapperra syn­
sists dominantly of lacustrine-fluvial deposits with clines, which are separated by the Gidgealpa­
local glaciofluvial and rare paraglacial aeolian sedi­ Merrimelia-Innamincka (GMI) and Murteree­
ments (see Kapel, 1 966, 1 972; Gatehouse, 1 972; Nappacoongee (MN) anticlinal trends (Fig. I)
 Isotope interpretation of siderite cement 463

(Thornton, 1 979; Apak, 1 994). The synclinal areas PALYNOLOGICA COOPER BASIN BASIN ��
AGE ZONE LITHOSTARTIGRAPHY DEVELOPMENT
contain up to 2500 m of Permian sediments, over­ HARLAN() El At • 1989 ��
lain by as much as 1 300 m of Jurassic to Tertiary 250 w
>- PP5 TOOLACHEE FM.
p
strata (Battersby, 1 976; Thornton, 1 979). fluvial :3 OARALINGIEFM

sandstones which occur at various levels within the J- PP4 )..,. UPLIFT
�
(�
ROSENEATHSHALE
Permian section represent the main petroleum res­
UNCON
260 EPSILON FM TECTONICALLY
STABLE PERIOD
z
ervoirs, including the fluvioglacial Tirrawarra Sand­ PP3
MURTEREESHAlE
l v

-
<
stone (Smyth, 1 979; Kantsler et al., 1 98 3 ; Heath, BASE LEVEL CHANGE
CONTROL THE

-'
> SEDIMENTATION
1 989; Hunt et al., 1 989; Yew & Mills, 1 989). About 270
:;: PP2.2 UPLIFT
95% of the Cooper basin oil occurs in the Tirrawarra 0: PATCHAWARAA FM REJUVENATION OF tv-v-
a: PRE-EXISTING FAULTS
Sandstone of the Tirrawarra Field (Heath, 1 989) " r
w
w
(Fig. 2). Additional oil reserves are found at the same 280
PP2.1
GENTLE
0.. OOWNWAAPING

stratigraphical interval in the Moorari and Fly Lake

).
Fields (Fig. 1 ). The two fields were discovered in �..__,____.
PP1.2.2 FLUVIO·DELTAIC
1 97 1 and are fault-bounded anticlinal structures. In 290 TIARAWAR � _..- · ""-A... DEPOSITION

�
<f)
::>
both, the Tirrawarra Sandstone reservoirs are char­ 0
0: PP1.2.1 GLACIAL DEPOSITION
ww MER AI MELIA F M
acterized by relatively low ambient pore porosities f-U.
)
:3�
(9-1 2%), low permeabilities (0.1 -1 5 mD in situ) 300 ID

PP1.1
rz
"
0:

and hence relatively low productivities (25-600 '" u

BOPD) (Rodda & Paspaliaris, 1 989; Yew & Mills,

1 989).
PREVIOUS INV ESTIGA TIONS
Many workers have commented on the depositional
environment of the Tirrawarra Sandstone. Thorn­
ton ( 1 979) suggested deposition in a braided river
system, whereas Williams & Wild ( 1 984) and Wild
( 1 98 7) proposed a low-sinuosity, bed-load domi­
nated fluvial channel origin. Seggie et al. ( 1 994)
proposed a braid-delta origin for the Tirrawarra
Sandstone in the Tirrawarra Field. Limited data are
available concerning the regional diagenetic evolu­
tion of the Tirrawarra Sandstone. Various authi­
genic minerals have been identified in Cooper basin
sediments, including quartz, carbonates, kaolinite,
dickite and illite, and locally clinochlore and pyro­
phyllite in the central Nappamerri Syncline (Stanley
& Halliday, 1 984; Schulz-Rojahn & Phillips, 1 989;
Schulz-Rojahn, 1 9 9 1 ; Stuart et al., 1 991 ). Carbon­
ate cement types include siderite, ankerite, dolo­
mite, ferroan dolomite and, rarely, calcite (Stuart et
al., 1 990; Schulz-Rojahn, 1 99 1 ). Based on a re­
gional database, Schulz-Rojahn ( 1 991 ) noted that
siderite is about 1 7 times more abundant by volume
than all other carbonate varieties together in the
Cooper basin sediments, and a major cause of
porosity reduction, including in the Tirrawarra
Sandstone. Martin (1 98 1 ) studied the Tirrawarra
Sandstone in six wells of the Tirrawarra Field and
interpreted the siderite cement to be an early
Fig. 2. Stratigraphic column for the southern Cooper
basin and main tectonic events. The fluvio-deltaic
Tirrawarra Sandstone was deposited during the early
subsidence history of the Cooper basin when Australia
was at high southern latitudes. Modified from Apak et
a/.(1993), Apak (1994) and Seggie et a/. (1994).
diagenetic event. Martin (1 984) investigated 40
Tirrawarra Sandstone samples in the Tirrawarra,
Moorari and fly Lake Fields, and concluded that
the siderite pre-dates quartz cementation. Bever et
al. ( 1 987, 1 98 8) investigated the Tirrawarra Sand­
stone in the Tirrawarra and Moorari Fields and also
concluded that the siderite is an early cement that
possibly provided early structural support against
grain compaction.
M ETHO DS
A total of 1 30 core samples from the Tirrawarra
Sandstone were collected from 1 4 wells of the
Moorari and fly Lake Fields. Sedimentological
descriptions of the cores were made and samples
from rocks deposited in a variety of depositional
environments were collected.
Thin sections were prepared for all samples fol­
lowing impregnation with blue-dyed epoxy resin to
facilitate the recognition of porosity. Quantitative
estimates of sandstone mineralogy, texture and
porosity were obtained by point counting (400-600
counts per thin section).
464 M.R. Rezaee and J.P. Schulz-Rojahn

Semiquantitative bulk-rock XRD analyses were samples were gently crushed in ethanol using an
carried out on 46 core samples (Table 1 ). The agate mortar and pestle, and then dried in an oven
at a temperature less than 600C to minimize clay
damage. Randomly oriented powders were pre­
Table 1. Derivation of the Tirrawarra Sandstone core pared by cavity-mounted pressing in an aluminium
samples. Mineralogical compositions were determined holder. The prepared samples were run in a Phillips
by semiquantitative bulk-rock XRD analysis. Siderite is
the only carbonate cement present, but occurs in varying PW 1 05 0 X-ray diffractometer at 50 kV and
proportions 3 5 rnA, using Co(KJ radiation, at a scan speed of
2' per minute. Semiquantitative mineral identifica­
Sample Depth tion was based on comparison with JCPDS files
Well name no. (mkB) Qtz Kao 1/M Sid
using the CSIRO software Xplot.
Fly Lake l F l-9397 2864.2 D M T D SEM studies were carried out on broken rock
Fly Lake l Fl-9417 2825.1 D M T M surfaces and polished sections coated with carbon
Fly Lake l F l-9431 2875.2 D D T s and gold/palladium using a Phillips XL20 electron
Fly Lake 2 F2-9554 2912.1 D T T T
Fly Lake 2 F2-9561 2914.2 D T T ?
microscope connected to a back-scattered electron
Fly Lake 2 F2-9568 2916.3 D T T T (BSE) detector. Energy-dispersive X-ray (EDX)
Fly Lake 2 F2-9570 2916.9 D s T T analysis was employed to study the composition of
Fly Lake 2 F2-9583 2920.9 D M T D representative authigenic minerals.
Fly Lake 2 F2-9590 2923.0 D D T s
Determination of quantitative elemental compo­
Fly Lake 2 F2-9598 2924.3 D s T M
Fly Lake 3 F3-9588 2921.5 D M T M sition of siderite cement was carried out on polished
Fly Lake 3 F3-9593 2923.9 D D T T thin sections covered with a thin layer of carbon
Fly Lake 4 F4-9441 2877.6 D T T D and using a CAMECA SX 5 1 electron microprobe
Fly Lake 5 F5-9401 2865.4 D T T T at 1 5 kV, with a 20 nA beam current and a 0.2 Jlffi
Fly Lake 5 F5-9454 2881.6 D T T
beam diameter. The BSE imaging system linked to
Fly Lake 6 F6-9398 2864.5 D T T
Fly Lake 6 F6-9401 2865.4 D T T T the electron microprobe was used to detect zonation
Moorari l M l -9420 2872.4 D T T M in the siderite cement, and composition analyses
Moorari l M l -9596 2924.9 D T T T were carried out for each zone (Table 2). Results
Moorari I Ml -9598 2925.5 D T T M
were normalized to 1 00 mol% Fe + Mn, Mg and Ca.
Moorari I Ml-9613 2930.0 D D T T
Moorari I M 1-9620 2936.1 D M T s The precision of the analyses was 1 00% ± 2. The
Moorari 2 M2- 10090 3075.4 D T T M standards used were: MgO for Mg, wollastonite for
Moorari 2 M2- 10116 3083.4 D T T M Ca, rhodonite for Mn, and Fe20 3 for Fe.
Moorari 2 M2- 10127 3086.7 D T T T Oxygen and carbon isotope analyses were carried
Moorari 2 M2-10145 3092.2 D T T T
M
out on 1 8 core samples which were selected using
Moorari 3 M3-9422 2868.8 D T T
Moorari 3 M3-9440 2874.6 D M T T optical and bulk XRD methods (Table 3). Only
Moorari 3 M3-9465 2884.9 D T T T samples containing major amounts of siderite that
Moorari 3 M3-9465 2885.2 D M T lack other carbonate cement types (as identified in
Moorari 3 M3-950 1 2895.9 D T T T
bulk XRD and under the electron microprobe) were
Moorari 3 M3-9503 2896.5 D M T T
Moorari 4 M4-9507 2897.7 D T T selected for stable isotope analysis. The samples
Moorari 4 M4-9523 2902.6 D T T were crushed to a fine dry powqer, and then left to
Moorari 4 M4-9531 2905.0 D T T react with I 00% phosphoric acid under vacuum at
Moorari 4 M4-9554 2912.1 D M T M 1 OO'C overnight (Rosenbaum & Sheppard, 1 986).
Moorari 5 M5-9458 2882.8 D D T T
M5-9463 2884.3 D T T
The resultant carbon dioxide was purified according
Moorari 5
Moorari 5 M5-9510 2898.6 D M T T to conventional techniques (McCrea, 1 950) and
Moorari 5 M5-9513 2899.6 D M T T analysed on a 6-inch dual-collector VG Micromass
Moorari 5 M5-9528 2904.1 D T T 602E mass spectrometer. The acid correction fac­
Moorari 5 M5-9583 2874.8 D T T T
tors of Rosenbaum & Sheppard ( 1 986) were used to
Moorari 6 M6-9780 2980.9 D T T T
Moorari 7 M7-9589 2922.7 D s T T compensate for the oxygen isotope fractionation.
Moorari 9 M9-9732 2966.3 D M T D Stable isotope values are reported in the 8 notation
in parts per thousand (%o). All oxygen isotope ratios
Qtz, quartz; Kao, kaolinite; 1/M, illite/muscovite; Sid,
are reported relative to standard mean ocean water
siderite; D, dominant (peak>1200 counts); S,
subdominant (800 <main peak>1200); M, minor (400 (sMow) (Craig, 1 96 1 ) and all carbon isotope values
<main peak>800); T, trace (peak <400). relative to PDB (Craig, 1 957).
 Isotope interpretation of siderite cement 465

Table 2. Microprobe results (mol%) for the different siderite cement generations, which can be further subdivided on
the basis of colour variations under the SSE microscope

Well name Sample no. FeO (%) MgO (o/o) CaO (%) MnO (%) Colour code

Fly Lake I Fl-9431 60.59 37.09 1.51 0.80 S2d

F l-9431 67.73 30.85 0.36 1.06 S2m
F l-9431 80.24 19.03 0.07 0.66 S31
F l-9431 71.89 27.41 0.04 0.66 S3m
F l-9431 68.39 29.65 1.22 0.74 S3d
F l-9431 77.79 21.32 0.15 0.74 S3m
Fly Lake 2 F2-9570 77.17 21.55 0.51 0.77 S31
F2-9583 96.90 0.24 1.16 1.69 Sll
F2-9583 75.88 22.92 0.20 1.00 S2m
Moorari I Ml-9613 63.72 34.45 1.03 0.80 S2d
Ml-9613 85.27 11.09 2.00 1.64 S21
Ml-9613 78.27 19.98 1.05 0.70 S2m
Ml-9620 94.81 1.09 2.50 1.61 Sll
Ml-9620 79.65 19.14 0.34 0.87 S2m
Moorari 2 M2-10116 70.66 27.92 0.46 0.96 S2d
M2-10116 65.88 32.78 0.26 1.08 S2m
M2-10116 63.48 34.82 0.48 1.22 S3d
M2-10116 81.82 16.83 0.35 1.00 S31
M2-10116 68.49 29. 78 0.56 1.1 7 S3m
M2-10145 70.56 28.71 0.12 0.61 S2d
M2-10145 77.15 20.18 0.96 I . 71 S21
M2-10145 69.92 28.96 0.31 0.80 S2m
M2-10145 84.97 9.07 ! .57 4.39 S31
M2-10145 82.78 15.48 0.73 1.01 S3m
Moorari 3 M3-9422 93.89 2.02 2.83 1.26 Sll
M3-9422 81.23 14.44 3.58 0.74 S2m
M3-9440 81.44 17.38 0.25 0.93 S2m
M3-9440 78.45 20.40 0.23 0.92 S3d
M3-9440 77.13 21.51 0.41 0.95 S3m
M3-9503 61.62 37.43 0.27 0.68 S2d
M3-9503 83.29 13.08 1.32 2.32 S21
M3-9503 76.10 22.37 0.40 1.13 S2m
Moorari 4 M4-9574 63.60 35.21 0.08 1.11 S2d
M4-9574 87.21 10.78 0.14 1.87 S21
M4-9574 74.39 23.78 0.49 1.34 S2m
M4-9574 80.82 17.41 0.42 1.35 S31
M4-9574 77.12 20.96 0.31 1.62 S3m
Moorari 5 M5-9553 63.83 34.54 0.26 1.38 S2d
M5-9553 82.49 14.93 0.37 2.21 S21
M5-9553 65.52 32. 77 0.21 ! .50 S2m
M5-9553 58.97 39.72 0.14 1.18 3D
M5-9553 72.84 25.87 0.25 1.04 S3m
M5-9553 69.46 29.34 0.22 0.98 S3d
Moorari 6 M6-9737 89.03 9.39 0.16 1.42 S31
M6-9737 75.07 23.55 0.29 1.08 S3d
M6-9737 86.32 12.32 0.12 1.23 S3m
Moorari 7 M7-9606 58.36 39.97 0.29 1.38 S2d
M7-9606 79.67 18.27 1.16 0.89 S2m
Moorari 9 M9-9732 97.05 0.85 0.93 1.18 Sll
M9-9732 68.05 30.30 1.06 0.59 S2d
M9-9732 82.82 15.99 0.21 0.98 S21
M9-9732 75.20 23.08 0.89 0.83 S2m
M9-9732 91.11 7.85 0.04 1.00 S31

I, light-coloured; m, moderately light-coloured; d, dark-coloured in BSE image.

466 M.R. Rezaee and J.P. Schulz-Rojahn

Table 3. Carbon and oxygen isotope data of the Tirrawarra Sandstone siderite cements. A good match is observed
between measured (o180measl and calculated oxygen isotope values (o180calcl, determined from image analysis results.
The isotope data reflect the varying proportions of the different generations of siderite cement

o'3c o'8o .S'80 meas SI S2 S3 o'80 calc

Well name Sample (PDB %a) (PDB %a) (SMOW %a) (%) (%) (%) (SMOW %a) Facies

Moorari 5 M5-9583 -10.37 -23.38 6.6 0 3 97 6.28 BD

Moorari 2 M2-10145 -11.13 -23.83 6.15 0 5 95 6.41 BD
Moorari 4 M4-9574 -10.68 -21.39 8.59 0 40 60 8.58 BS
Moorari 7 M7-9606 -10.49 -22.70 7.28 15 30 55 9.33 BM
Moorari 2 M2-10116 -8.13 -19.12 10.86 10 43 47 9.68 BS
Fly Lake I F l -9417 -3.83 -15.80 14.18 19 38 43 10.19 BD
Moorari 6 M6-9737 -9.70 -18.79 11.19 8 51 41 9.99 BS
Fly Lake I F l -9431 -4.22 -18.50 11.48 0 60 40 9.82 BD
Moorari I M l -9598 -7.07 -17.56 12.42 36 38 26 11.73 BD
Moorari 3 M3-9440 -5.98 -15.37 14.61 80 13 7 14.19 BS
Fly Lake 2 F2-9583 1.45 -15.84 14.14 92 3 5 14.66 MBB
Moorari 3 M3-9503 -7.97 -17.67 12.31 0 98 2 12.18 BD
Moorari I M l -9613 -9.22 -18.23 I 1.75 3 95 2 12.26 BD
Moorari 4 M4-9554 -8.95 -17.17 12.81 5 93 2 12.32 BS
Moorari 9 M9-9732 -4.99 -15.71 14.27 60 39 I 13.98 BM
Moorari 3 M3-9422 -3.87 -14.85 15.13 98 I I 15.08 BM
Moorari I M l -9620 -6.23 -16.70 13.28 15 85 0 12.74 BD
Fly Lake 4 F4-9441 1.46 -16.01 13.97 100 0 0 15.20 MBB

S I, early siderite cement; S2, middle generation siderite cement; S3, late generation siderite cement; BD, distal braid
delta; BM, medial braid delta; BS, beach-barrier sandstone; MBB, back-barrier marsh.

Twelve samples were selected for fluid inclusion
analysis. Cycling methods (Reynolds, 1 978) were
employed to carry out microthermometry on small
primary fluid inclusions. The fluid inclusion analysis
was carried out on a Leitz optical microscope with
a warming-cooling Reynolds stage. In all of the
siderite samples, isolated fluid inclusions of two­
phase liquid-vapour were present. The petrographic
features of the fluid inclusions are consistent with a
primary origin (see Goldstein & Reynolds, 1 994).
Because the fluid inclusions were small it was not
possible to observe the final melting of the ice. The
size of the fluid inclusions, which were mostly
equidimensional, suggesting that no stretching has
occurred, was between 3 and 6 !J.m for both the S2
and S3 siderite cement generations. The microther­
mometry measurement precision is estimated at
± 1 · c.
Different generations of siderite cement were
identified and characterized using a Phillips image
analysis system in conjunction with the SEM. The
BSE images of carbon-coated polished thin sections
were imported from the SEM in the form of
ment was characterized by a distinct range of
grey-scale values in any one sample, reflecting vary­
ing elemental compositions of the carbonate ce­
ment. Quantitative estimation of these different
siderite cement generations was accomplished by
assigning a unique colour code to the range of
grey-scale values representative of each cement
generation in each sample. For all samples, the
same analytical procedure was followed:
1 Acquisition of the BSE image of siderite cement.
Mg-poor (early-formed) siderite cement was found
to be relatively light-coloured, whereas more Mg­
rich siderite is darker in colour.
2 Adjustment of contrast and brightness of the
grey-scale image to enhance visual differentiation of
the various cement generations for each field of
view.
3 Production of an on-screen histogram showing
the frequency distribution of the range of grey-scale
values represented in each BSE image, and the
selection of grey-scale threshold values representa­
tive of each major cement generation for the pur­
pose of colour coding (e.g. red (Sl ) 0-80; green
=

grey-scale binary images (0-256 scale) using a video (S2) 8 1 -1 60; yellow (S3) 1 61 -256).
= =

camera and a Windows-based software program 4 Computerized colour conversion of the grey-scale
called Image Analysis. BSE image using the manually defined pixel thresh­
Generally, each main generation of siderite ce- old values. In the Tirrawarra Sandstone each major
 Isotope interpretation of siderite cement 467

generation of siderite cement was generally charac­

terized by a unique colour, reflecting different
chemical compositions. Interpreted Depositional Environment

5 Automatic determination of the relative abun­

dance of each user-defined colour zone, represent­
ing a different cement generation in this study. Meandering System
Machine readings were verified by random visual
inspection of some micrographs using a grid over­
lay, and in all cases the machine results were found
to be almost identical to the ones obtained by the
human operator. In order to obtain a statistically
meaningful set of results, between 20 and 30 fields
Beach Barrier Sand
of view were analysed for each sample (magnifi­
cation x 1 00). On average, about 2 minutes were Lacustrine

needed to analyse each field of view.

The potential shortcomings of the image analysis Beach Barrier Sand

technique relate to the fact that BSE images only Back Barrier Marsh
provide a two-dimensional representation of the
Distal Part of Braid delta
rock volume, which may lead to errors in the
statistical analysis of cement (colour) abundances.
In rare cases, owing to compositional similarities Medial Part of Braid delta
between different generations of siderite cement
and the resultant low grey-scale contrast, estimation
of the relative abundance of the cement generations
could not be carried out. Only samples for which Distal Part of Braid delta
the relative abundance of each generation of sider­
ite cement could be ascertained were integrated
with the oxygen and carbon isotope data in this
investigation. The image analysis technique cannot
Lower Shoreface
be applied in regions where different siderite ce­
ment generations have the same chemical composi­
tion, or overlapping compositions. �I
Fig. 3. Example of the different depositional
RESULTS environments and progradational and retrogradational
cycles of the Tirrawarra Sandstone in the Moorari and
Fly Lake Fields. The gamma ray log trace is derived
Palaeoenvironmental interpretations from the Moorari 9 well.

Various depositional palaeoenvironments are

recognized in the Tirrawarra Sandstone (Fig. 3),
including lacustrine, parallel beach-barrier, back­
barrier marsh with outwash beds, distal and medial
braid-delta, meandering system and aeolian deposi­
tional environments. The lacustrine deposits in­
clude both upper- and lower-shoreface clastics.
Parallel beach-barrier sandstones are chiefly
medium-grained, well-sorted quartzarenites. Back­
barrier marsh sediments consist of massive mud­
stones, fine-grained sandstones and thin coal beds.
The distal braid-delta sediments, which include
linguoid bars, interchannel bay and splay deposits,
are composed of dominantly medium to coarse-
grained, moderately sorted sandstones containing
some thin mudstone intercalations. The medial
braid-delta clastics include massive and trough
cross-bedded conglomerates and pebbly, trough and
planar cross-bedded, very coarse-grained, poorly
sorted sandstones. The meandering system is com­
posed of matrix-supported oligomictic gravel lag,
medium-grained well-sorted point-bar quartzaren­
ites, and floodplain mudstone and coals. The aeo­
lian beds are thin, medium-grained supermature
quartzarenites that formed on the point-bar sands
during times of low water discharge.
468 M.R. Rezaee and J. P. Schulz-Rojahn

Fig. 4. Petrographic, BSE and colour image characteristics of Tirrawarra Sandstone siderites. (A) Plane-polarized view
of the main siderite cement generations that can be distinguished under the optical microscope in this case, which is
the exception rather than the rule. S I has a brownish colour, whereas S2 and S3 are clear and colourless. S I is typically
engulfed by S2. Note the concentration of fluid inclusions in S2, and the irregular serrated boundary between S2 and
S3(arrow), implying some dissolution of S2 prior to precipitation of S3. Sample Ml-9598, Moorari I , 2925.5 m.
(B) More typical, homogeneous-looking siderite spar with irregular, serrated edges adjacent to kaolinite booklets
(arrow), indicating some siderite dissolution prior to kaolinite precipitation. Sample M3-9503, Moorari 3, 2896.5 m.
Plane-polarized view. (C) BSE image illustrating the cement stratigraphy. Light-coloured homogeneous S I ,
Continued
 Isotope interpretation of siderite cement 469

is thought to have formed as a replacement product

General diagenetic characteristics
Tirrawarra Sandstones in the Moorari and Fly Lake
Fields consist mainly of medium-grained, moder­
Siderite cement characteristics
ately sorted sublitharenites (mostly mica schist and
phyllite, shale and siltstone clasts) (classification of
Folk, 1 974). A variety of authigenic minerals are
recognized, including syntaxial quartz overgrowths,
minor illite, patchy kaolinite and siderite. Attention
is focused on the siderite and only a short descrip­
tion of the other diagenetic minerals is provided
here.
Quartz is the dominant pore-filling cement in
most samples. CL studies show at least six genera­
tions of quartz cement, although three main phases
are distinguished. The homogenization tempera­
tures of fluid inclusions entrapped within the quartz
cement indicate that this precipitated at tempera­
tures between 65 and 1 30 ° C, unless the fluid inclu­
sions re-equilibrated during burial (see Osborne &
Haszeldine; 1 993; Haszeldine & Osborne, 1 993).
Present-day reservoir temperature is about 1 24-
1 36oC in the study area. Quartz cementation was
initiated prior to major compaction, as evidenced
by the loose grain packing of detrital grains, but
probably continued until relatively recent times.
Pore-filling euhedral and vermiform kaolinite
booklets are common, and are sometimes inter­
grown with the outer margin of quartz overgrowths.
The kaolinite is believed to have formed mainly as
a replacement product of feldspars, and to a lesser
extent micas.
The authigenic nature of illite is evident from its
rare fibrous, lath-like habit. XRD analysis indicates
that the mineral is a dioctahedral 2M 1 variety with
a relatively broad base to the 1 0 A peak, indicative
of either a poorly crystalline nature or possible
interstratification with other minerals. The mineral
of chemically unstable rock fragments.
Siderite cement occurs in varying proportions in the
Tirrawarra Sandstone and constitutes up to about
30% of rock volume in some samples, as deter­
mined by point counting. It is the only carbonate
cement present in the samples (Table 1 ). Siderite
generally occurs as isolated sparry rhombs or as a
pore-filling cement, although a variety of different
crystal habits are apparent, including rhombohe­
dral, blocky and radial forms. The bulk-rock XRD
traces provide no clues to the presence of multiple
siderite cement generations. Rarely is the presence
of different generations of siderite cement evident
under the petrographic microscope (Fig. 4A). How­
ever, when viewed under the electron microprobe
and using the BSE imaging technique, three main
generations of siderite cement are identified, includ­
ing early (Sl ), middle (S2) and late (S3) (Figs 4C-F
and 5-S). The proportion of each generation quan­
tified by BSE image analysis varies from sample to
sample (Table 3). Fluid inclusion data indicate that
the different generations precipitated under dif­
ferent temperature conditions. Integration of the
BSE image analysis data with bulk-rock siderite
isotope results suggests that both oxygen and carbon
isotope characteristics are controlled by the relative
abundance of each cement generation.
Early generation of siderite cement (Sl)
Under the optical microscope, S l has a blotchy
appearance, displays a moderately light to dark
brown colour (Fig. 4A) and appears devoid of fluid
inclusions. In some samples Sl is the only siderite

Fig. 4. (Continued) characterized by an irregular outer edge, is engulfed by S2 displaying broad, uneven compositional
zoning, which in turn is surrounded by S3. Sample M9-9732, Moorari 9, 2966.3 m. (D) Colour-enhanced BSE image
of the same view as shown in (C)(see text). The method helps to differentiate the main generations of siderite cement.
Although the settings need to be adjusted from sample to sample, each main generation of siderite cement tends to be
characterized by a unique range of colours, which.facilitates volumetric estimation of the relative proportions of S I , S2
and S3 in each sample. Some compositional variation is evident in S I, the dissolution boundary between S I and S2 is
more enhanced, and the boundary between S2 and S3 can clearly be seen to be very sharp and irregular in nature.
Note the homogeneous texture of S3. (E) BSE image showing S2, which has a variably dark-grey colour and fringes the
edges of a pore that was subseqUently filled by a very homogeneous, light-grey S3 cement. S2 is of a pselidorhombic
nature. Sample M2-10116, Moorari 2, 3083.4 m. (F) The same view as shown in(E), colour enhanced. The
micrograph illustrates problems with the image analysis technique in some samples. Whereas S2 and S3 can clearly be
differentiated in most areas of the view shown, S2 locally displays a mixture of both colours(arrow), rendering
volumetric estimation of the different cement generations difficult in this sample. This problem is the exception rather
than the rule in the Tirrawarra Sandstones.
470 MR. Rezaee and J.P. Schulz-Roj ahn
Fig. 5. BSE image of the view shown in Fig. 4(A). S I
displays a light colour, whereas the surrounding S2
(medium grey) is characterized by a variable internal
composition and complex zoning. S3 is a relatively
homogeneous, late-generation pore-filling cement. Note
the irregular dissolution boundary between S l and S2,
and between S2 and S3(arrows). These cement
relationships are typical of Tirrawarra Sandstones.
Sample M l -9598, Moorari I, 2925.5 m.
cement, especially in fine-grained poorly sorted
marsh sediments rich in organic matter. In samples
where Sl cements the rock completely, quartz
Fig. 6. In this BSE image a homogeneous S3 cement is
the main pore-filling event; however, examination of
other micrographs shows that the relative proportion of
the different siderite cement generations can vary with
different fields of view (cf Figs 4A and 5). Again, notice
the dissolution boundary between S2 and S3, the
dissolution pits associated with S2, and the isolated
remnants of S l (arrow) within the S2 matrix. Further
observe the incipient euhedral rhombic terminations of
the S3 cement that grew on the S2 dissolution surface.
Sample M l -9598, Moorari I , 2925.5 m.
Fig. 7. In this BSE image Sl occurs as isolated patchy
white remnants(arrow) within S2 cement(medium
grey), which is characterized by an inhomogeneous
appearance and complex zoning. S2 is enclosed by a
relatively homogeneous, euhedral S3 cement displaying
some broad compositional zoning. Note the presence of
dissolution pits in S2 and the characteristic dissolution
boundary(arrows) between S2 and S3. Sample
M l -9598, Moorari I, 2925.5 m.
grains are characterized by a very loose grain
packing (high intergranular volume), with individ­
ual grains apparently 'floating' in the siderite matrix
(Fig. 9). Where other siderite cement generations
are present, S 1 generally represents the substrate or
nucleus for the middle generation of siderite cement
precipitation (S2) (Figs 5-8). The boundary be­
tween S 1 and S2 is not always distinct, but is
typically characterized by irregular and serrated
Fig. 8. BSE image showing remnant S I cement with
dissolution edges(white) engulfed by abundant S2
cement(dark grey) that displays complex compositional
variation. Sample M9-9732, Moorari 9, 2966.3 m.
 Isotope interpretation of siderite cement
Fig. 9. BSE image of a fine-grained, moderate to poorly
sorted back-barrier marsh sample completely cemented
by S I (white). Note the serrated nature of some quartz
grains and the very high intergranular volume(>50%),
which suggest the replacement of part of the margins of
quartz grains by siderite cement. Sample F4-9441, Fly
Lake 4, 2877.6 m. Scale bar= 500 Jlm.
edges, indicating some dissolution of S 1 prior to S2
cementation (Fig. 5). In the BSE image S 1 is light
coloured and appears homogeneous (Figs 5-9).
Electron microprobe analysis for S 1 shows a high
Fe/Mg ratio (Fig. 1 0). The S1 elemental composi­
tion ranges from (Fe97_7%M&.s%Ca0_7%Mn0_8%)
C03 to (Fe93. 4%Mg2%Ca3_3%Mnt.3%)C03, with the
average composition being (Fe96%Mg %Cat.7%
1
Mn13 %)C03. In samples dominated by Sl (92-
98%), oxygen isotope compositions range from
+ 1 4. 1 %o to+ 1 5. 1 %o, with o13C compositions vary­
ing between -3.8 and + l.45%o. The one sample in
Fig.lO. Ternary diagrams showing
the compositional ranges of
different generations of siderite
cement in the Tirrawarra
Sandstone. The early generation of
the siderite cement(S I ) is very
rich in Fe, whereas the middle(S2)
and late generations(S3) have
much higher substitution of Mg
and fall within the realm of
sideroplesite and pistomsite. S2
and S3 have almost identical
compositions except that S2 has a
slightly higher Ca content.
47 1
which S 1 constitutes 100% of the siderite cement
volume (sample F4-944 1 ) has a o180 value of
+ 1 3. 97%o and a o13C of+ l.46%o (Table 3).
Middle generation of siderite cement (S2)
In BSE images S2 has the appearance of rhombs,
which generally enclose S1 nuclei (Figs 5-8) and
which in turn are engulfed by S3. Differentiation
between S2 and S3 is difficult under the optical
microscope, except where small dissolution pits and
boundaries occur between S2 and S3 (Fig. 4A). S2 is
characterized by many small (3-5 J.l.m) primary
fluid inclusions, whereas S3 has fewer inclusions.
Locally, S2 is engulfed by quartz cement, indicating
that quartz cementation postdates S2 precipitation
(Fig. 1 1 ). In BSE image S2 displays a multiple
compositional zoning (Fig. 1 2).
Based on grey-scale characteristics, three main
zones of S2 precipitation are evident, namely dark-,
medium- and light-coloured zones. Electron micro­
probe analysis reveals different elemental composi­
tions and variable substitution of Mg for each of the
S2 subgenerations (Tables 2 and 5). S2 composi­
tions range from (Fe87_2%Mg9_5%Ca0_7%Mn2_6%)
C03 to (Fe56_7%Mg42_2%Ca0. 5%Mn0.95%)C03, with
1
the average composition being (Fe74%Mg24%Ca0_8%
Mnt.2%)C03 (Table 4), representing sideroplesite
for the light-coloured zone and pistomsite for the
medium- to dark-coloured zone (classification of
Deer et al., 1 992). In samples in which S2 is the
dominant carbonate cement phase (93-98%), oxy­
gen isotope values range from + 1 2 . 3%o to +12.8%o
and carbon isotope values between -9.2%o and
Fe+Mn
Fe+Mn
Fe+Mn
S1
472 M.R. Rezaee and J.P. Schulz-Rojahn

(/) 25
c
(j)

�:; 20
(/)
<1l
� 15
0
Jl10
E
::::>
z
5

Fig. 1 1. BSE image showing the intergrowth between S2
cement (white) displaying characteristic rhombs and
quartz overgrowths (medium grey). Locally the siderite
is completely engulfed by the quartz cement (arrow),
indicating that some siderite cementation preceded
quartz (Q) cementation. Sample F l -9433, Fly Lake I,
2875.2 m. Scale bar= 50 !!m.
0
Homogenisation Temperature {Th,'C)
Fig. 13. Fluid inclusion homogenization temperatures
for the middle (S2) and late generations (S3) of the
Tirrawarra Sandstone siderites. The fluid inclusions are
considered to be of primary origin, and did not
experience stretching. S2 formed at much lower
temperatures than S3.

-7.9%o. The mean o13C compositiOn is about

-8.6%o, and the average o180 composition is +1 2%o
for S2 (Table 4). Homogenization temperatures of
S2 fluid inclusions range from 66 to 76'C, with a
median of around 68'C (Fig. 1 3).
Late generation of siderite cement (S3)
Under the optical microscope S3 is a blocky, colour­
less, very clear cement (Fig. 4a), postdating S 1 and
S2. The boundary between S2 and S3 is character­
ized by an irregular serrated outline, implying some
dissolution of S2 prior to precipitation of S3
(Fig. 1 4). In some samples oil occurs in the bound­
ary zone between S2 and S3, indicating that hydro­
carbon migration occurred synchronous with or
after the dissolution event, but prior to S3 precipi­
tation. BSE images show that S3 is a relatively
homogeneous cement generation (compared with
S2), characterized by an initial high Mg content
(pistomsite) grading into a relatively thick, homo­
geneous sideroplesite cement (Figs 5-7). Electron
microprobe analyses indicate extensive substitution
of Mg, with an average composition of (Fe75.5%

Table 4. Summary table showing the average isotopic,

chemical composition and fluid inclusion characteristics
of the main siderite cement generations (S I , S2, S3)

Sl S2 S3

o180 (%o SMOW) +15 +12 +6

o13C (%o PDB) +1.45 -8.5 -II
Th (•C) 68 102
FeO (%) 96 74 75.5
MgO (%) 1.0 24 23
CaO (%) 1.7 0.8 0.5
Fig. 12. BSE image of the middle generation of siderite MnO (%) 1.3 1.2 1.0
cement (S2), showing compositional zoning and some
dissolution. Sample M9-9732, Moorari 9, 2966.3 m. Th, fluid inclusion homogenization temperature.
 Isotope interpretation of siderite cement 473

Table 5. Average elemental composition(mol%) o f different zones within the main siderite cement generations( S l , S2,
S3). Subdivisions of the main siderite cement generations are based on colour differences under the BSE microscope

Sll S21 S2m S2d S31 S3m S3d

FeO(%) 95.7 83.0 74.7 64.6 83.6 76.8 69.0

MgO(%) 1.0 14.3 23.6 34.0 14.4 21.8 29.6
CaO(%) 1.9 0.8 0.7 0.6 0.4 0.3 0.4
MnO(%) 1.4 1.8 1.0 0.9 1.5 1.1 1.0

I, light-coloured; m, moderately light-coloured; d, dark-coloured in BSE image.

Mg23%Ca0_5%Mn,%)C03 for S3 (Table 4). In
samples which contain the highest proportion of
S3 (95-97%) oxygen isotope compositions range
from +6.1 %o to +6.6%o, with o13C compositions
varying between - 1 1 . 1 %o and - 1 0.4%o. For these
samples mean oxygen and carbon isotope values
are about +6%o and - 1 1 %o, respectively (Table 4).
Fluid inclusion results for S3 indicate a homogeni­
zation temperature of between 98 and 1 1 4 · c, with
a median of about 1 0 2 "C (Fig. 1 3).
from samples with varying proportions of the dif­
ferent cement generations plot between these zones
in a broad scatter (Fig. 1 5). Samples deposited
under marsh environments are characterized by
relatively high o13C and o180 values, reflecting the
dominance of S 1 in these sediments. In contrast,
siderites that formed in rocks deposited under
different sedimentary conditions have generally
more 13C- and 180-depleted isotope signatures ow­
ing to the greater proportion of S2 and S3 in these
samples (Table 3).

Bulk o180 and o13C trends

A broad correlation exists between o180 and o13C 1

o 80(SMOW)
values for the different samples containing multiple
siderite cement generations (Fig. 1 5) . With increas­ 0 5 10 15 20
ing enrichment in o180, o13C values become less 5
negative. The isotope results of the samples domi­ 3 81
nated by a single siderite cement generation plot
@.
within narrow zones, whereas the ones derived
-1
m
0 -3 • •
(L II
-

C')o -5 •

• ••
•

(i\82
,....
-7
vo •

Fig. 14. Scanning electron micrograph showing euhedral,
blocky siderite thought to represent the S3 cement
generation that developed on an earlier, partly dissolved
siderite cement surface(S2). Sample M3-9405,
Moorari 3, 2847.8 m.
-9 83 •
u
-11 �·
-13
-15
Fig. 1 5. Cross-plot of carbon and oxygen isotope values
for the Tirrawarra Sandstone siderites. The bulk-rock
isotope signatures are the cumulative product of the
varying proportions of the different siderite cement
generations(S I, S2, S3). The pure end-member
compositions of the different siderite cement generations
can be estimated from samples which are dominated by
a single cement generation(circled areas). Only the
end-member isotope compositions of the different
cement generations should be taken into consideration
when evaluating multigeneration siderite cements.
474 M.R. Rezaee and J.P. Schulz-Rojahn
DISCUSS I O N
The integration o f bulk-rock isotope data with
electron microprobe and image analysis results has
led to the identification of a multigeneration pore­
filling siderite cement in the Tirrawarra Sandstone
of the Moorari and Fly Lake Fields. Although
previous workers had identified several siderite
cement morphologies, they had assumed the pore­
filling sparry siderite to be a single-generation pre­
cipitate in Cooper basin sediments (Martin &
Hamilton, 1 98 1 ; Schulz-Rojahn & Phillips, 1 989).
Results from the present investigation show that
the pore-filling siderite precipitated in three main
stages in the Moorari-Fly Lake area. An early,
homogeneous Fe-rich siderite (S I ) was followed by
a generally more extensive cement generation char­
acterized by a complex compositional zoning (S2).
This in turn was engulfed by a late-generation,
relatively homogeneous (compared with S2) siderite
cement (S3). The relative proportion of each ce­
ment generation varies between samples, and in
some rocks one or two of these siderite cement
generations are absent (Table 3). The identification
of multiple siderite cement generations is of impor­
tance for the interpretation of bulk-rock oxygen and
carbon isotope results in the Cooper basin, and
probably also in other geological provinces.
In the Tirrawarra Sandstone each generation of
siderite cement has a distinct isotope signature, as
revealed by samples which are dominated by a
single cement generation ("" 9 5%). Only a small
proportion of the samples fall into this category
(Table 3). The majority of the isotope results are the
cumulative product of the varying proportions of
the different siderite cement generations. Failure
to recognize this fact may lead to erroneous inter­
pretations of the isotope data. For example, the
broad trend of overall more negative o 1 3C values
with increasing depletion in 1 80 (Fig. 1 5) could be
interpreted as a continuous (gradual) evolution in
carbonate isotope composition during burial dia­
genesis. The trend is commonly observed in car­
bonate cements in a variety of clastic provinces
(Fritz et al., 1 97 1 ; Irwin et a!., 1 977; Irwin, 1 980;
Schulz-Rojahn, 1 99 1 ; Mozley & Carothers, 1 99 2;
Spot! et a!. , 1 993). However, results from the
present investigation show that the pattern can also
be produced by sample heterogeneity. The data
demonstrate that major changes in pore fluid iso­
tope composition occurred between the precipita­
tion of each major cement generation in the
Tirrawarra Sandstone (Fig. 1 5). The conclusion is
supported by the presence of a dissolution bound­
ary between each major cement generation (Figs 4A
and 5).
Further, non-recognition of the isotopically het­
erogeneous nature of most Tirrawarra Sandstone
samples could lead to erroneous perceptions of the
major source(s) of carbon for siderite precipitation.
In the study area, a high proportion of o 1 3C values
for the siderites fall within the range of about -3
and -8%o (Fig. 1 5). Without a knowledge of the
influence of differential cement development on
bulk-rock isotope signatures, the values could be
attributed mainly to mixing of carbon derived from
marine limestones (+2 to - 2%o; Hudson, 1 977) and
volcanic or geothermal sources (on average between
-5 and -7%o; Deines, 1 986). However, as the isotope
values within the range of -3 and -8%o are derived
from samples that contain multiple siderite cement
generations in varying proportions (Table 3), these
hybrid data provide no useful clues to the condi­
tions under which the different generations formed.
The approach differs from other investigations of
diagenesis, in which there is no discrimination of
the bulk-rock isotope data and where all bulk-rock
isotope values are used to derive a model of carbon­
ate cementation. Difficulties in estimating the vol­
ume of individual cement generations generally
prohibit a greater qualitative control on isotope
interpretations. However, the present study shows
that BSE image analysis can provide an efficient
means of quantifying the influence of multigenera­
tional cement development on bulk-rock isotope
signatures, if it is assumed that BSE-derived cement
generations are isotopically homogeneous.
Method for enhanced isotope interpretation
In the study area the influence of variable cement
proportions on o 1 80 can be determined by plotting
the relative abundances of S 1 , S2 and S3 for each
sample on a ternary diagram (Fig. 1 6a). The sam­
ples with the highest proportion of a single cement
generation give the closest approximation to end­
member 8 1 80 values, and provide the basis for the
labelling of each corner of the diagram. IsosMow
lines can then be drawn which allow the prediction
of 8 1 80 values for the remainder of the samples
(Fig. 1 6b). In cases where there is no end-member
representative, the IsosMow lines established from
existing values can be extrapolated to the end­
member locations at the corners of the ternary
 Isotope interpretation of siderite cement
Fig. 1 6 . (a) Ternary diagram
showing the relative abundance
(%) of the different siderite cement
generations in the Tirrawarra
Sandstone, as determined from
statistical evaluation of electron
microprobe and image analysis
results. (b) The samples with the
highest proportion of a single
cement generation give the closest
approximation to end-member
8180 values(+15%o, + 12%o
and +6%o were assumed for 100%
pure S l , S2 and S3 respectively,
based on the data shown in Table
3). IsoSMOW lines can then be
constructed for the remainder of
the samples. The application of
the method is the calculation of
isotope compositions for
individual siderite cement
generations in samples which
contain more than one siderite
cement generation.
diagram, in order to estimate end-member o 1 80
values. The relationship can be expressed by the
equation
o ' so( bul k ) ( Vs l X 0 1 80s ! )
=
+ ( Vs2 x O I 80 s2) + ( VsJ x o i SOsJ)
where o 1 80(buik) is the oxygen isotope result of the
bulk-rock analysis; V5 1 is the proportion of S 1 ; V5 2
is the proportion of S2; V5 is the proportion of S3;
3
o 1805 is the oxygen isotope value of S 1 ; o 1805 2 is
1
the oxygen isotope value of S2; and 0 1805 is
3
oxygen isotope value of S3.
Table 3 shows that the measured and calculated
bulk-rock 8 1 80 values generally exhibit a good
2
correlation ( r 0.82 ). Minor discrepancies be­
= pure or nearly pure samples of carbonate cement
tween the two data sets are probably due to small
errors in the estimation of the volume of individual
cement generations in some samples. The bulk-rock
o 1 3C composition of each sample is also controlled
by the relative proportion of each generation of
siderite cement, and the same mathematical for­
mula as shown above (substituting o 1 3C for o1 80)
can be applied. However, the correlation coefficient
2
for calculated o1 3C values ( r 0.64) is lower than
= lization during burial diagenesis. No unstable
the one for calculated o 1 80 compositions. We are
uncertain about the reason(s) for this phenomenon,
but believe it may be due to the fact that o 1 80 in a
cement, at any given pore-water isotope composi­
tion, is a strongly temperature-dependent variable,
475
81
1 5 (o'"O %o )
(b) 13
81 12
( 1 00%)
11
10
(a) •
.
. . .. • 7
.
- · ·_··�-L�--�_u��3
�--���-�
•
•
•
83
82 UL,_-,--,-�-( 1-r-��--.--T� 00%)
( 1 00%)
whereas o 1 3C is independent of temperature.
Both o 1 80 and o 1 3C values for each end-member
cement generation plot within narrow zones
(Fig. 1 5), allo'fing a precise definition of the condi­
tions under which the individual cement genera"­
tions formed, albeit based on a small data set. The
data show that the quantitative approach to isotope
analysis provides added p �cision to routine bulk­
rock isotope interpretation methods in the study
area. The technique used in this study may be
applicable to other geological provinces, provided
different cement generations do not have the same
chemical composition. The image analysis tech­
nique is particularly valuable for rocks in which
end-members do not exist.
Precipitation temperatures of siderite cement
In the study area compositional zoning is evident in
both the S2 (Figs 8 and 1 2) and S3 (Fig. 6) cement
generations, indicating that the cements precipi­
tated from solution and did not undergo recrystal­
precursor for siderite is known, and there is no
documented case of siderite recrystallization (Moz­
ley & Carothers, 1 99 2). For these reasons, fluid
inclusions are thought to provide a genuine record
of the temperatures at which the Tirrawarra Sand-
476 M.R. Rezaee and J.P. Schulz-Rojahn
stone siderites crystallized, unless resetting of the
inclusions (McLimans, 1 987; Prezbindowski &
Larese, 1 987; Prezbindowski & Tapp, 1 99 1 ) did
occur. However, consistent differences in homoge­
nization temperatures between S2 and S3 (Fig. 1 3)
suggest that the fluid inclusions did not undergo
re-equilibration, in view of the fact that appreciable
differences in the size of the fluid inclusions were
not detected between S2 and S3. Further, present
temperatures at reservoir level exceed the maxi­
mum homogenization temperatures of S3 by at
least 20-30 ° C, indicating that the fluid inclusions
did not reset during recent geological times. Geohis­
tory modelling shows that Cooper basin sediments
did not undergo major subsidence in the last few
million years in the Patchawarra syncline (Tupper
& Burckhardt, 1 990). Therefore, we conclude that
S2 precipitated at a mean water temperature of
about 68 ° C, whereas S3 formed at about 1 02 ° C on
average (Fig. 1 3).
No fluid inclusion data are available for S 1 ;
however, the cement stratigraphy would suggest
that S 1 crystallized at temperatures lower than those
for S2, i.e. less than about 68 oc. During the Early
Permian the palaeolatitude of central Australia was
about 70-75 ° south (McElhinny, 1 969; Embleton &
McElhinny, 1 982; Veevers, 1 984). The il 1 80 of re­
cent meteoric water is between - 1 5 and - 1 6o/oo at this
latitude (Dansgaard, 1 964). Accordingly, if we as­
sume a il 1 80 value of - 1 5. 5o/oo for Early Permian
pore water, then S1 must have precipitated at a
temperature of about 30"C, which does indeed
correspond to relatively early diagenesis.
Sources of carbon
In the Tirrawarra Sandstone, carbon isotope values
show that major changes in conditions occurred
between the precipitation of S 1 and the later sider­
ite cement generations, S2 and S3. Whereas the
sample where S 1 is the only siderite cement present
has a i> 1 3C composition of about + 1 .45%o, those
samples dominated by S2 or S3 are much more
depleted in ' 3C (Fig. 1 5).
The il 1 3C character of S1 is consistent with a
major source of carbon involving methanogenesis
(Curtis & Coleman, 1 986). Although a marine
source of carbon (-2 to +2%o; Hudson, 1 977) could
also theoretically explain the observed il 1 3C compo­
sition of S 1 , this explanation is inconsistent with
the fluvio-lacustrine nature of the Cooper basin
sediments. Locally, marine Warburton basin lime-
stones of Cambrian to Early Ordovician age uncon­
formably underlie the Cooper basin sediments
(Battersby, 1 976; Gatehouse, 1 986). However, it is
most unlikely that any mineral components were
derived from this limestone source during S I precip­
itation, because there is an almost complete absence
of calcite cement in the Cooper basin. The minor
calcite cement that is present in Permian sediments
tends to concentrate in the shallowest Cooper basin
sediments, furthest away from the Warburton Basin
limestones (Schulz-Rojahn, 1 99 1 ). Further, it is un­
clear why upwelling fluid movement from these
limestones should have coincided with S 1 precipita­
tion but not S2 and S3 precipitation, and how this
very late fluid movement could have been accom­
plished in view of the fact that more than 1 80 million
years separate the deposition of the Cambro­
Ordovician limestones and the Tirrawarra Sand­
stone. Therefore, a major source of carbon
involving methanogenesis is the most probable
explanation for the i> 1 3C character of S 1 .
During methanogenesis, strongly 1 3C-depleted
methane and 1 3C-enriched carbon dioxide is pro­
duced by microbial activity (Rosenfeld & Silver­
man, 1 959; Hudson, 1 977). Acetate fermentation is
the most likely cause of methanogenesis in fresh­
water depositional environments (Whiticar et al.,
1 986). In modern marsh sequences, highly 1 3C­
enriched values are produced by methanogenic
processes during early diagenesis (Moore et al.,
1 992). The carbonate cements produced from bac­
terial fermentation reactions start to precipitate at
some depth below the sediment-water interface
(o;; 1 0 m) from pore waters supersaturated with 1 3C­
rich bicarbonate (.;:; +1 5o/oo), and the i> 1 3C composi­
tion of the pore water becomes progressively poorer
in 1 3C with increasing burial depth (� 1 000 m)
(Irwin et al., 1 977). In the present study the concen­
tration of S 1 in marsh sediments, coupled with the
petrographic evidence and the fact that the elemen­
tal composition of S 1 is similar to that described by
Mozley ( 1 989) for early fresh-water siderite cement
in different geological provinces, suggests that this
cement generation formed in the relatively shallow
diagenetic realm. However, S 1 probably precipi­
tated below the initial zone of pore water supersat­
urated with 1 3C-rich bicarbonate as described by
Irwin et a!. ( 1 977), because S1 has a il 1 3C composi­
tion of +1 .45o/oo rather than +1 5o/oo.
Preferential removal of 1 2C by organic materials
(Schidlowski et al., 1 975, 1 976; Faure, 1 986) can
also contribute toward the 1 3C enrichment of sider-
 Isotope interpretation ofsiderite cement
ite in organic-rich marsh sediments. In the Tirra­
warra Sandstone, both this process and methano­
genesis probably produced the positive o 1 3C
character of the S I cement.
The later siderite cement generations (S2 and S3)
are relatively closely related in terms of their carbon
isotope signatures, having o 1 3 C values of -8.5 and
- I I %o, respectively (Fig. 1 5). They also have exten­
sive substitution by Mg, which was probably de­
rived from the alteration of Mg-rich minerals (such
as micas), which are abundant in nearly all of the
Tirrawarra Sandstones, including in metamorphic
rock fragments. This source of Mg was suggested by
Macaulay et a!. ( 1 993) for siderite cements in the
Magnus Sandstone, North Sea. Mg could also have
been released from kerogen after burial (Desbor­
ough, 1 978). According to Desborough ( 1 97 8), the
higher content of magnesium with respect to cal­
cium in kerogen-rich rocks is probably due to the
preferential concentration of magnesium by blue­
green algae, whose remains released magnesium
during kerogen maturation. The o 1 3C compositions
of both S2 and S3 are consistent with a major
source of carbon involving the thermal decarboxy­
lation of organic matter, which produces a strongly
1 2C-enriched carbon (see Hudson, 1 977; Irwin et
a!., 1 977; Carothers & Kharaka, 1 980; Kharaka et
a!., 1 98 3). The same source of carbon was proposed
by Morad et a!. ( 1 9 94) for Mg-rich siderites in
fluvial Triassic sandstones from southern Tunisia.
Significance of dissolution events
The results from this investigation show that sider­
ite can undergo repeated cycles of precipitation and
dissolution during the diagenetic history of a basin,
and that secondary porosity which is produced as a
result of siderite leaching can be a temporary
phenomenon in clastic provinces. Because it cannot
be determined how much siderite cement was
dissolved during each dissolution event, it is uncer­
tain whether the secondary porosity was volumetri­
cally significant in the past. However, the rather
subtle nature of the dissolution boundaries would
suggest relatively minor dissolution events. The
dissolution boundaries almost certainly mark a
substantial time gap between the precipitation of
each major generation of siderite cement, as is
indicated by the different isotope compositions and
the fluid inclusion data of the various cement
generations.
The first dissolution event (D 1 ) occurred between
477
the precipitation of S 1 and S2, at reservoir temper­
atures probably between about �30 and 6 8 · c, as
suggested by the stable isotope and fluid inclusion
data. The second dissolution phase (D2) took place
after the formation of S2 but before S3 cementa­
tion, in the temperature range of about 68- 1 02 · c
(Fig. 1 7).
The origin of the dissolution events is uncertain,
but in the case of D I may be related to the invasion
of low-pH meteoric pore waters during kaolinite
precipitation (see Bj0rlykke & Brendsdal, 1 986).
The presence of oil at the dissolution boundary
between S2 and S3 may point to the role of organic
processes in triggering carbonate cement dissolu­
tion, prior to or synchronous with petroleum migra­
tion. The second dissolution phase (D2) broadly
coincides with the temperature window for peak
hydrocarbon generation (see Tissot & Welte, 1 978).
It is possible that organic acids which accompanied
kerogen maturation (Schmidt & McDonald, 1 979;
Surdam et a!., 1 984; Burley, 1 986) triggered the
dissolution event. According to Curtis ( 1 983}, ma­
turing kerogen can generate Al-bearing acidic pore
water that produces late-generation kaolinite. In the
S1-
D1 ooo o · · · · · ·
Q uartz ......
Feldspar ?= = = .:.:.:.:.:.: :.:.;.:.;.: ?
Diss ol u ti o n 7:7: ·
Kao linite ? - - - ;:±.��?
S2- ::::: :::::::
D2c}�¥.
l l litisati o n
? ��::�
: ::�:::�:::�:....
-5-3-... ?
0 20 40 60 80 1 00 1 20 140
Temperature (C)
Fig. 17. Generalized paragenetic sequence for the
Tirrawarra Sandstone in the Fly Lake-Moorari area,
Cooper basin. The interpretation is based on the
integration of petrographic, isotope and fluid inclusions
results. The estimated timing of oil generation and
migration is indicated(shaded zone). S l , early
generation of siderite cement; S2, middle generation of
siderite cement; S3, late generation of siderite cement;
D l and 02, first and second phases of siderite
dissolution, respectively.
478 MR. Rezaee and J.P. Schulz-Rojahn
Tirrawarra Sandstone, either meteoric invasion or
source-rock maturation probably accounts for the
association of authigenic kaolinite patches with
siderite spar displaying dissolution (Fig. 4B).
C O N C L US I O NS
Multigenerational siderite in the Tirrawarra Sand­
stone of the Moorari and Fly Lake Fields highlights
the important role of differential cement develop­
ment on bulk-rock isotope signatures. The results
from this investigation show that caution must be
exercised in the interpretation of bulk-rock isotope
signatures, even when only a single carbonate ce­
ment phase is indicated by semiquantitative bulk
XRD analysis. Failure to recognize isotopically
heterogeneous samples may lead to erroneous inter­
pretations of the isotope results. Only the end­
member isotope compositions of the different
cement generations should be taken into consider­
ation when interpreting the genesis of multigenera­
tion siderite cements. Generally, the different
generations are not readily identifiable under the
optical microscope, highlighting the importance of
BSE image analysis for siderite characterization. In
particular, the integrated use of video capture and
image analysis software provides an efficient means
of quantifying the different siderite cement genera­
tions seen under BSE. As the method can be
semi-automated, the technique provides a poten­
tially powerful tool for improved bulk-rock isotope
interpretations in clastics containing multigenera­
tional carbonate cements. It allows the determina­
tion of end-member o180 and o13C compositions of
individual cement generations in cases where pure,
or nearly pure, samples of end-member carbonate
cement generations are not available for isotope
analysis, provided that a statistically representative
number of BSE images is analysed, and that the
various cement generations have different chemical
compositions.
In the Tirrawarra Sandstone of the Moorari and
Fly Lake Fields, the application of the BSE image
analysis technique, together with bulk-rock isotope
and fluid inclusion studies, has led to the identifi­
cation of three main generations of siderite cement.
The first and second generations were each followed
by at least one dissolution event. The first genera­
tion of siderite cement is a homogeneous Fe-rich
siderite with a o13C signature of + 1 .45o/oo, which
probably formed during low-temperature methano-
genesis (:c;; 3 Q0C). The second generation is an Mg­
rich inhomogeneous siderite characterized by a
complex zoning, with a o 1 3C signature of -8. 5o/oo. It
is thought to have formed mainly by the decarbox­
ylation of organic matter at temperatures between
64 and 76 oc. The third and final precipitation
event produced an Mg-rich, relatively homoge­
neous pore-filling siderite with a o 1 3C character of
- 1 1 o/oo. This is also interpreted to have formed
during kerogen maturation, albeit at more elevated
temperatures (98- 1 1 0°C). The results from this
study show that organic processes controlled sider­
ite cementation over a range of different burial
conditions in the study area.
A C K N O W L E D G E M E N TS
The authors thank Drs Nick Lemon (NCPGG) and
John Collen (Victoria University of Wellington) for
their constructive criticisms of the draft manu­
script. The manuscript was greatly improved by the
comments of lAS reviewers Drs Sadoon Morad
(University of Uppsala), Earle F. McBride (Univer­
sity of Texas at Austin) and Richard Worden
(Queen's University of Belfast). The authors grate­
fully acknowledge the financial support of the
NCPGG, the Australian Research Council (ARC)
and SANTOS Ltd.
R E F E R E N C ES
AL-AASM, I.S., TAYLOR, B.E. & Sou TH , B. (1990) Stable
isotope analysis of multiple carbonate samples using
selective acid extraction. Chern. Geol., 80. 1 19-125.
APAK, S.N. (1994) Structural de velopmenl and comrol on
stratigraphy and sedimentati on i n the Cooper Basin
Austral ia. PhD thesis, University of Adelaide. 105 pp.
APAK, S.N., STUART, W.J. & LEMON, N.M. ( 199 3 )
Structural-stratigraphic development o f the Gidgealpa­
Merrimelia-Innamincka Trend with implications for
petroleum trap styles, Cooper Basin, Australia. A ust.
Petrol. Explor. Ass. J., 33, 94-l 04.
ARMSTRONG, J.D. & BARR, T.M. (1986) The Eromanga
Basin. An overview of exploration and potential. In:
Eromanga Bas in Sympos ium (Eds Moore, P.S. &
Mount, T.J.). Geol. Soc. Aust. Spec. Pub!., 1 2. 25-38.
BAH RIG, B. ( 1989) Stable isotope composition of siderite
as an indicator of the paleoenvironmental history of oil
shale lakes. Palaeogeog., Palae ocl imatol.. Palaeoecol.,
70, 139-151.
BATTERSBY, D.G. (1976) Cooper Basin oil and gas fields.
In: Econom ic Ge ology ofAustral ia and Ne»· Guinea. 3
(Eds Leslie, R.B., Evans, H.J. & Knight, C.L). Petrol.
Australasia Inst. Mineral. Metal., 7. 321-370.
 Isotope interpretation ofsiderite cement
BEVER, J.M. CARROL, P.G., WILLIAMS, B.P.J. & WILD, E.K.
( 1987) Petrography and core study, T irrawarra Sand­
stone. Moorar i and T irrawarra Fields, 2 vols. Unpub­
lished report, SAGASCO Ltd, 1 38 pp.
BEVER, J.M., CARROLL, P.G., WILD, E. & WILLIAMS, B.P.J.
( 1 988) Core facies, petrology, and permeability of
Tirrawarra Sandstone, Moorari Field, Cooper Basin,
South Australia (abstract). Bull. Am. Ass. Petrol. Geol. ,
72, 1 62.
BJ0RLYKKE, K. & BRENDSDAL, A. ( 1 986) Diagenesis of the
Brent Sandstone in the Statfjord Field. In: Spec. Pub!.
Soc. Econ. Paleont. Miner., Tulsa, 38, 1 57- 1 67.
BURLEY, S.D. ( 1 986) The development and destruction of
porosity within Upper Jurassic reservoir sandstones of
the Piper and Tartan Fields, Outer Moray Firth, North
Sea. Clay M in eral. , 2 1 , 649-694.
CAROTHERS, W.W. & KHARAKA, Y.K. ( 1980) Stable carbon
isotopes of HC03 in oil-field waters-implications for
the origin of C02. Geoch im. Cosmoch i m . Acta, 44,
323-332.
CARPENTER, S.J., ERICKSON, J.M., LOHMANN, K.C. & OWEN,
M.R. ( 1988) Diagenesis of fossiliferous concretions
from the Upper Cretaceous Fox Hills Formation, North
Dakota. J. sedi m ent. Petrol. , 58, 706-723.
CRAIG, H. ( 1 957) Isotope standards for carbon and oxygen
and correction factors for mass spectrometric analysis
of carbon dioxide. Geoch i m . Cosmoch i m . Acta, 1 2,
1 33- 1 49.
CRAIG, H. ( 1 96 1 ) Standards for reporting concentrations
of deuterium and oxygen- 1 8 in natural waters. Sc i e nce,
133, 1 833- 1 834.
CuRTIS, C.D. ( 1 983) Link between aluminium mobility
and destruction of porosity. Bull. Am. Ass. Petrol. Geol. ,
63, 380-384.
CuRTIS, C.D. & COLEMAN, M.L. ( 1986) Controls on the
precipitation of early diagenetic calcite, dolomite, and
siderite concretions in complex depositional sequences.
In: Roles ofOrgan ic Matter in SedimentD i agenes is (Ed.
Gautier, D. L.). Spec. Pub!. Soc. Econ. Paleont. Miner.,
Tulsa, 38, 23-33.
CURTIS, C.D., PEARSON, M.J. & SOMOGYI, V.A. ( 1 975)
Mineralogy, chemistry, and origin of a concretionary
siderite sheet (clay-ironstone band) in the Westphalian
of Yorkshire. M in er. Mag., 40, 385-393.
DANSGAARD, W. ( 1964) Stable isotopes in precipitation.
Tel/us, 16, 436-468.
DEER, W.A., HowiE, R.A. & ZusSMAN, J. ( 1 992) An
Introduction to the Rock-Form ing M in erals. Longman,
Harlow, 695 pp.
DEINES, P. ( 1986) The isotopic composition of reduced
organic carbon. In: Handbook ofEnv i ronmental Isotope
Geoche m i stry (Eds Fritz, P. & Fontes, J.Ch.), pp. 329-
406. Elsevier Scientific, New York.
DESBOROUGH, G.A. ( 1978) A biogenic-chemical stratified
lake model for the origin of oil shale of the Green River
Formation: an alternative to the playa-lake model. Geol.
Soc. Am. Bull., 89, 961-97 1 .
DICKSON, J.A.D., SMALLEY, P.C., RAHEIM, A . & STIJF­
HOORN, D.E. ( 1990) Intercrystalline carbon and oxygen
isotope variations in calcite revealed by laser microsam­
pling. Geology, 18, 809-8 1 1 .
EMBLETON, B.J.J. & McELHINNY, M.W. ( 1982) Marine
magnetic anomalies, palaeomagnetism and the drift
479
history of Gondwanaland. Earth Planet. Sc i. Lett. 58,.
1 4 1- 150.
ExoN, N.F. & SENIOR, B.R. ( 1 976) The Cretaceous of the
Eromanga and Surat Basins. Bur. M in er. Resour. J. Aust.
Geol. Geophys. , 1 , 33-50.
FAIRBURN, W.A. ( 1989) The geometry of Toolachee unit
'C' fluvial sand trends, Moomba Field, Permian Cooper
Basin, South Australia. Aust. Petrol. Explor. Ass. 1. 29.
.
239-250.
FAURE, G. ( 1 986) Pri n c iples of Isotope Geology, 2nd edn.
John Wiley & Sons, Chichester, 589 pp.
FOLK, R.L. ( 1 974) Petrology ofSedi mentary Rocks. Hemp­
hill, Austin, TX, 1 82 pp.
FRITZ, P., BINDA, P.L., FOLINSBEE, F.E. & KROUSE, H.R.
(I 97 1 ) Isotopic composition of diagenetic siderites from
Cretaceous sediments in Western Canada. J. sediment.
Petrol. , 4 1 , 282-288.
GATEHOUSE, C.G. ( 1 972) Formations of the Gidgealpa
Group in the Cooper Basin. Aust. O il Gas J. , 18, I 0- 1 5.
GATEHOUSE, C.G. ( 1 986) The geology of the Warburton
Basin in South Australia. Aust. J. Earth Sc i., 33, 1 6 1 -
1 80.
GAUTIER, D.L. ( 1 982) Siderite concretions: indicators of
early diagenesis in the Gammon Shale (Cretaceous). J.
sediment. Petrol., 52, 859-87 1 .
GOLDSTEIN, R.H. & REYNOLDS, T.J. ( 1 994) Systematics o(
Flu id Inclus ions in D i agenetic M in erals. Soc. Econ.
Paleont. Miner. Short Course, 31, 1 99 pp.
GouLD, K.W. & SMITH, J.W. ( 1 979) The genesis and iso­
topic composition of carbonates associated with some
Permian Australia coals. Chern. Geol., 24, 137-1 50.
HASZELDINE, R.S. & OSBORNE, M. ( 1 993) Fluid inclusion
temperatures in diagenetic quartz reset by burial: impli­
cations for oil field cementation. In: D i agenes is and
Bas in Development (Eds Horbury, A.D. & Robinson.
A.G.). Am. Ass. Petrol. Geol. Studies in Geology, 36,
35-46.
HEATH, R. ( 1 989) Exploration in the Cooper Basin. Aust.
Petrol. Explor. Ass. J. , 29, 366-378.
HuDSON, J.D. ( 1977) Stable isotopes and limestone lithifi­
cation. J. Geol. Soc. London, 133, 637-660.
HuNT, J.W., HEATH, R.S. & McKENZIE, P.F. ( 1 989) Ther­
mal maturity and other geological controls on the
distribution and composition of Cooper Basin hydro­
carbons. In: The Cooper and Eromanga Bas ins. Austra­
lia (Ed. O'Neil, B.J.), pp. 509-523. Proceedings of the
Cooper and Eromanga Bas ins Conference. Adela ide.
26-27 June 1 989. Petroleum Exploration Society of
Australia, Society of Petroleum Engineers, Australian
Society of Exploration Geophysicists (SA branches).
IRWIN, H. ( 1 980) Early diagenetic carbonate precipitation
and pore fluid migration in the Kimmeridge Clay of
Dorset, England. Sedimentology, 27, 577-59 1 .
IRWIN, H., CuRTIS, C.D. & COLEMAN, M. ( 1 977) Isotopic
evidence for source of diagenetic carbonates formed
during burial of organic-rich sediments. Nature , 269,
209-213.
KANTSLER, A.J., PRUDENCE, T.J.C., COOK, A.C. & ZWIGU­
LIS, M. ( 1 983) Hydrocarbon habitat of the Cooper/
Eromanga Basin. Aust. Petrol. Explor. Ass. 1., 23, 373-
389.
KAPEL, A.J. ( 1 966) The Cooper's Creek Basin. Aust. Petrol.
Explor. Ass. J. , 6, 7 1-75.
480 M.R. Rezaee and J. P. Schulz-Rojahn
KAPEL, A.J. ( 1972) The geology of the Patchawarra area
Cooper Basin. Aust. Petrol. Explor. Ass. J., 1 2, 53-57.
KHARAKA, Y.K., CAROTHERS, W.W. & RosENBAUER, R.J.
(1983) Thermal decarboxylation of acetic acid: implica­
tions for origin of natural gas. Ge och im. Cosmoch im.
Acta, 47, 397-402.
LAWS, R.A. (1989) Preface. In: The Cooper andEromanga
Bas ins, Austral i a (Ed. O'Neil, B.J.), pp. v-vi. Proceedings
of the Cooper and Eromanga Bas ins Conference, Ade­
la ide, 26-27 June 1 989. Petroleum Exploration Society
of Australia, Society of Petroleum Engineers, Australian
Society of Exploration Geophysicists (SA branches).
MACAULAY, C.!., HASZELDINE, R.S. & FALLICK, A.E. (1993)
Distribution, chemistry, isotopic composition and ori­
gin of diagenetic carbonates: Magnus Sandstone, North
Sea. J. sediment. Petrol., 63, 33-43.
MARTIN, K.R. (1981) Petrology and D i agenes is of the
T irrawarra Sandstone in Tirrawarra Field. Unpublished
report, Santos Ltd, I 7 pp.
MARTIN, K.R. (1984) Petrology and D i agenes is and Reser­
vo ir Qual ity ofthe T i rrawarra Sandstone in T i rrawarra
Fly Lake and Mo orar i F i e lds. Unpublished report,
Santos Ltd, 31 pp.
MARTIN, K.R. & HAMILTON, N.J. (1981) Diagenesis and
reservoir quality, Toolachee Formation, Cooper Basin.
Aust. Petrol. Explor. Ass. J., 2 1 , 143-154.
MATSUMOTO, R. & hJIMA, A. (I 981) Origin and diagenetic
evolution of Ca-Mg-Fe carbonates in some coalfields of
Japan. Sedimentology, 18, 239-259.
McCREA, J.M. (1950) On the isotopic chemistry of carbon­
ates and a palaeotemperature scale. J. Chern. Phys. , 18,
849-857.
McELHINNY, M.W. (1969) The palaeomagnetism of the
Permian of southeast Australia and its significance
regarding the problem of intercontinental correlation.
In: Geol. Soc. Aust. Spec. Publ., 2, 61-67.
McLIMANS, R.K. (1987) The application of fluid inclusions
to migration of oil and diagenesis in petroleum reser­
voirs. Appl. Geochem., 2, 585-603.
MOORE, S.E., FERRELL, R.E. & AHARON, P. ( 1992) Diage­
netic siderite and other ferroan carbonates in a modern
subsiding marsh sequence. J. sediment. Petrol., 62, 357-
366.
MoRAD, S., BEN ISMAIL, H.N., DE Ros, L.F., AL-AASM, I.S.
& SERRHINI, N.-E. (1994) Diagenesis and formation
water chemistry of Triassic reservoir sandstones from
southern Tunisia. Sedimentology, 4 1 , 1253-1272.
MOZLEY, P.S. ( 1989) Relation between depositional envi­
ronment and the elemental composition of early diage­
netic siderite. Geology, 17, 704-706.
MOZLEY, P.S. & CAROTHERS, W.W. ( 1992) Elemental and
isotopic composition of siderite in the Kuparuk Forma­
tion, Alaska: effect of microbial activity and water/
sediment interaction on early pore-water chemistry. J.
sediment. Petrol., 62, 681-692.
OSBORNE, M. & HASZELDINE, R.S. (1993) Evidence for
resetting of fluid inclusion temperatures from quartz
cements in oilfields. Mar. Petrol. Geo/., 10, 27 1-278.
PEARSON, M.J. ( 1 985) Some chemical aspects of diagenetic
concretions from the Westphalian of Yorkshire, Eng­
land. Chern. Geo/., 3 1 , 225-244.
PREZBINDOWSKI, D.R. & LARESE, R.E. ( 1987) Experimental
stretching of fluid inclusions in calcite-implications for
diagenetic studies. Geology, 15, 333-336.
PREZBINDOWSKI, D.R. & TAPP , J.B. (199 1 ) Dynamics of
fluid inclusion alteration in sedimentary rocks: a review
and discussion. Org. Geoche m., 17, 131- 1 42.
REYNOLDS, T.J. (1978) Flu id Inclus i on Adapted USGS Gas
Flow Heating/Freez ing Instructi on Manual. Fluid Incor­
porated, Denver, CO.
RICIPUTI, L.R. & PATERSON, B.A. (1994) High spatial­
resolution measurement of 0 isotope ratios in silicates
and carbonates by ion microprobe. Am. Mineral. , 79.
1227-1230.
RODDA, J.S. & PASPALIARIS, T.G. ( 1989) Tirrawarra and
Moorari Oil Fields enhanced oil recovery schemes­
further developments. Aust. Petrol. Explor. Ass. J.. 29.
121-130.
ROSENBAUM, J. & SHEPPARD, S.M.F. (1986) An isotopic
study of siderites, dolomites and ankerites at high
temperatures. Ge och im. Cosmoch im. Acta. 50. 114 7-
1150.
ROSENFELD, W.O. & SILVERMAN, S.R. (1959) Carbon iso­
tope fractionation in bacterial production of methane.
Sc i e nce, 1 30, 1658-1659.
SCHIDLOWSKI, M., EICHMANN, R. & JUNGE. C. ( 1 975)
Precambrian sedi!Tlentary carbonates: carbon and oxy­
gen isotope geochemistry and implication for the terres­
trial oxygen budget. Precambr ian Res . 2, 1-69.
.
SCHIDLOWSKI, M., EICHMANN, R. & JUNGE, C. (1976)
Carbon isotope geochemistry of the Precambrian Loma­
gundi carbonate province, Rhodesia. Geoch im. Cosmo­
ch im. Acta, 40, 449-455.
SCHMIDT, V. & Mc DONA L D , D.A. ( 1 979) The role of
secondary porosity in the course of sandstone diagene­
sis. In: Aspects of D i agenes is (Eds Scholle. P.A. &
Schluger, P.R.). Spec. Publ. Soc. Econ. Paleont. M iner..
Tulsa, 26, 175-207.
ScHULZ-ROJAHN, J.P. (1991) Or igin, evolut i on and controls
of Perm ian reservo ir sandstones in the southern Cooper
Bas i n , South Austral ia. PhD thesis. National Centre for
Petroleum Geology and Geophysics. University of Ade­
laide, 187 pp.
SCHULZ-ROJAHN, J.P. & PHILLIPS, S.E. ( 1989) Diagenet ic
alteration of Permian reservoir sandstones in the Nap­
pamerri Trough and adjacent areas. southern Cooper
Basin. In: The Cooper and Eromanga Basins. A usrralia
(Ed. O'Neil, B.J.), pp. 629-645. Proceedings of" rhe
Cooper and Eromanga Bas ins Con/erence. Adelaide.
26-27 June 1 989. Petroleum Exploration Society of
Australia, Society of Petroleum Engineers. Australian
Society of Exploration Geophysicists (SA branches).
SEGGIE, R.J., LANSOM, P.B., HAMLIN, H.S. & JOHNSON, G.A.
( 1994) The Tirrawarra oil field: field revitalisation
through reservoir description and characterisation.
Aust. Petrol. Explor. Ass. J., 34, 33-54.
SENIOR, B.R., MOND, A. & HARRISON. P.L. ( 1978) Geology
of the Eromanga Basin. Bur. M in er. Resour. Bull. , 167.
102.
SMALLEY, P.C., MAILE. C.N., COLEMAN, M.L. & ROUSE. J.E.
( 1992) Lassie (laser ablation sampler for stable isotope
extraction) applied to carbonate minerals. Chern. Geol. ,
1 0 1 , 43-52.
SMYTH, M. (1979) Hydrocarbon generation in the Fly
Lake-Brolga area of the Cooper Basin. Ausr. Petrol.
Explor. Ass. J. , 19, 108-114.
 Isotope interpretation of siderite cement
SPIRO, B., GIBSON, P.J. & S HAW, H.F. (1993) Eogenetic
siderite in lacustrine oil shales from Queensland, Aus­
tralia, a stable isotope. Chern. Geo/., 106, 415-427.
SP6TL, C., MATTER, A. & BREVART, 0. (1993) Diagenesis
and pore water evolution in the Keuper Reservoir, Paris
Basin (France). J. sediment. Petrol. , 63, 909-928.
STANLEY, D.J. & HALLIDAY, G. (1984) Massive hydraulic
fracture stimulation of Early Permian gas reservoirs, Big
Lake Field, Cooper Basin. Aust. Petrol. Exp/or. Ass. J. ,
24, 180- 1 95.
STUART, W.J. ( 1976) The genesis of Permian and Lower
Triassic reservoir sandstones during phases of southern
Cooper Basin development. Aust. Petrol. Explor. Ass. J. ,
16, 37-47.
STUART, W.J., KENNEDY, S. & THOMAS, A.D. (1988) Influ­
ence of structural growth and other factors on the
configuration of fluviatile sandstones, Permian Cooper
Basin. Aust. Petrol. Explor. Ass. J. , 28, 255-265.
STUART, W.J., FARROW, 8.8., LEMON, N.L. & PHILLIPS, S.E.
( 1990) Porosity and Permeability in Permian Sand­
stones: Southern Cooper Basin. Interim report,
NERD DC Project No. 117 5, National Centre for Petro­
leum Geology and Geophysics, University of Adelaide,
104 pp.
STUART, W.J., FARROW, B.B., TINGATE, P.R. e t a/. (199 1 )
Porosity and Permeability in Permian Sandstones:
Southern Cooper Basin. End-of-grant report, NERD DC
Project No. 1407, National Centre for Petroleum Geol­
ogy and Geophysics, University of Adelaide, 198 pp.
SURDAM, R.C., BOESE, S.W. & CROSSEY, L.J. ( 1984) The
chemistry of secondary porosity. In: Clastic Diagenesis
(Eds McDonald, D.A. & Surdam, R.C.). Mem. Am. Ass.
Petrol. Geol., Tulsa, 37, 1 27-149.
THORNTON, R.C.N. ( 1979) Regional stratigraphic analysis
of the Gidgealpa Group, Southern Cooper Basin, Aus­
tralia. Geo/. Surv. South Aust. Bull. , 49, 140 pp.
48 1
TISSOT, B.P. & WELTE, D.H. (1978) Petroleum Forma/ion
and Occurrence-a New Approach to Oil and Gas Explo­
ration. Springer-Verlag, Berliri, 538 pp.
TUPPER, N.P. & BURCKHARDT, D.M. (1990) Use of the
methylphenanthrene index to characterise expulsion of
Copper and Eromanga Basin oils. Aust. Petrol. Exp/or.
Ass. J., 30, 373-385.
VEEVERS, J.J. (1984) Phanerozoic Earth History o(Austra­
/ia. Oxford Geological Science Series, Oxford Univer­
sity Press, New York, 418 pp.
WHITICAR, M.J., FABER, E. & ScHOELL, M. ( 1 986) Biogenic
methane formation in marine and fresh water environ­
ments: CO, reduction vs. acetate fermentation-isotope
evidence. Geochim. Cosmochim. Acta, 50, 693-709.
WILD, E.K. ( 1987) The sedimentology and reservoir qualil r
of the Kinnerton Sandstone Formation, UK, and the
Tirrawarra Sandstone, S. Australia. PhD thesis, Depart­
ment of Geology, University of Bristol.
WILLIAMS, B.P.J. & WILD, E.K. ( 1 984) The Tirrawarra
Sandstone and Merrimelia Formation of the Southern
Cooper Basin, South Australia-the sedimentation and
evolution of a glaciofluvial system. Aust. Petrol. Exp/or.
Ass. J., 24, 37 7-392.
WILLIAMS, B.P.J., WILD, E.K. & SUTTILL, R.J. ( 1985)
Paraglacial aeolianites: potential new hydrocarbon res­
ervoirs, Gidgealpa Group, southern Cooper Basin. Aust.
Petrol. Exp/or. Ass. J. , 25, 291-310.
YEw, C.C. & MILLS, A.A. (1989) The occurrence and
search for Permian oil in the Cooper Basin, Australia.
In: The Cooper and Eromanga Basins, Australia (Ed.
O'Neil, B.J.), pp. 339-359. Proceedings of the Cooper
andEromanga Basins Conference, Adelaide. 26-2 7 June
1 989. Petroleum Exploration Society of Australia. Soci­
ety of Petroleum Engineers, Australian Society of Ex­
ploration Geophysicists (SA branches).
Spec. Pubis int. Ass. Sediment. ( 1998) 26, 483-499

Carbonate cement dissolution during a

cyclic C02 enhanced oil recovery treatment

L. K. S M I T H
Institute for Energy Research, PO Box 4068,
Laramie, WY 82071, USA,
e-maill ksmith@uwyo.edu

ABSTRACT

Diagenetic and reservoir studies usually require original, uncontaminated formation water. However,
abundant information can be obtained using water produced from enhanced oil recovery projects. In
this study of cyclic C02 well treatments in six oilfields in Wyoming, USA, produced water chemistry
was monitored for several weeks after the C02 treatment. In all cases, increased concentrations of
calcium, magnesium, bicarbonate, silica and aluminium indicated mineral dissolution. The ratio of
calcium to magnesium provided information about the relative dissolution rates between calcite and
dolomite, assuming congruent dissolution. At very high PC02 the dolomite dissolution rate may
approach that of calcite. Comparing post-treatment mineral saturation indices for the produced water
indicated that the ratio of calcite to water in the reservoir is more important to dissolution rates than
PC02, because reservoirs with higher carbonate mineral/water ratios were more oversaturated with
respect to calcite than reservoirs with higher PC02. Also, from the mineral saturation indices it was
found that injected C02 can come to equilibrium with carbonate minerals in the reservoir in
approximately 6 months (at P 18 MPa). This type of information is useful in scaling up laboratory
=

dissolution experiments to field conditions. For the aluminosilicate minerals dissolution and/or
alteration is more difficult to discern from the water chemistry. However, aluminium concentrations
peaked later than silica concentrations in the two wells where such analyses were performed. From this,
it was surmised that a more silica-rich mineral (such as feldspar) dissolved/altered first, followed later
by the dissolution/alteration of a more aluminium-rich mineral (such as clay). The shape of the
concentration-time profiles also provided information about the reservoir. In a good treatment, where
oil production increased after the C02 injection, the concentration for all the ions, except aluminium,
peaked immediately after production recommenced and then followed a steady decline back to
pretreatment levels. In unsuccessful treatments ionic concentrations peaked much later, or peaked and
did not decline. This information from produced water was used to assess reservoir heterogeneity,
which led to guidelines for choosing the wells most likely to be good candidates for cyclic C02
treatments.

INTRODUCTION

Oil and gas field waters have long been used in
diagenetic studies, primarily as vehicles for under­
standing basin fluid flow, oil migration pathways,
evolution of water composition, and reservoir com­
partmentalization. However, water samples used
for these purposes must be original formation wa­
ter, not mixed with drilling mud or with injected
fluids such as fracture fluid, waterflood water, or
other fluids used for enhanced oil recovery (EOR),
as such contaminants generally render the sample
useless for studying many aspects of the formation.
However, this is not always true. There are notable
exceptions of studies that specifically used forma­
tion water subjected to some human-induced per­
turbation. Several studies have examined waters
from steam flood injection and in situ combustion
projects. These studies examined the water-rock
interaction caused by the perturbation and applied

Carbonate Cementation in Sandstones Edited by Sadoon Morad 483

© 1998 The International Association of Sedimentologists
ISBN: 978-0-632-04777-2
484 L .K Smith
that knowledge to our understanding of diagenesis
from natural geological processes (Hutcheon et a!.,
1981, 1988, 1990, 1995; Gunter & Bird, 1988;
Hutcheon & Abercrombie, 1990; Abercrombie,
1991; Perkins et a!., 1992). In another type of EOR,
Bowker & Shuler (1989, 1991) looked at produced
water from the miscible carbon dioxide flood at the
Rangely field in Colorado, USA. Their primary
concern was the prediction of barite scale in the
production wells. However, their study also pro­
vided information about how the rocks altered in
contact with injected C02 and river water.
This study provides additional new information
on using produced water from a different type of
EOR project-immiscible cyclic C02. The pro­
duced water chemistry is used to ascertain well-bore
scale potential, relative rates of mineral dissolution
and reservoir heterogeneity. Much of the informa­
tion and methodology contained herein applies
equally well to other types of EOR or to other
reservoirs where produced water chemistry is mon­
itored regularly. Also, the information about min­
eral dissolution has applications for any study
involving mineral dissolution and diagenesis caused
by low pH and to the upscaling of experimental
laboratory results.
METHODOLOGY
Water sample collection and analysis
Pre-C02 water compositions were determined for
the fields to provide a baseline for comparison. This
was done either by direct sampling of the well prior
to treatment (Crooks Gap), by sampling an adjacent
well (Bonanza), or by using historical analyses from
operator well files (Cole Creek, Grass Creek, Beaver
Creek and North Grieve). After C02 injection the
water composition was monitored for several weeks
after the well was returned to production. Except
for two samples, Crooks Gap and Bonanza water
samples were collected and treated at the well site
following the guidelines of Lico et a!. (1982). Alka­
linity and pH were measured at the site. In the
laboratory these samples were analysed for Na, K,
Ca, Mg, Fe, AI and Si02 by inductively coupled
plasma mass spectrometry and for Cl and S04 by
ion chromatography. Samples for the other fields
were taken by the operators and were not treated at
the well site. In the laboratory these samples were
filtered and acidified, and analyses included only
major cations and sulphate. The Beaver Creek well
had a report only on total dissolved solids (TDS) for
pretreatment water. To facilitate comparison be­
tween pre- and post-C02 treatment, TDS were
estimated for the post-treatment water (Smith,
1993). Because complete analyses were performed
on Crooks Gap and Bonanza samples, they received
the most extensive study and are shown in Table I. 1
Chemical modelling
The chemical modelling performed for this study
used SOLMINEQ.88 (Kharaka et a!., 1988) to
calculate the mineral saturation indices. This is an
equilibrium thermodynamic model which calcu­
lates the distribution of species by solving a set of
mass-action and mass-balance equations based on
the ion-association model (Garrels & Christ, 1965;
Helgeson, 1967). To model C02 injection into the
reservoir, the program's C02 option was used with
a specified PC02. Also, the mass transfer option
was used to predict changes in water chemistry.
These changes would result from mineral dissolu­
tion or precipitation caused by the change in the
mineral saturation state of the water.
CYCLIC C02 ENHANCED
OIL RECOVERY TREATMENTS
Cyclic C02 EOR treatments consist of three stages.
In the first, carbon dioxide is injected into a reser­
voir over a period of hours to a couple of days,
depending upon the amount injected. The second
stage consists of shutting the well in and allowing
the C02 to 'soak' in the reservoir for between 2 and
4 weeks. In the third stage, production is recom­
menced from the well into which the C02 was
injected. Cyclic C02 treatments. are performed at
pressure and temperature (P-T) conditions that are
below the minimum miscibility point between oil
and C02, placing the C02 density below approxi­
mately 0.25 g/cm3. This leaves a C02 gas phase in
the reservoir that occupies approximately 50% of
the pore space of the treated reservoir volume (T.
Monger-McClure, Marathon Oil Co., personal com­
munication). In contrast, well-to-well C02 floods
are usually at oii-C02 miscible conditions. There
are two mechanisms by which the cyclic process is
1 Analyses for Beaver Creek, Cole Creek, Grass Creek and
North Grieve can be obtained from the author.
 Carbonate cement d issolution 485

Table I. Pre- and post-C02 treatment water analyses for Crooks Gap and Bonanza Fields

Na K Ca Mg Si02 Fe AI Cl so. HC03 pH

Crooks Gap Field

pre-C02 2460 27 14 4 34 0.1 0.1 3110 44 1182 7.8
post-C02 day 3 4276 175 30 77 3.5 0 5853 70 1879 8.3
post-C02 day 11 4303 184 31 100 0.1 0 5961 81 1776 8.2
post-C02 day 12 4095 155 29 75 0.1 0 5605 71 1751 8.5
post-C02 day 14 4016 146 28 73 0.2 0 5575 73 1556 7.7
post-C02 day 16 3798 136 27 70 0.1 0 5166 72 1647 7.6
post-C02 day 17 3534 37 122 25 72 8.8 0.7 4754 68 1629 7.8
post-C02 day 25 2426 88 30 33 0.2 0.4 2771 86
post-C02 day 25 2711 142 31 40 21 0.2 3467 85
post-C02 day 26 2405 65 15 43 3.3 0 2873 48 1647 7.4
post-C02 day 31 3744 14 24 64 0.1 0 4983 70 1440 7.5

Bonanza Field
pre-C02 25 25 258 108 11 0.12 0.0 7.1 774 588 7.2
post-C02 day 1 20 13 796 194 34 1.2 0.0 3.6 543 2282 6.9
post-C02 day 4 17 13 828 200 33 9.2 0.1 3.7 580 2361 6.7
post-C02 day 10 18 12 819 191 32 8.0 0.2 6.3 572 2282 6.8
post-C02 day 20 19 15 722 176 27 6.4 0.2 4.5 674 1989 6.7
post-C02 day 26 20 16 698 164 26 5.4 0.2 4.2 708 1818 6.7
post-C02 day 38 18 18 712 168 25 7.6 0.1 3.8 727 1776 6.5
post-C02 day 50 24 17 628 144 21 6.4 0.3 3.7 682 1544 6.5

thought to enhance oil production. One is a volume 9600 ft) (Smith, 1993; Smith & Surdam, 1993b).
increase or 'swelling' of the oil, and the other is a The Cretaceous fields are all similar in their
reduction of the oil's viscosity (Monger et al., 1988). framework grain composition and in the authigenic
Water injection is not involved in a cyclic C02 phases present (Table 2). There is one important
treatment as it is in well-to-well floods. The excep­ difference to note, however, in comparing the two
tion to this is when a small amount of water is used Dakota Formation producing fields. At Crooks
to displace into the formation any C02 that remains
in the well bore at the end of the injection. Typi­
cally, a well is treated two or three times. The
amount of increased oil recovery from each succes­
sive treatment generally declines from the previous
treatment until it is no longer economical to inject
more C02.

CYCLIC C02-TREATED FIELDS

IN THIS STUDY

Comparison and contrast among the treated fields

was enhanced by their being alike in some parame­
ters and different in others (Table 2). The six fields
represent four basins in Wyoming, USA (Fig. 1).
Five of the fields produce from Cretaceous forma­
tions (two of them being the same formation, i.e.
Dakota), and one produces from a Pennsylvanian
formation (Tensleep). Oil gravity is roughly the
same in all the fields, even though the present-day Fig. I. Map of Wyoming showing the location of the
depth ranges between 335 and 2900 m (1100 and fields where cyclic C02 treatments were performed.
 ""'"
00
0'1
Table 2. Summary data for the C0 -treated fields
2
Beaver Creek Bonanza Cole Creek Crooks Gap Grass Creek North Grieve

Field data
Formation 2nd Cody Sst Tensleep Dakota Dakota Frontier Muddy
Nature of trap Asymmetric Asymmetric Anticline and Asymmetric faulted Asymmetric faulted Stratigraphics
anticline anticline stratigraphic anticline anticlines
pinchout
Drive mechanism Solution gas drive Water drive Pump Water drive1 Solution gas drive Gas expansions
Secondary None None Waterflood None Waterflood Water injection
Recovery for pressure
maintenance
Reservoir Goods Continuous Variable Variables Continuous Limited
Continuity
Lithology Shaley sandstone Sandstone, sandy Sublitharenite3 Sublitharenite2•3 Shaley sandstone, Subarkosic
subarkosic2 sandstone6
Cements Calcite, siderite, Dolomite, calcite2 Calcite, dolomite, Calcite, dolomite, Calcite, feldspar, Quartz, dolomite,
quartz, chlorite1 quartz, chlorite, quartz, chlorite, quartz, calcite,
kaolinite3 kaolinite2•3 montmorillonite4•5 kaolinite2• 7
r<
Well data �
Well number BC 37 BON 8 CC 42X-26G CG4 GC 11-33 NG 1-36
Date
Depth
1955
1056 m
1952
777 m
1970
2620 m
1945
1497 m
1987
336 m
1986
2920 m
�s.
Thickness 43 m 56 m 5m 18 m 17 m 12 m
Net pay 5m 42 m 5m 18 m 9m 12 m
Completion Open hole Perf. 42 m, from Open hole Perf. 1497-1515 m Perf. 336-346 m, Perf. 2920-2932 m
1056-1099 m 777 to 828 m 2620-2625 m sand frac
Porosity 13% 19.8% from adjacent 11.7% 23% 27% 13%
well
Permeability No data 314 mD from 55 mD 232 mD 21 mD from nearby 47 mD
adjacent well well
Temperature 46"C 29"C 93"C 64"C 27"C 66"C
Pressure 7.6 MPa 2.66 MPa 17.9 MPa 8.14 MPa 1.7-2.2 MPa 14 MPa

Perf., perforated.
1 Hansley & Whitney (1990).
2Smith ( 1993).
3Reisser & Blanke ( 1989).
4Surdam et a/. ( 1989).
5Tillman & Almon (1979).
6Von Dehrle ( 1975).
7Webb (1974).
swyoming Geological Association ( 1989).
 Carbonate cement d issolution
Fig. 2. Thin-section photomicrograph of Dakota
sandstone from Crooks Gap field, 1539 m (5048 ft),
showing small, dispersed nature of the calcite (arrow).
Scale bar 144 Jlm.
=
Gap, calcite cement occurs in small (100-200 Jlm
across), dispersed patches that make up approxi­
mately 2% of the rock volume (Fig. 2). In the
Dakota Formation at Cole Creek, calcite is more
abundant, in places making up to 5% o� the rock
volume (Reisser & Blanke, 1989; Smith, 1993). In
the treated well at Cole Creek, the calcite was
commonly observed in thin section lining fractures,
making it more accessible to injected C02 (Smith,
1993; Smith & Surdam, 1993a,b).
It is also important to contrast the Cretaceous
fields with the Pennsylvanian Tensleep. Core was
not taken at the Bonanza field, but Tensleep miner­
alogy elsewhere in the Bighorn basin and in the
Wind River basin consists of quartz and K-feldspar
(Mankiewicz & Steidtmann, 1979; Smith, 1993).
Dolomite is the dominant authigenic phase, making
up as much as 38% of the rock in dolomitic
sandstones, but more commonly ranging between 5
and 18% (Fig. 3) (Smith, 1993). Calcite occurs as
scattered poikilotopic nodules and rarely as calcite­
cemented bands; it makes up 2-3% of the rock
volume (Smith, 1993). The average calcite/
dolomite ratio is 0.29, which is used later to make
inferences about the relative dissolution rates be­
tween calcite and dolomite.
PRODUCTION RESULTS O F
CYCLIC C02 TREATMENTS
One well in each field was injected with C02. Only
487
Fig. 3. Thin-section photomicrograph with partially
crossed nicols of Tensleep sandstone from Big Sand
Draw field, Wyoming, 2211 m (7255 ft), showing
euhedral dolomite crystals <lJlm (arrows) rimming
framework grains. Scale bar= 38 Jlm.
one of these, Crooks Gap, responded significantly to
this treatment. From a pretreatment production
level of 20 BOPD, oil production peaked at
176 BOPD and plateaued at approximately
40 BOPD, until the well was treated with C02 a
second time (Deans, 1990; Deans & King, 1991).
Production increased again, but to a lower maxi­
mum. Water composition was not monitored after
the second C02 treatment. Grass Creek exhibited
increased production for 2 days, quickly returning
to its baseline level of 6-7 BOPD (Deans et al.,
1992). After 1 month of production the well had
produced back only about 5% of the injected C02
(Deans et al., 1992). Beaver Creek had enhanced
production for only 3 weeks and, as with Grass
Creek, only 25% of the injected C02 was produced
back (Deans et al., 1992). At Cole Creek, oil-cut
increased from 2% to 36% early in the post­
treatment production. However, problems with the
downhole pump caused the well to be shut in for
extended periods beyond the 4-week soak period
(Deans et al., 1992). Eventually, this treatment
project was terminated because of severe scaling
problems. North Grieve was somewhat successful,
with approximately 200 BBLS of incremental oil,
but this amount was far less than what was required
to make the project an economic success (Deans et
al., 1993). As with Grass Creek and Beaver Creek,
not all of the injected C02 was produced back.
488 L.K Sm ith

MINERAL DISSOLUTION water over their pretreatment levels in all of the

AMOUNTS AND RATES fields. This is also shown in Table I for Crooks Gap
and Bonanza fields. Beaver Creek (Fig. 4f) exhibits
increased TDS, which is presumed to be due prima­
Carbonate
rily to increases in calcium, magnesium and bicar­
bonate, as in the other fields.
Ev i d ence for d issolut ion: elevat ed Ca, Mg
The amount of calcium and magnesium increase
and HC03
varies between fields. The largest percentage in­
Figures 4 and 5 show that calcium and magnesium crease in calcium is in Cole Creek field (Fig. 4d),
exhibit increased concentration in the produced which was over 2000%. The largest absolute in-

a. Crooks Gap b. Bonanza

1• 00
200 1000
00
I
Calcium •o Calcium
0 • Magnesium 800 00 • 0 Magnesium
"[ 150 00 0 oo 0
s 0
c 0 600
0
� 100
0
c 400
Ql
() 0
c 50
0
u 0 l!l!JcJ 0 8
200 Do o o 0 0
[DJ
_.�
50 100 50 100

c. Grass Creek d. Cole Creek

400 -1---'--"---'---'-.....J.--'-.-'--t-
0
' _=-...,.. ...,.. _ _' 0 0
E' _
e o c a lciu m
Cl. 60 • 0 Magnesium 0 300
s
c 0 (ll)
�
0 00 0
40 0 200 8

c
0
c
Ql
()
0
0
100
8 • 00
•
Calcium
Magnesium
20
u
§ bl
0 []
50 100
50 100

e. North Grieve f. Beaver Creek

12ooo -t--'--"---'--;=='==='= =='=='====::::;-l
0 I T"l Total Dissolved Solids I
E'2oo oo �
Cl. 10000
s
.Q 150
"§ 0
8000
c 100
Ql
()
c
6000
8 50
DO
0
4000 -+---,-.--,-,--,-t-
50 100 150 200 50 100
Day Since Production Recommenced Day Since Production Recommenced

Fig. 4. Concentration before and after C02 treatment. Solid symbols are pretreatment concentrations. Open symbols
are post-treatment concentrations through time. (a) Crooks Gap calcium and magnesium. (b) Bonanza calcium and
magnesium. (c) Grass Creek calcium and magnesium. (d) Cole Creek calcium and magnesium. (e) North Grieve
calcium and magnesium. (f ) Beaver Creek total dissolved solids. Note the different x-axis scale for North Grieve.
 Carbonate cement dissolution 489

a. Crooks Gap HC0 to a maximum of 2361 ppm. At Crooks Gap

3
2000 --h;---,L_--'-----'
--L --'--
1 ----L--'------'----
-+ --' the increase was from 1182 to 1879 ppm. Also in
0
'[ 1500 � 0
0
these wells, pH either rose slightly or stayed approx­
imately the same (Table I).
s
c
0
� 1000
Evidence for dissolution: interpretation
c
Q)
(,)
:5 500 The elevated calcium, magnesium and bicarbonate
u levels are an indication that calcite and/or dolomite
I + 0 Bicarbonate I is dissolving from the reservoir. The increased
o�-.-.--.-.,-.-�
l ==r==r��+,
0 50 100 bicarbonate is interpreted not to be due to the
injected C02; if this had been the case the pH of the
b. Bonanza post-treatment water would have declined signifi­
2400 -b'I-'---'--------'--------"-
_L_----L
__.__
__._
____,
_ +- cantly. Also, additional bicarbonate in the water is
0 balanced by the additional calcium and magne­
'[2000 sium. If calcite or dolomite were to dissolve congru­
s
c ently, then calcium plus magnesium would charge­
0 0
� 1600 balance bicarbonate, which would probably not be
c 0
Q) the case if other non-carbonate minerals were dis­
(,)
:5 1200 solving and contributing cations to the water. In the
u
I+ 0 Bicarbonate I pre-C02 water at Crooks Gap, mea>+ + mMg>+ does
800 +-----r-r---r--.-.-----r-t- not balance mHeO]• but the post-injection water
0 100 � does (Fig. 6). Most of the points in Fig. 6a cluster
Day Since Production Recommenced
around the mea>+ + mMg>+ mHeo3 line, which
=

Fig. 5. Concentration of bicarbonate before and after
CO, treatment for (a) Crooks Gap and (b) Bonanza.
Solid symbols are pre-C02 concentrations. Open
symbols are post-C02 concentrations through time.
crease was at Bonanza field (Fig. 4b), where calcium
increased to a maximum of 828 ppm from a pre­
injection average of 331 ppm. There are also vari­
able increases in magnesium. At Cole Creek, magne­
sium increased to over 55 ppm from a pretreatment
level of only 6 ppm. At Bonanza field, magnesium
increased from 121 ppm (average) to 200 ppm.
Crooks Gap (Fig. 4a) shows increases of calcium and
magnesium from 13 ppm to 184 ppm and 4 ppm to
31 ppm, respectively. All of the fields except North
Grieve showed a decline in the concentrations of
calcium and magnesium back to, or near, pretreat­
ment levels. The return to pretreatment concen­
trations levels is interpreted as water producing
from uncontacted portions of the reservoir. The
significance of the shape of the concentration-time
profile will be discussed later, under reservoir
heterogeneity and treatment success.
In addition to increased calcium and magnesium
in the post-C02 water, bicarbonate also increased at
Crooks Gap and Bonanza where this analysis was
performed (Fig. 5; Table 1). At Bonanza field, bicar­
bonate increased from an average of 446 ppm
strongly suggests carbonate mineral dissolution.
Those points not lying on the line are from later
during the production phase, when original forma­
tion water that was outside the C02-contacted area
began to produce. For the Bonanza well (Fig. 6b),
all of the post-C02 samples plot along the I : I line,
which is what would result if those ions came only
from the congruent dissolution of carbonate miner­
als. At Bonanza, water composition was not moni­
tored long enough to see a return to original
formation water as at Crooks Gap.
The composition of the produced waters after the
C02 injection at Crooks Gap indicates that dolo­
mite is present in the reservoir, although none was
observed in thin-section examination. The excess
Ca/Mg molar ratio varies between 3.3 and 3.8. If
calcite was the only carbonate mineral that had
dissolved from the reservoir, and even if the calcite
contained abundant Mg, the Ca/Mg molar ratio
could only be as low as about 6. Another source of
magnesium must have dissolved from the reservoir,
and the most likely and abundant source would be
dolomite, which also became undersaturated after
the injection of C02• Clay minerals might also have
contributed magnesium to the water, but dissolu­
tion rates for clays are considerably slower than for
carbonate minerals. This is discussed more fully in
the section on aluminosilicate minerals.
490 L.K Smith

a. Crooks Gap volume of calcite was assumed to be constant at 2%

15 (Smith, 1993) (Fig. 2), the volume of the reservoir
contacted by the C02 was approximately 4200 m3
(148 000 ft3) (Smith et a!., 1991; Smith, 1993), and

'a-
the calcite dissolution was evenly distributed
QJ
X
10 X throughout the contacted area. The contacted area
.s
Cl
X was estimated from the volume of C02 injected and
�
+ the porosity and thickness of the reservoir. The
"'
u calculation shows that the 28 000 cm3 of calcite
"' 0"'""
"'
QJ
5
x,+(J
dissolved would represent less than 0.1o/o of that
u
X � available in the reservoir. This would result in a
UJ � X
(;'If negligible increase in porosity and permeability. A
ve.
"'"'X second calculation was performed to determine
<v"'i'
whether permeability would increase if all the dis­
5 10 15 solution took place immediately adjacent to the well
Excess HC03 (meq/1)
bore. If all the dissolution took place within 0.3 m
(1.0 ft) of the well bore the porosity would increase
b. Bonanza by only 1.0%. As it takes, on average, a 4% porosity
40
increase to effect a one-order-of-magnitude change
in permeability (Archie, I 950), the I o/o porosity
increase results in a negligible increase in perme­
_30
ability. This contradicts the suggestions of Craw­
'a-
QJ
ford et a!. (1963), Ramsay & Small (1964) and
.s
Cl Wiseman (1981) that permeability enhancement
� 20 due to rock dissolution is one mechanism by which
"'
u immiscible cyclic C02 treatments work.
"'
"'
QJ
u
� 10
Modelling calcite dissol ution

Modelling calculations were performed for Crooks

0
Gap and Bonanza to determine how much calcite
0 10 20 30 40
Excess HC03 (meq/1) could dissolve given sufficient time to reach equilib­
rium between calcite and the added C02• The
Fig. 6. Excess bicarbonate vs excess calcium plus
purpose of the calculations was to determine how
magnesium for (a) Crooks Gap and (b) Bonanza. Excess
concentrations are concentrations in the post-C02 waters far from equilibrium the natural systems were, and
with the original, pre-C02 concentrations subtracted out. to assess the potential for using a thermodynamic
equilibrium modelling program to predict well bore
scale (discussed in the next section).
For Crooks Gap the simulation was run with
Amo unt of dissol ution
a partial pressure of C02 "(PC02) 8.9 MPa =

For Crooks Gap, the amount of calcite that actually
dissolved was estimated using the excess calcium
concentrations, the percentage of water in total fluid
produced, and fluid flow rates. The total volume of
2
calcite dissolved was approximately 2.8 x I 0- m3
( 1.0 ft3). Although this seems a small amount, if it
had all reprecipitated downhole as scale in the well
bore, it could cause serious damage for a submersible
(1300 psi), which was the maximum anticipated
total pressure during the C02 injection. Before
injection the reservoir was producing no gas. There­
fore, with no other gas phase present, the assump­
tion that the partial pressure of C02 was equal to
the reservoir pressure is valid. With C02 in the
reservoir, the water becomes greatly undersaturated
with respect to calcite (Sicaic -2.8) and pH is
=

pump with tolerances measured in micrometres. expected to drop to about 4.3. Under these circum­
An order-of-magnitude calculation was per­ stances, calcite in the reservoir should dissolve. The
formed to determine the effect of that amount of amount of calcite that could dissolve is 3.0 x I o-3 g
dissolution on permeability. For this calculation the calcite for every 1.0 g of water contacted by the C02•

This amount of dissolution would raise the calcium
and bicarbonate concentrations to approximately
1100 and 4400ppm, respectively. For Bonanza,
with PC02 2. 7 MPa (385 psi), the potential dis­
3
=
solved amount is 3.2 x 1 0- g calcite/g H20. The
actual amount of carbonate mineral in the reser­
voirs is 0.29 and 0.40 g calcite/g H20 for Crooks
Gap and Bonanza, respectively. Therefore, both res­
ervoirs have more than enough carbonate minerals
to accommodate all of the potential dissolution. As
the post-treatment calcium and bicarbonate concen­
trations never reached the amounts predicted for
saturation at Crooks Gap, the system never reached
equilibrium between calcite and C02.
Although the mass transfer portion of the model
overpredicts calcite dissolution, the calculations of
pre- and post-C02 calcite saturation indices are still
useful. Subtracting one from the other yields a
modelled, or predicted, change in the Slcaic (Fig. 7),
and it correlates with the observed increases in the
calcium concentration. Before C02 injection the
calcium concentrations in the five fields were very
similar. However, the maximum calcium concen­
tration observed after the C02 injection, which sug­
gests calcite dissolution, varied considerably. When
these numbers are compared with the original con-
25 +-----�--�--�
c:
0 /
Q.)�
(/).;,
<ll Q)
/
Q) c:
u g
E o
Q)o
.£;
E examining the ratio of excess Ca to excess Mg in the
<ll :::J Crooks Gap
m ·u X post-treatment produced water. This ratio is 2.6
a:Cii North Grieve
0 G rass C reek /X
.S )( //
Bonanza
X sulted from the dissolution of 1.6 mol of calcite and
0 +------.--r---�
2.5 3.0 3.5
Modeled Change n i tion, the Ca/Mg ratio would result in a calcite/
Calcite Saturation I
ndex
Fig. 7. Modelled change in the calcite saturation index
vs the relative increase observed in the calcium
concentration. The modelled change in the calcite
saturation index (Sicaicl refers to the Slcalc calculated for
pretreatment water minus the Slcalc that was calculated
based on predicted C02 injection conditions. The
relative increase in calcium concentration is the
maximum post-treatment concentration observed
divided by the original concentration.
Carbonate cement dissolution 491
centration, maximum calcium concentration at Cole
Creek was 21x the original concentration, whereas
the other fields ranged between 4x and 6x . Figure 7
indicates that the thermodynamic equilibrium
model does have potential for simple modelling of
these dissolution processes, even though the systems
do not reach equilibrium.
Relative rates of dissolution
The Tensleep at Bonanza has more carbonate min­
erals than the Dakota at Crooks Gap, but the more
abundant carbonate minerals in the Tensleep did
not lead to more dissolution. Crooks Gap water
showed a larger percentage increase in calcium than
did Bonanza water. However, the more abundant
carbonate minerals at Bonanza did lead to more
rapid dissolution, which brought the system closer
to equilibrium in approximately the same amount
of 'soak' time. At Crooks Gap, the sample with the
highest calcium concentration after the soak phase
had Slcalc -1.5, whereas at Bonanza Slcalc -0.7.
= =
Bonanza field was nearer calcite saturation after the
C02 injection than was Crooks Gap, even though
Crooks Gap had a higher reservoir pressure, higher
PC02, and a larger change in the calcium concen­
tration. This confirms for the field what has been
observed in the laboratory: that even small differ­
ences in carbonate mineral/water ratios (0.29 at
Crooks Gap and 0.40 at Bonanza) can make a big
X difference in dissolution rates. Also, this ratio is
Cole Creek more important than a large C02 pressure differ­
ence.
The relative dissolution rates between calcite and
dolomite were also estimated for Bonanza field by
(Fig. 8). Assuming pure calcite (CaC03) and pure
dolomite (CaMg(C03h), this could only have re­
1 mol of dolomite, a calcite/dolomite dissolution
4.0 ratio of 1.6. Assuming an impure calcite composi­
dolomite dissolution ratio of approximately 1.9.
Even though calcite is a minor component in the
Tensleep reservoir, its relative contribution to the
post-C02 injection water is great. This is also
despite the fact that the dolomite is more finely
crystalline than the calcite, and would thus be
expected to have a larger surface area exposed to the
carbonated water.
This may not be the case at much higher PC02•
492
10 15 20 25 30
Excess Ca (meq/1)
Fig. 8. Excess calcium vs excess magnesium for
post-C02 waters from Bonanza. Excess concentrations
are the amounts in the post-C02 waters with the
original, pre-C02 concentrations subtracted out.
The absolute dissolution rate of dolomite has pre­
viously been shown to increase with increasing
PC02 at low temperature (Busenberg & Plummer,
1982). Additional information about the relative
dissolution rate of dolomite comes from comparing
dissolution at Bonanza with that of the experimen­
tal C02 corefloods of Ross et a/. (1982). The core
material used in one of Ross et al.'s experiments
was dolomitic Rotliegendes Sandstone from the
Indefatigable Field (Ross et a/., 1982), which is an
aeolian sandstone (Glennie, 1972) similar to the
Tensleep. Ross et a/. did not report the presence of
any mineral other than dolomite in their core.
However, Pearson et a/. (1991) reported both cal­
cite and anhydrite in cores from this field. Also, a
Ca!Mg ratio greater than 1 : 1 in the produced water
of the C02 coreflood experiment could only result
from the dissolution of another Ca-bearing mineral
besides dolomite, or from precipitation of an Mg­
bearing phase. Precipitation of an Mg-bearing clay
is unlikely, because Ross et a/. reported an increase
in permeability in the experiment. Clay precipita­
tion during C02 experiments would result in either
a permeability decrease, as reported by Sayegh et a/.
(1987, 1990) and by Shiraki (in Dunn, 1995, 1996)
or no change in permeability (Bowker & Shuler,
1991).
Data from Ross eta/.'s (1982) Fig. 13 was used to
determine a Ca/Mg ratio of 1.4 for the effluent of
their experiment (Fig. 9). Assuming that all the
L.K Smith
E: 4
0"
Q)
.s
Ol
:::!'
VJ
VJ
Q)
� 2
w
2
Excess Ca (meq/1)
Fig. 9. Excess calcium vs excess magnesium for
post-C02 waters from C02 coreflood experiments (data
from Ross et a/., 1982, their Fig. 13). Excess
concentrations are the amounts in the post-C02 waters
with the original, pre-C02 concentrations subtracted out.
magnesium came from dolomite dissolution, then
only 0.4 mol of other calcium-bearing minerals was
dissolving for every mol of dolomite. It is likely that
calcite is less abundant than anhydrite, because
calcite is seldom reported to be present at all in
other Rotliegendes fields in the area. This would
result in a very small calcium contribution from
calcite-probably less than half of the 0.4 mol that
had to be coming from non-dolomite calcium­
bearing minerals. Therefore, in Ross et al.'s experi­
ment dolomite dissolution was proceeding more
rapidly compared with calcite than it was in the
Bonanza field case. The most likely explanation for
this relative rate difference is the different temper­
ature and pressure conditions between the experi­
mental and field cases. The Bonanza field pressure
is only 2.7 MPa at a temperature of only 29·c,
whereas the coreflood experiments were run at a
pressure qf 13.8 MPa (2000 psi) and a temperature
of 80·c (Ross et a/., 1982). This suggests that at
very high PC02 the rate of dissolution of dolomite
approaches, and may equal, that of calcite.
Aluminosilicate minerals
Ev idence for dissolution: elevated AI and S i02
Where analysis for aluminium and silica was per­
formed (Crooks Gap and Bonanza) it showed in-
 Carbonate cement dissolution 493

a. Co
r oks Gap abundant aluminosilicate mineral present in the
reservoir and the one most likely to be affected by
90 the added C02• Carbon dioxide destabilizes the
80 0.6 plagioclase by the formation of carbonic acid. The
"[ 70
a.
modelled Slalbite changes from +4.43 before the
� 60 C02 to -1.96 with C02 added to the system. The
� 50 added C02 has less effect on the existing clays in the
40 reservoir. Most of the calculated clay saturation
-.A.6.Silica
30 +<?Aluminium- indices remained near zero, and kaolinite, which is
the most abundant clay phase present at Crooks
50 Gap, remained stable in the presence of C02 with
Slkaolinite l. 76.
b.
=
Bonanza
At Bonanza similar mechanisms may be opera­
tive. Shiraki (reported in Dunn, 1996) observed an
30 initial increase in potassium and silica in C02
"[ 25 0.2 � coreflood experiments on the Tensleep sandstone.
.e, u He concluded that concentrations of these compo­
� 20 "0
2. nents were controlled by hydrolysis of K-feldspar to
Ui
15 0.1 form kaolinite, which was observed in cores after
-A6.Silica the experiments. However, in the field example
10
+<?Aluminium- presented here no increase in potassium was ob­
5 0 served. There are three possible explanations for
0 � 100
Day Since Production Recommenced this, but the data are insufficient to determine which
ones are operative. A potassium- or sodium­
Fig. 10. Concentrations of silica and aluminium before deficient aluminosilicate, such as altered biotite or
and after C02 treatments at (a) Crooks Gap and (b) staurolite, may have dissolved instead of or in
Bonanza. Closed symbols are pre-C02. Open symbols
are post-C02 concentrations through time.
addition to K-feldspar. These minerals have been
observed elsewhere in the Tensleep, although only
in trace amounts (Smith, 1993). Likewise, there
may have been a sink for any potassium released by
creased concentration after the C02 treatment K-feldspar dissolution. The sink might be the for­
coincident with the increases in calcium, magne­ mation of a new K-rich clay phase, such as a
sium and bicarbonate (Fig. 10; Table I). At Crooks smectite, or the formation of a well-bore scale such
Gap the silica concentration increased from 34 to as jarosite (KFe3(S04h(OH)6). Jarosite has been
100 ppm. At Bonanza it increased from 10 to observed forming in wells at the C02 operation at
34 ppm. Aluminium concentration also increased Rangely field in Colorado (T.L. Dunn, Institute for
in both fields, although the peak increase came after Energy Research, University of Wyoming, personal
the peak in concentration of the other dissolved communication), which produces from a Tensleep
components. At Crooks Gap aluminium increased equivalent (Weber Sandstone). A third possibility is
·
from 0.2 to 0. 7 ppm, and at Bonanza from 0.04 to that the sodium and potassium did actually increase
0.29 ppm. but that the analysis measured small increases in a
very large number. All these proposed mechanisms
may be operating simultaneously.
Evidence for dissolution: interpretation

The increases in silica and aluminium are indica­

tions of aluminosilicate alteration and/or dissolu­
tion, although the data are insufficient to determine
specific mechanisms. Several mechanisms involv­
ing more than one aluminosilicate mineral may be
operating, and they may be different in the two
fields. At Crooks Gap, the aluminium probably
comes from sodic plagioclase which is the most
Dissolution rates
It is interesting to note the different concentration
profiles exhibited by silica and aluminium, because
these provide indirect evidence of the possible
dissolution and/or precipitation rates of different
aluminosilicate minerals in the reservoir when C02
is added to the system. Silica exhibits the same
494
Calcite
Stable
Precipitation
300 Possible
�
'* 600
.§.
.<:::
c.
Q)
0 900
1200 Calcite
Unstable
No Scale
Precipitation
1500 -'1"----.---,--f----+
-1.5 0.0 1.5
Calcite Saturation Index
Fig. 11. Calcite saturation index vs depth for Crooks
Gap. The water analysis used for the calculations was
the original water analysis plus the amount of added
calcium and bicarbonate determined from Fig. 7.
pattern of increased concentration after the C02
injection as the calcium and magnesium (compare
Fig. lOa with 4a, and lOb with 4b). The peak Si02
concentration occurs at the same time as the peak
Ca and Mg concentrations. Aluminium concentra­
tions, however, do not follow the same pattern. At
Crooks Gap, aluminium peaks 7 days after the
other cations, and at Bonanza the delay is 50 days.
The delay in the peak aluminium concentration
relative to calcium and magnesium makes sense
when comparing the dissolution rates of carbonate
and aluminosilicate minerals. Comparing dissolu­
tion rate data on calcite (Morse, 1983) with those
on albite (Chou & Wollast, 1985; Knauss & Wolery,
1986) shows that at a pH of 3 calcite dissolves
considerably faster, with a difference of several
orders of magnitude.
The delay in the aluminium concentration by
several days to weeks relative to silica is more
difficult to explain. The silica increase followed by
an aluminium increase cannot be explained by
simple incongruent dissolution of feldspar. Such a
scenario would suggest the formation of an alumi­
nous residual layer. However, incongruent dissolu­
tion of albite at low pH leads to the formation of a
siliceous, not an aluminous, residual layer (Chou &
Wollast, 1985). Also, the Si/AI ratios in the pro­
duced waters after the C02 injection are much
L .K Sm ith
higher than could be explained by feldspar dissolu­
tion (between 46 and 124 for the first 6 days after
the C02 treatment at Crooks Gap). One mechanism
that may contribute to, although not completely
explain, the Si/Al ratio is different dissolution rates
between feldspar and some of the kaolinite and
mica present. Chou & Wollast (1985) and Knauss &
Wolery (1986) reported albite dissolution rates that
are approximately twice those of kaolinite (Nagy et
a/., 1990) or of muscovite (Knauss & Wolery, 1989)
at the same pH and temperature (pH= 3, 70-80°C).
More silica-rich phases dissolving first and more
aluminium-rich phases dissolving more slowly and
contributing their dissolved components to the
water later would result in different silica and
aluminium profiles, as was observed. However, the
limited field data, particularly the lack of rock
samples after the C02 injection, preclude the deter­
mination of specific controlling reactions.
WELL-BORE SCALE
AND EQUILIBRIUM IN THE
NATURAL LABORATORY
The potential for well-bore scale during these C02
treatments is determined by first predicting how
much carbonate mineral dissolution will take place.
After a calcite saturation index has been calculated
for a water analysis for both pre- and post-C02
treatment, then Fig. 7 can be used to determine how
much calcium (and therefore bicarbonate) will have
been added to treated waters from calcite dissolu­
tion. These calcium and bicarbonate values can be
added to the original formation water, which is then
used to calculate a new calcite saturation index for
the new water.
This was done for Crooks Gap using nine sets of
calculations at different pressures to represent the
declining pressure conditions up the production
tubing to surface lines. The results are shown in
Fig. l l , which is a plot of Sicaic vs depth. Even with
extra calcium in the water from calcite dissolution,
the water in the subsurface production tubing re­
mains undersaturated with respect to calcium car­
bonate, and therefore scale would not be expected
to precipitate. After 230 days of production the
downhole pump at Crooks Gap was retrieved in
preparation for another C02 treatment, and indeed
no scale was observed on the pump.
Only one of the treated wells, Cole Creek, expe­
rienced well-bore scale, and in that well the scaling
 Carbonate cement dissolution
was so severe as to completely destroy the downhole
pump. Two factors contributed to this scale. The
pressure drop from the formation into the well bore
was estimated to be over 1 3 MPa (2000 psi), which
was considerably higher than in the other wells. A
more important factor, however, was the numerous
shut-in periods: during the 8-month soak and pro­
duction phases of the project the well was shut in,
off and on, for approximately 4 months. These
numerous shut-in periods allowed additional time
for the C02 and calcite in the reservoir to react.
Modelling calculations performed for Cole Creek
water suggest that at Cole Creek P-T conditions,
equilibrium between C02 and calcite in the reser­
voir was reached at between 4 and 8 months. This
represents the time between when the last water
sample was taken (shown in Fig. 4d) and the time
the well-bore scale was observed. The modelling
that suggests that equilibrium was reached is shown
in Fig. 1 2, which plots calcite saturation indices for
a series of five modelling steps. In step 1 the
pre-C02 conditions are modelled using the original
formation water composition. In step 2 the system
is perturbed by adding C02 at reservoir pressure.
This causes the water to be undersaturated with
calcite. In step 3 the system is forced to equilibrium
by dissolving calcite. Using the new water compo-
3 ,------.
c
0 tion-time profile will have high concentration for
u the earliest production and lower concentration
Q)
N "C
0
"'
0
1500+-------'--''--+
E� 0 0
.2 E 1000
u a.
8 .eo 500
Fig. 12. Modelling of Cole Creek water to predict
well-bore scale. Upper graph shows modelled calcite
saturation index (SicaJcl for various modelling
conditions. Lower graph shows anticipated calcium
concentrations. Solid black is original, pre-C02 calcium
concentration. Open circles are predicted calcium
concentrations. Diamond is actual post-C02
concentration observed 4 months prior to intense
well-bore scale precipitation in this well.
495
sition from step 3 with added calcium and bicar­
bonate from calcite dissolution, the system is
perturbed again in step 4. In step 4, production in
the well is simulated by dropping the pressure. The
system now becomes oversaturated with calcite.
Finally, in step 5 the system is forced back to
equilibrium, this time by precipitating calcite
(scale). For each step the predicted_calcium concen­
tration is shown on the lower graph of Fig. 1 2. In
the final step the predicted calcium concentration,
after scale precipitation, is 500 ppm. This compares
well with the maximum concentration that was
observed 4 months prior to the scale precipitation,
which was 377 ppm (diamond).
RESERVOIR GEOMETRY,
TREATMENT SUCCESS AND
P RODUCED WATER CHEMISTRY
The shape of the water composition vs time profiles
for each of the COrtreated wells (Figs 4 and 5) can
provide information about the geometry or hetero­
geneity of the reservoir. In any cyclic C02 treatment
the greatest amount of dissolution will take place
where the most C02 resides. In a homogeneous
reservoir most of the C02 resides near the well bore,
as shown by the simulation of a cyclic C02 treat­
ment in Fig. 1 3. Therefore, most of the dissolution
will tak� place near the bore, and less will take place
further away. During production, water near the
well bore with abundant dissolved components will
produce back first, followed by water with fewer
dissolved components. The resulting concentra­
1 . o r'--'--'--'-..___.-'-;=�==z:::==�
0 End of Injection
c 0.8 0 End of Soak
0
t:,. End of Production
·�
:; 0.6
<;;
� 04
«<
<!J
0.2
20 40 60 80
Distance from Wellbore (m)
Fig. 13. C02 gas saturation vs distance from the
injection well bore for a simulated cyclic C02 treatment
on a homogeneous reservoir. After Hsu & Brugman
( 1 986).
496 L.K Sm ith

later, and will mimic the shape of the saturation­ Watertlood C02 Treated
Injector Well Well a. Injection
distance profile in Fig. 1 3. Such was the case for the
C02 treatment at Crooks Gap (compare Figs 4a and
1 3). Most of the dissolution at Crooks Gap was near
the well bore, indicating that most of the C02 was
also near the well bore. Also, as the simulation
(Fig. 1 3) indicates that high C02 saturation near the
well bore results from treatment in a homogeneous
reservoir, it can be concluded that the Crooks Gap
reservoir was probably fairly homogeneous. In ad­
dition, among all the C02-treated wells in this
b. Soak
study, Crooks Gap exhibited the greatest success in
terms of enhanced oil recovery. One of the keys to
treatment success, then, is confinement of the C02
to the near-well-bore region. If the injected C02 gets
too far away, it and the contacted residual oil
cannot be produced back when production is re­
commenced. Crooks Gap, then, can be considered a
reference case, and concentration-time profiles of
the other fields can be compared with it to assess
reservoir heterogeneity and treatment success. c. Production
There are a variety of reasons why the C02 may
not stay in the near-well-bore region. One might be
the presence of high permeability zones or fractures.
Although it did not alter the shape of the profile,
this was apparently the problem at Beaver Creek,
because C02 was detected in production from an
adjacent well (Deans et a!. , 1 9 92).
Another reason for a lack of C02 confinfiment is
the geometry of waterflood injectors, the C02-
treated well and the oil-water contact. This was the Fig. 14. Schematic representation of Grass Creek C02
problem at Grass Creek, which did not produce well treatment for (a) injection, (b) soak and (c)
production phases, depicting how the updip waterflood
back all of the C02 that was injected, and the C02 water injector pushed the C02 into the waterleg during
that did produce back did so approximately 3 the soak period, resulting in a delay in the increased
months after production recommenced. The water calcium concentration (see Fig. 4c) and in the production
chemistry profile corroborates this. of the injected C02 during the production phase.
At Grass Creek (Fig. 4c), the calcium concentra­
tion peaked considerably later than at Crooks Gap
(Fig. 4a)-about 3 months after the well was put vertically from the treated well perforations (Smith,
back on production. At Grass Creek, high cation 1 99 3). When post mortem analysis was performed
concentrations 3 months after production are an on the well to determine the reasons for the lack of
indication that the area of the reservoir affected by success, the produced water chemistry provided
the C02 was some distance away from the treated valuable information that could not have been
well. The water composition profile and the fact surmised from the other available data alone.
that the well produced very little incremental oil is Another reason for C02 not staying near the well
an indication that the C02 bypassed the area bore is the lack of a water drive-mechanism to help
around the well bore and went directly into the push the C02 back to the production bore. At North
water leg, where it resided during the shut-in period. Grieve, confinement was not considered to be a
This is illustrated by the schematic scenario in problem (Deans et a!., 1 992) in the sense that it did
Fig. 1 4, showing how updip waterflood injectors not break through and produce from another well
helped push the C02 downdip into the water leg, bore. However, the post-injection produced water
which was only 1 4 m (46 ft) laterally and 3 m (9 ft) chemistry at North Grieve indicates that not all of
 Carbonate cement dissolution
the C02 stayed in the near-well-bore region during
the soak period. The fact that cation concentrations
remained high for such an extended period indi­
cates that cement dissolution took place within a
very large volume around the treated well. Even
after 6 months, fluids from the uncontacted and
therefore unaltered portion of the reservoir had not
yet been produced. Therefore, it can be concluded
that the C02, which caused the dissolution, diffused
to fill a large volume around the well. This may have
resulted from the lack of a water-drive support that
otherwise would have helped push the C02 back to
the well bore during production. Confinement was
probably achieved but over a broad area, owing to
the fact that the Muddy Formation tends to form
compartmentalized reservoirs (Surdam et a!. ,
1 993). Again, this picture of the reservoir would not
have been drawn without the information gained
from the produced water chemistry.
These models for the C02-treated reservoirs
drawn from the produced water chemistry and the
information gained through chemical modelling of
scale formation led to the development of guide­
lines for selecting wells for cyclic C02 treatment
(Smith & Surdam, 1 99 3a,b; Surdam et a!. , 1 994).
Such wells should be in homogeneous reservoirs
without high permeability streaks or abundant frac­
tures. The reservoir should have a moderate to
strong water drive, and the well should not be near
any waterflood injectors, unless perhaps it is sur­
rolJnded by injectors. To reduce the risk of scale
formation bottomhole pressure should be kept high,
as close to reservoir pressure as possible.
CONCLUSIONS
Carbonate cement dissolution caused by cyclic
C02-enhanced oil recovery treatments can provide
information in several areas. From this study,
which examined mineral dissolution, well-bore
scale and reservoir heterogeneity/geometry, the fol­
lowing conclusions are drawn.
1 At very high PC02, dolomite dissolution rates
approach those of calcite.
2 Small differences in mineral/water ratios in the
reservoir can result in large differences in carbonate
mineral dissolution rates.
3 At Cole Creek, where reservoir pressure was high,
equilibrium between the injected C02 and calcite in
the reservoir was achieved in 4-8 months, whereas
at Bonanza (where reservoir pressure was low but
497
the carbonate mineral/water ratio was high) equilib­
rium between the injected C02 and calcite in the
reservoir may nearly have been achieved during the
! -month soak period.
4 Permeability enhancement is not the mechanism
of enhancing oil recovery in these cyclic C02
treatments, with the possible exception of 'cleaning
up' scale that might have been clogging perfora­
tions.
5 The injected C02 appears to cause high Si/AI
aluminosilicate minerals to dissolve more rapidly
than low Si/AI aluminosilicate, causing silica con­
centrations in the post-C02 waters to peak before
aluminium concentrations.
6 Thermodynamic equilibrium chemical modelling
programs can be used to predict non-equilibrium
processes when calibrated with real data.
7 Well-bore scale is not a problem in cyclic C02
EOR treatments, except in situations where the
'soak' phase becomes extended enough to allow the
C02 to equilibrate with minerals in the reservoir.
8 Post-C02-treatment waters can provide informa­
tion about the types of minerals present in the rock,
and about reservoir geometry/heterogeneity, that
can help explain treatment success or failure.
ACKNOWLEDGEMENTS
Funding for this study was provided by the En­
hanced Oil Recovery Institute at the University of
Wyoming, BP Exploration, K&N Energy, Kerr­
McGee Corporation, Marathon Oil Co., and the
Stripper Well Violation Fund. Thanks are due to
R.C. Surdam, H.A. Deans, T. Monger-McClure, R.
King, M. Calody, D.B. MacGowan and T.L. Dunn
for helpful discussions. Marathon Oil Co., Amoco
Production Co., Timberline Production Co., Koch
Production Co. and G.G. Nicolaysen Co. provided
data for the study. Water analyses were performed
by Dr S. Boese. A. Deiss helped with draughting
some of the figures. The manuscript was greatly
improved by reviews by Ian Hutcheon and Richard
Worden.
REFERENCES
ABERCROMBIE, H.J. ( 1 991) Reservoir processes in ste?.m­
assisted recovery of bitumen, Lemin Pilot, Cold Lake,
Alberta, Canada: composition, mixing and sources of
co-produced waters. App/. Geochem. , 6, 495-508.
498 L.K Smith
ARCHIE, G.E. ( 1 9 50) Introduction to petrophysics of reser­
voir rocks. Bull. Am. Ass. Petrol. Geol. , 34, 943-96 1 .
BowKER, K.A. & SHULER, P.J. ( 1 989) Carbon dioxide
injection and resultant alteration of the Weber sand­
stone (Pennsylvania-Permian), Rangely field, Colorado.
SPE paper 1 9875 presented at the 64th Annual Technical
Conference of the Society of Petroleum Engineers, 8-1 1
October 1 989, San Antonio, TX.
BowKER, K.A. & SHULER, P.J. ( 1 9 9 1 ) Carbon dioxide
injection and resultant alteration of the Weber sand­
stone, Rangely field, Colorado. Bull. Am. Ass. Petrol.
Geol. , 75, 1 489- 1 499.
BusENBERG, E. & PLUMMER, LN. ( 1 982) The kinetics of
dissolution of dolomite in COrH20 systems at 1 . 5 to
6 5 " C and I atm PC02. Am. J. Sci., 282, 45-78.
CHOU, L. & WOLLAST, R. ( 1 9 8 5 ) Steady-state kinetics and
dissolution mechanisms of albite. Am. J. Sci., 285,
963-993.
CRAWFORD, M.R., NEILL, G.H., BUCY, B.J. & CRAWFORD,
P.B. ( 1 963) Carbon dioxide-a multipurpose additive
for effective well stimulation. J. Petrol. Tech. , 15, 2 3 7-
242.
DEANS, H.A. ( 1 990) Progress through December 31, 1 990
on single-well cyclic C02 stimulation of depleted wells in
Wyoming. Enhanced Oil Recovery Institute, Report to
the Wyoming Legislature.
DEANS, H.A. & KING, R.A. ( 1 99 1 ) The Crooks Gap huff 'n'
puff project. Proceedings of the Sixth Wyoming En­
hanced Oil Recovery Symposium, Casper, WY, 3-4 May
1 990, pp. 29-42.
DEANS, H.A., JoHNSON, L. & SATCHWELL, R. ( 1 992) Update
on the EORI C02 Huff 'n' Puff program. Presented at
the 7th Wyoming EOR Symposium, Casper, WY, 1 May
1 991.
DEANS, H.A., JOHNSON, L. & SATCHWELL, R. ( 1 993) The
economics of C02 cyclic stimulation of oil wells in
Wyoming 1 98 9 - 1 992. Presented a t the 8th Wyoming
EOR Symposium, Casper, WY, 1 May 1 992.
DuNN, T.L. ( 1 99 5 ) Anisotropy and Spatial Variation of
Relative Permeability and Lithologic Character of Ten­
sleep Sandstone Reservoirs in the Bighorn and Wind
River Basins, Wyoming. 2nd Annual Technical Report
for United States Department of Energy contract no.
DE-AC22-93BC 1 4897, 1 1 4 pp.
DuNN, T.L. ( 1 996) Anisotropy and Spa tial Variation of
Relative Permeability and Lithologic Character of Ten­
sleep Sandstone Reservoirs in the Bighorn and Wind
River Basins, Wyoming. Final Technical Report for
United States Department of Energy contract no. DE­
AC22-9 3BC 1 48 9 7 , 1 73 pp.
GARRELS, R.M. & CHRIST, C.L. ( 1 965) Solution, Minerals,
and Equilibria. Freeman-Cooper, San Francisco,
450 pp.
GLENNIE, K.W. ( 1 972) Permian Rotliegendes of north­
west Europe interpreted in light of modern desert
sedimentation studies. Bull. A m . Ass. Petrol. Geol. , 56,
1 048- 1 07 1 .
GuNTER, W.G. & BIR D , G. W. ( 1 9 88) C02 production i n tar
sand reservoirs under in-situ steam temperatures: reac­
tive calcite dissolution. Chern. Geol. , 70, 3 0 1 -3 1 1 .
HANSLEY, P.L. & WHITNEY, C.G. ( 1 990) Petrology, diagen­
esis, and sedimentology of oil reservoirs in Upper
Cretaceous Shannon Sandstone beds, Powder River
Basin, Wyoming. US Geol. Surv. Bull. , 19 17-C, 3 3 pp.
HELGESON, H. C. ( 1 967) Thermodynamics of complex dis­
sociation in aqueous solutions at elevated temperatures.
J. Phys. Chern., 71, 3 1 2 1 -3 1 36.
Hsu, H.H. & BRUGMAN, R.J. ( 1 986) C02 huff-puff simu­
lation using a compositional reservoir simulator. SPE
paper 15503 presented at the 6lst Annual Technical
Conference and Exhibition of the Society of Petroleum
Engineers, 5-8 October, New Orleans, LA.
HUTCHEON, I. & ABERCROMBIE, H.J. ( 1 990) Fluid-rock
interaction in thermal recovery of bitumen, Tucker
Lake Pilot, Cold Lake, Alberta. In: Prediction of Reser­
voir Quality Through Chemical Modeling (Eds Meshri,
I.D. & Ortoleva, P.J). Mem. Am. Ass. Petrol. Geol.,
Tulsa, 49, 1 6 1 - 1 70.
HUTCHEON, 1., 0LDERSHAW, A. & NAHNYBIDA, C. ( 1 98 1 )
The effect of in-situ steam injection on Cold Lake oil
sands: Part II: Petrophysical and petrological consider­
ations. Bull. Can. Petrol. Geol. , 29, 465-478.
HUTCHEON, I . , ABERCROMBIE, H., SHEVALIER, M. & NAHNY­
BIDA, C. ( 1 9 88) A comparison of formation reactivity in
quartz-rich and quartz-poor reservoirs during steam
assisted recovery. In: UNITARIUNDP Fourth Interna­
tional Conference, Edmonton, 3, paper 2 3 5 , pp. 1 - 1 2.
HUTCHEON, I., ABERCROMBIE, H. & KROUSE, H.R. ( 1 990)
Inorganic origin of carbon dioxide production in low
temperature thermal recovery of bitumen: chemical and
isotopic evidence. Geochim. Cosmochim. Acta, 54, 1 6 5-
171.
HUTCHEON, I . , SHEVALIER, M., NAHNYBIDA, C. & KROUSE,
H.R. ( 1 99 5 ) Chemical and isotopic examination of
produced waters from the BP-Wolf Lake in-situ com­
bustion pilot. Appl. Geochem. , 10, 64-8 3 .
KHARAKA, Y .K., GUNTER, W.D., AGAR WALL, P. K. , PERKINS,
E.H. & DEBRAAL, J.D. ( 1 9 8 8 ) SOLMINEQ.88: A Com­
puter Program Code for Geochemical Modeling of
Water-Rock Interactions. US Geol. Surv. Water Inves­
tigations Rep. 88-4227.
KNAUSS, K.G. & WOLERY, T.J. ( 1 986) Dependence of
albite dissolution kinetics on pH and time at 25 ·c and
7 0 " C. Geochim. Cosmochim. Acta, 50, 248 1 -2497.
KNAUSS, K.G. & WOLERY, T.J. ( 1 989 ) Muscovite dissolu­
tion kinetics as a function of pH and time at 7 0 " C.
Geochim. Cosmochim. Acta, 53, 1 493- 1 50 1 .
LICO, M.S., KHARAKA, Y.K., CAROTHERS, W.W. & WRIGHT,
V.A. ( 1 982) Methods .for Collection and Analysis of
Geopressured Geothermal and Oi(field Waters. US Geol.
Surv. Water-Supply Paper 2 194, 2 1 pp.
MANKIEWICZ, D. & STEIDTMANN, J.R. ( 1 979 ) Depositional
environments and diagenesis of the Tensleep Sand­
stone, eastern Big Horn basin, Wyoming. In: Aspects of
Diagenesis (Eds Scholle, P.A. & Schluger, P.R.). Spec.
Pub!. Soc. Econ. Paleont. Miner., Tulsa, 26, 3 1 9-336.
MONGER, T.G., RAMOS, J.C. & THOMAS, J. ( 1 98 8) Light oil
recovery from cyclic C02 injection: influence of low
pressures, impure C02> and reservoir gas. 63rd Annual
Technical Conference and Exhibition of the Society of
Petroleum Engineers, Houston, TX, October 1 988.
MORSE, J.W. ( 1 98 3 ) The kinetics of calcium carbonate
dissolution and precipitation. In: Carbonates: Mineral­
ogy and Chemistry (Ed. Reeder, R.J.). M ineral. Soc.
Am. Rev. Mineral., 1 1 , 227-264.
NAGY, K.L., STEFFEL, C. I., BLUM, A.E. & LASAGA, A.C.
 Carbonate cement dissolution
( 1 990) Dissolution and precipitation kinetics of kaolin­
ite: Initial results at 8 0 " C with application to porosity
evolution in a sandstone. In: Prediction of Reservoir
Quality Through Chemical Modelling (Eds Meshri, I. D.
& Ortoleva, P.J.). Mem. Am. Ass. Petrol. Geol., Tulsa,
49, 8 5- 1 02 .
PEARSON, J.F.S., YoUNGS, R.A. & SMITH, A . ( 1 99 1 ) The
Indefatigable Field, blocks 491 1 8, 491 1 9, 49/23, 49/24,
UK North Sea. In: United Kingdom Oil and Gas Fields.
25 Years Commemorative Volume (Ed. Abbotts, l.L.).
Mem. Geol. Soc. London, 14, 443-450.
PERKINS, E.H., YOUNG, B.K., GUNTER, W.D. & TURNER,
J.C. ( 1 992) Evaluation of in-situ geochemical tracers
and their implication for production at a south Midway
Sunset heavy oil reservoir. Proceedings of the 6 7th
Annual Technical Conference ofthe Society ofPetroleum
Engineers, Washington, DC, SPE Paper 24946, pp. 94 1 -
954.
RAMSAY, H.J., Jr. & SMALL, F.R. ( 1 964) Use of carbon
dioxide for water injectivity improvement. J. Petrol.
Tech., 16, 25-3 1 .
REISSER, K.D. & BLANKE, S.J. ( 1 989) Paleostructural con­
trol of Dakota hydrocarbon accumulations on the
southern Moxa Arch, southwest Wyoming and north­
east Utah. In: Petrogenesis and Petrophysics of Selected
Sandstone Reservoirs of the Rocky Mountain Region
(Eds Coalson, E.B., Kaplan, S.S., Keighin, C.W.,
Oglesby, C.A. & Robinson, J.W.), pp. 1 1 7- 1 34. Rocky
Mountain Association of Geologists, Denver, CO.
Ross, G.D., ToDD, A.C., TwEEDIE, J.A. & WILL, A.G.S.
( 1 982) The dissolution effects of COz-brine systems on
the permeability of UK and North Sea calcareous
sandstones. SPE/DOE paper 10685 presented at the
Third Joint Symposium on Enhanced Oil Recovery,
Tulsa, OK, 4-7 Apri/ 1 982.
SAYEGH, S.G., KRAUSE, F.F., GIRARD, M. & DEBREE, c.
( 1 98 7 ) Rock-carbonated brine interactions Part I Car­
dium Formation cores. Presented at the 38th Annual
Technical Meeting of the Petroleum Society of Canada,
Calgary, Alberta, 7- 1 0 June 1 9 8 7 . Paper 87-38-78.
SAYEGH, S.G., KRAUSE, F.F., GIRARD, M. & DEBREE, c.
( 1 990) Rock/fluid interactions of carbonated brines in a
sandstone reservoir. Proceedings of the SPE Formation
Evaluation meeting, Pembina Cardium, Alberta, Can­
ada, December 1 990, pp. 399-405.
SMITH, L.K. ( 1 993) Aspects ofoilfield water chemistry. PhD
Dissertation, University of Wyoming, 290 pp.
SMITH, L.K. & SuRDAM, R.C. ( l 993a) Geological and
geochemical considerations of cyclic C02 enhanced oil
recovery: Cretaceous fields of Wyoming. Proceedings of
499
the 8th Wyoming Enhanced Oil Recovery Symposium,
Casper. WY, 20-21 May 1 992.
SMITH, L.K. & SuRDAM, R.C. ( 1 993b) Cyclic C02 en­
hanced oil recovery in Cretaceous fields. Wyoming
Geological Association 49th Annual Field Conference
Guidebook, pp. 1 45- 1 66.
SMITH, L.K., MACGOWAN, D.B. & SURDAM, R.C. ( 1 99 1 )
Scale prediction during C02 huff'n'puff enhanced re­
covery, Crooks Gap field, Wyoming. Proceedings of the
SPE Rocky Mountain Regional Meeting, Denver. April
1 991, SPE Paper 2 1 8 3 8 .
SURDAM, R.C., DUNN, T.L., MACGOWAN, D.B. & HEASLER,
H.P. ( 1 9 89) Conceptual model for the prediction of
porosity evolution with an example from the Frontier
Sandstone, Bighorn Basin, Wyoming. In: Petrogenesis
and Petrophysics ofSelected Sandstone Reservoirs of the
Rocky Mountain Region (Eds Coalson, E.B., Kaplan,
S.S., Keighin, C.W., Oglesby, C.A. & Robinson, J.W.).
pp. 7-28. Rocky Mountain Association of Geologists,
Denver, CO.
SURDAM, R.C., JAIO, Z.S. & MARTINSEN, R.S. ( 1 99 3 ) The
regional pressure regime in Cretaceous sandstones and
shales in the Powder River Basin. In: Pressure Compart­
mentalization in Sedimentary Basins (Eds Ortoleva, P.J.
& Al-Shaieb, Z.). Mem. Am. Ass. Petrol. Geol., Tulsa,
6 1 , 2 1 3-234.
SURDAM, R.C., DUNN, T.L., MACGOWAN, D.B. & SMITH,
L.K. ( 1 994) Development of Technical Assistance .for the
Implementation ofC02 Treatments in Clastic Reservoirs
of Wyoming: A Geological, Geochemical and Petrophys­
ical Study. Final report submitted to Wyoming Depart­
ment of Economic and Community Development,
administers of the Stripper Well Petroleum Violation
Fund, 1 44 pp.
TILLMANN, R.W. & ALMON, W.R. ( 1 979) Diagenesis of
Frontier Formation offshore bar sandstones, Spearhead
Ranch field, Wyoming. In: Aspects of Diagenesis (Eds
Scholle, P.A. & Schluger, P.R.). Spec. Pub!. Soc. Econ.
Paleont. Miner., Tulsa, 26, 3 3 7- 3 7 8 .
VoN DERHLE, W.F. ( 1 9 7 5 ) Amos Draw field, Campbell
County, Wyoming. Wyoming Geological Association
36th Annual Field Conference Guidebook, pp. 1 1 - 1 5 .
WEBB, J.E. ( 1 974) Relation of oil to secondary clay
cementation, Cretaceous sandstones, Wyoming. Bull.
Am. Ass. Petrol. Geol. , 58, 2245-2249.
WISEMAN, B.W., Jr. ( 1 9 8 1 ) US Patent 4250965.
Wyoming Geological Association ( 1 989) Symposium on
Wyoming Oil and Gas Fields, Bighorn and Wind River
Basins, Oil and Gas Fields Symposium Committee,
5 5 8 pp.

Page numbers in bold refer to tables,and in
italic to figures.
Adori Sandstone, 353
Afro-Brazilian depression, I 09, I II, I 12
albite
authigenic, 90,126
detrital, 304
diagenetic,71
albitization, 236
of plagioclase, 18, 71, 73,76,136,
188-9,304
of K-feldspar,71, 82,126,136
aluminium concentrations, delayed,after
cyclic C02 treatments,494
aluminosilicate materials, evidence for
dissolution after cyclic C02
treatment, 497
dissolution rates,493-4
elevated AI and Si02,492-3
interpretation,493
anatase,72,73
Angel Field, 328
carbonate cement volume, 350--1
controls on carbonate cement
distribution, 351-6
implications for petroleum
exploration, 356-7
gamma ray log correlations, 344
hydrocarbon types and reserves,334-5
Miocene tectonic activity-hydrocarbon
migration-dolomite precipitation
coincidence, 352
over-all paragenetic sequence,
338-9,341
reservoirs and seals, 333, 334
seismic mapping, 34i-50
seismic response,345-7
summary of reservoir characteristics,
351,352
tectonic settings and regional geology,
329-33
burial subsidence and temperature
history, 330
traps, 333-4
wireline log responses, 342, 344
Angel Formation, 329, 330
deposition of,331
low 1>180 value of carbonate cements,
353-4
migration of gas from Locker Shale, 332
anhydrite cement,126-7,149,455,492
Angel Field, 337, 338, 355
anisotropic tortuosity, 187
ankerite cement,5-6,135, 406,422
Magnus Field,405
mesogenetic, 137
temperature of formation, 79-80
Carbonate Cementation in Sandstones Edited by Sadoon Morad
© 1998 The International Association of Sedimentologists
ISBN: 978-0-632-04777-2
Index
Middle Jurassic, 400,403
Middle and Lower Lunde Members,67-9
Oseberg Formation, 29/,292, 293, 304
formation temperature, 296
REE patterns, 300, 301, 302
Sr content,87Sr/86Sr ratios, 1)180 and
li13C values,298-9
two-phase fluid inclusions, 294
as a replacement,14
SerrariaFormatioi1,121-2,123,128
1)180 values of post-rift mesogenetic
phase, 136
ankerite overgrowths, 147
ankerite precipitation, Breathitt and Lee
Formations, 102
anorthite, 304
anoxic seawater,10
Antognola Formation/Marls, 223, 244,248
Appalachian thrust zone, 88
aquifers, Tertiary,Swiss Molasse basin,
divisions of, 153
aragonite, 184
bioclastic,dissolution of, 204,205
Arbuckle Group, saddle dolomite from, 451
arenites, 311
hybrid,244,254-6
carbonate intrabasinal clasts in, 245
246,247
deposition of, 241-2
synorogenic hybrid and lithic, diagenetic
evolution of,241-58
turbiditic,244, 248,254-6
arid/semi-arid regions,carbonate
precipitation in,7
Avalon/Ben Nevis Sandstone Formation,
366,374,377
early calcite cementation,369
secondary porosity, 381,389-90
1)13C value,386
Ballycastle-Murlough Bay outcrop
dolomite-cemented fractures,416-23
sandstone petrology and diagenesis,
414-16
sedimentology and facies associations,
412
stratigraphy, 413
banded iron formations, 405
Banquereau Formation, 367
Bare Formation, 349
bariocelestite, 149
barite cement, 72,73,90, 126, 149
baroque dolomite see entries for saddle
dolomite
basinal fluids
from Viking Graben, 304
Rathlin basin,429
see also brines, basinal
501
Beaver Creek Field, 488
C02 did not remain near well-bore, 496
enhanced production, 487
bicarbonate
increased after C02 treatment, 489
marine, 399,400
bioclasts,199
Bismantova-Termina succession,237,
245,246,247,248-50,257
burial dissolution of, I 0
Namorado Sandstone, 314, 315, 320
biogenic carbonate, 304
heterogenous distribution of, 187
in shallow marine sandstone, I 0
as source for calcite cement, 183-4,
184-5
a source for carbon in central San
Joaquin basin calcites, 275-6
biotite, 91,269
detrital, 270
kaolinization of,76
Birkhead Formation, 332, 344-5
Bismantova Formation, 223,226
calcite, 235,237
Bismantova-Termina succession, 244
carbonate cements, 248-50
other cements, 250--3
paragenetic evolution and porosity
destruction,254-6
petrography and provenance, 245-8
sources and processes of carbonate
diagenesis, 256-8
bitumen envelopes, 73
Bonanza Field
calcite dissolution modelled, 491
carbonate minerals in Tensleep
Formation,491
dissolution of aluminosilicate minerals,
492-4
dolomite dissolution rate, 492
evidence for carbonate dissolution, 485,
488,489
increased bicarbonate, 489
relative dissolution rates between calcite
and dolomite estimated, 491, 492
Borello Formation, 234
calcite, 235,237
spherical concretions,225-6
Bowen-Gunnedah-Sydney basin system,
1)13C values for dawsonite,406
Brae Field, submarine fan sandstones,car­
bon isotopic compositions, 405-6
Breathitt Formation, 88
marine shales in, 88-,9
petrography and geochemistry of
carbonates,90--9
brecciation, 43, 44
Brent Group,286, 396
502
Brent Group (Cont.)
buriaVthermal history of, 287
deltaic sandstone reservoirs, 396,400,
405
late diagenetic calcite, 299
brines
basinal
ascending,causing carbonate
cementation,18-19
highly evolved, 450
evaporative,introduction into Oseberg
reservoir, 304
upwelling
Angel Formation,and salinity of
formation waters,355-6
North West Shelf,Australia,355
Burdigalian Sea,transgression of,143
burial diagenesis,replacement of one
carbonate with another,14
calcite,dissolved,transport by convection,
183
calcite-ankerite-magnesite stability
relationships,15
calcite cement,90,406
ancient marine sediments,10--11
Bismantova-Termina succession,
248-50
differences between hybrid and
turbiditic arenites,248
blocky,257
blocky spar, Zia Formation,37, 39
Breathitt and Lee Formations,early
and late,92-6,97
continental,7-9
diagenetic
Swiss Molasse basin, 146,147, 148
volume in Oseberg Formation, 303
and dolomite cement
alternating bands in calcretes and
dolocretes,75-76
eogenetic,11
marine,9-11
eogenetic,9
fibrous,62
Hibernia Field, 374,376,377
intergranular
Namorado Sandstone, 314,315,318
Swiss Molasse basin,147
Lower Namur Sandstone, 339, 341
Luxembourg Sandstone Formation,
199-200
marine,9-11
micritic,39
microcrystalline aggregations, 315
Middle Jurassic,carbon isotopic
composition,400,401
Middle and Lower Lunde members,
62,66,67
mosaic calcite, 270, 314, 315,317
from redistribution of biogenic
carbonate,159
related to organic C02, influences
hydraulic properties of
sandstone,158
Narnorado Sandstone,315-23
chemical composition, 315,319
Index
main sources, 309-10
types of, 313-15
northern Apennines (sandstones),219
geochemistry of,220-32
sources of Ca, Mg,Fe and Mn, 236-7
Oseberg Formation,292-3
calcium source for, 303-4
covariation in calcite isotopic
compositions, 300, 302
fluid inclusions,.294-5
REEs in, 300,30i,302
Sr content,87Sr/86Sr ratios, 8180 and
o13C values,299-300
patchy/microconcretionary,182
poikilotopic,14, 37, 62,114,118,200,
292-3,299,304,315,339
precompactional
precipitation temperature,79
8180 values of,78-9
San Joaquin basin
basin centre,269-76
basin margin,276-9,277-8
cement zones in margins,276
central,composition of, 273-6
Serraria Formation,114, 119,128,134
in shallow marine sandstones,179-90
nucleation and growth of,185,186
occurrence of,180--2
sources of,182-5
source of under burial conditions,207
sparry,147,200,201,377
syncompactional,mesogenetic euhedral,
8180 values of,79
see also ferroan calcite cement; low-Mg
calcite cement
calcite cementation
diverse patterns of,northern
Apennines,216
early,Avalon/Ben Nevis sandstone, 369
from evolved pore waters,159
late,mechanism of,207-8
patchy,188-9
types of,Zia Formation,34-41
calcite-cemented layers/lenses/zones,180,
188
Luxembourg Sandstone Formation,
196-8
described,196-8
inhibiting fluid flow,194
stratabound,197,198
predicting the geometry of,189
calcite doggers,407
calcite-dolomite-magnesite stability
relationship,16
calcite growth,expansive,45-6
calcite inclusions, in saddle dolomite, 445,
447
calcite lenses
early cementation,206
Luxembourg Sandstone Formation,
196-8
mouldic pores in,201
tightly cemented,203
calcite precipitation
depth and timing of,Namorado
Sandstone, 315-20
early,Hibernia Field, 374, 377
late,102
microbially mediated, 236
repeated events,San Joaquin basin,280,
281
temperature constraints from textural
data, 155, 159
Tertiary sandstones,northern Apennines,
geochemical history of,213-38
calcite rims, 254,257, 314, 374
calcite saturation index,491,494,494
calcite veins,northern Apennine sandstone
units,226
calcium (Ca),76
Breathitt and Lee Formations, source of
uncertain,102-3
increased in produced water after cyclic
C02 treatment,488-9
Serraria Formation,source uncertain,131
sources of
for North Sea poikilotopic cements,
303-4
northern Apennine Sandstones,236-7
calcrete,7,28,54,67
calcite cement in,75
carbon isotopic composition, 81
and dolocrete,influencing fluid flow,53
phreatic,130
caliche, 28
Campos basin,309
geological setting, 311
oil migration and accumulation,313
stratigraphy,312
see also Namorado Sandstone
carbon
derived from alteration of organic
matter, 81
derived from within methanogenesis
zone,81
dissolved,from C3 plants,130
isotopically depleted,Chaunoy
Formation dolomite,174
organogenic,428
sources of,Middle and Lower Lunde
members,81, 82
carbon dioxide (C02)
in enhanced oil recovery,may not remain
in near-well-bore region, 496-7
upward migration,Texas Gulf oil
province,406
carbon dioxide concentration,regional pat­
terns, Cooper basin reservoirs, 355
carbon dioxide corefloods,492
experiments in Tensleep Formation,493
carbon isotope values,saddle dolomite,449
carbon isotopes,northern Apennines,
general trends,220-5
carbon isotopic compositions
calcite cements,central San Joaquin
basin, 275-6
calcretes and dolocretes, 81
early marine cements,San Joaquin basin
margins, 278
reservoir sandstones, Jurassic,northern
North Sea,398,400--6
carbonate cement dissolution,13-14
during a cyclic C02-enhanced oil
recovery treatment,483-97

triggering organic processes, 477
carbonate cementation
affecting reservoir properties of
sandstones,2
controlling factors, I
diagenetic conditions of,constraints from
isotopes and fluid inclusions,78-81
geochemical zones of,2-7
mesogenetic,patterns of fluid flow,
16-19
phreatic zone,131-2
seismic scale,structural controls on,
327-58
concentration near crests of fields,
347,349
migration-related control,352,356-7
carbonate cements
Angel and Gidgealpa fields
controls on distribution,351-6
distribution,342,344-51
anomalous, 180-enrichment of,6
assessment of distribution in sandstones
using petrophysical logs,163-76
Bismantova-Termina succession,
248-50
diagenetic,equilibrium relationships
among, 15-16
early,Serraria Formation,palaeogeo­
graphical and palaeoclimatic
imprints, 127
early precompactional
Middle and Lower Lunde members,
()180 values anomalously low, 78-9
survival of, Lee and Breathitt
Formations, 99
eogenetic, 53-4
dissolution and reprecipitation of,18
precipitation of,78,137
and telogenetic,75-6
facies-related distribution of,7-12
continental calcite and dolomite,7-9
marine calcite and dolomite,9-11
mixed marine-meteoric water
carbonates,11-12
siderite,12
from bacterial sulphate reduction, 4-5
from microbial methanogenesis, 5-6
from thermal decarboxylation of organic
matter,6-7
marine,dissolution of,386
mesogenetic, 76,81-2,137
northern North Sea,396
oxic,2,4
pedogenic,28,43,47
phreatic,4,28,47
recrystallization and replacement of,14
suboxic,4
in Tirrawarra Sandstone, 463
vadose,non-pedogenic,28
see also ankerite cement; calcite cement;
dolomite cement; siderite cement
carbonate diagenesis
in non-marine foreland sandstones,
87-103
sources and processes of,
Bismantova-Termina succession,
256-8
Index
carbonate dissolution,204,386
after cyclic C02 treatment,488-92
amount of dissolution,490
calcite dissolution modelled, 490-1
elevated Cp., Mg and HC03 as
evidence,488-9
interpretation of evidence,489
by carbonic acid, 389
mesogenetic, 13
carbonate grain accumulation,and fringe
cement,206-7
carbonate grains
detrital,216
as nuclei for calcite cement,219
dissolution of, 7,208
Luxembourg Sandstone Formation, 199
as preferential sites for cement
precipitation,256
carbonate leaching, 7
carbonate precipitation
deltaic sandstones, Brent Group, 400,
405
eogenetic, 78,137
from mixed waters,11-12
late-diagenetic, 455
vadose zone
arid/semi-arid regions, 7
marine, 9
carbonate rims,precompactional,256
carbonate rock fragments (CRFs),368
northern Apennine sandstones
as major source of carbon in calcite
cement,235,237
reservoir of calcium in,236
trace elements in,226,227-8, 228
sites for precipitation of ferroan
dolomite, 96
source of calcite cement,184
carbonate rocks,saddle dolomite in,443
carbonate-cemented zones,Angel and
Gidgealpa fields,328,334,
351,357
carbonates
detrital, 34,41,236
diagenetic, 396
inorganic,precipitated from sea water,
10-11
marine,suboxic, o 13 values of,4
vadose, 4
carboxylic acid(s)
generation of,389
in sedimentary basin pore waters,405
thermal degradation of,6
Carnavon basin see Dampier sub-basin
cataclasis,suggested mechanics of,424
cataclastic slip bands (CSBs)
concentration of dolomite cementation
round, 420,424
extensional reactivation of,424-5,
426,431
formation of, 424,430
cataclastic slip zones, 424
Catalina Sandstone Member, 366,369-70,
374,377
intraparticle porosity,381,382
secondary porosity,381,384
origin of,390
503
cement fabrics,marine calcite and
dolomite,9-10
cement petrology,San Joaquin basin, 270-2
cementation
advective,18
or by ionic diffusion,18
by Fe-poor calcite and dolomite,during
eodiagenesis, 75
dominance of phreatic over vadose, 75
mixed vadose and phreatic, 45-6
northern Apennine sandstones, time,
depth and temperature of,232-4
phreatic,130
characteristics of, 45
Zia Formation,45,47
San Joaquin basin,timing of,265-9
syncompactional,258
vadose
characteristics of,43
ZiaFormation,44
cementation episodes,San Joaquin basin,
280
origin of, 281
'cemented fractures' see dolomite­
cemented fractures,
Ballycastle-Murlough Bay outcrop
cemented layers,recent subtidal sands,
205-6
cements
composition of,San Joaquin basin
margins,277-9
environments of formation,Zia
Formation, 43-6
micritic,41.
vadose,criteria for identification of, 9
chalcedony,replacing bioclasts,250,253,
255
Chanac Formation,263
Charlie Gibbs fracture zone, 365
Chaunoy Formation
core analysis data,168
deposited as a minor transgressive­
regressive cycle,164
differing effects of quartz and dolomite
cementation on the pore network,
170-1
dolomite cement
amount and distribution of in the
sandstone,173
effects of on reservoir properties of the
sandstone,173-4
origin of distribution pattern,174-5
origin of,174
grain size data,166
petrographic analysis,167-8
wireline log analysis,168-71
chert,33
chlorite,91,235
Serraria Formation,134-5
chlorite cement,71,76
authigenic, 77
Hibernia Field,380
Ranzano Formation,219
chlorite coatings,380
chlorite rims,71,124,250,254
chlorite/smectite,pore-lining, !58
chloritization, 71,82,124
504
chloritization (Cont.)
Middle Lunde Member, 77
chloritized coatings,Serraria Formation,
124
circumgranular cracking, 44,45
clay coatings,374
clay drapes, Luxembourg Sandstone
Formation,196,205
clay fraction,Swiss Molasse basin,147
closure development,Angel Field,348-50
Cole Creek Field
calcite cement,487
evidence for carbonate dissolution,
488-9
severe well-bore scale, 494-5
compaction
chemical
Bismantova-Termina succession, 254
Hibernia Field,372
Serraria Formation,128-9
see also pressure dissolution
differential,Madeleine Trend, 349-50
mechanical
Bismantova-Termina succession, 254
Hibernia Field, 368-72
Luxembourg Sandstone Formation,
201-3
Oseberg Formation,293
Serraria Formation,128
Namorado Sandstone,318
northern Apennine sandstones,218-19,
232
and occurrence of saddle dolomite, 443
reducing porosity,San Joaquin basin,
269
compositional zoning
saddle dolomite, 444-5
Tirrawarra Sandstone,470,472,474,475
compression
caused by northward movement of
Australian plate,331
Cooper basin,462
North Antrim region, 411,412
concretions
aligned along faults, Loiano and
Bismantova formations,226
Bismantova-Termina succession,248
calcite cement is polycrystalline,185
calcrete/dolocrete, 7, 9
coalescence of, 187,314
early formed,222
Namorado Sandstone,calcite, 309,
313-14,314
origin of, 315,317
tabular,320
northern Apennine sandstones, 214
comparisons of isotopes within and
among concretions, 225-6
Mg, Fe and Mn content variable,
228-32
nucleation and growth model,180
range of influence concept,185,186
scattered,180,182,187
shape of,187-8
spacing between,186-7
stratabound, 180,186, 187
and continuously cemented layers,188
Index
Zia Formation
nodules,34
ovoid and elongate,34-7,45
platy, 37-8, 44
rod, 38
tabular cemented units, 38-41,45,
47-8
conglomerate, 467
continental collision, 242-3
Cooper basin
geological setting,462-3
hydrocarbon migration influenced by
seals, 333
oil expulsion from Permian sequence,
333
sediments, 329
tectonic history,331-2
Cooper-Eromanga basin system,329
petroleum systems,332-3
coseismic elastic rebound,426
covariation
Mg and Fe, Breathitt and Lee
Formations, 96,98, 101
I) lBO and 813C,Breathitt and Lee
formations, 97-9,101
Crooks Gap Field
amount of dissolution,490
calcite cement,485,487
calcite dissolution modelled, 490--1
calcite saturation index,494,494
dissolution of aluminosilicate minerals,
492-4
enhanced production, 487
evidence for carbonate dissolution,485,
488,489
greatest success in enhanced oil
recovery,496
increased bicarbonate,489
cross-bedding,tabular cemented units,
Zia Formation,39
cross-formational flow,compaction-
driven, 280
crustal doming, 411
crustal thinning,differential,135
cyclic C02-enhanced oil recovery treat­
ment,carbonate dissolution
during, 483-97
cyclic C02-treated fields studied,485-7
mineral dissolution amount and rates,
488-94
oil recovery treatments, 484-5
production result of treatments, 487
reservoir geometry,treatment success
and produced water chemistry,
495-7
well-bore scale, 494-5
Dakota Sandstone Formation, 485
carbonate minerals in, 491
spheroidal concretions, 180,182
see also Cole Creek Field; Crooks Gap
Field
Dampier sub-basin, 328
petroleum systems,332
stratigraphy of,329, 330
tectonic history,331
Dawson Canyon Formation,367
dawsonite,o13C values for,
Bowen-Gunnedah-Sydney basin
system,406
deformation
brittle, 410,420,424
compressional, 88
ductile,254
of ductile grains,201
diagenesis
basinal,timing of Ca-plagioclase
alteration,103
Hibernia reservoir,380
meteoric waters,Sr isotope ratios, 80
models of,397
organic,producing hydrocarbon
accumulations, 269
see also carbonate diagenesis; silicate
diagenesis
diagenetic evolution,little difference
between Upper Marine Molasse and
Lower Freshwater Molasse, 157-8
dickite cement, 71,76,124,135,201
Dingo Claystone, 332,352
dissolution
of bioclastic aragonite,204,205
calcite cement,Hibernia Sandstone,
384-5
of early ferroan calcite cement,Catalina
Sandstone Member,381,384
Hibernia Field,381
partial,122,201,250,251
of dolomite/ankerite cements,133
Serraria Formation,135,136
telogenetic,74-5
an unlikely explanation for carbonate
cement distribution, 351
see also carbonate dissolution
dissolution events,significance of,
Tirrawarra Sandstone, 477-8
dissolution porosity, 129
dissolved inorganic carbon,406
in meteoric waters, 399
dolocrete, 7,53,54, 67
carbon isotopic composition, 81
groundwater,137
microcrystalline dolomite in, 75
Middle and Lower Lunde Members,
precipitation temperatures from
meteoric waters, 79
phreatic,131
isotopic composition of,132
dolocrete pedogenesis,174
dolomite
detrital,223,250
Swiss Molasse basin, 145
early overgrowths,254
dolomite-ankerite,stability relationship,16
dolomite cement, 9,62,169,269
Angel Field, 337-9
carbon isotopic values, 354
derived from kerogen maturation, 355
poikilotopic, 337-8,340
Angel Formation, 357,358
Bismantova-Termina succession,250,
251
Chaunoy Formation sandstone,173-5
effects on reservoir properties,173-4

continental,7-9
diagenetic,147
Bismantova-Termina succession,250,
251
early,central San Joaquin basin,270,
279-80
formed in zone of methanogenesis,
272
eogenetic
precipitation of,131
o13 values,Serraria Formation,132
fracture-related,in porous sandstone,
origin and significance of,409-32
Hibernia Field, 377
intergranular,420, 422, 430
enhanced precipitation of,429
late,central San Joaquin basin,273
marine, 9-l l
mesogenetic,137
temperature of formation,79-80
microcrystalline,67,75,167
Middle and Lower Lunde Members,
67-9,81
Middle Jurassic,carbon isotopic
composition,400, 402,405
Serraria Formation,121-2,123,128,135
()180 values of post-rift mesogenetic
phase,136
Tensleep Formation, 487
see also ferroan dolomite cement
dolomite-cemented fractures,
Ballycastle-Murlough Bay
outcrop, 413, 416-23
argument in favour of post-cataclasis
cementation,425-6
banded cement fills, 422-3,427
dilational central veins
characteristics of, 425
fine crystalline dolomite filling,420,
421,422-3
genesis of,423-6
geometry of,420
intergranular volume (IGV),414, 416
morphology and field relationships,
416-20
multiple veins, 417, 418, 423
origin of pore fluids and source of
cement,427-9
overall width, 417,419
petrography and geochemistry, 420-3
pore fluid pressure and controls on
cementation,426-7
timing of fracturing and cementation,
429-30
trends,414,419
walls of,416,418
dolomite-cemented zones
Angel Field, 344
San Joaquin basin margins,276
dolomite dissolution,200,20l
after cyclic C02 treatment,492
dolomite precipitation,219,237
dolomite-cemented fractures, 426-7
in a marine environment,7
dolomite recrystallization,14
dolomite texture,planar and
non-planar, 438
Index
dolomitization,of calcite cement,14
dolostone,250,251
domain overgrowth,quartz, 374,375
Downing Formation, 366
ductile grains,deformation of,201
Dunlin Group, 400
Egret Member,organic shales, 366
Eifel depression,195
elemental composition
saddle dolomite,in carbonates and
sandstones,444-7
siderite,controlled by depositional water
chemistry,12
eodiagenesis
climatic and palaeogeographical con­
trols,Serraria Formation,130-2
and telodiagenesis,role of on diagenetic
evolution of Middle and Lower
Lunde Members, 73-6
Epi-Ligurian Miocene succession see
Bismantova-Termina succession
Eromanga basin, 328
calcite-cemented zones and oil pools,
357
hydrocarbon migration and entrapment,
333
reactivation of Late Triassic structures,
331
see also Cooper-Eromanga basin system
Eurydice Formation, 366
extension,57
Grand Banks, 365
hydraulic fracturing and net upward
flow,426
leading to Grand Banks-Galicia margin
separation, 365
North Antrim region, 411,412,430
and strain cycling,429
faults
affecting fluid flow,17
Ballycastle-Murlough Bay outcrop,414,
419
basin margin,56
concretions aligned along, 226
intrabasinal, 56-7
normal, influencing fluid flow,426
reactivation of, 412
rejuvenation of, Cooper basin, 462
reservoir level, Angel Field, 347,349
shallow,Gidgealpa Field, 350
transfer,365
see also namedfaults
feldspar, 167, 169
albitized,61
detrital
dissolution and replacement of,90
localization of ferroan
dolomite/ankerite and calcite
on,103
see also K-feldspar; plagioclase
feldspar alteration,207
feldspar dissolution
eogenetic,and kaolinization,76-7
Oseberg Formation,299
rates differ for kaolinite and mica,494
505
San Joaquin basin,269
Serraria Formation,129
Fensfjord Formation,prediction of
calcite-cemented zones, 189
fermentation
acetate, 476
bacterial, 399,400,476
ferroar\ calcite cement, 248, 292-3
early and late,Hibernia Field, 374,376,
377,378,380,381
Avalon/Ben Nevis Sandstone, 381
Catalina Sandstone Member,
dissolution of, 381,384
Hibernia Sandstone, 384, 386
possible sources of, 386-9
ferroan carbonate cement, 81
formed during mesodiagenesis,15-16
see also entriesfor ankerite; siderite
ferroan dolomite cement,5-6,90,147,
357,410
Ballycastle sandbody fractures, 410
Breathitt Formation, 96,97,99
Chaunoy Formation,174,175
Hibernia Field, 384
early and late,377
isotopic composition, 388
as overgrowths,147,167,219
zoning within,96
ferroan dolomite overgrowths, 67
ferroan dolomite precipitation, Breathitt
Formation,96,102
floating grain texture,128, 167,339,
381,470
fluid flow,182
influenced by normal faulting, 426
permeability barriers to,285-6
subsurface,evidence for,16-19
through sandstone reservoirs,194
fluid inclusions, 422
Chaunoy Formation,temperature of,174
Hibernia Field late ferroan dolomite
cement, 377,379
Middle and Lower Lunde members,61,
67,69
northern Apennine sandstone units,226,
234
Oseberg Formation,289,294-6,299
saddle dolomite,451-4
Swiss Molasse basin
analysis·of,150,152-3
homogenization temperatures,Upper
Marine Molasse, 152-3
one phase, Lower Freshwater
Molasse,150
Fly Lake Field,462
Tirrawarra Sandstone iii,469
foreland basins,northern Apennines, 214,
243
estimates of burial depth/temperature,
233-4
introduction of meteoric waters,235
formation waters
Angel Field, ��fected by brine-upwelling,
355-6
isotopically eyol�ep,as source of carbon,
northern Apennine sandstones,236
Swiss Molasse basin
506
formation waters (Cont.)
evidence for early mixing,157-8,
159-60
modern, 153
more depleted in 13C than calcite
cements, 153-4
reasons for o13C shift,154-5
Foyle Fault,411
fracture porosity, 381
fracture systems,fluid flow along,17-18
fractures
precipitation of cements along, 18
see also dolomite-cemented fractures
framework compositions,modified by
feldspar dissolution,129
framework grains
diagenetic modification of,61-2
replacement by ferroan calcite,374
replacement by saddle dolomite
cement,443
framework mineralogy,influencing fringe
cementation,206-7
fringe cementation
in early marine CaC03 cementation,
203-4,205
influenced by framework mineralogy,
206-7
Frio Formation, 390,406
gas migration, from Locker Shale to Angel
Formation, 332
Gidgealpa Field,328
carbonate cement volume, 350--1
controls on carbonate cement
distribution, 351-6, 356-7
diagenetic observations,microscopic
scale, 341-2, 343, 344
hydrocarbon types and reserves, 335
production of gas and oil from, 332-3
reservoirs and seals,334
seismic mapping,350
seismic response, 345-7
summary of reservoir characteristics,
351,352
tectonic settings and regional geology,
329-33
burial subsidence and temperature
history, 330
traps, 333-4
wireline log responses,342,344-5
Gidgealpa-Merrimelia-Innamincka (GMI)
anticlinal trend, 329, 350,462
glaucony, 244
intrabasinal,248
glaucony pellets, 338
grain alteration
mobilization during, 237
producing, 180 through water-rock
interaction,235
grain packing,414
loose
Angel Formation, 337,340, 352-3
Serraria Formation,130
grain rotation,420,421
Grand Banks,subdivisions of, 365-6
Grass Creek Field
increased production short-lived,487
Index
late calcium concentration peak,496
gravel lag,467
Great Gaw Fault,412,428,429
groundwater flow,meteorically driven,and
patchy calcite cementation,188-9
haematite,73
hardgrounds,203-4
Heather Field,calcite cements,406,407
heavy minerals,62, 368
Bismantova-Termina succession,248
Hegre Group,57
heulandite,254, 255
Hibernia Field, 366
cement precipitation,372-80
sandstone diagenesis, 368-86
sandstone petrography and framework
grains, 368
sequences of diagenetic events, 380
Hibernia Sandstone Formation, 366, 374
intraparticle porosity, 381,382
secondary porosity, 384-6
origin of, 390
high-Mg-calcite,184,205
'Highland Border Ridge',411
Hopeman Sandstone,426
Hutton Formation, 332
hydrocarbon reservoirs,affected by calcite
cemented zones,180
hydronamic flow,205
controlling cementation,206
IA Pliocene sandstones
concretions,225
Mg- and Fe-enriched calcites,236-7
shell-rich layers not preferentially
cemented, 237-8
igneous activity,extrusive,Antrim,411-12
illite,169, 235
authigenic,469
filamentous, 147
illite cement, 135, 380
illite fibres,71
illite/smectite (liS) mixed-layer clays,71,
389
illitization,124
of kaolinite,135-6
of smectite,207
index mineral,saddle dolomite as,438, 454
infiltrated clays
Serraria Formation,122,132, 136
smectitic,71,135
intergranular porosity
Angel Field, 338
Ballycastle sandbody,primary and
secondary,414-16
Hibernia Field, 381
intergranular pressure dissolution,128-9
intergranular volume (IGV)
Bismantova-Termina succession,254,
256,256
decreased by burial carbonate
cementation,18
dolomite-cemented fractures,414,416
high,182
Lunde cements,75
Lee and Breathitt Formations, 90
early calcites,93
Middle and Lower Lunde Members,
indicative of early cementation, 77
San Joaquin basin
basin margins,276-7
central, 273
Swiss Molasse basin,149-50
values,of northern Apennine sandstones,
233,235,237
interparticle porosity see intergranular
porosity
intraclasts,carbonate,microcrystalline,76
intracrystalline boundaries, corroded, 62,
66
intracrystalline dissolution,67
intracrystalline dissolution pores,69,70,
122
intraformational clasts,reworked from
hardgrounds and lenses,204
intragranular porosity,129
intraparticle porosity, 381
iron carbonates, 6
iron (Fe),5
in calcite,northern Apennine sandstones,
226-32
in dolomite,80
in oxic carbonates,4
replacing magnesium, 61,69
in saddle dolomite,444, 446
sources of
for ferroan calcite cements, 387-8
northern Apennine Sandstones, 236-7
iron-sulphides, 5
iron-Ti oxides,75
Serraria Formation,127
Iroquois Formation, 366
isostatic rebound, 367
isotope and REE geochemistry, Oseberg
Formation,289-90,296-303
jarosite, Rangely Field,493
Jeanne d'Arc basin,Grand Banks
depositional environments and
stratigraphy, 366-8
geological setting, 364-6
Jeanne d'Arc Formation, 366, 386
compaction in, 372
K-feldspar,269
albitized,71
eogenetic,precipitation of,132
partial albitization of,126,136
K-feldspar overgrowths,250,254,255
Oseberg Formation, 303-4
kaolin, 54-5
kaolinite, 169
byproduct of plagioclase dissolution,
San Joaquin basin,269,270
from hydrolysis of K-feldspar, 493
kaolin(ite) booklets, 76,122, 125, 377,
378,469
kaolin(ite) cement, 71,90,201,384
Angel Field, 353
authigenic, 77, 377
eogenetic,74
replacing feldspar,mica and
pseudomatrix,71

Serraria sandstones,124
telogenetic,74-5
kaolinite precipitation,62
invasion of meteoric pore wares
during, 477
kaolinite vermicules, 147
diagenetic,293
kaolinization,71
eogenetic,Middle and Lower Lunde
Members,81
Serraria Formation,124
in Upper Lunde sandstones,76
Kern River Formation,263
kerogen maturation,355, 429,477
Kimmeridge Clay Formation,major source
rock,390,396,399,400,405,407
Lee Formation,88
marine shales in,88-9
petrography and geochemistry of
carbonate components,90-9
limestone grains and clasts,northern
Apennine sandstones,219
litharenites
feldspathic,33,145
siliclastic feldspathic see arenites,
turbiditic
Locker Shale,gas source,332,352
Loiano Sandstone Formation,223,226,
236,244
early-formed concretions, 222
limited degree of rock alteration,235
Mg- and Fe-enriched calcite,237
low-Mg calcite cement
concretions,7
Luxembourg Sandstone Formation,
199-200
replacive during late diagenesis,208
two generations,199-200
precipitation of,205
Lower Freshwater Molasse,143
pore water isotopic composition,156-7
Lower Lunde Member see Middle and
Lower Lunde Members
Lower Marine Molasse,143
Lower Namur Sandstone,339-42,350
Lower N appamerri Group seal,breaching
of,352
Lunde Formation
deposition of,57
depositional environments,57-8
provenance of,Tampen Spur,61
Luxembourg,Gulf of,195
Luxembourg Sandstone Formation,
193-209
calcite-cemented lenses and layers
described,196-8
deposition of and facies,196, 197
detrital composition of sandstones,199
diagenetic minerals,199-203
early marine CaC03,sources for,204-6
early marine CaC03 cementation
differential,206-7
timing of,203-4
geological setting,195-6
late calcite cementation mechanism,
207-8
Index 507
later diagenesis enhancing early perme­ Luxembourg Sandstone Formation,
ability patterns,208 208
Serraria Formation,136,137
Macae Formation,317 introduced into northern Apennine
carbonate rocks a possible calcite source, basins,235,237
318,320 limited telogenetic circulation of,258
Madeleine Trend,329,331 mixing with fault-transported fluids,406
differential compaction over,349-50 Namorado Sandstone,320
magnesite cement,eogenetic,12 methane, 455
ferroan,16 biogenic,reduction of, 400
magnesium (Mg),76 oxidation of, 236
Breathitt and Lee Formations, 97 methanogenesis,5-6
in calcite,northern Apennine sandstones, precipitation in zone of,272,280,388
226-32 providing carbon for Tirrawarra
high in siderite,75 Sandstone siderite cements,476
increased in produced water after cyclic mica
C02 treatment,488-9 detrital,11,61-2
released from kerogen, 477 kaolinitized,71,124
Serraria Formation,source uncertain,131 mica flakes,expansion of,67
sources of,northern Apennine micritic envelopes, 205
sandstones,236-7 micritization, 254
magnesium-rich calcite,11 microcodium,43
magnesium-rich minerals,alteration of to microtextures
provide carbon for siderite cement, alveolar, 44
477 and fenestral,40
Magnus Field,submarine fan sandstones, floating grain,45
carbon isotopic compositions, radial spar,44
405-6 Middle and Lower Lunde Members
manganese (Mn) burial history,58-9
in calcite,northern Apennine sandstones, depositional environments,57-8
226-32 petrography and chemistry of diagenetic
in dolomite,80 materials,62-73
in oxic carbonates, 4 role of eodiagenesis and telodiagenesis
in saddle dolomite,444, 446 on diagenetic evolution of,73-6
sources of,northern Apennine sources of eogenetic calcite/dolomite
sandstones,236-7 cements,76
manganese/iron ratios, Breathitt and Lee timing of main diagenetic processes, 73,
Formations,93, 97 74
marine CaC0 3 cementation,early Middlesboro syncline,89
differential,206-7 minus-cement porosity,293
internal and external sources for cement, Hibernia Field,368,372,374,377
204-6 mixed-layer clays,147
timing of,203-4 see also illite/smectite (liS) mixed-layer
Marnoso-arenacea Formation, 223,226 clays
Upper,Mg- and Fe-enriched calcite,237 Monte Piano Marls,244
mass transport,diffusional,185 Monterey Formation,279-80
matrices,micritic,44 Moorari Field,462
megasequences,Hegre Group,57 Tirrawarra Sandstone in,469
mesodiagenesis mouldic pores,201,314
role of eogenetic minerals and Muddy Formation, 497
temperature,76-7 mudrocks
Serraria Formation calcite in,northern Apennines,220
and post-rift subsidence,135-6 northern Apennines,226
and syn-rift subsidence,132-5 Murlough Bay coals,Mg-rich pore fluids
meteoric water flushing,183,286,431 from, 429
meteoric waters,298,386 Murre Fault,365-6
and dissolution of carbonate cements, Murteree-Nappacoongee (MN) anticlinal
13-14 trend,462-3
in dissolved inorganic carbon,399 Murteree Shale,333
effects of infiltration into Serraria
Formation,135 Namorado Sandstone,309-24
eogenetic interaction with detrital deep-water turbidites,311
minerals,74 depositional environment,313
incursion into San Joaquin basin depth and timing of calcite precipitation,
margins,264 315-20
influx of,Hibernia Field,387,389 burial history,315,320
influx of and dissolution calcite types,313-15,319,320
508
Namorado Sandstone (Cont.)
oil emplacement,influence on carbonate
cement distribution, 321-3
sandstone petrography, 311-15
Namur Sandstone Formation, 329,330,
33/,332,334,339
broad paragenetic sequence, 341-2,344
calcite-cemented intervals,344,345,346
carbon isotopic value, 336,354,355
possible sources for calcite cement, 355
Nappamerri Group,seal integrity of, 333,
350,355
Nappamerri syncline, 462
Newfoundland fracture zone, 365
Newman Limestone, late calcites, 102
nodular cementation, 207
North Alpine Foreland Basin, 141
North Coles Levee reservoir
burial history plot, 264
calcite cements, 273
fluctuating carbon isotope values, 276
late calcite cements, 274-5
carbon source in, 280-1
North Grieve Field, 489
enhanced production, 487
information from post-injection­
produced water chemistry, 496-7
North Helvetic Flysch, 142-3
northern Apennine sandstones
oxygen isotopic composition of
cementing fluids, 234-5
sources
of Ca, Mg, Fe and Mn in calcite,236-7
of carbon,235-6
spatial variation of carbon and oxygen
isotopes within concretions and
beds,224-5
stratigraphical units, 214,215
time,depth and temperature of
cementation,232-4
northern Apennines
stratigraphy,depositional and burial
history of the Epi-Ligurian
succession,243-5,243
tectonic evolution, 242-3
northern North Sea
Jurassic reservoir sandstones, 395-407
comparison with other sedimentary
basins, 406-7
oil
in Namorado Sandstone turbidites,311
see also petroleum
oil charging, influencing diagenetic evolu­
tion of Namorado Sandstone, 322-3
oil emplacement
Namorado Sandstone,influence on car­
bonate cement distribution, 321-3
North Coles levee,and timing of late
carbonate cementation, 280-1
oil migration
and accumulation,Campos basin, 313
Oseberg structure, 287-8
oil-water interface, Namorado Sandstone,
separating two calcite types, 322
oil fields,northern North Sea, calcite­
cemented intervals,285-6
Index
ooids,Luxembourg Sandstone, 205
organic acids,269,389,429
carbon derivation from breakdown of,
405
triggering dissolution events, 477
organic carbon
contribution to cement,222
isotopically negative, 399
reaction paths for sequestration in
cements, 236
organic material/matter
carbon from as a source for calcite
cement, 184
oxidation of, 81
2
preferential removal of 1 C by, 476-7
thermal decarboxylation of, 81
thermal degradation of,236
organic shales,Egret Member, 366
orogeny, post-collisional, 242-3
Oseberg Formation
fluid inclusions, 289, 294-6,299
geological setting, 286-8
isotope and REE geochemistry,296-303
petrography and spatial distribution of
carbonate cements, 290-4
source of calcium, 303-4
overpressure, Rathlin basin, 429
oxidation
bacterial, 399
microbial, not appropriate for Angel and
Gidgealpa Fields, 354-5
of organic matter,205,428-9
oxygen isotope values
and i513 C values
dolomite-cemented fractures, North
Antrim,425, 428
Oseberg Formation, 296-300
saddle dolomite, 449,455
oxygen isotopes, 78-80
northern Apennines,general trends,
220-5
oxygen isotopic compositions
calcite cements,San Joaquin basin
margins, 277-8
cementing fluids,northern Apennine
sandstones,234-5
reservoir sandstones, Jurassic,northern
North Sea, 406
packing proximity index, 315,323
palaeosols,Serraria Formation,127-8
Paris Basin
geological setting, 164,195-6
Liassic source rocks, 174
Patchawarra Formation, 332
Patchawarra syncline, 462
pedogenesis
dolocrete, in interchannel facies, 174
and phreatic cementation, 45
permeability
Chaunoy Formation, 163
derived from porosity and mineralogy
data, 168
derivation from porosity, 170
vadose zone, 43
permeability anisotropy,187
petroleum
biodegradation of, 303
a source of C02,399
petroleum migration conduits,late-stage
origin of carbonates along, 352
phreatic zone, cementation in,28,75
phyllosilicates,detrital, 91
picotite grains, 248
piggy-back basins,northern
Apennines,214
calcite precipitation in, 235
temperature of cementation,233
Pine Mountain overthrust, 88
pistomsite cement, 471,472
plagioclase, 76,313
detrital,269
albitization of, 71,73,136
kaolinitized, 76
plagioclase alteration, importance of,
San Joaquin basin, 273, 274,281
plagioclase dissolution, 428
San Joaquin basin
central, 273
eastern margin, 278-9
kaolinite a byproduct, 269, 270
source of calcite cement,184
Plateau (Mittelland) Molasse,141
Poolowanna Formation,332
pore fluid pressure
and controls on cementation,
Ballycastle-Murlough Bay section,
426-7,431-2
higher in fracture conduits, 424
pore fluids, origin of and source of dolomite
cement, Ballycastle-Murlough Bay
section, 427-9
pore waters, 54,386
and alteration of detrital magnesian
minerals, 75
evaporative increase in ionic
concentration, 131-2,132
evolved,Swiss Molasse basin, calcite
cementation from,159
isotopic ratios from calcite, 155-7
marine
Angel Formation, 353
central San Joaquin basin,264,269,
280
northern Apennine sandstones,237,
258
modern, Swiss Molasse basin,
relationship with diagenetic
calcites, 153-5
oi80 compositions, 78
saline, 455
San Joaquin basin,264
sedimentary basin,carboxylic
acids in, 405
87Sr-enriched, 450
porosity
Bismantova-Termina successions,
destruction of,254, 256
and permeability in sandstones, 193
key controlling factors, 163
primary, 381
redistribution of, 77
Serraria Formation, 129,136
wireline-derived data,168-70

see also intragranular porosity;
secondary porosity
porosity enhancement,by carbonate
dissolution, 137
porosity evolution,compaction vs.
cementation and reservoir
implications,77
porosity loss
Namorado Sandstone,321, 323
Oseberg Formation,293
primary,by silica cementation, 385-6
through compaction, 201
reduced when fluid is oil, 321
precipitation,syncompactional,122, 123
pre-Muribeca unconformity,129
pressure dissolution,136,235
Bismantova-Termina succession,251,
254,255,257
of carbonate rock fragments,237
of detrital carbonate particles,219-20
pressure-solution contacts, Hibernia
Sandstone, 372
pyrite cement, 72, 73,149,269
Bismantova-Termina succession,253,
254,255
early and late,Hibernia Field, 380
Serraria Formation,126,135
pyrite cubes, 338
quartz arenites,135
quartz cement,90, 126,147
Angel Field, 337
Chaunoy Formation,164
Hibernia Field,372,374,375
mesogenetic,76
precipitation during burial diagenesis,71
Serraria Formation,124,126
containing bitumen enclosures,136
Tirrawarra Sandstone,469,471, 472
quartz cementation
Breathitt and Lee Formations, 93
Luxembourg Formation, 200--1
quartz overgrowths,62,76-7,200,341,469
eogenetic,128
Hibernia Field, 372, 374, 375, 380
Serraria Formation,124,126,132,135
quartzarenites,113, 368,414,467
Lee Formation,90
Rangely Field,prediction of barite scale,
484
Rankin Formation, 366
Rannoch Formation, predictions of concre­
tionary cementation confirmed,189
Ranzano Formation,244,248
chlorite cement in, 219
spherical concretions, 225
rare earth elements (REE), Oseberg
Formation, 300,301, 302
Rathlin basin
evolution and structure, 410-12
possible source of dolomite-precipitating
fluids,429
recrystallization,of carbonate cements,14
regional reflectors, Australia,used in
isochron mapping,Angel Field,
348-50
Index
reservoir rocks,porosity degraded,163
reservoir sandstones,significance of 813C of
carbonate cements in, 395-407
carbon sources, 399-400
interpretation,400-6
sampling representativeness, 396, 397-8
reservoirs
characteristics of,Angel and Gidgealpa
Fields,summarized, 351, 352
inferences,Serraria Formation,136-7
porosity
reduced by poikilotopic carbonate
cements, 357
in sandstone reservoirs,evolution
controlled by carbonate
cementation, 363, 386
properties of
affected by fracture/fault compart­
mentalization of sandstones,
409,431
Chaunoy Formation sandstone,effects
of dolomite cement on,173-4
control by diagenetic alteration, 193-4
quality of
Chaunoy Formation,176
enhanced,Serraria Sandstone,135
implications of dolomite-cemented
fractures, 430-1
potential greatly reduced,47-8
rhizocretions,and vadose cementation, 43
rhodochrosite,4,12
rift tectonics,North Sea,286
rock fragments,ductile, 368, 369
root mats,44-5
Roseneath Shale, 333
Rotliegendes Sandstone,Indefatigable
Field,492
saddle dolomite cement, 377
in carbonates and sandstones,437-56
CL characteristics,443-4
elemental composition,444-7
fluid inclusions,451-4
a high-temperature diagenetic index
mineral,454
little information on name-giving
crystal distortion,455-6
overview of published work, 438,
439-41
petrographic features,438, 443
precipitating fluid reducing,455
quantitative constraints on formation
of,454-6
stable isotopic composition,447-9
strontium isotopic composition,450-1
saddle dolomite precipitation
coincident with chemical compaction,
443
fluids involved,455
temperature of,455
San Joaquin basin,261-81
basin pore water,264
calcite cement
early marine,280
late,carbon source in,280-1
cementation episodes,origin of,281
cements
509
basin centre,269-76
basin margin,276-9
diagenetic history, 269
dolomite, early marine, 279-80
geological history, 262-4
central region, 263
eastern margin,263-4
western margin,264
timing of cementation,265-9
carbonate cement studies and data,
266-8
Sand Hill Fault,29
sandstone sequences, in carbonates, and
reservoir quality, 53
sandstone(s)
calcite-cemented,strata-bound, I 0
deltaic,Middle Jurassic, 396,400,
405,407
lithic,287
nodular cemented, I0
northern Apennines
detrital and authigenic carbonate in,
219-20
petrography,218-20
shallow marine,10
calcite cement in,179-90
Luxembourg Formation,193-209
stacked sequences,Angel Formation, 334
submarine fan sandstones,Upper
Jurassic,396,405-6, 407
Santa Fe Group,28
Santa Margarita Formation, 263
sea level rise,eustatic,411
sea water
anoxic,upwelling of,10
for CaC03 cementation,206
supersaturated,an external CaC03
source, 205
seals, regional, 333, 334, 358
breaching of,and positions of
hydrocarbon pools and carbonate­
cemented zones,333, 352, 353
secondary oil recovery, 2
secondary porosity, 62,337,363
due to dissolution of carbonate cements,
criteria for recognition of,14
Hibernia Field, 372, 380-6
origin of, 389-90
intragranular, 90
Luxembourg Sandstone Formation,
201-3,205,208
sedimentary basins,patterns of fluid
flow in,17,17
sediments
terrestrial,early diagenetic alteration in,
27-8
volcaniclastic,alteration of,10
seismic pumping,17-18,425
seismic valving,429
Sergipe-Alagoas Basin, Brazil
rainfall rate during exposure of Serraria
sandstones, 135
structural components of, 109
structural evolution of marginal basips,
111-12
Serraria Formation
compaction,128-9
510
Serraria Formation (Cont.)
detrital composition and provenance,
113-14
diagenetic domains,108
diagenetic evolution and burial history,
129-30
diagenetic minerals,114-27
eodiagenesis,climatic and palaeogeo­
graphical controls,130-2
geological setting
palaeogeography and palaeoclimate,
109-11
stratigraphy and depositional
evolution, 111-13
lithofacies,112-13
palaeosols, 127-8
porosity,129
post-rift subsidence and
mesodiagenesis, 135-6
post-rift uplift and telodiagenesis,135
recent exposure and telodiagenesis,136
reservoir inferences,136-7
syn-rift subsidence and
mesodiagenesis,132-5
shales,marine,organic-rich,
Ballycastle-Murlough Bay
section,428
shaliness,defined by gamma Jog,165
shell debris,dissolution of, 399
shrinkage porosity,129,381,383
siderite,4
elemental composition controlled by
depositional water chemistry,12
siderite cement,90,269
Breathitt and Lee Formations,91-2
source for, !OJ
Hibernia Field, 372,373, 380
possible origins, 388-9
high-Mg,16,75
Magnus Sandstone, 405
Middle and Lower Lunde Members,69,
70,75,81
enrichment in 160,79
Middle Jurassic, 404, 405
Oseberg Formation,290-2,293, 304
REE patterns,300,301, 302
Sr content,87Sr/86Sr ratios, 8180 and
li13C values,297-8
San Joaquin basin,270
Tirrawarra Sandstone,isotope
interpretation of, 461-78
bulk 8180 and li13C, 473, 475
cement characteristics, 469
early generation,468, 469-71
late generation,468, 470,472-3
method for enhanced isotope
interpretation,474-5
middle generation,470,471-2
multiple generations of identified,
466-7,474
significance of dissolution events,
477-8
sources of carbon,476-7
siderite-magnesite,stability relationship
and extent of solid solution
between,16
siderite precipitation,5-6,78,297
Index
Tirrawarra Sandstone cements,475-6
siderite spar, 341
sideroplesite cement,47 1,472
silica cementation,porosity Joss by, 385-6
silica (Si)
authigenic,associated with calcrete and
dolocrete,9
concentration profile,493-4
silicate diagenesis,235,269
silicates
eodiagenesis and telodiagenesis,73-5
Luxembourg Sandstone Formation,
200-1
mesogenetic,76
smectite,9,73,147
pore-lining,158
smectite coatings,transformed to
chlorite,71
smectite to illite diagenesis,207,388
San Joaquin basin,269
Snorre fault block,55
uplift and tilting of,59
Snorre Field
geological setting,55-9
structure,stratigraphy and
palaeoclimate,55-7
Snorre Field reservoir, 55
source rocks
Kimmeridge Clay Formation, 390, 396,
399,400,405,407
Liassic, Paris Basin,174
shale,San Joaquin basin,276
South Coles Levee reservoir
effects of plagioclase alteration,273-4
low Sr content of calcites,274
southern Appalachian Basin,carbonate
diagenesis in non-marine foreland
sandstones,87-103
geological setting,88-9
sphalerite,90, 126
stability relationships
calcite,dolomite and ankerite,16
calcite-ankerite-magnesite,15
dolomite and ankerite,16
siderite and magnesite,16
stable isotopic composition,saddle
dolomite, 447-50
Statfjord Formation,57
Stevens sandstone,263, 264
dolomite in,272,279-80
receiving Ca and Sr from plagioclase
alteration, 273
strain cycling,426,429
strain hardening,424
strontium (Sr)
radiogenic,100,297
saddle dolomite,444,447
strontium concentration,calcite and ferroan
dolomite/ankerite,99
strontium content
cements from San Joaquin basin
margin,279
Oseberg Formation,296-300
strontium isotope composition
central San Joaquin basin cements,
269-70
saddle dolomite,450-1
strontium isotopes, Middle and Lower
Lunde Members,80
strontium isotopic ratios
guide to cementation timing,269
typical of meteoric water diagenesis, 80
87strontium/86Sr analyses,San Joaquin
basin margin cements,278
87stronium/86Sr ratios
calcite and ferroan dolomite/ankerite,99
Lunde Formation,61
Oseberg Formation,296-300
saddle dolomite,449
Stevens sandstones,272
Subalpine Molasse,141
subarkoses,61,1 13
lithic,145
quartzose,113
sublitharenites,90,368,469
subsidence
Rathlin basin, 411,429
Serraria Formation, 129-30
post-rift and mesodiagenesis,
135-6,138
syn-rift and mesodiagenesis,
132-5,137
sulphate reduction
bacterial,4-5, 399; 400
diagenetic,386
sulphate reduction zone,280,297
sulphides
late-stage,455
and sulphates,Serraria Formation, 126-7
sulphur isotopic ratios, Lower Freshwater
Molasse,150
sutures,stylolitic, 372
Swiss Molasse basin
authigenic minerals, 145-9
minor cements,147, 148,149
detrital composition,145
history of formation waters,evidence for
early mixing, 157-8, 159-60
major stratigraphic units,142-3
origin of calcite cements,158-9
pore water isotopic ratios from calcites,
155-7
relationship between modern pore waters
and diagenetic calcites,153-5
stable isotope geochemistry,149-53
tectonic framework,141, 142
Talisman Sandstone,ferroan
dolomite in, 357
Tampen Spur,55
alluvial reservoir rocks in,56-7
tectonic pulsing, 428, 432
teepee structures,43, 44
telodiagenesis
and eodiagenesis,role of on diagenetic
evolution of Middle and Lower
Lunde members,73-6
Serraria Formation
and post-rift uplift,135
and recent exposure,136
Tennapperra syncline,462
Tensleep Formation, 485,487,493
carbonate minerals in,491
see also Bonanza Field

Texas Gulf Coast oil province, 406
thermallthermocatalytic decarboxylation,
174,386
of organic matter, carbonates from,6-7
thermallthermocatalytic decarboxylation
zone, 298, 387
thermobaric fluids, 429
tight beach rocks, 9
tight layers,and concretions, 182
tight-sandstone zones, 369,377,3 8 1
Tirrawarra Sandstone, 332
depositional environment,463
isotope interpretation of siderite
cement, 461-78
palaeoenvironmental interpretations,467
Toolachee Formation, 332, 350
total organic carbon (TOC),in Hibernia
Sandstone, 390
Tow Valley Fault,411,412,429
trace element partitioning,Breathitt and
Lee Formations,99-100
trace elements
northern Apennine sandstones, 236-7,
238
saddle dolomite, 444
Trealla Limestone, 349
uplift
Miocene, 89
Index
post-rift and telodiagenesis,Serraria
Formation, 135,137-8
Rathlin basin,411
Upper Freshwater Molasse,143
Upper Lunde Member, 58,76
Upper Marine Molasse,143
pore water isotopic composition,155-6
USA
mid-continent,saddle dolomite related to
localized MVT lead-zinc deposits,
454
Wyoming, cyclic C02-treated fields,
485-6
Veslefrikk Field,calcite cements, 397-8,
400,406,407
volcanic grains,euhedral calcites from
alteration of,158
volcanic rock fragments,unstable,alter­
ation of, 254
Voyageur Formation, 366
Walcott Formation, 348
water-rock interactions, 235, 483-4
Weber Sandstone,493
well-bore scale, potential for during C02
treatments,494-5 • vadose cementation,44-5, 48
Western Canada Sedimentary Basin, saddle
511
dolomites, 449-50
white sparry dolomite see entriesfor saddle
dolomite
Wilcox Formation,406
wireline data,advantages for derivation
of porosity,mineralogy and
permeability,171-3
zeolite crystals,250,254
Zia Formation
controls on spatial distribution of
cementation, implications for
groundwater and petroleum
resources,47-8
environments of cement formation, 43-6
facies associations, 32
geological setting,28-9, 31
isotope geochemistry, 42-3,46-7
pedogenic cementation,47
phreatic zone cementation, 45,47, 48
sandstone petrography, 33-4
stratigraphical columns, 30,36
timing of cementation, 47
types of calcite cementation, 34-41
concretions, 34-8
tabular cemented units, 38-41
lower 8180 values,46
Plate I. Cathodoluminescence (CL) and ultraviolet (UV)
fluorescence images of diagenetic carbonates in Lower
and Middle Lunde Members: (A) UV image of brightly
yellow-fluorescing microcrystalline calcite in calcrete
with vugs filled by zoned non-fluorescing to dull green
coarse blocky calcite; (B) CL image of calcrete with
microcrystalline red to reddish-brown luminescing
calcite and vugular pore filling by coarse, blocky,
non-luminescing bright orange zoned calcite; (C) C L
image o f a sandstone cemented b y poikilotopic calcite
Carbonate Cementation in Sandstones Edited by Sadoon Morad
© 1998 The International Association of Sedimentologists
ISBN: 978-0-632-04777-2
with overall orange luminescence and dark red to brown
zones; (D) CL image of a dolocrete with abundant
displacive microcrystalline dolomite with red zonation;
(E) CL image of red-luminescing microcrystalline
dolomite rims and bright orange blocky pore-filling
calcite; (F) UV image of brightly fluorescing
collomorphic microcrystalline dolomite rims, with some
alternations of non-luminescing to brown calcite; coarse
pore-filling calcite is non-luminescing.
[facing page 62)
Plate I. Photographs showing the characteristic
authigenic carbonates in the studied sandstones. (A,B)
Cross-polarized and UV-fu l orescence photomicrographs
showing early dolomite overgrowth on a detrital
dolomite grain (note the inclusion-free thin authigenic
rim, arrows), Tiefenbrunnen, 664.7 1 m, U MM. Scale bar
200 )lm. (C) Cathodoluminescence photomicrograph
showing the pervasive mosaic cement, Gurten, 229.8 m,
U M M. Scale bar 300 )lm. (D) Cathodoluminescence
photomicrograph showing the euhedral calcite cement
(arrow), Altishofen, 955.7 m, L F M. Scale bar, 300 )lm.
(E) Photomicrograph showing pervasive mosaic calcite
from Murgental, 18. 1 m, L F M. Scale bar 500 )lm. ( F)
Photomicrograph showing early dolomite overgrowth.
Note the inclusion-free overgrowth and the postdating
ferroan calcite, stained darker. Tiefenbrunnen,
596.92 m, U MM. Scale bar 300 )lm.
[facing page !58]
Plate I. Laterally extensive continuously Plate 3. Scattered calcite-cemented concretions in low
calcite-cemented layers and layers of stratabound angle-laminated beach to upper shoreface sandstones of
calcite-cemented concretions in the Lower Jurassic the Middle Jurassic Valtos Formation at Valtos on Skye,
Bridport Sands, south of Burton Bradstock in Dorset, Scotland. Scale is I m long.
southern England. The calcite-cemented layers extend
laterally for several kilometres through the extensively
bioturbated lower shoreface sandstones. Calcite­
cemented sandstone sdnds out in relief because of its
greater resistance to weathering. Scale is 4 m long.

Plate 2. Close-up view of calcite cementation in the

Bridport Sands south of Burton Bradstock. Two layers of
stratabound flattened concretions are present below a
continuously calcite-cemented layer where an early
concretionary growth stag:' can be seen by its lighter
weathering colour. Note c'.i.fferences in concretion shape
and spacing between the three layers. Scale is 2 m long.

Plate 4. Patchy calcite cementation merging downwards

to form an almost pervasively cemented interval. Mouth
bar sandstone from the Middle Jurassic Oseberg
Formation in the Veslefrikk Field, Norwegian North
Sea. Width of core is 14 em; calcite-cemented areas are
light grey.

[facing page J82j

Plate 5. Bedding plane view of stratabound calcite­
cemented concretions in the top of a mouth bar
sandstone in the Cretaceous Helvetiafjellet Formation at
Kvalvagen, east coast of Spitsbergen. Calcite-cemented
areas have a brown weathering colour, the surrounding
quartz-cemented sandstone is light grey. Note how in
places concretions have merged to form small
calcite-cemented layers. Scale is l m long.

Plate 6. Giant calcite-cemented concretions in the low

angle-laminated beach to upper shoreface sandstones of
the Middle Jurassic Valtos Formation at Valtos on Skye,
Scotland. Scale is I m long.

Plate 7. Perfectly spherical calcite-c �mented concretion

in turbiditic sandstones of the Paleogene Frigg
Formation, 2072.5 mRKB, well 25/2-9, Norwegian
North Sea. Width of core is 14 em.
Plate 8. Large calcite-cemented concretion formed by Plate 9. Calcite-cemented foresets and horizontal layers
merging of several concretions, plus a flattened in the otherwise unconsolidated sandstone of the
spheroidal concretion in the lower left-hand corner. estuarine Lower Jurassic Luxembourg Sandstone at
Perfectly preserved hummocky lamination is seen within Vance, Belgium. Scale is 4 m long.
the concretions, which were eroded from the otherwise
unconsolidated shallow marine hummocky-laminated
sandstone of the Upper Jurassic Bencliff Grit at
Osmington Mills in Dorset, southern England. Feet for
scale.

Plate 10. Three strongly dissolved carbonate shells; the

shell in the middle has been almost totally dissolved,
leaving only a thin rind of insoluble material (arrow).
Note how the sand grains move into the volume
previously occupied by the carbonate fossils. Upper
Jurassic Ula Formation, well 71 12-2, 3432.69 mRKB,
Norwegian North Sea.
Plate 1. (A) Medium-grained Avalon Sandstone
completely cemented by early, poikilotopic ferroan
calcite cement (C), which corrodes quartz (Q) grains
(arrows). 0-35 well, 2 191.5 m, scale bar= 0. 1 mm.
(B) Boundary between calcite-cemented and
uncemented horizons in Avalon Sandstone, showing
no evidence of dissolution. Straight crystal faces of
the poikilotopic calcite (left) at the boundary (centre)
indicate the presence of a cementation front, rather than
a dissolution front. Note that framework grains in the
porous zone are coated with thin clay rims, which are
absent in the cemented zone. The former is also slightly
more compacted than the latter. Same locality as
Fig. 14A, 2 197.9 m, scale bar= 0. 1 mm, stratigraphical
top left. (C) Late ferroan calcite cement associated with
< 15% minus-cement porosity. Note concavoconvex
grain contacts and corroded quartz grain boundaries
(arrow). Hibernia Sandstone, B-27 well, 3905.0 m, scale
bar= 0.05 mm. (D) Completely shattered quartz grain
between two other larger grains cemented with
poikilotopic calcite (C). The widespread occurrence of
healed microfractures in siliciclastic rocks has been
largely overlooked until very recently (Milliken, 1994).
Hibernia Sandstone, B-08 well, 3496.4 7 m, scale
bar= 0.05 mm. All photomicrographs plane-polarized
light; A,B,D, alizarin-S stained.
(facing page 374)
Plate 2. (A) Irregularly shaped secondary pores (P)
resulting from dissolution of framework grains and
pore-filling and replacive cements. Hibernia Sandstone,
B-27 well, 3850.65 m. (B) Intraparticle microporosity
of secondary origin resulting from partial dissolution
of calcite in recrystallized portion of fossil fragment.
Arrows indicate uncorroded crystal faces of calcite.
Incipient pressure solution between quartz and calcite
shell. Avalon Sandstone, 0-35 well, 2 185.8 m. (C) Large
mouldic pore resulting from dissolution of carbonate
fragment. The prismatic ferroan-calcite cement crystals
that had grown on the fragment are less affected by
dissolution and relatively well preserved. Their colour
varies as a function of distally increasing iron content.
Avalon Sandstone, B-27 well, 2578.84 m, scale
bar= 0.05 mm. (D) Fossil-rich fine-grained sandstone
lacking carbonate cement despite pressure solution
(arrows) between quartz grains and fossil fragments.
Porosity is primary. Same locality as Fig. 178, 2 195.4 m.
All photomicrographs plane-polarized light. A,B,D, scale
bar=0.5 mm.

Plate 1. Thin-section photomicrographs in
cross-polarized light showing characteristic features of
saddle dolomite and ankerite in sandstones and
carbonate rock. (A) Ferroan saddle dolomite crystal
marginally replacing quartz framework grains (arrow).
Intergranular quartz cement postdated the formation
of saddle dolomite. Upper Carboniferous Spiro
Sandstone, Arkoma basin, USA. Scale bar� 200 Jlm.
(B) Quartzarenite pervasively cemented by saddle
ankerite and minor pyrobitumen (opaque material).
Individual dolomite crystal outlines are poorly
developed but sweeping extinction is widespread. Same
unit as (A). Scale bar� 500 Jlm. (C) Fracture-fill saddle
ankerite composed of crystals with well-developed
boundaries and strong curvature. Note authigenic
quartz (arrows) intergrown with saddle ankerite.
Upper Carboniferous carbonate rock, Ouachita
'
Mountains, USA. Scale bar� 500 Jlm. (D) Void-filling
coarse-crystalline saddle dolomite postdating authigenic
quartz. Triassic limestone with epigenetic dolomite
and sphalerite-galena-fluorite mineralization, Eastern
Alps (Wanneck area, Tyrol). Scale bar� 500 Jlm.
(E) Coarse-crystalline saddle dolomite replacing host
dolomicrospar. Note optical zoning of dolospar due to
variable inclusion densities. Triassic limestone with
epigenetic dolomite and sphalerite-galena-fluorite
mineralization, Eastern Alps (Wanneck area, Tyrol).
Scale bar� 500 Jlm. (F) Host carbonate rock completely
replaced by non-planar dolospar. Note that about half
of the crystals in the field of view show regularly curved
crystal boundaries and/or incipient sweeping extinction,
i.e. saddle dolomite. Triassic limestone with epigenetic
dolomite and sphalerite-galena-fluorite mineralization,
Eastern Alps (Wanneck area, Tyrol). Scale bar� 500 Jlm.
!.facing page 4461